Self-assembling microsphere-based dextran hydrogels … · 2020. 3. 4. · Self-assembling...
Transcript of Self-assembling microsphere-based dextran hydrogels … · 2020. 3. 4. · Self-assembling...
Self-assembling microsphere-based dextran hydrogels for pharmaceutical applications
Sophie Van Tomme
The research presented in this thesis was financially supported by OctoPlus N.V., Leiden,
The Netherlands and SenterNovem (IS042016).
Publication of this thesis was financially supported by:
OctoPlus N.V., Leiden, The Netherlands
Organon Bioscience N.V., Oss, The Netherlands
TA Instruments, Etten-Leur, The Netherlands
Utrecht Institute for Pharmaceutical Sciences, Utrecht, The Netherlands
UIPS Ut recht Inst itute forPharmaceutical Sciences
Self-assembling microsphere-based dextran hydrogels for pharmaceutical applications
Sophie Van Tomme PhD Thesis with summary in Dutch
Department of Pharmaceutics, Utrecht University, Utrecht, The Netherlands September 2007
© 2007 Sophie Van Tomme
All rights reserved. No part of this thesis may be reproduced or transmitted in any form or by any means without written permission of the author.
ISBN: 9789039346341
Cover design: Rob Sens
Printed by PrintPartners Ipskamp, Enschede, The Netherlands
Self-assembling microsphere-based
dextran hydrogels
for pharmaceutical applications
Op microsferen gebaseerde, fysisch gecrosslinkte dextraanhydrogelen voor farmaceutische toepassingen
(met een samenvatting in het Nederlands)
Proefschrift
ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof. dr. W.H. Gispen, ingevolge het besluit van het college voor promoties in het
openbaar te verdedigen op woensdag 19 september 2007 des middags te 2.30 uur
door
Sophie Rolande Van Tomme
geboren op 25 oktober 1980 te Brugge, België
Promotoren: Prof. dr. ir. W.E. Hennink
Prof. dr. S.C. De Smedt
Co-promotor: Dr. C.F. van Nostrum
A nos amours, à nos faiblesses
Table of contents
1 General introduction
1
2 Biodegradable dextran hydrogels for protein delivery applications
13
3 Self-gelling hydrogels based on oppositely charged dextran microspheres
39
4 Mobility of model proteins in hydrogels composed of oppositely charged
dextran microspheres studied by protein release and fluorescence recovery after photobleaching
53
5 Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
71
6 Effect of particle size and charge on the network properties of microsphere-
based hydrogels
89
7 Tuning the degradation behavior of ionically crosslinked dextran hydrogels by variation of the cationic groups
109
8 Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
121
9 Possible applications and future perspectives
139
10 Summary
151
Appendix: Samenvatting in het Nederlands List of abbreviations
List of publications Curriculum Vitae
Dankwoord/Acknowledgement
159 167
169 173
175
1 General introduction
& Aim and outline of this thesis
Chapter 1
2
General introduction
3
1. General introduction
1.1. Hydrogels
Hydrogels are defined by Peppas et al. as ‘three-dimensional, hydrophilic, polymeric networks
capable of imbibing large amounts of water or biological fluids’.[1] Since the introduction of hydrogels as soft contact lenses in the 1960s,[2] their use has increased tremendously and
nowadays they are favored in a broad range of pharmaceutical and biomedical applications,[1, 3-5] of which protein delivery and tissue engineering are relevant to this thesis. The polymers that are used to create macroscopic networks can be divided into 2 groups, according to their origin:
natural or synthetic. Most common synthetic polymers, used as building blocks for hydrogels, are poly(ethylene glycol) (PEG)[6-9] and poly(hydroxyethyl methacrylate) (pHEMA).[2, 10, 11] The use of
synthetic polymers generally results in hydrogels with high mechanical strength and offers the opportunity to tailor the network properties and for instance release behavior, in a reproducible
way. Natural polymers have gained more interest the past decades because of their biocompatibility and the presence of biologically recognizable groups to support cellular activities. Two major drawbacks, however, are their generally weak mechanical properties and the batch to
batch variability. Furthermore, they can be derived from various sources and may contain impurities or pathogens as a result. Natural polymers frequently used for the preparation of
hydrogels are, amongst others, chitosan,[12-14] hyaluronic acid,[15-17] alginate,[18-20] gelatin[21] and, as will be discussed below, dextran.
1.2. Crosslinking strategies to design hydrogels
The hydrogels are crosslinked to prevent dissolution of the polymer chains when the hydrogel is in
an aqueous environment. To provide degradability of the network, degradable linkers can be incorporated.[22] A wide range of crosslinking methods has been used that can be categorized into
chemical and physical crosslinking (Fig. 1).[23] Chemical crosslinking leads to permanent covalent bonds and can be accomplished by e.g. radical,[24] high-energy[25] or enzyme-mediated
polymerization.[26] Physical crosslinking involves the introduction of non-permanent, reversible crosslinks via e.g. hydrophobic interactions,[27] hydrogen bonding,[28] stereocomplexes[29] or ionic
interactions.[30] Whereas chemical crosslinking leads to networks with relatively high mechanical strength, physical crosslinking generally results in weaker hydrogels. On the other hand, the physical crosslinks can be reversibly broken by e.g. shear forces, which may allow injectability of the
gels. Additionally, the crosslinking conditions during physical crosslinking are mild, while the use of chemical crosslinking agents might lead to denaturation of entrapped proteins or could be toxic
to cells. Depending on the specific application foreseen for the hydrogels system, the above mentioned issues need to be taken into consideration.
Chapter 1
4
1.3. In situ gelation
Hydrogels can be designed in different geometries, e.g. slabs or cylinders,[31] films,[32] microspheres[33] or nanospheres.[34] While the latter two can be administered by injection,
macroscopic cylinders need surgical administration. Therefore, nowadays, interest has shifted to hydrogels that can be administered as liquid formulation and are formed in situ, at the site of
injection.[35] In situ gelation via self-assembly can occur by physical interactions that develop in time (e.g. stereocomplexed gels,[36] ionically crosslinked gels[14]), or in response to a certain trigger
(e.g. temperature[37]). UV-polymerization can also be applied for in situ gelation if the polymer solution is injected in a cavity from where it cannot leak out before polymerization is finalized.[38] Recently, in situ gelation was accomplished via Michael-addition resulting in chemically crosslinked
hydrogels.[39-41] Currently, much research is done on ‘stimuli-responsive’ hydrogels[42] that release their content in reaction to a stimulus, such as glucose concentration,[43] pH,[44] temperature,[45, 46]
ionic strength[47] or binding to antigens.[48]
+ +
chemical crosslinkingphysical crosslinking
hydrophilicpolymer chainscrosslinks
+ +
chemical crosslinkingphysical crosslinking
hydrophilicpolymer chainscrosslinks
Figure 1: Schematic representation of physical and chemical crosslinking.
1.4. Dextran hydrogels
Dextran is one of the most promising polysaccharides of the past few decades for the preparation
of hydrogels for controlled protein release.[49] It can be produced either chemically or by bacteria
from sucrose. Bacterial dextran mainly consists of linear α-1,6-linked glucopyranose units, with
some degree of 1,3-branching (Fig. 2). Several coupling reactions have been described that
provide the opportunity to modify the numerous OH- groups along the dextran chains.
General introduction
5
O
O
OHO
HOHO
O
O
OH
HOHO
O
OH
HOHO
O
O
O
OHO
HO
α-1,3 linkage
α-1,6 linkage
glucopyranose unit
Figure 2: Chemical structure of dextran.
In our group, extensive research on dextran hydrogels as biodegradable protein delivery systems
has been performed. Dextran (dex) was modified with methacrylate (MA),[50] hydroxethyl methacrylate (HEMA)[51] and oligolactate[52] for the formation of both chemically and physically
crosslinked macroscopic hydrogels. Dex-MA hydrogels can be enzymatically degraded, while the other formulations are mainly degraded by chemical hydrolysis.
A novel preparation method for dex-MA and dex-HEMA microspheres was developed, without the
use of organic solvents.[53, 54] The approach is based on the immiscibility of aqueous solutions of dextran and PEG, resulting in a water-in-water emulsion. Next, radical polymerization leads to the formation of dex-MA or dex-HEMA microspheres (Fig. 3). The microspheres can be loaded with
therapeutic proteins and subsequently injected intramuscularly or subcutaneously.[55, 56] These microspheres are very promising protein delivery vehicles, but the necessity of radicals for the
polymerization reactions remains an obstacle, especially when the microspheres are loaded with oxidation-sensitive proteins.[57] To circumvent this hurdle, we are interested in physically
crosslinked, injectable, macroscopic dextran hydrogels.
dexPEG PEG
dex+
stirring radical
polymerization
water-in-water emulsion
dexPEG PEG
dex+
stirring radical
polymerization
water-in-water emulsion Figure 3: Preparation of dex-MA or dex-HEMA microspheres via a water-in-water emulsion and subsequent
radical polymerization.
Chapter 1
6
2. Aim of this thesis
The aim of this thesis was the synthesis and physicochemical characterization of self-assembling
microsphere-based networks, arising through physical crosslinking of dextran microgels, for drug delivery purposes.
Additionally, as Figure 4 illustrates, we aimed to evaluate whether proteins and/or cells can be entrapped in the pores between the individual microspheres. Two different approaches to self-
assemble the microspheres were investigated: ionic interactions and hydrophobic interactions combined with stereocomplexation.
In the first approach dex-HEMA was copolymerized with methacrylic acid (MAA) or N,N-
dimethylamino ethyl methacrylate (DMAEMA) resulting in, respectively, negatively and positively
charged microspheres at physiological pH. Because ‘opposites attract’, hydrogel formation occurred when aqueous dispersions of both types of microspheres were mixed. In the second
approach, hydrogels were obtained through hydrophobic self-assembly of and stereocomplexation between dex-HEMA microspheres, grafted with oligolactates of opposite chirality.
This thesis aimed (a) to reveal the network properties of these self-assembled microsphere-based
hydrogels through rheological analysis, (b) to develop various strategies to tailor the network properties and the degradation behavior of the matrices, (c) to evaluate the applicability of the
hydrogels as protein delivery systems by monitoring the in vitro release of proteins.
type 1 microspheres type 2 microspheres
proteins
macroscopic hydrogel
type 1 microspheres type 2 microspheres
proteins
macroscopic hydrogel
Figure 4: Schematic representation of the self-assembling microsphere-based hydrogels in which proteins are
entrapped by simply mixing the protein solution with the microsphere dispersions.
General introduction
7
3. Outline
Chapter 2 reviews the current state-of-the-art of dextran hydrogels for pharmaceutical
applications. Both chemically and physically crosslinked hydrogels are described. Special attention is given to the medical applicability of the systems.
In Chapter 3, a novel self-assembling hydrogel is reported based on ionic interactions between
oppositely charged dextran microspheres. Dex-HEMA was copolymerized with methacrylic (MAA)
acid or N,N-dimethylamino ethyl methacrylate (DMAEMA) yielding negatively and positively charged microspheres, respectively. Equal amounts of aqueous dispersions of both microsphere types were mixed and their network forming capacity was investigated. Rheological analysis was
applied to monitor the network properties.
Chapter 4 describes the mobility of model proteins (lysozyme, BSA and IgG) in the ionically self-
assembled microsphere-based dextran hydrogels. Protein release was studied and mathematical modeling of the release curves was done to obtain a better understanding of the mechanisms that are involved. Fluorescence recovery after photobleaching experiments were applied to gain
insight into the mobility of proteins in the hydrogels on a microscopic level.
The degradation behavior of both the charged dextran microspheres as well as the ionically crosslinked macroscopic gels is in focus in Chapter 5. Besides swelling experiments, confocal
imaging and rheology on degrading hydrogels were performed. For the swelling experiments, the ratio positively/negatively charged particles was varied to evaluate the charge effect on the
degradation behavior of the microspheres.
In Chapter 6 the effect of the particle size (distribution) and charge of the dextran microspheres
on the network properties of the self-assembled hydrogels is demonstrated. The ratio (MAA or
DMAEMA)/HEMA was varied to obtain microspheres with different zeta (ξ)-potentials. Different
microsphere preparation procedures were explored to alter the particle size and size distribution.
Positively charged microspheres were prepared with degradable cationic monomers based on 2-hydroxypropyl methacrylamide (HPMA), i.e. HPMA-DMAE (N,N-dimethyl 2-aminoethanol) and HPMA-DMAPr (N,N-dimethyl 3-aminopropanol) instead of DMAEMA in Chapter 7. Introductory
degradation experiments were performed to investigate the effect of charge loss in time on the network properties of the macroscopic gels.
Chapter 8 deals with a second approach to create self-gelling microsphere-based dextran
hydrogels. For this purpose, dex-HEMA microspheres were grafted with oligolactates of opposite
chirality. The potential of hydrophobic interactions or stereocomplexation between the lactate grafts on the individual microspheres to create a macroscopic network was evaluated rheologically.
Finally, in Chapter 9, potential applications and future perspectives of the self-assembling
microsphere-based dextran hydrogels are outlined, with a special focus on biocompatibility of the
materials. Some pilot cytotoxicity data are reported.
Chapter 1
8
References
1. Peppas NA, Bures P, Leobandung W, Ichikawa H. Hydrogels in pharmaceutical formulations. Eur. J. Pharm.
Biopharm. 2000;50:27-46.
2. Wichterle O, Lim D. Hydrophilic gels for biological use. Nature 1960;185:117-118.
3. Hoffman AS. Hydrogels for biomedical applications. Adv. Drug Deliver. Rev. 2002;43:3-12.
4. Kashyap N, Kumar N, Ravikumar MNV. Hydrogels for pharmaceutical and biomedical applications. Crit. Rev. Ther.
Drug 2005;22:107-150.
5. Peppas NA. Hydrogels and drug delivery. Curr. Opin. Colloid In. 1997;2:531-537.
6. Quick DJ, Anseth KS. DNA delivery from photocrosslinked PEG hydrogels: encapsulation efficiency, release profiles,
and DNA quality. J. Control. Release 2004;96:341-351.
7. Van de Wetering P, Metters AT, Schoenmakers RG, Hubbell JA. Poly(ethylene glycol) hydrogels formed by conjugate addition with controllable swelling, degradation, and release of pharmaceutically active proteins. J.
Control. Release 2005;101:619-627.
8. Mahoney MJ, Anseth KS. Three-dimensional growth and function of neural tissue in degradable polyethylene
glycol hydrogels. Biomaterials 2006;27:2265-2274.
9. Hiemstra C, Zhong ZY, Van Tomme SR, Hennink WE, Dijkstra PJ, Feijen J. Protein release from injectable stereocomplexed hydrogels based on PEG-PDLA and PEG-PLLA star block copolymers. J. Control. Release
2006;116:e19-e21.
10. Lu S, Anseth KS. Photopolymerization of multilaminated poly(HEMA) hydrogels for controlled release. J. Control.
Release 1999;57:291-300.
11. Garcia DM, Escobar JL, Noa Y, Bada N, Hernaez E, Katime I. Timolol maleate release from pH-sensible poly(2-
hydroxyethyl methacrylate-co-methacrylic acid) hydrogels. Eur. Polym. J. 2004;40:1683-1690.
12. Obara K, Ishihara M, Ishizuka T, Fujita M, Ozeki Y, Maehara T, Saito Y, Yura H, Matsui T, Hattori H. Photocrosslinkable
chitosan hydrogel containing fibroblast growth factor-2 stimulates wound healing in healing-impaired db/db
mice. Biomaterials 2003;24:3437-3444.
13. Akbuga J, Ozbas-Turan S, Erdogan N. Plasmid-DNA loaded chitosan microspheres for in vitro IL-2 expression. Eur. J.
Pharm. Biopharm. 2004;58:501-507.
14. Berger J, Reist M, Mayer JM, Felt O, Peppas NA, Gurny R. Structure and interactions in covalently and ionically
crosslinked chitosan hydrogels for biomedical applications. Eur. J. Pharm. Biopharm. 2004;57:19-34.
15. Peattie RA, Nayate AP, Firpo MA, Shelby J, Fisher RJ, Prestwich GD. Stimulation of in vivo angiogenesis by cytokine-
loaded hyaluronic acid hydrogel implants. Biomaterials 2004;25:2789-2798.
16. Masters KS, Shah DN, Leinwand LA, Anseth KS. Crosslinked hyaluronan scaffolds as a biologically active carrier for
valvular interstitial cells. Biomaterials 2005;26:2517-2525.
17. Shu XZ, Liu Y, Palumbo FS, Luo Y, Prestwich GD. In situ crosslinkable hyaluronan hydrogels for tissue engineering.
Biomaterials 2004;25:1339-1348.
18. Gombotz WR, Wee SF. Protein release from alginate matrices. Adv. Drug Deliver. Rev. 1998;31:267-285.
19. Rowley JA, Madlambayan G, Mooney DJ. Alginate hydrogels as synthetic extracellular matrix materials. Biomaterials
1999;20:45-53.
20. Leonard M, De Boisseson MR, Hubert P, Dalencon F, Dellacherie E. Hydrophobically modified alginate hydrogels as
protein carriers with specific controlled release properties. J. Control. Release 2004;98:395-405.
21. Sutter M, Siepmann J, Hennink WE, Jiskoot W. Recombinant gelatin hydrogels for the sustained release of proteins.
J. Control. Release 2007;119:301-312.
22. Peppas NA, Hilt JZ, Khademhosseini A, Langer R. Hydrogels In biology and medicine: from molecular principles to
biotechnology. Adv. Mater. 2006;18:1345-1360.
23. Hennink WE, van Nostrum CF. Novel crosslinking methods to design hydrogels. Adv. Drug Deliver. Rev. 2002;54:13-36.
24. Chung JT, Vlugt-Wensink KDF, Hennink WE, Zhang Z. Effect of polymerization conditions on the network
properties of dex-HEMA microspheres and macro-hydrogels. Int. J. Pharm. 2005;288:51-61.
25. Leach JB, Schmidt CE. Characterization of protein release from photocrosslinkable hyaluronic acid-polyethylene
glycol hydrogel tissue engineering scaffolds. Biomaterials 2005;26:125-135.
26. Ferreira L, Gil MH, Cabrita AMS, Dordick JS. Biocatalytic synthesis of highly ordered degradable dextran-based
hydrogels. Biomaterials 2005;26:4707-4716.
27. Li J, Li X, Ni X, Wang X, Li H, Leong KW. Self-assembled supramolecular hydrogels formed by biodegradable PEO-
PHB-PEO triblock copolymers and [alpha]-cyclodextrin for controlled drug delivery. Biomaterials 2006;27:4132-4140.
General introduction
9
28. Noro A, Nagata Y, Takano A, Matsushita Y. Diblock-type supramacromolecule via biocomplementary hydrogen
bonding. Biomacromolecules 2006;7:1696-1699.
29. Li S. Bioresorbable hydrogels prepared through stereocomplexation between poly(L-lactide) and poly(D-lactide)
blocks attached to poly(ethylene glycol). Macromol. Biosci. 2003;3:657-661.
30. Kuo CK, Ma PX. Ionically crosslinked alginate hydrogels as scaffolds for tissue engineering: Part 1. Structure,
gelation rate and mechanical properties. Biomaterials 2001;22:511-521.
31. Franssen O, Vandervennet L, Roders P, Hennink WE. Degradable dextran hydrogels: controlled release of a model
protein from cylinders and microspheres. J. Control. Release 1999;60:211-221.
32. Draye J-P, Delaey B, Van de Voorde A, Van Den Bulcke A, Bogdanov B, Schacht E. In vitro release characteristics of
bioactive molecules from dextran dialdehyde cross-linked gelatin hydrogel films. Biomaterials 1998;19:99-107.
33. Sinha VR, Trehan A. Biodegradable microspheres for protein delivery. J. Control. Release 2003;90:261-280.
34. Van Thienen TG, Lucas B, Demeester J, De Smedt SC. On the synthesis and characterization of biodegradable
dextran nanogels with tunable degradation properties. J. Control. Release 2006;116:e12-e13.
35. Hatefi A, Amsden B. Biodegradable injectable in situ forming drug delivery systems. J. Control. Release 2002;80:9-
28.
36. Bos GW, Jacobs JJL, Koten JW, Van Tomme SR, Veldhuis TFJ, van Nostrum CF, Den Otter W, Hennink WE. In situ crosslinked biodegradable hydrogels loaded with IL-2 are effective tools for local IL-2 therapy. Eur. J. Pharm. Sci.
2004;21:561-567.
37. Ruel-Gariepy E, Leroux J-C. In situ-forming hydrogels--review of temperature-sensitive systems. Eur. J. Pharm.
Biopharm. 2004;58:409-426.
38. Anseth KS, Metters AT, Bryant SJ, Martens PJ, Elisseef JH, Bowman CN. In situ forming degradable networks and
their application in tissue engineering and drug delivery. J. Control. Release 2002;78:199-209.
39. Elbert DL, Pratt AB, Lutolf MP, Halstenberg S, Hubbell JA. Protein delivery from materials formed by self-selective
conjugate addition reactions. J. Control. Release 2001;76:11-25.
40. Lutolf MP, Raeber GP, Zisch AH, Tirelli N, Hubbell JA. Cell-responsive synthetic hydrogels. Adv. Mater. 2003;15:888-892.
41. Hiemstra C, vanderAa LJ, Zhong Z, Dijkstra PJ, Feijen J. Rapidly in situ-forming degradable hydrogels from dextran
thiols through Michael Addition. Biomacromolecules 2007;8:1548-1556.
42. Miyata T, Uragami T, Nakamae K. Biomolecule-sensitive hydrogels. Adv. Drug Deliver. Rev. 2002;54:79-98.
43. Tanna S, Joan Taylor M, Sahota TS, Sawicka K. Glucose-responsive UV polymerised dextran-concanavalin A acrylic
derivatised mixtures for closed-loop insulin delivery. Biomaterials 2006;27:1586-1597.
44. Gupta P, Vermani K, Garg S. Hydrogels: from controlled release to pH-responsive drug delivery. Drug Discov. Today
2002;7:569-579.
45. Huang X, Lowe TL. Biodegradable thermoresponsive hydrogels for aqueous encapsulation and controlled release
of hydrophilic model drugs. Biomacromolecules 2005;6:2131-2139.
46. Vermonden T, Besseling NAM, van Steenbergen MJ, Hennink WE. Rheological studies of thermosensitive triblock
copolymer hydrogels. Langmuir 2006;22:10180-10184.
47. Zhang R, Tang M, Bowyer A, Eisenthal R, Hubble J. A novel pH- and ionic-strength-sensitive carboxy methyl
dextran hydrogel. Biomaterials 2005;26:4677-4683.
48. Lu Z, Kopeckova P, Kopecek J. Antigen responsive hydrogels based on polymerizable antibody Fab' fragment.
Macromol. Biosci. 2003;3:296-300.
49. Coviello T, Matricardi P, Marianecci C, Alhaique F. Polysaccharide hydrogels for modified release formulations. J.
Control. Release 2007;119:5-24.
50. Van Dijk-Wolthuis WNE, Franssen O, Talsma H, van Steenbergen MJ, Kettenes-van den Bosch JJ, Hennink WE.
Synthesis, characterization and polymerization of glycidyl methacrylate derivatized dextran. Macromolecules
1995;28:6317-6322.
51. Van Dijk-Wolthuis WNE, Tsang SKY, Kettenes-van den Bosch JJ, Hennink WE. A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer
1997;38:6235-6242.
52. De Jong SJ, De Smedt SC, Demeester J, van Nostrum CF, Kettenes-van den Bosch JJ, Hennink WE. Biodegradable hydrogels based on stereocomplex formation between lactic acid oligomers grafted to dextran. J. Control. Release
2001;72:47-56.
53. Franssen O, Hennink WE. A novel preparation method for polymeric microparticles without the use of organic
solvents. Int. J. Pharm. 1998;168:1-7.
Chapter 1
10
54. Stenekes RJH, Franssen O, van Bommel EMG, Crommelin DJA, Hennink WE. The use of aqueous PEG/dextran
phase separation for the preparation of dextran microspheres. Int. J. Pharm. 1999;183:29-32.
55. De Groot CJ, Cadée JA, Koten JW, Hennink WE, Den Otter W. Therapeutic efficacy of IL-2-loaded hydrogels in a
mouse tumor model. Int. J. Cancer 2002;98:134-140.
56. Vlugt-Wensink KDF, de Vrueh R, Gresnigt MG, Hoogerbrugge CM, van Buul-Offers SC, de Leede LGJ, Sterkman
LGW, Crommelin DJ, Hennink WE, Verrijk R. Preclinical and clinical in vitro in vivo correlation of an hGH dextran
microsphere formulation. 2007 submitted.
57. Cadée JA, van Steenbergen MJ, Versluis C, Heck AJR, Underberg WJM, den Otter W, Jiskoot W, Hennink WE.
Oxidation of recombinant human interleukin-2 by potassium peroxodisulfate. Pharm. Res. 2001;18:1461-1467.
General introduction
11
2 Biodegradable dextran hydrogels for protein delivery
applications
Sophie R. Van Tomme, Wim E. Hennink
Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS),
University Utrecht, Utrecht, The Netherlands
Expert Review of Medical Devices 4 (2007) 147-164
Chapter 2
14
Summary
The rapid development of protein-based pharmaceuticals during the past decades has tremendously increased the need for suitable delivery systems, guaranteeing a safe and controlled
delivery of proteinacous drugs. Hydrogels offer good opportunities as protein delivery systems or tissue engineering scaffolds because of an inherent biocompatibility. Their hydrophilic, soft and
rubbery nature ensures minimal tissue irritation and a low tendency of cells and proteins to adhere to the hydrogel surface. A variety of both natural and synthetic polymers have been used for the design of hydrogels in which network formation is established by chemical or physical crosslinking.
This review will introduce the general features of hydrogels and will further on focus on dextran hydrogels in particular. Chemically and physically crosslinked systems will be described and their
potential suitability as protein delivery systems as well as tissue engineering scaffolds will be discussed. Special attention will be given to network properties, protein delivery, degradation
behavior and biocompatibility.
Keywords: hydrogels, dextran, chemical crosslinking, physical crosslinking, protein delivery, tissue
engineering, biodegradation, biocompatibility
Biodegradable dextran hydrogels for protein delivery applications
15
1. Introduction
Nobel Prize winner Paul Berg’s invention of recombinant DNA in 1972 opened the door to a whole new world of biopharmaceuticals.[1] Indeed, modern biotechnology made it possible to produce virtually any protein that could be of therapeutic use in a variety of diseases, including cancer,
immune disorders and infectious diseases. The number of recombinant proteins entering clinical studies over the past 30 years is enormous and it is predicted that their number will keep on
growing the next decade. In 2004 and 2005 the FDA approved 16 new protein therapeutics.[2] However, even more than the traditional drug molecules, these novel therapeutics do suffer from
a lack of suitable delivery systems.[3] Proteins are fragile molecules sensitive to both physical degradation induced by, among others, pH and temperature and chemical degradation, such as oxidation or enzymatic cleavage. Denatured and degraded proteins do not only endure a loss of
biological activity, more importantly, they might induce an immune response.[4] Therefore, a thorough understanding of protein stability, immunogenicity and pharmacokinetic profile is
indispensable to develop a successful protein-based therapeutic.[5] Oral delivery, the most preferred, non-invasive route of administration is unattainable due to the harsh conditions in the
gastrointestinal tract. In the stomach, low pH and enzymatic degradation affect the protein integrity. Additionally, proteins have difficulties to pass the intestinal epithelium. After parenteral
administration, most proteins are rapidly cleared from the circulation, requiring the need of frequent injections or continuous infusions. In order to extend the plasma half-life of proteins and increase their safety and efficacy by avoiding peak concentrations and prolonging the presence in
specific tissues and organs, specific formulation strategies such as controlled release systems, are needed. Alternatively, conjugation of proteins and peptides with poly(ethylene glycol) (PEG) has
also shown to lead to improved clinical properties such as increased solubility and stability, extended circulation time and reduced immunogenicity and susceptibility to enzymatic
degradation.[6] As a result of sustained release or prolonged circulation of proteins, the injection frequency is reduced and patient comfort and compliance are increased.[7, 8] Polymeric systems such as hydrogels,[9, 10] microspheres[11] and nanoparticles,[12] as well as lipid-
based systems such as liposomes[13] and water-in-oil emulsions[14] have been studied extensively for the delivery of pharmaceutical peptides and proteins. This review will focus on biodegradable
hydrogels, in particular dextran-based systems, designed for protein delivery and tissue engineering applications.
2. Hydrogels
Hydrogels are three-dimensional hydrophilic matrices, capable of absorbing large quantities of water. They have been used in a variety of applications, e.g. as wound dressings, transdermal
patches, drug delivery devices, contact lenses or in reconstructive surgery. Their soft and rubbery nature provokes minimal tissue irritation and makes them particularly attractive to incorporate
proteins and cells. Natural (e.g. collagen,[15] hyaluronate,[16] alginate,[17] starch,[18] chitosan[19]) as well as synthetic polymers (e.g. poly(ethylene glycol),[20] poly(vinyl alcohol),[21] poly(hydroxyethyl
methacrylate)[22]) can be used to prepare hydrogels, of which the latter are better defined and
Chapter 2
16
provide more control over physical and chemical properties. Additionally, natural polymers can be
derived from various sources with possible contaminations as a result. Depending on the application not only biocompatibility, i.e. the ability of a material to perform with an appropriate
host response in a specific application,[23] but also biodegradability will be requested. The degradation properties under physiological conditions can be tailored by incorporation of degradable linkers.[10, 24] Physical and chemical methods can be applied to obtain crosslinking of
polymers into hydrogel structures in order to prevent dissolution of the hydrophilic polymer chains in the aqueous environment. Chemical crosslinking can be accomplished by radical
polymerization, high-energy irradiation, via enzymes or by chemical reaction of complementary groups. These methods result in a network with a relatively high mechanical strength and,
depending on the nature of the chemical bonds in the building blocks and the crosslinks, in tailor-made degradation times.[25] However, chemical crosslinking can possibly damage the entrapped
bioactive substance, leading to a loss of activity. Moreover, the crosslinking agents are mostly toxic and removal needs to be ensured before in vivo application. In recent years there is a growing interest in physically crosslinked hydrogels. In these systems non-permanent bonds, based on
physical interactions between the polymer chains, such as ionic, hydrophobic or antigen-antibody interactions, hydrogen bonding and crystallization, are created.[25] An attractive class of physically
crosslinked systems is those where gel formation occurs after a certain trigger (e.g. temperature, pH, ionic strength).[26, 27] These stimuli-responsive hydrogels, as well as those that are formed after a
certain time (e.g. stereocomplexes[28]) or after UV-irradiation,[29] are particularly interesting as injectable, in situ-forming matrices for drug delivery and tissue engineering applications.[30-34] Table 1 summarizes the properties an ‘ideal’ hydrogel should have to be suitable as protein
delivery system or as tissue engineering scaffold. As will be discussed in the next section, dextran meets most of these requirements, making it an attractive polymer to design hydrogels.
Table 1: ‘Ideal’ properties of hydrogels suitable for biomedical applications.
Protein delivery system Tissue engineering scaffold
Biocompatible and biodegradable
Injectable
Self-assembling
Non-toxic degradation products
Mild inflammatory reaction
Non-immunogenic
Tailorable release properties Tailorable mechanical strength Possibility to target specific sites Tailorable degradation time
Controllable shape and size
3. Dextran based hydrogels
Table 2 gives an overview of the various strategies developed to create dextran hydrogels. The
possible formulations as well as the specific in vitro and in vivo evaluations carried out are listed.
Biodegradable dextran hydrogels for protein delivery applications
17
Table 2: Overview of dextran derivatives, crosslinking methods, formulation options and type of in vitro/in vivo
evaluations carried out.
Polymer X-linking Formulations In vitro In vivo Ref
Chemical crosslinking
Dex-MA
Dex-HEMA
Radical
polymerization
(KPS/TEMED)
Hydrogel
implants
Microspheres Scaffolds
IgG and rhIL-2
release
Tissue regeneration
Release of rhIL-
2 and hGH
Biocompatibility
39-54
Functionalized
dextran (e.g. sulfate)
STMP Particles rhTGF-β1 and
rhBMP-2 release
Cell support
Biocompatibility 55-56
CM-dextran EDC/NHS Membranes pH and ionic
strength
dependent drug
transport
57
Dex-AI or
dex-MA
copolymerized with NiPAAm
UV-
polymerization
Thermosensitive
hydrogels and
Particles
58-61
Dex-AE
copolymerized with PEGDA
UV-
polymerization
pH-sensitive
hydrogel implants
BSA release 62
Dex-MA
copolymerized with con A-MA
UV-
polymerization
Glucose-
responsive hydrogel
implants
Insulin release 63-64
Oxidized dextran Oxidized dextran
+ gelatin
AAD Hydrogels Hydrogel films
EGF release Biocompatibility 65 66-68
Dex-DVA Enzyme Hydrogel implants
Biocompatibility 69-70
Physical crosslinking
Dextran 6000 Crystallization Hydrogels Microspheres
71
Dex-PEG or dex-
PPG
+ cyclodextrins
Hydrogen
bonding
Temperature
sensitive
hydrogels
72-73
Dex-PLL + cyclodextrins
Hydrogen bonding
Temperature and pH-sensitive
hydrogels
74
Dex-L-lactate + dex-D-lactate
Stereo-complexes
Injectable hydrogels
Lysozyme, IgG and rhIL-2 release
rhIL-2 release Biocompatibility
27, 75-80
Dextran-sulfate
+ chitosan
Ionic
interactions
Nanoparticles Insulin release 81
Chapter 2
18
3.1. Introduction
Dextran is an exocellular bacterial polysaccharide predominantly consisting of linear α-1,6-linked
glucopyranose units, with some degree of 1,3-branching (Fig. 1A). This highly water-soluble
polymer is produced in a sucrose-rich environment by Lactobacillus, Leuconostoc and Streptococcus and is commercially available with different molecular weights. Both the degree of branching and
the molecular weight distribution affect the physicochemical properties.[35] Native dextran has a high molecular weight and a high degree of polydispersity, which can be decreased by partial
hydrolysis and subsequent fractionation.[36] Dextrans with an average molecular weight of 1000 to 2 million g/mol are commercially available
for research purposes.[37] Clinically, dextran, in particular low molecular weight (40000 and 70000 g/mol), has been used for over 50 years in plasma volume expansion, thrombosis
prophylaxis, peripheral blood flow enhancement and for the rheological improvement of, for instance, artificial tears.[36, 37] After parenteral administration, low molecular weight dextran
(40000 g/mol or smaller) has a half-life of 8 hours and is secreted by the kidneys.[38] Dextrans of a higher molecular weight exhibit longer half-lives and are subsequently degraded by the
reticuloendothelial system.[39] Additionally, dextrans are metabolized by different dextranases (α-1-
glucosidases) in various parts of the body, including liver, spleen and colon.[36] Besides their favorable characteristic of being highly water-soluble, dextrans are stable under mild
acidic and basic conditions. Furthermore, these polymers contain a large number of hydroxyl groups, making them suitable for derivatization and subsequent chemical or physical crosslinking.[36]
The past decades research interest has focused on the use of dextran as macromolecular carriers, e.g. hydrogels, in which the drug can be incorporated. Dextran hydrogels can be obtained in
various ways, based on either chemical or physical crosslinking (as listed in Table 2).
3.2. Chemically crosslinked dextran hydrogels
3.2.1. Methacrylate-derivatized dextrans
Edman et al. pioneered in the research on polymerizable dextran by reaction of dextran with
glycidylacrylate in water.[40] Hydrogels were formed after addition of the initiator system N,N,N’,N’-tetramethylethylenediamine (TEMED) and ammonium peroxydisulfate (APS) to an aqueous
solution of acryldextran in the presence of N,N-methylenebisacrylamide. Enzyme-loaded microspheres were obtained using an emulsion polymerization technique during which the enzyme activity was completely retained. In our department methacrylate-derivatized dextrans
(dex-MA) (Fig. 1B) have been crosslinked in the presence of potassium peroxydisulfate (KPS) and TEMED as initiating system, leading to hydrogels, insensitive to hydrolysis at physiological
conditions.[41] Introduction of hydrolysis-sensitive bonds between the dextran backbone and the methacrylate side chains resulted in hydrogels that are fully biodegradable under physiological
conditions. In this way hydroxyethyl methacrylate-derivatized dextran (dex-HEMA) (Fig. 1C) hydrogels were created in which carbonate esters are the hydrolyzable units. Additional control
over the degradation profile was obtained after incorporation of a lactate spacer (dex-lactate-
Biodegradable dextran hydrogels for protein delivery applications
19
HEMA) (Fig. 1D).[42] The degradation time could be varied from one day to more than three
months, depending on the crosslink density, water content, type of ester group in the crosslinks and length of the lactate spacer. Both the in vitro and in vivo biocompatibility and biodegradability
of these dextran hydrogels has been demonstrated and it was shown that the in vivo degradation rate correlates well to the in vitro situation.[43, 44] Not only macroscopic gels but also microspheres could be prepared from these dextran derivatives. An all-aqueous preparation method was
developed, based on the phase separation between PEG and dextran solutions.[45, 46] Controlled release of a model protein (IgG) from enzymatically degrading macroscopic dex-MA
hydrogels could be accomplished by incorporation of an endo-dextranase, responsible for degradation of the dextran backbone.[47] The release was dependent on the crosslink density of the
gels and the amount of dextranase incorporated. Release of IgG from dex-HEMA gels with a low crosslink density (DS 3, degree of substitution, i.e. number of HEMA side chains per 100
glucopyranose units) and moderate water content (70 % (w/w)) occurred over 20 days in a nearly zero-order fashion. Hydrogels with a higher DS and lower water content exhibited a delayed release, of which the delay time was dependent on the crosslink density of the gels (e.g. 35 days for
DS 11). Eventually, around 65 % of the entrapped IgG was released after 15 days. Release of IgG from dex-HEMA microspheres also showed a delay time, however, considerably shorter than for
macroscopic gels. Interestingly, a zero-order IgG release from enzymatically degrading dex-MA microspheres was observed, while the dex-HEMA microspheres exhibited a biphasic release
profile. The authors state that these differences might result from the fact that during degradation of dex-MA microspheres, the dextran backbone is enzymatically hydrolyzed, while in dex-HEMA microspheres, hydrolysis of the side chains is the cause of degradation. In macroscopic hydrogels
on the other hand, the protein remains entrapped as long as the microscopic cages are not yet connected with each other, resulting in a delayed release from both dex-MA and dex-HEMA
gels.[48] Besides model proteins the release of therapeutically relevant proteins from dextran hydrogels was also studied. The release of the recombinant human cytokine interleukin-2 (IL-2), an
important mediator of the immune response, was investigated by Cadée et al.[49] Release from the non-degradable dex-MA gels occurred in a diffusion-controlled fashion and was affected by the
water content and crosslink density of the network. HPLC and radioactivity measurements showed that rhIL-2 was not quantitatively released with 6 and 80 % of the initial amount still present after 35 days in gels with a water content of 90 and 70 %, respectively. Decreasing the water content of
degradable dex-lactate-HEMA and dex-HEMA gels also decreased the rhIL-2 release rate. The release from highly hydrated dex-lactate-HEMA gels (90 % water content) followed Fickian
diffusion. Gels with a water content of 70 % or lower, released the protein in an almost zero-order profile during 5 to 15 days. Cadée et al. attribute this release profile to two factors compensating
each other: firstly, the diffusion coefficient of the protein increases due to swelling of the matrix in time, secondly, the hydrogel size increases resulting in a decreased concentration gradient and a decreased release rate. RhIL-2 release from dex-HEMA gels was shown to be slower than from dex-
lactate-HEMA gels with comparable network characteristics, ascribed to a faster degradation of the dex-lactate-HEMA gels. Importantly, it was shown that rhIL-2 was mainly released in its monomeric
form from all the gel types and maintained 50 to 70 % of its original activity. Release of rhIL-2 from non-biodegradable dex-MA and biodegradable dex-HEMA microspheres was studied in vivo in
tumor-bearing mice by de Groot et al.[50] The required amount of rhIL-2 was gradually released over
Chapter 2
20
O
O
OHO
HOOOO
O
O
O
O
OHO
HOO
O
O
OH
HOOO
O OO
OO
O
O
O
O
(A) (B)
(C)
n
O
O
O O
HOHO
O
OHO
O
O
O O
HOHO
NHO
O
O
O
HOHO
O
O
OHO
HOO
O
O
O
O
OHO
HOOHN
ONH
OO
O
OHO
HOOHN
ONH
O
OO
O
O
OHO
HOO
ONH
H
NH2O
OH
HOOO
O OO
OO
O
O
O
n
O
n
n n
(D) (E)
(F)
(I)
(K)
(J)
(L)
O
O
OHO
O O
(G) (H)
O
O
OHO
HOHO
O
O
OH
HOHO
O
OH
HOHO
O
O
O
OHO
HO
α-1,3 linkage
α-1,6 linkage
glucopyranose unit
Figure 1: Building blocks of various dextran-based hydrogels:
(A) dextran (B) dex-MA (methacrylate) (C) dex-HEMA (D) dex-lactate-HEMA (E) dex-AI (F) dex-MA (maleic acid)
(G) oxidized dextran (H) dex-DVA (I) dex-PEG (J) dex-PPG (K) dex-PL (L) dex-lactate.
Biodegradable dextran hydrogels for protein delivery applications
21
a period of 5 to 10 days. The therapeutic effect of one injection of rhIL-2-containing microspheres
was comparable to the effect of free rhIL-2 injections for 5 consecutive days. Vlugt-Wensink et al. studied the release of human growth hormone (hGH) from dex-HEMA microspheres
subcutaneously injected in Dwarf mice and healthy human volunteers.[51] In dwarf mice, a single injection of the hGH-containing microspheres resulted in a significant, dose-dependent increase in body length and weight. Daily injection of a single dose hGH gave the same results, indicating the
hGH released from the microspheres was fully bioactive. Administration of hGH-loaded microspheres to human healthy volunteers led to increased hGH serum concentrations from day 2
on, with peak concentrations after 7-8 days. The serum concentration of the biomarkers insulin-like growth factor-I (IGF-1) and IGF binding protein response-3 (IGFBP-3) followed the hGH profile,
again demonstrating that the released hGH was bioactive. Additionally, a good in vitro - in vivo correlation of the hGH release was found.
De Geest et al. used charged microspheres based on dex-HEMA copolymerized with methacrylic acid (MAA) or dimethylamino ethyl methacrylate (DMAEMA) as templates for layer-by-layer (LbL) assembly of the polyelectrolytes poly(sodium 4-styrenesulfonate) (PSS) and poly(allylamine
hydrochloride) (PAH) for the design of self-rupturing microcapsules for pulsed drug delivery (Fig. 2).[52, 53] Sequential adsorption of polyelectrolytes led to membranes that could be controlled
within nanometers. It was found that the permeability of the coating was pH dependent. Three bilayers of PSS/PAH rendered the microcapsules impermeable to 20000 g/mol FITC-dextran at pH
9 while they were still permeable at pH 7. Six polyelectrolyte bilayers were needed to make the membranes impermeable to 20000 g/mol FITC-dextran at pH 7. Upon degradation of the microgels at pH 9, dextran degradation products of 19000 g/mol are formed, unable to penetrate
the surrounding membrane. It was found that sudden rupture of the membrane occurred as a result of an increased osmotic pressure due to the presence of degradation products inside the
capsule. The self-rupturing of the microcapsules is completely controlled by the degradation kinetics of the gels, independent of any external stimulus. Self-exploding dextran-based
microcapsules appear to be promising systems for pulsed delivery of antigens for vaccination purposes. However, further research is needed to evaluate the applicability of the system under
physiological conditions.
Figure 2: Schematic presentation of the coating, swelling and explosion of a dex-HEMA microgel. Reprinted
from Journal of Controlled Release 116, 159-169 (2006) De Geest et al. with permission from Elsevier.[52]
Chapter 2
22
Lévesque et al.[54] reported on the use of dex-MA for the preparation of hydrogel scaffolds, also
taking advantage of the immiscibility of aqueous solutions of PEG and dextran (Fig. 3).[45] Dex-MA was polymerized in glass molds in the presence of an excess of water and PEG by which porous
structures were formed. It was found that the concentration of PEG dramatically influenced the pore structure of the scaffolds, ranging from microporous to macroporous gel-wall to macroporous with interconnected beads. It was possible to predict the porosity of the scaffolds by
making use of dex-MA/PEG/H2O phase diagrams. The authors pointed out that macroporous gel scaffolds with interconnected beaded structure can be used in tissue engineering since cell
penetration, nutrient diffusion and tissue regeneration are facilitated. Recently, the same research group described macroporous scaffolds of dex-MA copolymerized with aminoethyl methacrylate
(AEMA).[55] The primary amines were introduced to allow grafting of extracellular matrix (ECM)-derived peptides on the dex-MA-co-AEMA hydrogels to promote specific cellular interactions. It
was found that the introduction of primary amines (AEMA) as such, without additional grafting of ECM-derived peptides, improved cell adhesion already. Importantly, modification of the scaffolds with peptides further enhanced the cell adhesion and moreover a significant increase in cellular
activity, as measured by neurite outgrowth, was observed.
Figure 3: Preparation of dex-MA scaffolds making use of the PEG/dextran phase separation. Reprinted from
Biomaterials 26, 7436-7446 (2005) Lévesque et al. with permission from Elsevier.[53]
3.2.2. Functionalized dextrans bearing negatively charged groups
Functionalized dextrans (FD), bearing carboxylate, benzylamide and sulfate groups, were described
by Maire et al. as potential building blocks for hydrogels which are able to bind and release
transforming growth factor β1 (TGF-β1).[56] Hydrogels were obtained through chemical
crosslinking of dextran and FD with sodiumtrimetaphosphate (STMP) as a crosslinking agent under
alkaline conditions. After drying, the hydrogels were crushed and sieved to obtain particles of 1.0 to 1.6 mm in diameter. Protein loading was achieved by soaking of the particles in buffer
containing rhTGF- β1 for 30 min at 4 °C. Negatively charged phosphate groups, produced during
Biodegradable dextran hydrogels for protein delivery applications
23
the crosslinking reaction as well as the negative charge of the FD resulted in electrostatic binding
of rhTGF-β1 to the hydrogel matrix. A burst release, ascribed to a rapid desorption of the protein
was observed during the first six hours, after which the protein was slowly released over 3 days. The porous particles provided a high exchange surface leading to a fast release. Furthermore,
loading of hydrogels by soaking in protein solution mostly results in an inhomogeneous distribution of the protein in the matrix and subsequent burst release. The retaining capacity of the
gels was dependent on the FD and crosslinker content. In another study the functionalized dextran-based hydrogels were evaluated as bone morphogenic protein (BMP) carriers to enhance
bone formation.[57] The material is injectable making it suitable for repair of irregularly shaped skeletal defects. In view of this application a relatively long BMP retention capacity is important.
Maire et al. found that rhBMP-2 was less retained than rhTGF-β1, investigated in the first study, due
to a different interaction pattern (ion-pairing, Van der Waals, hydrogen bonding) between the proteins and the matrix. Nevertheless, significant bone formation was seen with only half of the
amount of BMP required when compared to a collagen sponge. An irreproducible calcification of the hydrogels was observed in vivo, which was explained by interaction of the anionic chemical groups with Ca2+ ions, causing precipitation of phosphocalcic mineral structures. Calcification and
bone formation occurred independently of each other but it was shown that calcification improved the biocompatibility of the hydrogels as evidenced by the absence of a major
inflammatory reaction. A limitation of these scaffolds is their inability to allow cell ingrowth into the core of the system, solely leading to bone formation between the particles and at the exterior of
the scaffold.
3.2.3. Stimuli-responsive dextran hydrogels
3.2.3.1. pH, ionic strength- and temperature-sensitive materials
A pH- and ionic strength-sensitive dextran hydrogel was developed by the group of Hubble.[58] Hydrogel formation was accomplished by the intermolecular crosslinking between hydroxyl and
carboxyl groups of carboxymethyl (CM) dextran using 1-ethyl-(3-3-dimethylaminopropyl) carbodiimide (EDC) in the presence of N-hydroxysuccinimide (NHS). The reaction of dextran and
sodium chloroacetate, catalyzed by NaOH, was varied from 15 min at 62 °C to 1 h at 70 °C, to
introduce sufficient crosslinkable groups to obtain a less or more pH-sensitive material, respectively. SEM images revealed a reversible pH-dependent morphology of the membranes,
with a more compact interior structure at pH 5.5 than at pH 7.4. At higher pH the COOH groups dissociate, initiating electrostatic repulsion between the polymer chains and as a consequence increasing the porosity of the matrix. Consequently, diffusion of lysozyme through the hydrogel
membranes at pH 5.5 was significantly slower than at pH 7.4 and the transport rate changed reversibly when the pH was switched repeatedly between acidic and neutral. The lysozyme
transport also proved to be influenced by the ionic strength of the buffer, with the highest diffusion rate in ionic strengths of 0.15 M and 0.2 M at pH 5.5 and 7.4, respectively. A further
increase in ionic strength led to a reduced lysozyme transport due to shielding of the COO- groups by Na+ ions resulting in a reduced swelling. Neither a pH- nor an ionic strength-dependency was
observed when hydrogel membranes were used with a lower carboxylic acid substitution. As a result of a lower amount of COOH groups in the network, less repulsive forces are active with a
Chapter 2
24
reduced swelling as a consequence. This offers the possibility to tailor the hydrogels for specific
applications. Copolymerization of NIPAAm with dextran derivatives was used by Zhang et al. to create
thermosensitive hydrogels.[59, 60] Dextran was substituted with allyl isocyanate (dex-AI) (Fig. 1E) and photopolymerized with NIPAAm. Differential scanning calorimetry showed an increasing LCST (=
lower critical solution temperature) from 32 to 36.8 °C when the amount of dextran moieties was
increased from 0 to 50 % (w/w), as a result of a shift in the hydrophilic/hydrophobic balance. No
LCST was found in the studied temperature range (25-50 °C) in hydrogels containing even more
dextran units due to the extensive number of non-thermosensitive polymers. SEM analysis of
freeze-dried hydrogels showed that the gels were denser when the dextran content increased
from 20 to 80 %, with pore sizes from 3.5 to 0.8 μm, respectively. Due to the presence of the
NIPAAm groups in the gels the swelling of the gels decreased when the temperature was elevated
above the LCST. Furthermore, below the LCST, the swelling ratio was reduced when the dextran content was augmented, owing to a denser network, higher pore density and less
thermosensitivity. Depending on specific requirements of possible applications, the response to external temperature changes of these gels can be tailored by varying the
hydrophilic/hydrophobic balance of the gels. The same group reported on temperature and pH-sensitive hydrogels, again composed of dextran
and NIPAAm.[61] In this study, the dextran hydroxyl groups were derivatized with maleic anhydride, yielding a polymerizable macromer with carboxylic acid groups (Fig. 1F). As for the dex-
AI/PNIPAAm systems, an LCST increase from 35.9 to 39.1 °C was observed when the ratio (w/w)
dex-MA/PNIPAAm was increased from 0.25 to 4. These slightly higher LCST values when compared to the Dex-AI composites were due to a different hydrophilicity level of the dextran derivatives.
SEM imaging revealed a change in pore structure from irregular round and loose to well-defined honey-comb like, when dex-MA was copolymerized with NIPAAM (Fig. 4). Increasing amounts of dex-MA resulted in smaller pores. As in the first study, Zhang et al. explain the relationship between
pore size and dextran content by the presence of the crosslinkable groups along the dextran backbone, affecting the crosslinking level. A higher dextran content leads to more crosslinks and a
denser network as a result. The crosslinking level also greatly affected the swelling ratio with less swelling with increasing dex-MA content. Moreover, the gel properties were pH dependent. In
alkaline medium the gels exhibited a swelling ratio that was almost double of that in acidic medium. The free carboxylic groups in dex-MA are deprotonated at neutral and alkaline pH,
causing electrostatic repulsion between the polymer chains and thus an increased swelling. The dex-AI/PNIPAAm hybrid hydrogels were also used for the preparation of temperature-sensitive dendrite-shaped particles.[62] Microgels (average diameter of 1 mm) were obtained after
precipitation polymerization of dex-AI and NIPAAm in water at 75 °C in the presence of
ammonium persulfate (APS) as initiator. DSC measurements showed an LCST around 26 °C, which
is significantly lower than the LCST of the corresponding macroscopic gels. Additionally a faster
and larger swelling of the particles was observed when compared to macroscopic gels due to a lower crosslinking and higher surface area.
A major disadvantage of the PNIPAAm-containing hydrogels described above is the non-biodegradability of the thermosensitive polymer. Extensive in vitro and in vivo degradation studies
need to be conducted to evaluate the biodegradability of the dex/PNIPAAm hybrid networks.
Biodegradable dextran hydrogels for protein delivery applications
25
Figure 4: SEM images of dex-MA/PNIPAAm hydrogels: Figure (A) is pure PNIPAAm. Figures (B) to (E) show hydrogels with increasing dextran content. Reprinted from
Biomaterials 25, 4719-4730 (2004) Zhang et al. with permission from Elsevier.[60]
Sun et al. reported on a photopolymerized pH-sensitive hydrogel composed of dextran-allyl
isocynate-ethylamine (dex-AE) and poly(ethylene glycol)-diacrylate (PEGDA)[63]. The influence of the dex-AE/PEGDA ratio on the pore size of the gels was investigated with SEM. A denser and more compact interior morphology was observed with increasing content of dex-AE from 0 to
30 %. Further increase of the dex-AE content to 50 and 70 % led to a more open and looser structure. This phenomenon was attributed to an incomplete crosslinking, also evidenced by a
reduced conversion efficiency due to the lower activity of the C=C bonds in the allyl isocyanate of dex-AE when compared to the C=C bonds in PEGDA. Swelling studies confirmed these findings
with a lowest swelling ratio for gels with the intermediate dex-AE content (30 %). Furthermore, the swelling ratio was pH dependent, with a higher swelling at pH 3 than at pH 10, especially
significant for those hydrogels containing higher amounts of dex-AE. This effect was ascribed to electrostatic repulsion between the protonated amine groups in dex-AE in acidic conditions, leading to expansion of the network and increase in swelling. The hydrogels were loaded with BSA
by soaking of the gels in BSA solution for 48 h at room temperature. Release of BSA took place over 10 to 30 days from hydrogels with 10 to 70 % dex-AE. All hydrogels exhibited a burst release
during the first 8 h, attributed to release of BSA located near the surface of the gels, as a result of the inhomogeneous loading of the gels after soaking in BSA solution. Contrary to the results of the
swelling experiments and the SEM images, a more sustained release was observed from gels with a higher dex-AE content. This indicates that the BSA release is not only dependent on the network structure but also on electrostatic interactions between the positively charged amine groups in
dex-AE and negatively charged BSA molecules.
D
CBA
E
Chapter 2
26
3.2.3.2. Glucose-responsive dextran hydrogels
Glucose-responsive UV-polymerized dextran-based hydrogels were developed by Tanna et al. [64] for the design of a self-regulating insulin delivery system. Dextran and concanavalin A (con A) were
derivatized with methacrylic anhydride to introduce polymerizable groups. Con A is a glucose-selective lectin, i.e. a protein containing moieties that specifically interact with glucose. Rheological analysis of the crosslinked mixtures revealed that although there are covalent bonds present, the
complex viscosity decreased with increasing glucose concentration. This effect can be attributed to the competition between free glucose and glucopyranose units of dextran to interact with
specific receptor sites of con A. To be of physiological relevance, the system should have a reduced complex viscosity, enabling the release of insulin, when the glucose concentration is 0.1-1 %,
which corresponds to ∼ 5.5-55 mmol/L. Tanna et al. found that a dex-MA (DS 3)-con A-MA gel,
which was irradiated with UV-light (365 nm, 10 mJ cm-2) for 50 min, showed an 80 % drop in viscosity when the relevant glucose concentration was reached. A longer irradiation time resulted
in more permanent crosslinks, leading to a viscosity drop of only 50 % for gels that were irradiated for 70 and 100 min. However, sufficient crosslinking is essential to minimize component leach, i.e.
non-bonded con A and dex-MA. In vitro insulin release studies revealed an increasing graded response to glucose concentrations between 0.1 and 1 %. Additionally, insulin release was
reversibly affected by the glucose concentration and could be triggered repeatedly (Fig. 5). In a subsequent paper Tanna et al. investigated the effect of the derivatization degree of both dextran and con A on the glucose-dependent insulin release.[65] As would be expected, the presence of
more polymerizable acrylic groups on dextran led to a tighter crosslinked network and related material properties. Surprisingly, non-polymerized dextran and con A derivatives also showed a
glucose-dependent complex viscosity profile. This was ascribed to the competition of glucose units in dextran and free glucose with the glucose-binding sites on con A, causing a rupture of the
three-dimensional network. Apparently, besides covalent chemical crosslinks, physical entanglements are important for the network properties of the mixture. A slower glucose response time was found when the substitution degree of dextran was raised from 3 to 5.8 %. The
derivatization degree of con A had an optimum at 60 %, with the lowest component leach and satisfactory glucose sensitivity. A higher substitution degree resulted in partial denaturation of con
A leading to a loss of glucose binding capacity and less effective physical crosslinking with dextran chains. To design a closed-loop insulin delivery device, a compromise between the glucose
response time and the glucose sensitivity of the material is indispensable. Therefore, the crosslink density of the hydrogel should be sufficient to prevent an early drop of the complex viscosity but
the network should not hamper the insulin diffusion outwards.
Biodegradable dextran hydrogels for protein delivery applications
27
Figure 5: Insulin release from dex-MA-con A-MA hydrogels in response to altering glucose concentrations.
Reprinted from Biomaterials 27, 1586-1597 (2006) Tanna et al. with permission from Elsevier.[63]
3.2.4. Oxidized dextrans
Maia et al. took advantage of the self-assembling properties of oxidized dextrans (dexOx) (Fig. 1G)
in the presence of adipic acid dihydrazide (AAD) to form injectable hydrogels.[66] Dextran was oxidized with sodium periodate and subsequent reaction of the aldehyde groups with hydrazide groups of AAD resulted in crosslinking at room temperature overnight. Rapid conversion of a
viscous solution to an elastic gel was observed within 4 minutes after which a long curing process (above 3 h) took place before complete gelation was accomplished. Swelling and degradation of
the hydrogels was dependent on the number of intermolecular crosslinks and as a consequence on the AAD content. At physiological conditions, the degradation time could be varied from 9 to
15 and 23 days, for gels consisting of 15 % oxidized dextrans with, respectively, 5, 10 and 20 % of AAD. The influence of degradation on the pore-size was studied by means of SEM and mercury
intrusion porosimetry (MIP). Maia et al. found that the pore size increased during degradation of the network from 1.5 to 6.3 μm after 11 days. In tissue engineering applications this might assist the ingrowth of cells and result in a better integration of the matrix with the surrounding tissue.
Hydrogels prepared by reaction of aldehyde groups of oxidized dextran with amino residues in gelatin were described by Schacht et al.[67] Both chemical crosslinks and physical gelation of gelatin
were responsible for the gel strength. It was found that the storage modulus (G’) increased with increasing degree of oxidation of dextran. An oxidation degree lower than 20 % resulted in weaker
networks when compared to control gelatin hydrogels, probably due to partial separation between the incompatible gelatin and dextran phases. Sustained release of epidermal growth factor (EGF) from hydrogel films was observed up to 7 days.[68] Schacht et al. emphasize that the in
vivo release profiles might differ considerably from the in vitro results due to the biodegradable nature of gelatin and dextran. In vivo biocompatibility studies of gelatin-oxidized dextran
composites showed a moderate foreign body reaction around the subcutaneous implants.[69]
Time (min)
Insu
lin c
once
ntra
tion
(mg/
ml)
1 % w/v glucose trigger
0.5 % w/v glucose trigger
0.2 % w/v glucose trigger
Δ 0.1 % w/v glucose trigger
Control – no glucose added
GLUCOSE ADDED
GLUCOSE REMOVED
Chapter 2
28
3.2.5. Enzymatically synthesized hydrogels
An elegant enzymatic synthetic route toward dextran hydrogels was developed by Ferreira et al. [70, 71] They made use of a single-step biocatalytic transesterification reaction between dextran and divinyladipate (DVA) in dimethylsulfoxide (DMSO) (Fig. 1H). Different proteases and lipases were tested on their conversion capability, of which an alkaline protease from B. subtilis (Proleather FG-F)
performed the best. The elasticity modulus of the enzymatically produced hydrogels ranged from 1.4 kPa to 5.8 kPa for hydrogels with a degree of substitution (DS, i.e. number of DVA molecules per
100 glucopyranose residues) of respectively 20 and 47 %. Presumably, the Proleather catalysis favors the formation of intermolecular crosslinks due to a high efficiency in promoting the
attachment of both of the terminal vinyl ester groups of DVA to dextran. Additionally, SEM images and MIP measurements of the biocatalytic hydrogels showed that the total porosity was at least 80 % with a unimodal, narrow and relatively homogenous pore size distribution with average
diameters of 0.4 to 2 μm, depending on the crosslink density of the network. Histological examination of surrounding tissue after subcutaneous implantation of the hydrogels in rats
indicated a good biocompatibility of the materials. The inflammatory response was mild with granulocyte and lymphocyte cells disappeared by day 10. Macrophages that had phagocytosed
hydrogel particles were observed 5 to 40 days after implantation, depending on the DS. This phagocytosis process begins earlier for low DS hydrogels since less crosslinks are present and
degradation is more rapid. Only for hydrogels with the highest DS (31 and 47 %) collagen, deposited by the fibroblasts surrounding the macrophages and foreign body giant cells, organized into a fibrous capsule by day 10. For the DS 31 % gel no mature capsule could be formed due to
degradation of the gel by that time, while for the DS 47 % a discontinuous fibrous capsule was observed. Ferreira et al. state that these gels are suitable for tissue engineering and drug delivery
applications because of their superior mechanical properties, no substantial fibrous capsule formation after implantation and porosities above 80 %.
3.3. Physically crosslinked dextran hydrogels
As pointed out above, hydrogels created by physical interactions are especially attractive
candidates for protein delivery and tissue engineering applications since they are formed through self-assembly, without the aid of crosslinking agents.
3.3.1. Crystallization of dextran in aqueous solution
Crystallization was used by Stenekes et al. to produce physically crosslinked dextran hydrogels.[72] Sol-gel conversion of low-molecular weight dextran (Mw 6000 g/mol) in aqueous solution was
observed caused by intermolecular hydrogen bonding. Rheology confirmed the formation of
mainly elastic gels with tan(δ) values of 0.05 to 0.03 for 40 and 60 % dextran 6000 solutions,
respectively. For more concentrated solutions a shorter lag time and a higher G’ plateau value was observed, demonstrating that precipitation and gelation were more rapid. The authors hypothesize that the dextran chains associate through hydrogen bonding, after which nucleation
and growth of the crystals occurs. A low degree of hydration in more concentrated solutions as well as the presence of salt ions that require water molecules to dissolve, facilitate association of
Biodegradable dextran hydrogels for protein delivery applications
29
the chains and thus crystallization. Stirring of the solution induces orientation of the chains also
resulting in accelerated crystallization.
3.3.2. Polymer inclusion complexes of cyclodextrins with dextran derivatives
The group of Yui reported on supramolecular assembly of hydrophobically aggregated crystalline
domains creating injectable polymer inclusion complexes (PICs). They made use of the favorable property of cyclodextrins (CDs) to selectively include a wide range of guest molecules. CDs are
cyclic water-soluble oligosaccharides with internal hydrophobic cavities. PEG[73] or poly(propylene
glycol) (PPG),[74] of which it is known to form inclusion complexes with α-CDs and β-CDs,
respectively, was grafted to dextran (Fig. 1I and 1J). Addition of the graft copolymers to a CD
solution rendered the solution opaque en eventually led to the formation of a gel as a result of physical crosslinks between crystalline inclusion complexes, induced by hydrogen bonding
between CDs threaded along the PEG/PPG chains. The time required for complete gel formation could be tailored by varying the concentration of the graft copolymer and the PEG/PPG content,
i.e. the graft density on the dextran chains. Contrary, when PEG/PPG was only mixed with CDs, crystalline precipitates were formed, as a result of intermolecular hydrogen bonding between the CDs threaded along the PEG/PPG chains. When hydrophilic dextran chains are present, the
hydrophobic PIC domains act as physical junctions, creating a hydrogel (Fig. 6). It was further found that above a certain temperature (Tgel-melting) the supramolecular assembly dissociates
yielding a viscous solution, but after cooling down to a specific temperature (Tgelation) an opaque gel was reformed. Significant differences were observed between the Tgel-melting and Tgelation and
both transitions could be controlled by altering the concentration, the molar ratio [EG/PG]/[CD]
and the PEG/PPG content of the graft copolymer. For dex-PEG-α-CD complexes with a molar
[EG]/[CD] ratio of 2 the Tgel-melting and Tgelation typically ranged from 50 to 60 °C and from 35 to 45 °C,
respectively, depending on the PEG content. A dramatic increase of the Tgel-melting and Tgelation was found with an increasing [EG]/[CD] ratio from 1 to 2, but a further increase when the ratio was above 2 was hardly seen.[73]
Figure 6: Schematic presentation of the self-assembly of dex-PEG with α-CDs into supramolecular structures.
(A) Uncomplexed state before inclusion complexation occurs (B) α-CDs threaded along PEG chains resulting in
inclusion complexes. Reprinted from Macromolecules 34, 8657-8662 (2001) Huh et al. with permission from ACS.[72]
Chapter 2
30
For dex-PPG-β-CD complexes with a molar [PG]/[CD] ratio of 2 the Tgel-melting and Tgelation was 60 to
65 °C and 80 to 85 °C respectively, also depending on the PPG content (Fig. 7). Both Tgel-melting and
Tgelation gradually decreased when the [PG]/[CD] ratio exceeded 2.[74] It is known that the stoichiometry of [EG/PG]/[CD] is 2, meaning that at higher ratios almost all the PEG/PPG grafts and
CD moieties already participated in the inclusion complexation with no significant additional change in the gel properties as a result. In addition to the thermoreversible gelation based on physical interactions between hydrophobic
inclusion complexes a pH-sensitive functionality was introduced.[75] In this study dextran was
grafted with a cationic polymer poly(ε-lysine) (PL) (Fig. 1K). At high pH, the primary amines of PL
are deprotonated, allowing the CDs to be threaded onto the PL chain. Again, the concentration, molar feed ratio and grafting density were determining for the phase transition behavior. At pH 4, a rapid gel-to-sol transition occurred, due to dissociation of the inclusion complexes in the
protonated state of PL. This is due to the energetically unfavorable situation of protonated amines in the hydrophobic cavities of the CDs.
Figure 7: Melting and gelation temperatures of dex-PPG/β-CD mixtures as a function of the PG/CD ratio and
the number of PPG graft. Reprinted from Macromolecular Bioscience 2, 298-303 (2002) Choi et al. with
permission from Wiley Interscience.[73]
3.3.3. Stereocomplexation
Upon mixing of two polymer enantiomers, possessing opposite chirality, the formation of crystals occurs, referred to as stereocomplexes. De Jong et al. made use of stereocomplex formation to
create physically crosslinked hydrogels.[76] Lactic acid oligomers were synthesized and coupled to dextran (Fig. 1L). Both L- and D-lactide were used resulting in dex-L- and dex-D-lactate. Rheological analysis revealed that upon mixing aqueous solutions of both polymers a hydrogel was formed. In
time, a substantial increase of the elasticity modulus (G’) was observed whereas the dex-L-lactate solutions did not show any change. Creep experiments showed an almost elastic behavior of the
mixture while the dex-L-lactate solution behaved as a visco-elastic material (Fig. 8). The presence of stereocomplex crystals, creating the physical junctions between the dextran
chains, was confirmed by X-ray diffraction.[77] It was further found that the length of the oligolactate chains played a crucial role in the network formation. When oligolactate chains of
DPav 5 were used, i.e. on average 5 repeating lactate units, only a weak network was formed upon mixing of dex-L- and dex-D-lactate, comparable to a dex-L-lactate solution alone.
Biodegradable dextran hydrogels for protein delivery applications
31
0
500
1000
1500
2000
2500
3000
0 20 40 60 80 100 120
time (s)
stra
in (%
)
0,00,51,01,52,02,53,03,54,04,5
0 20 40 60 80 100 120time (s)
stra
in (%
)
Figure 8: Creep experiments illustrating the visco-elastic behavior of a dex-L-lactate solution (A) and the elastic
behavior of a mixture of dex-L- and dex-D-lactate solutions (B). Reprinted from Macromolecules 33, 3680-3686
(2000) de Jong et al. with permission from ACS.[75]
Apparently, longer oligolactate chains are required to result in sufficient interactions. DSC analysis showed that in blends of enantiomeric lactic acid oligomers stereocomplex formation could only
occur when the DPav was ≥ 7. In the case of monodisperse oligomers, coupled to dextran, the DP
should at least be 11 to create hydrogels. Stronger hydrogels could be obtained by increasing the DPav, the lactate substitution degree and the solid content of the gels (i.e. the initial polymer
fraction in the gel).[78] Along with a higher network density, stronger hydrogels and longer degradation times are related. Degradation of the gels is caused by OH- driven hydrolysis and
could be varied at pH 7 from 1 to 3.5 days for hydrogels of 70 % water content, DPav 9, DS 3 and DPav 12 DS 6, respectively. A higher pH resulted in accelerated degradation whereas at pH 4 the hydrogels remained stable for more than 1 month. Longer degradation times up to approximately
60 days could be obtained with stereocomplexed hydrogels in which the dextran backbone was replaced by 2-hydroxypropyl methacrylamide (HPMA).[79] In these hydrogels, the oligolactate side
chains were acetylated, preventing rapid chain end scission (backbiting) making slow random chain scission the main action of degradation. A diffusion-controlled release from dex-lactate
hydrogels of the model protein lysozyme during approximately 5 days was observed, whereas the larger protein IgG was mainly released subsequent to degradation of the matrix in about 8 days,
from hydrogels with 70 % initial water content and DS 6. Low polydisperse lactate grafts, i.e. DP 11-14 instead of DP 1-30, led to denser network structures, as evidenced by rheology, and resulted in a slightly retarded release of both proteins.[80]
In a succeeding study by Bos et al. the dex-lactate hydrogels were used as in situ gelling system for the release of the therapeutic protein recombinant human interleukin-2 (rhIL-2).[28] From hydrogels
containing 82 % water a rapid in vitro release was followed by a more gradual release the next days, with 65 % of the protein released within 3 days. An in vivo release study was conducted in
SL2 lymphosarcoma-bearing mice. RhIL-2 is a broad acting T cell-derived cytokine with anti-tumor activity after local administration. Dex-L-lactate and dex-D-lactate solutions, both containing rhIL-2, were mixed prior to injection in the peritoneal cavity where they gelled in situ. All mice treated
with the rhIL-2-loaded gel, were cured, whereas the mice in the negative control groups (buffer and empty hydrogel) all died and the mice in the positive control groups (free rhIL-2, 1 and 5
injections) had a cure rate of 60 %. Since the tumor had infiltrated the abdominal muscles and metastasized in lungs, liver and other organs by the day the hydrogels were injected, it is
noteworthy that the local rhIL-2 treatment led to a systemic effect. At day 60 of the study, all cured
A B
Chapter 2
32
mice were rechallenged with SL2 cells and appeared to be immune to the tumor. It was
concluded that these stereocomplexed dex-lactate hydrogels enhance the clinical applicability of rhIL-2 therapy as it was shown that the therapeutic efficacy of one injection of the rhIL-2-
containing hydrogels was as least as good as the free rhIL-2 injections on 5 consecutive days. Bos et al. also reported on the in vivo biocompatibility and tissue reactions of the dex-lactate gels after subcutaneous implantation in rats.[81] At the outset of the study, several sterilization techniques
were exploited and their influence on the network properties of the gels was investigated. Dry heat sterilization was the preferred method, since it neither degrades the lactate side chains
(caused by autoclaving) nor the dextran backbone (caused by gamma irradiation). After subcutaneous implantation of the hydrogels the rats were sacrificed at various time points and the
surrounding tissue was examined. It was found that at day 1 polymorph nuclear cells (PMN) had infiltrated near the gels. By day 3, PMN were no longer present, indicating that possible cytotoxic
or complement activating substances were not released from or present in the gels. After 5 days, gel particles were actively phagocytosed by macrophages. By day 15 and 30, the gel had disappeared and was replaced by connective tissue. These results demonstrate that the
stereocomplexed dex-lactate hydrogels showed a good biocompatibility, evoking only a mild foreign body reaction, mainly directed to the degradation of the gels. The immune system was
hardly triggered by the gel or degradation product, as evidenced by the low number of lymphocytes.
3.3.4. Ionically crosslinked dextran-based hydrogels
Nanoparticles composed of negatively charged dextran-sulfate ionically crosslinked with positively charged chitosan were described by Sarmento et al.[82] Complex coacervation occurred after
dropwise addition of chitosan solution of pH 5.0 to dextran-sulfate solution of pH 3.2, resulting in spherical 500 nm-sized particles with a smooth surface. To obtain insulin-loaded particles, the
protein was dissolved in the dextran-sulfate solution prior to chitosan addition, leading to an association efficiency of 85 %. Insulin release from these nanoparticles showed to be pH dependent. At pH 1.2, 4.5 and 5.2 no insulin was released, attributable to the overall positive
charge of insulin at pH values lower than its pI (5.3), retaining the protein at the negatively charged dextran-sulfate sites. At pH 6.8 sustained insulin release was observed with 40 % released after
15 min followed by a slower release up to 70 % after 5 h. Sarmento et al. states that these release profiles suggest a dissociation-driven, pH-dependent release mechanism. This was further
illustrated by a slower release from particles with a higher dextran-sulfate/chitosan ratio. Increasing the ratio from 1:1 to 2:1 decreased the total insulin release after 24 h from 76 to 59 %. Both ELISA and HPLC analysis showed that the released insulin was intact, indicative of a protein-friendly
nanoparticle preparation method. Another approach to design ionically crosslinked hydrogels, based on ionic interactions between
oppositely charged dextran microspheres is reported in this thesis.
Biodegradable dextran hydrogels for protein delivery applications
33
4. Expert Commentary
A large variety of polymeric hydrogels, both of synthetic and natural origin, have been developed and used for drug delivery purposes. Many of them are based on non-biodegradable polymers requiring surgical removal of the device when drug release is completed. Consequently, interest
has grown in polymers that are biodegraded into harmless products. Their degradation rate should be tailorable to meet the requirements of specific applications in drug delivery and tissue
engineering. Biocompatibility is another feature of candidate polymers that is of the utmost importance. As discussed in this review, dextran possesses both favorable characteristics,
biodegradability and biocompatibility. Numerous approaches have been exploited for the design of dextran-based hydrogels. The major part of them involves chemical crosslinking of derivatized dextran during which encapsulated cells or proteins might be adversely affected. On the other
hand, chemical crosslinking results in matrices of high mechanical strength with reproducible properties. Strategies are available to create stimuli-responsive hydrogels that react upon changes
in temperature, pH, ionic strength and even glucose concentration. In the field of the physically crosslinked hydrogels composed of dextran derivatives, many paths are not yet tread on. Most
promising are those strategies based on supramolecular self-assembly, leading to injectable matrices that gellify at the site of injection. Ideally, hydrogels will be designed that can act both as
scaffolds to support cell growth and as delivery devices to release proteins over a controlled period of time to assist in the formation of new tissue. Giving the excellent performance of dextran hydrogels as protein delivery systems and tissue engineering scaffolds, as pointed out in this
review, in the future various clinical applications can be foreseen.
5. Five-year view
Extensive research has been done on dextran hydrogels designed for biomedical applications. The various approaches, discussed throughout the text, have been listed in Table 2. Strikingly, only few of them have been investigated on their behavior in vivo. Furthermore, in those cases where
protein delivery is envisioned, mainly the in vitro release of model proteins has been monitored. It is clear that research focus should shift to a more in depth application-driven investigation of the
dextran hydrogels developed up till now. Special attention should be given to their performance as carriers of pharmaceutically relevant proteins and the in vitro-in vivo correlation should be
thoroughly addressed. In vivo biocompatibility of the devices is of the utmost importance and has for many of the current dextran gels not been sufficiently tackled thus far. Although dextran is a
biocompatible polymer, dextran derivatives or substances used for the preparation of the devices can influence their in vivo faith. It can be anticipated that most of the hydrogel systems discussed in this review will be further
exploited on their potentials in pre-clinical evaluations as well as in clinical trials. Likely, five years from now, a number of dextran-based hydrogels will have entered clinical trials with protein
delivery as major application. The favorable properties of dextran hydrogels as tissue engineering scaffolds have also been summarized in this review. Chances are high that clinical evaluation will
follow after the successful introduction of protein releasing hydrogels.
Chapter 2
34
6. Key issues
• Recombinant DNA technology has lead to a whole new generation of protein-based pharmaceuticals. Since traditional pharmaceutical dosage forms (tablet, capsule) are not suitable
to formulate proteins, there is a need for applicable delivery systems.
• Delivery systems should release the entrapped protein in a controlled manner, with full
preservation of its bioactivity. Biodegradability and biocompatibility of the release device should be ensured.
• Hydrogels exhibit favorable characteristics as protein delivery matrices, such as high water
content, tailorable network properties (and as a consequence controllable release profiles) and degradation behavior.
• Dextran is a non-toxic and highly water-soluble polysaccharide that has been clinically used for over 50 years as plasma volume expander. The hydroxyl groups along the dextran backbone
render it particularly suitable for derivatization and subsequent crosslinking to yield hydrogels.
• In recent years, there is a growing interest in self-assembling hydrogels that are injectable and
gellify in situ.
• The soft and rubbery nature of hydrogels and the low tendency of cells to adhere to the
hydrogel surface certify minimal tissue irritation and make them attractive candidates as tissue
engineering scaffolds.
Biodegradable dextran hydrogels for protein delivery applications
35
References
1. Pavlou A, Reichert J. Recombinant protein therapeutics - success rates, market trends and values to 2010. Nat. Biotechnol. 2004;22:1513-1519.
2. Reichert J. Trends in US approvals: new biopharmaceuticals and vaccines. Trends Biotechnol. 2006;24:293-298.
3. Orive G, Hernandez RM, Rodriguez Gascon A, Dominguez-Gil A, Pedraz JL. Drug delivery in biotechnology: present and future. Curr. Opin. Biotechnol. 2003;14:659-664.
4. Hermeling S, Crommelin DJA, Schellekens H, Jiskoot W. Structure-Immunogenicity relationships of therapeutic proteins. Pharm. Res. 2004;21:897-903.
5. Frokjaer S, Otzen DE. Protein drug stability: a formulation challenge. Nat. Rev. Drug Discov. 2005;4:298-306.
6. Bailon P, Berthold W. Polyethylene glycol-conjugated pharmaceutical proteins. Pharm. Sci. Technol. To. 1998;1:352-356.
7. Beals JM, Shanafelt AB. Enhancing exposure of protein therapeutics. Drug Discov. Today 2006;3:87-94.
8. http://www.in-pharmatechnologist.com/news/ng.asp?n=69488-octoplus-octodex-protein-formulation-polyactive-pegylation (last visited on 31.07.2007)
9. Hoffman AS. Hydrogels for biomedical applications. Adv. Drug Deliver. Rev. 2002;43:3-12.
10. Peppas NA, Hilt JZ, Khademhosseini A, Langer R. Hydrogels In biology and medicine: from molecular principles to biotechnology. Adv. Mater. 2006;18:1345-1360.
11. Freiberg S, Zhu XX. Polymer microspheres for controlled drug release. Int. J. Pharm. 2004;282:1-18.
12. Panyam J, Labhasetwar V. Biodegradable nanoparticles for drug and gene delivery to cells and tissue. Adv. Drug Deliver. Rev. 2002;55:329-347.
13. Metselaar JM, Mastrobattista E, Storm G. Liposomes for intravenous drug targeting. Mini-Rev. Med. Chem. 2002;2:319-329.
14. Jorgensen L, Moeller EH, van de Weert M, Nielsen HM, Frokjaer S. Preparing and evaluating delivery systems for proteins. Eur. J. Pharm. Sci. 2006;29:174-182.
15. Friess W. Collagen- biomaterial for drug delivery. Eur. J. Pharm. Biopharm. 1998;45:113-136.
16. Masters KS, Shah DN, Leinwand LA, Anseth KS. Crosslinked hyaluronan scaffolds as a biologically active carrier for valvular interstitial cells. Biomaterials 2005;26:2517-2525.
17. Leonard M, De Boisseson MR, Hubert P, Dalencon F, Dellacherie E. Hydrophobically modified alginate hydrogels as protein carriers with specific controlled release properties. J. Control. Release 2004;98:395-405.
18. Woo BH, Jiang G, Jo YW, DeLuca PP. Preparation and characterization of a composite PLGA and poly(Acryloyl Hydroxyethyl Starch) microsphere system for protein delivery. Pharm. Res. 2001;18:1600-1606.
19. Berger J, Reist M, Mayer JM, Felt O, Peppas NA, Gurny R. Structure and interactions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications. Eur. J. Pharm. Biopharm. 2004;57:19-34.
20. Van de Wetering P, Metters AT, Schoenmakers RG, Hubbell JA. Poly(ethylene glycol) hydrogels formed by conjugate addition with controllable swelling, degradation, and release of pharmaceutically active proteins. J. Control. Release 2005;101:619-627.
21. Martens PJ, Anseth KS. Characterization of hydrogels from acrylate modified poly(vinyl alcohol) macromers. Polymer 2000;41:7715-7722.
22. Lu S, Anseth KS. Photopolymerization of multilaminated poly(HEMA) hydrogels for controlled release. J. Control. Release 1999;57:291-300.
23. Bowen WR, Liang Y, Williams PM. Gradient diffusion coefficients- theory and experiment. Chem. Eng. Sci. 2000;55:2359-2377.
24. Peppas NA, Bures P, Leobandung W, Ichikawa H. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 2000;50:27-46.
25. Hennink WE, van Nostrum CF. Novel crosslinking methods to design hydrogels. Adv. Drug Deliver. Rev. 2002;54:13-36.
26. Ruel-Gariepy E, Leroux J-C. In situ-forming hydrogels--review of temperature-sensitive systems. Eur. J. Pharm. Biopharm. 2004;58:409-426.
27. Gupta P, Vermani K, Garg S. Hydrogels: from controlled release to pH-responsive drug delivery. Drug Discov. Today 2002;7:569-579.
28. Bos GW, Jacobs JJL, Koten JW, Van Tomme SR, Veldhuis TFJ, van Nostrum CF, Den Otter W, Hennink WE. In situ crosslinked biodegradable hydrogels loaded with IL-2 are effective tools for local IL-2 therapy. Eur. J. Pharm. Sci. 2004;21:561-567.
29. Park YD, Tirelli N, Hubbell JA. Photopolymerized hyaluronic acid-based hydrogels and interpenetrating networks. Biomaterials 2003;24:893-900.
30. Hatefi A, Amsden B. Biodegradable injectable in situ forming drug delivery systems. J. Control. Release 2002;80:9-28.
Chapter 2
36
31. Packhaeuser CB, Schnieders J, Oster CG, Kissel T. In situ forming parenteral drug delivery systems: an overview. Eur. J. Pharm. Biopharm. 2004;58:445-455.
32. Anseth KS, Metters AT, Bryant SJ, Martens PJ, Elisseef JH, Bowman CN. In situ forming degradable networks and their application in tissue engineering and drug delivery. J. Control. Release 2002;78:199-209.
33. Cohn D, Sosnik A, Garty S. Smart hydrogels for in situ generated implants. Biomacromolecules 2005;6:1168-1175.
34. Jeong B, Gutowska A. Lessons from nature: stimuli-responsive polymers and their biomedical applications. Trends Biotechnol. 2002;20:305-311.
35. Walker GJ. Dextrans. University Park Press: Baltimore, 1978; Vol. 16, p 75-125.
36. Mehvar R. Dextrans for targeted and sustained delivery of therapeutic and imaging agents. J. Control. Release 2000;69:1-25.
37. http://www.dextran.net (last visited on 31.07.2007)
38. Arturson G, Wallenius G. The renal clearance of dextran of different molecular sizes in normal humans. Scan. J. Clin. Lab. Inv. 1964;1:81-86.
39. Arturson G, Wallenius G. The intravascular persistence of dextran of different molecular sizes in normal humans Scan. J. Clin. Lab. Inv. 1964;1:76-80.
40. Edman P, Ekman B, Sjöholm I. Immobilization of proteins in microspheres of biodegradable polyacryldextran. J. Pharm. Sci. 1980;69:838-842.
41. Van Dijk-Wolthuis WNE, Hoogeboom JAM, van Steenbergen MJ, Tsang SKY, Hennink WE. Degradation and release behavior of dextran-based hydrogels. Macromolecules 1997;30:4639-4645.
42. Van Dijk-Wolthuis WNE, Tsang SKY, Kettenes-van den Bosch JJ, Hennink WE. A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer 1997;38:6235-6242.
43. De Groot CJ, Van Luyn MJA, Van Dijk-Wolthuis WNE, Cadee JA, Plantinga JA, Den Otter W, Hennink WE. In vitro biocompatibility of biodegradable dextran-based hydrogels tested with human fibroblasts. Biomaterials 2001;22:1197-1203.
44. Cadée JA, van Luyn MJA, Brouwer LA, Plantinga JA, van Wachem PB, de Groot CJ, den Otter W, Hennink WE. In vivo biocompatibility of dextran-based hydrogels. J. Biomed. Mater. Res. 2000;50:397-404.
45. Franssen O, Hennink WE. A novel preparation method for polymeric microparticles without the use of organic solvents. Int. J. Pharm. 1998;168:1-7.
46. Stenekes RJH, Franssen O, van Bommel EMG, Crommelin DJA, Hennink WE. The preparation of dextran microspheres in an all-aqueous system: effects of the formulation parameters on particle characteristics. Pharm. Res. 1998;15:557-561.
47. Franssen O, Vos OP, Hennink WE. Delayed release of a model protein from enzymatically-degrading dextran hydrogels. J. Control. Release 1997;44:237-245.
48. Franssen O, Vandervennet L, Roders P, Hennink WE. Degradable dextran hydrogels: controlled release of a model protein from cylinders and microspheres. J. Control. Release 1999;60:211-221.
49. Cadée JA, de Groot CJ, Jiskoot W, den Otter W, Hennink WE. Release of recombinant human interleukin-2 from dextran-based hydrogels. J. Control. Release 2002;78:1-13.
50. De Groot CJ, Cadée JA, Koten JW, Hennink WE, Den Otter W. Therapeutic efficacy of IL-2-loaded hydrogels in a mouse tumor model. Int. J. Cancer 2002;98:134-140.
51. Vlugt-Wensink KDF, de Vrueh R, Gresnigt MG, Hoogerbrugge CM, van Buul-Offers SC, de Leede LGJ, Sterkman LGW, Crommelin DJ, Hennink WE, Verrijk R. Preclinical and clinical in vitro in vivo correlation of an hGH dextran microsphere formulation. 2007 submitted.
52. De Geest BG, Déjugnat C, Sukhorukov GB, Braeckmans K, De Smedt SC, Demeester J. Self-rupturing microcapsules Adv. Mater. 2005;17:2357-2361.
53. De Geest BG, Dejugnat C, Verhoeven E, Sukhorukov GB, Jonas AM, Plain J, Demeester J, De Smedt SC. Layer-by-layer coating of degradable microgels for pulsed drug delivery. J. Control. Release 2006;116:159-169.
54. Lévesque SG, Lim RM, Shoichet MS. Macroporous interconnected dextran scaffolds of controlled porosity for tissue-engineering applications. Biomaterials 2005;26:7436-7446.
55. Lévesque SG, Shoichet MS. Synthesis of cell-adhesive dextran hydrogels and macroporous scaffolds. Biomaterials 2006;27:5277-5285.
56. Maire M, Logeart-Avramoglou D, Degat M-C, Chaubet F. Retention of transforming growth factor-β1 using functionalized dextran-based hydrogels. Biomaterials 2005;26:1771-1780.
57. Maire M, Chaubet F, Mary P, Blanchat C, Meunier A, Logeart-Avramoglou D. Bovine BMP osteoinductive potential enhanced by functionalized dextran-derived hydrogels. Biomaterials 2005;26:5085-5092.
58. Zhang R, Tang M, Bowyer A, Eisenthal R, Hubble J. A novel pH- and ionic-strength-sensitive carboxy methyl dextran hydrogel. Biomaterials 2005;26:4677-4683.
Biodegradable dextran hydrogels for protein delivery applications
37
59. Zhang X, Sun G, Wu D, Chu C. Synthesis and characterization of partially biodegradable and thermosensitive hydrogel. J. Mater. Sci. Mater-M. 2004;15:865-875.
60. Zhang X-Z, Wu D-Q, Sun G-M, Chu C-C. Novel biodegradable and thermosensitive Dex-AI/PNIPAAm hydrogel. Macromol. Biosci. 2003;3:87-91.
61. Zhang X, Wu D, Chu C-CC-C. Synthesis and characterization of partially biodegradable, temperature and pH sensitive Dex-MA/PNIPAAm hydrogels. Biomaterials 2004;25:4719-4730.
62. Zhang X-Z, Sun G-M, Chu CC. Temperature-sensitive dendrite-shaped PMIPAAm/Dex-AI hydbrid hydrogel particles: formulation and properties. Eur. Polym. J. 2004;40:2251-2257.
63. Sun G, Chu C-C. Synthesis, characterization of biodegradable dextran-allyl isocyanate-ethylamine/polyethylene glycol-diacrylate hydrogels and their in vitro release of albumin. Carbohyd. Polym. 2006;65:273-287.
64. Tanna S, Joan Taylor M, Sahota TS, Sawicka K. Glucose-responsive UV polymerised dextran-concanavalin A acrylic derivatised mixtures for closed-loop insulin delivery. Biomaterials 2006;27:1586-1597.
65. Tanna S, Sahota TS, Sawicka K, Taylor MJ. The effect of degree of acrylic derivatisation on dextran and concanavalin A glucose-responsive materials for closed-loop insulin delivery. Biomaterials 2006;27:4498-4507.
66. Maia J, Ferreira L, Carvalho R, Ramos MA, Gil MH. Synthesis ans characterization of new injectable and degradable dextran-based hydrogels. Polymer 2005;446:9604-9614.
67. Schacht EH, Bogdanov B, Van Den Bulcke A, De Rooze N. Hydrogels prepared by crosslinking of gelatin with dextran dialdehyde. React. Funct. Polym. 1997;33:109-116.
68. Draye J-P, Delaey B, Van de Voorde A, Van Den Bulcke A, Bogdanov B, Schacht E. In vitro release characteristics of bioactive molecules from dextran dialdehyde cross-linked gelatin hydrogel films. Biomaterials 1998;19:99-107.
69. Draye J-P, Delaey B, Van de Voorde A, Van Den Bulcke A, De Reu B, Schacht E. In vitro and in vivo biocompatibility of dextran dialdehyde cross-linked gelatin hydrogel films. Biomaterials 1998;19:1677-1687.
70. Ferreira L, Gil MH, Dordick JS. Enzymatic synthesis of dextran-containing hydrogels. Biomaterials 2002;23:3957-3967.
71. Ferreira L, Gil MH, Cabrita AMS, Dordick JS. Biocatalytic synthesis of highly ordered degradable dextran-based hydrogels. Biomaterials 2005;26:4707-4716.
72. Stenekes RJH, Talsma H, Hennink WE. Formation of dextran hydrogels by crystallization. Biomaterials 2001;22:1891-1898.
73. Huh KM, Ooya T, Lee WK, Sasaki S, Kwon IC, Jeong SY, Yui N. Supramolecular-structured hydrogels showing a
reversible phase transition by inclusion complexation between poly(ethylene glycol) grafted dextran and α-cyclodextrin. Macromolecules 2001;34:8657-8662.
74. Choi HS, Kontani K, Huh KM, Sasaki S, Ooya T, Lee WK, Yui N. Rapid induction of thermoreversible hydrogel formation based on poly(propylene glycol)-grafted dextran inclusion complexs. Macromol. Biosci. 2002;2:298-303.
75. Choi HS, Yamamoto K, Ooya T, Yui N. Synthesis of poly(ε-lysine)-grafted dextrans and their pH- and thermosensitive hydrogelation with cyclodextrins. ChemPhysChem 2005;6:1081-1086.
76. De Jong SJ, De Smedt SC, Wahls MWC, Demeester J, Kettenes-van den Bosch JJ, Hennink WE. Novel self-assembled hydrogels by stereocomplex formation in aqueous solution of enantiomeric lactic acid oligomers grafted to dextran. Macromolecules 2000;33:3680-3686.
77. De Jong SJ, van Nostrum CF, Kroon-Batenburg LMJ, Kettenes-van de Bosch JJ, Hennink WE. Oligolactate-grafted dextran hydrogels: detection of stereocomplex crosslinks by X-ray diffraction. J. Appl. Poly. Sci. 2002;86:289-293.
78. De Jong SJ, De Smedt SC, Demeester J, van Nostrum CF, Kettenes-van den Bosch JJ, Hennink WE. Biodegradable hydrogels based on stereocomplex formation between lactic acid oligomers grafted to dextran. J. Control. Release 2001;72:47-56.
79. Van Nostrum CF, Veldhuis TFJ, Bos GW, Hennink WE. Tuning the degradation rate of poly(2-hydroxypropyl methacrylamide)-graft-oligo(lactid acid) stereocomplex hydrogels. Macromolecules 2004;37:2113-2118.
80. De Jong SJ, van Eerdenbrugh B, van Nostrum CF, Kettenes-van den Bosch JJ, Hennink WE. Physically crosslinked dextran hydrogels by stereocomplex formation of lactic acid oligomers: degradation and protein release behavior. J. Control. Release 2001;71:261-275.
81. Bos GW, Hennink WE, Brouwer LA, den Otter W, Veldhuis TFJ, van Nostrum CF, van Luyn MJA. Tissue reactions of in situ formed dextran hydrogels crosslinked by stereocomplex formation after subcutaneous implantation in rats. Biomaterials 2005;26:3901-3909.
82. Sarmento B, Ribeiro A, Veiga F, Ferreira D. Development and characterization of new insulin containing polysaccharide nanoparticles. Colloid Surface B 2006;53:193-202.
3 Self-gelling hydrogels based on oppositely charged dextran
microspheres
Sophie R. Van Tommea, Mies J. van Steenbergena, Stefaan C. De Smedtb, Cornelus F. van Nostruma, Wim E. Henninka
a Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS),
University Utrecht, Utrecht, The Netherlands b Laboratory of General Biochemistry and Physical Pharmacy, Department of Pharmaceutics,
Ghent University, Ghent, Belgium
Biomaterials 26 (2005) 2129-2135
Chapter 3
40
Abstract
This paper presents a novel self-gelling hydrogel potentially suitable for controlled drug delivery and tissue engineering. The macroscopic gels are obtained by mixing dispersions of oppositely
charged crosslinked dextran microspheres. These microspheres in turn were prepared by crosslinking of dextran derivatized with hydroxyethyl methacrylate (HEMA) emulsified in an
aqueous poly(ethylene glycol) (PEG) solution. Negatively or positively charged microspheres were obtained by addition of methacrylic acid (MAA) or dimethylaminoethyl methacrylate (DMAEMA) to the polymerization mixture. Rheological analysis showed that instantaneous gelation occurred
when equal volumes of oppositely charged microspheres, dispersed in buffer solutions of pH 7, were mixed. The shear modulus of the networks could be tailored from 30 Pa to 6500 Pa by
varying the water content of the system. Moreover, controlled strain and creep experiments showed that the formed networks were mainly elastic. Importantly for application of these
systems, e.g. as controlled matrix of pharmaceutically active proteins, it was demonstrated that the hydrogel system has a reversible yield point, meaning that above a certain applied stress, the system starts to flow, whereas when the stress is removed, gel formation occurred. Further it was
shown that the network structure could be broken by either a low pH or a high ionic strength of the medium. This demonstrates that the networks, formed at pH 7 and at low ionic strength, are
held together by ionic interactions between the oppositely charged dextran microspheres. This system holds promise as injectable gels that are suitable for drug delivery and tissue engineering
applications.
Keywords: injectable hydrogels, dextran microspheres, ionic interactions, viscoelasticity, drug
delivery, tissue engineering
Self-gelling hydrogels based on oppositely charged dextran microspheres
41
1. Introduction
Hydrogels are an important class of materials that have been studied extensively in the last
decades for the controlled release of pharmaceutical proteins, and for tissue engineering applications.[1-5] Formation of hydrogels can be achieved by both chemical and physical
crosslinking.[6] By chemical crosslinking covalent bonds between the different polymer chains are introduced. Chemical crosslinking results in a network with a relatively high mechanical strength
and, depending on the nature of the chemical bonds in the building blocks and the crosslinks, in relatively long degradation times. However, chemical crosslinking can possibly damage the entrapped bioactive substance, leading to a loss of activity. Moreover, the crosslinking agents are
mostly toxic and removal needs to be ensured before in vivo application. In recent years there is a growing interest in physically crosslinked hydrogels. In such systems non-permanent bonds, based
on physical interactions between the polymer chains, are created. Different methods have been investigated to prepare physically crosslinked hydrogels. An attractive class of physically
crosslinked gels is those where gel formation is not instantaneous, but occurs a certain time after mixing the hydrogel components (e.g. stereocomplex gels[7-10]) or after a certain trigger (e.g. temperature[3, 11-15]). Such systems can be administered by injection as liquid formulation and gellify
in situ. Gel formation through chemical crosslinking can also occur using UV-light as a trigger.[16, 17] In our Department, both chemically and physically crosslinked dextran hydrogels have been
developed in recent years. An organic solvent free approach to obtain crosslinked microspheres has been described where preparation occurs in an all-aqueous environment.[18, 19] The in vivo
biocompatibility of dextran-based hydrogels and microspheres has been demonstrated as well as the relation between in vitro and in vivo degradation behavior.[20-22]
In this paper a novel injectable hydrogel system, as schematically outlined in Figure 1, is investigated. The macroscopic hydrogels are designed by combining the injectability of microspheres with physical crosslinking through ionic interactions. Anionically or cationically
charged microspheres were prepared and gels were obtained by mixing aqueous dispersions of the oppositely charged microspheres. Gel formation was studied by rheological experiments and
special attention was given to the reversibility of the system.
mixing
network formation
in situ gelling at site of injection
shear
--
Figure 1: The concept of the physically crosslinked hydrogel system. The hydrogel is obtained after mixing
aqueous dispersions of negatively charged dex-HEMA-MAA and positively charged dex-HEMA-DMAEMA
microspheres. When shear is applied, the interactions between the microspheres are broken and the sample
flows. Upon removal of the shear stress, the network rebuilds itself.
Chapter 3
42
2. Materials and methods
2.1. Materials
Dextran T40 (from Leuconostoc ssp.), N,N,N’,N’-tetramethylethylenediamine (TEMED) and 2-
hydroxyethyl methacrylate (HEMA) were obtained from Fluka (Buchs, Switzerland). Poly(ethylene glycol) (PEG) 10000 and potassium peroxodisulfate (KPS) were provided by Merck (Darmstadt,
Germany). N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes) was purchased from Acros Chimica (Geel, Belgium). Methacrylic acid (MAA) and dimethylaminoethyl methacrylate (DMAEMA) were provided by Sigma-Aldrich (Zwijndrecht, The Netherlands).
2.2. Hydroxyethyl methacrylate-derivatized dextran (dex-HEMA)
Dextran was derivatized with hydroxyethyl methacrylate (dex-HEMA) (Figure 2) as described previously.[23] The degree of substitution (DS, i.e. the number of HEMA groups per 100
glucopyranose units) used in this study was 6.
O
O
OHO
HO OOO
O
O
Figure 2: Chemical structure of dex-HEMA
2.3. Preparation of charged microspheres
The dextran microspheres with a water content of 70 % were obtained through radical
polymerization of dex-HEMA, emulsified in an aqueous PEG solution.[18, 24] In short: aqueous solutions of PEG (40 % (w/w)) and dex-HEMA (20 % (w/w)) were prepared in Hepes buffer (100 mM pH 7.0). PEG, dex-HEMA and buffer solution, 197.6 g, 18.3 g and 284.1 g respectively (total weight
500 g) were transferred into a 500 mL glass cylinder. Subsequently, either 12.5 mmol of MAA (Fig. 3A), or 12.5 mmol of DMAEMA (Fig. 3B) was added to the two-phase system (molar ratio (MAA
or DMAEMA)/HEMA=19). The two-phase system was flushed with nitrogen and intensively mixed (30 min, 11000 rpm, IKA Ultra-Turrax® T 25 basic, IKA-®WERKE GMBH & CO.KG, Staufen, Germany). In
this way, a water-in-water emulsion was created that was allowed to stabilize for 15 min. Next, a TEMED solution (10 mL, 20 % v/v, adjusted to pH 7 with 4 M HCl) and a KPS solution (18 mL, 50
mg/mL), both freshly prepared, were added to the mixture. The emulsified droplets were allowed to polymerize for 30 minutes at ambient temperature. Under these conditions the HEMA conversion is > 90 %.[25] Two types of microspheres were prepared, containing either MAA (dex-
HEMA-MAA) or DMAEMA (dex-HEMA-DMAEMA). The crosslinked particles were collected and purified by multiple washing and centrifugation steps (thrice with reversed osmosis water, 15 min,
3000 rpm). Ultimately the microspheres were lyophilized. The particle size distribution of the microspheres was determined using a Coulter Counter
Multisizer® 3 (Beckman Coulter Nederland B.V., The Netherlands) with a 100-μm orifice.
Self-gelling hydrogels based on oppositely charged dextran microspheres
43
O
ON
O
ON
H
O
OH
O
O
A
B
Figure 3: Chemical structures of methacrylic acid (A) and dimethylaminoethyl methacrylate (B).
2.4. Formation of macroscopic gels with charged microspheres
Lyophilized microspheres (dex-HEMA-MAA or dex-HEMA-DMAEMA) were dispersed in Hepes
buffer (100 mM pH 7; solid content between 10 % and 25 %). The dispersions were stored at 4 °C
for 2 hours to allow full hydration of the microspheres. The equilibrium water content of the rehydrated microspheres was determined using the blue dextran exclusion assay.[26] To study
possible gel formation, equal volumes (200 μL) of the two different microsphere dispersions were
mixed.
2.5. Rheological experiments
The rheological measurements were performed using a controlled stress rheometer (AR1000-N, TA
Instruments, Etten-Leur, The Netherlands), equipped with an acrylic flat plate geometry (20 mm
diameter) and a gap of 500 μm. Immediately after mixing equal volumes of both dispersions (see
section 2.4), the sample was placed between the plates. A solvent trap was used to prevent
evaporation of the solvent. The viscoelastic properties of the sample were determined by
measuring the G’ (shear storage modulus) and G” (loss modulus) at 20 °C with a constant strain of
1 % and constant frequency of 1 Hz. Also frequency sweep and strain sweep experiments were
performed. Creep experiments were performed to evaluate the extent of recovery of the material after deformation. In the creep experiment a shear stress of 1 Pa was applied while the strain was
monitored. After 1 min the stress was removed and the recovery of the sample was monitored by measuring the strain during 2 min. As a control, the same rheological experiments were performed
on dispersions containing only dex-HEMA-MAA or dex-HEMA-DMAEMA microspheres.
To determine the yield point of the system, stress sweep experiments were performed at 20 °C.
During these experiments the G’ and G” were monitored while the stress was increased. The
frequency was kept constant to 1 Hz. The experiment was performed 4 times in a row using the same sample. After each experiment the sample was allowed to recover for 1 h.
Most experiments were performed on hydrogels containing 15 % (w/w) of freeze-dried microspheres. Controlled strain and creep experiments were also performed on hydrogels with different percentages (10 % - 25 % w/w) solid content. The influence of pH and ionic strength on
the systems was studied by using different buffers (phosphate buffer (100 mM, pH 3) or Hepes buffer (100 mM, pH 7) with variable ionic strengths (NaCl, from 17- 1000 mM)).
Chapter 3
44
3. Results and discussion
3.1. Preparation of charged dex-HEMA microspheres
Charged dextan particles were obtained by radical copolymerization of dex-HEMA with either
MAA or DMAEMA. At physiological pH both MAA and DMAEMA are mainly ionized (pKa MAA= 4.7;[27] pKa DMAEMA= 8.4[28]), resulting in charged microspheres at this pH. The mean
volume diameters of the dex-HEMA-MAA and the dex-HEMA-DMAEMA microspheres were comparable (8.3 μm and 7.5 μm respectively; 90 % < 12.5 μm (Fig. 3)). The equilibrium water content of the rehydrated microspheres was 70 %, as determined with the blue dextran exclusion
assay.[26]
0 5 10 15 20 25 30 35 400.00
0.25
0.50
0.75
1.00
1.25
particle size (µm)
volu
me
(%)
Figure 4: Volume diameter distribution of dex-HEMA-MAA (—) and dex-HEMA-DMAEMA ( ) microspheres.
3.2. Gel formation through ionic interactions between microspheres
Addition of buffer to freeze-dried microspheres created a homogenous opalescent dispersion. It should be noted that the lyophilized microspheres absorbed water resulting in a dispersion of hydrated microspheres in a continuous aqueous phase. The percentage of free water depends on
the amount of dried particles dispersed in the aqueous phase and amounts 50 % when the solid content of the dispersion was 15 %. When the solid content was 10 % or less, the dispersions were
freely flowing. Increasing the solid content, and so decreasing the amount of free water, yielded more viscous dispersions. With a solid content above 25 % (free water is < 16 %) dispersions with a
very high viscosity were obtained. When equal volumes of dex-HEMA-MAA and dex-HEMA-DMAEMA microsphere dispersions were
mixed, gelation clearly occurred instantly. However, the obtained gel could be easily handled by a positive displacement pipette. This implicates that the network can be easily broken and rebuilt when exposed to stress and deformation, as expected for physical crosslinking. This aspect is
studied in more detail is sections 3.3 and 3.5.
Self-gelling hydrogels based on oppositely charged dextran microspheres
45
3.3. Rheological characterization of the system
The viscoelastic properties of dex-HEMA-MAA/dex-HEMA-DMAEMA microsphere dispersions (15 % solid content) were investigated with controlled strain experiments.
Figure 5 shows that when a strain of 1 % is applied the sample is still in the linear viscoelastic deformation range. As Figure 6A shows, after mixing the anionic and cationic dex-HEMA
microspheres, the storage modulus (G’) increased gradually in time while the loss modulus (G”)
remained low. The G”/G’ ratio or tan(δ) was lower than 0.1, which indicates that the obtained
material is mainly elastic. Figure 6B shows the rheological characteristics of a dex-HEMA-MAA
microsphere dispersion with the same solid content. Compared with the mixture of oppositely
charged microspheres Figure 6B shows that the tan(δ) was substantially higher (> 4). It indicates
that, as expected, an elastic network does not exist in a dispersion of negatively charged dex-
HEMA spheres. Positively charged dex-HEMA microspheres showed comparable results (data not shown).
0.01 0.1 1 10 1000
100
200
300
400
500
0
1
2
3
4
5
% strain
G' (
Pa)
tan(
δ)
Figure 5: Storage modulus (G’) (—) and tan(δ) ( ) of a dex-HEMA-MAA/dex-HEMA-DMAEMA microsphere
dispersion (solid content 15 % (w/w)) at 20 °C as a function of the % strain.
0 5 10 15 200
100
200
300
400
500
600
0.00
0.02
0.04
0.06
0.08
0.10
time (min)
G' a
nd G
" (P
a)
tan(
δ )
0 5 10 15 200
1
2
3
4
5
6
0
1
2
3
4
5
6
time (min)
G' a
nd G
" (P
a)
tan(
δ )
Figure 6: Storage modulus (G’) (—), loss modulus (G”) (⎯) and tan(δ) ( ) at 20 °C as a function of the time.
(A) dex-HEMA-MAA/dex-HEMA-DMAEMA microsphere dispersion (solid content 15 % (w/w)). G’, G” and tan(δ)
were followed in time after mixing the anionic and cationic dex-HEMA microspheres. (B) dex-HEMA-MAA microsphere dispersion (solid content 15 % (w/w)).
Figure 7A shows the results of a creep experiment on a mixture of dex-HEMA-MAA and dex-HEMA-
DMAEMA microspheres. Upon applying the shear stress (1 Pa), the system deformed, evolving to 0.15 % strain. When the stress was removed, the sample recovered almost completely, confirming
A B
Chapter 3
46
the almost fully elastic properties of the material, and indicating the presence of a network. In
contrast, the dex-HEMA-MAA microsphere dispersion showed mainly viscous behavior (Fig. 7B). The deformation in the retardation phase was more than a 10000 fold stronger than the one in Figure 7A. Also, after removal of the stress the sample did not recover, indicating that the
dispersion is not elastic. These results are in full agreement with the high value of tan(δ) for this
dispersion (Fig. 6B).
0 50 100 150 2000.00
0.05
0.10
0.15
time (s)
% s
train
0 50 100 150 2000
250500750
100012501500175020002250
time (s)%
stra
in
Figure 7: Creep experiment (applied stress 1 Pa, 20 °C). (A) dex-HEMA-MAA/dex-HEMA-DMAEMA microsphere
dispersion. (B) dex-HEMA-MAA microsphere dispersion. The solid content of the dispersions was 15 % (w/w).
3.4. Influence of solid content, pH and ionic strength on gel properties
Figure 8 shows the G’ and tan(δ) of dex-HEMA-MAA/dex-HEMA-DMAEMA microsphere dispersions
as a function of the solid content of the dispersions. When the solid content of the mixture was
10 %, the microspheres did not establish a network structure as evidenced from the high tan(δ)
(~ 1.6). Obviously, due to the high water content, anionic and cationic dex-HEMA microspheres are too far separated from each other to form a network. From 12.5 % solid content (58 % free water)
on, the microspheres do interact and create a network, clearly illustrated by the increase in G’ and
the low tan(δ). For dispersions with a solid content of 25 %, G’ equaled 6500 Pa while tan(δ) was
0.09.
5 10 15 20 25 300
1000200030004000500060007000
0.0
0.5
1.0
1.5
2.0
% solid
G' (
Pa)
tan(
δ)
Figure 8: Storage modulus (G’) (—) and tan(δ) ( ) as a function of the solid content of dex-HEMA-MAA/dex-
HEMA-DMAEMA microsphere dispersions. The data are shown as sample mean ± the standard deviation (n=3).
A B
Self-gelling hydrogels based on oppositely charged dextran microspheres
47
For a 25 % dispersion of dex-HEMA-MAA microspheres, G’ equaled 700 Pa while tan(δ) was 0.3. The
high G’ of the dex-HEMA-MAA dispersion can be explained by the low free water content (16.5 %),
which forces the microspheres to be closely packed despite their negative charge. The higher
tan(δ) indicates that there is less elasticity in these dispersions when compared to the dex-HEMA-
MAA/dex-HEMA-DMAEMA system. Table 1 shows the rheological properties of a dex-HEMA-
MAA/dex-HEMA-DMAEMA dispersion (solid content 15 %), prepared at respectively pH 3 and pH 7. Interestingly and in contrast to pH 7-dispersions, at pH 3 the system shows mainly viscous
behavior (tan(δ) > 2), comparable to the results of the dex-HEMA-MAA dispersion (Fig. 6B). These
results can be explained by the fact that the dex-HEMA-MAA microspheres essentially loose their negative charge at pH 3.
Table 1: Storage modulus (G’), loss modulus (G”) and tan(δ) of a dex-HEMA-MAA/dex-HEMA-DMAEMA
microsphere dispersion as a function of the pH. The hydrogel solid content was 15 % (w/w). All data are shown
as sample mean ± the standard deviation (n=3).
pH of the buffer G’ (Pa) G” (Pa) tan(δ)
3 11 ± 1 29 ± 5 2.7 ± 0.2
7 509 ± 18 29 ± 2 0.06 ± 0.00
Figure 9 shows the influence of the ionic strength of the buffer on the rheological properties of
dex-HEMA-MAA/dex-HEMA-DMAEMA microsphere dispersions. Increasing the concentration of
NaCl significantly decreases G’ and increases tan(δ). It indicates that a higher salt concentration
inhibits network formation, which is explained by the shielding of the microsphere charges at high ionic strength. Taken together the results presented in Table 1 and Figures 8 and 9 demonstrate that the
observed network formation is indeed due to electrostatic interactions between oppositely charged microspheres, creating a physically crosslinked hydrogel network.
0 100 200 300 400 500 6000
100
200
300
400
500
600
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
ionic strength (mM)
G' (
Pa)
tan(
δ)
Figure 9: Storage modulus G’(—) and tan(δ) (---) of a dex-HEMA-MAA/dex-HEMA-DMAEMA microsphere
dispersion as a function of the ionic strength of the buffer. The hydrogel solid content was 15 % (w/w). The
data are shown as sample mean ± the standard deviation (n=3).
Chapter 3
48
3.5. Determination of the yield point
In view of the possible application as an injectable dispersion which gellifies in situ, it is important to know whether the material flows when shear forces are applied. The dispersions should flow
during injection, however the network structure should be established at the place of injection (Fig. 1). The shear at which flow starts is referred to as the yield point.[29] Because the obtained yield
stress value is dependent on the applied technique, it is preferable to use the term ‘apparent yield stress’. [30, 31]
To determine the apparent yield point, stress sweep experiments were performed on a hydrogel
formed at pH 7 at 20 °C by mixing dex-HEMA-MAA and dex-HEMA-DMAEMA dispersions (Fig. 9).
With increasing stress (from 0.1 Pa to 50 Pa), the G’ gradually decreased and the tan(δ) increased
simultaneously. However, when the applied stress exceeds 10 Pa the G’ dramatically dropped from
300 Pa to 3 Pa whilst the tan(δ) increased from 0.08 to 5. Next, the stress was removed and the
system was allowed to recover for 1 h. When an increasing stress was put on the gel, a similar
rheogramme as shown in Figure 10 was observed. Four consecutive stress sweep experiments
were performed, each giving comparable values for G’, G” and tan(δ). The results of Figure 10
shows that the system of oppositely charged dextran microspheres is plastic in a rheological sense,
meaning the ionic interactions between the microspheres can be broken by mechanical stress and that the network rebuilds itself when the stress is removed. As expected, the yield stress was
dependent on the gel composition and amounted 150 Pa for 25 % systems.
0 10 20 30 40 500
50100150200250300350400450
0.00.51.01.52.02.53.03.54.04.55.05.56.06.5
oscillatory stress (Pa)
G' a
nd G
" (Pa
)
tan(
δ)
Figure 10: Storage modulus G’ (—), loss modulus G” (—) and tan(δ) ( ) of a dex-HEMA-MAA/dex-HEMA-
DMAEMA microsphere dispersion (solid content 15 % (w/w)) at 20 °C as a function of the oscillatory stress.
Self-gelling hydrogels based on oppositely charged dextran microspheres
49
4. Conclusion
This paper reports on a novel method to design macroscopic hydrogels, combining injectability of
hydrogel microspheres with physical crosslinking through ionic interactions. The ionic interactions between the cationic and anionic microspheres and so creating a physical network, can be broken
when exposed to stress. The gel forms again when the stress is removed, indicating the reversible character of the system. A number of possible applications for this novel system can be foreseen
among which controlled delivery of pharmaceutically active proteins and entrapment of living cells for tissue engineering. At present, we are studying the release of proteins from these systems.
Acknowledgement
The authors like to thank J.F.W. Nijsen and S.W. Zielhuis from the Department of Nuclear Medicine, University Medical Center, Utrecht, The Netherlands for assisting in the particle size measurements.
Chapter 3
50
References
1. Gombotz WR, Pettit DK. Biodegradable polymers for protein and peptide drug delivery. Bioconjugate Chem. 1985;6:332-351.
2. Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chem. Rev. 2001;101:1869-1879.
3. Hatefi A, Amsden B. Biodegradable injectable in situ forming drug delivery systems. J. Control. Release 2002;80:9-28.
4. Hoffman AS. Hydrogels for biomedical applications. Adv. Drug Deliver. Rev. 2002;43:3-12.
5. Drury JL, Mooney DJ. Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials 2003;24:4337-4351.
6. Hennink WE, van Nostrum CF. Novel crosslinking methods to design hydrogels. Adv. Drug Deliver. Rev. 2002;54:13-36.
7. De Jong SJ, De Smedt SC, Wahls MWC, Demeester J, Kettenes-van den Bosch JJ, Hennink WE. Novel self-assembled hydrogels by stereocomplex formation in aqueous solution of enantiomeric lactic acid oligomers grafted to dextran. Macromolecules 2000;33:3680-3686.
8. Bos GW, Jacobs JJL, Koten JW, Van Tomme SR, Veldhuis TFJ, van Nostrum CF, Den Otter W, Hennink WE. In situ crosslinked biodegradable hydrogels loaded with IL-2 are effective tools for local IL-2 therapy. Eur. J. Pharm. Sci. 2004;21:561-567.
9. Li SM, Vert M. Synthesis, characterization and stereocomplex-induced gelation of block copolymers prepared by ring opening polymerization of L(D)-lactide in the presence of poly(ethylene glycol). Macromolecules 2003;36:8008-8014.
10. Fujiwara T, Mukose T, Yamaoka H, Yamane H, Sakurai S, Kimura Y. Novel thermo-responsive formation of a hydrogel by stereocomplexation between PLLA-PEG-PLLA and PDLA-PEG-PDLA block copolymers. Macromol. Biosci. 2001;1:204-208.
11. Zentner GM, Rathi R, Shih C, McRea JC, Seo M, Oh H, Rhee BG, Mestecky J, Moldoveanu Z, Morgan M, Weitman S. Biodegradable block copolymers for delivery of proteins and water-insoluble drugs. J. Control. Release 2001;72:203-215.
12. Jeong B, Kim SW, Bae YH. Thermosensitive sol-gel reversible hydrogels. Adv. Drug Deliver. Rev. 2002;54:37-51.
13. Jeong B, Bae YH, Lee DS, Kim SW. Biodegradable block copolymers as injectable drug-delivery systems. Nature 1997;388:860-862.
14. Kim S, Healy KE. Synthesis and characterization of injectable Poly(N-isopropylacrylamide-co-acrylic acid) hydrogels with proteolytically degradable cross links. Biomacromolecules 2003;4:1214-1223.
15. Cellesi F, Tirelli N, Hubbell JA. Towards a fully-synthetic substitute of alginate: development of a new process using thermal gelation and chemical cross-linking. Biomaterials 2004;25:5115-5124.
16. Anseth KS, Metters AT, Bryant SJ, Martens PJ, Elisseef JH, Bowman CN. In situ forming degradable networks and their application in tissue engineering and drug delivery. J. Control. Release 2002;78:199-209.
17. Park YD, Tirelli N, Hubbell JA. Photopolymerized hyaluronic acid-based hydrogels and interpenetrating networks. Biomaterials 2003;24:893-900.
18. Franssen O, Hennink WE. A novel preparation method for polymeric microparticles without the use of organic solvents. Int. J. Pharm. 1998;168:1-7.
19. Franssen O, Vandervennet L, Roders P, Hennink WE. Degradable dextran hydrogels: controlled release of a model protein from cylinders and microspheres. J. Control. Release 1999;60:211-221.
20. De Groot CJ, Van Luyn MJA, Van Dijk-Wolthuis WNE, Cadee JA, Plantinga JA, Den Otter W, Hennink WE. In vitro biocompatibility of biodegradable dextran-based hydrogels tested with human fibroblasts. Biomaterials 2001;22:1197-1203.
21. Cadée JA, van Luyn MJA, Brouwer LA, Plantinga JA, van Wachem PB, de Groot CJ, den Otter W, Hennink WE. In vivo biocompatibility of dextran-based hydrogels. J. Biomed. Mater. Res. 2000;50:397-404.
22. Cadée JA, Brouwer LA, den Otter W, Hennink WE, van Luyn MJA. A comparative biocompatibility study of microspheres based on crosslinked dextran or poly(lactic-co-glycolic) acid after subcutaneous injection in rats. J. Biomed. Mater. Res. 2001;56:600-609.
23. Van Dijk-Wolthuis WNE, Tsang SKY, Kettenes-van den Bosch JJ, Hennink WE. A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer 1997;38:6235-6242.
24. Stenekes RJH, Franssen O, van Bommel EMG, Crommelin DJA, Hennink WE. The preparation of dextran microspheres in an all-aqueous system: effects of the formulation parameters on particle characteristics. Pharm. Res. 1998;15:557-561.
25. Stenekes RJH, Hennink WE. Polymerization kinetics of dextran-bound methacrylate in an aqueous two phase system. Polymer 2000;41:5563-5569.
Self-gelling hydrogels based on oppositely charged dextran microspheres
51
26. Stenekes RJH, Hennink WE. Equilibrium water content of microspheres based on cross-linked dextran. Int. J. Pharm. 1999;189:131-135.
27. Grant DH, McPhee VA. Determination of methacrylic acid by coulometric titration Anal. Chem. 1976;48:1820-1820.
28. Van de Wetering P, Zuidam NJ, van Steenbergen MJ, van der Houwen OAGJ, Underberg WJM, Hennink WE. A mechanistic study of the hydrolytic stability of poly(2-(dimethylamino)ethyl methacrylate). Macromolecules 1998;31:8063-8068.
29. Martin A. Physical Pharmacy. 4th ed.; Lea & Febiger: Philadelphia, 1993.
30. Semancik JR. Yield stress measurements using controlled stress rheometry. TA Instruments. Publication RH-058.
31. Barnes HA, Hutton JF, Walters K. An introduction to rheology. Elsevier: 1991.
4 Mobility of model proteins in hydrogels composed of
oppositely charged dextran microspheres studied by protein release and fluorescence recovery after photobleaching
Sophie R. Van Tommea, Bruno G. De Geestb, Kevin Braeckmansb, Stefaan C. De Smedtb, Florence Siepmannc, Juergen Siepmannc, Cornelus F. van Nostruma, Wim E. Henninka
a Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS),
University Utrecht, Utrecht, The Netherlands b Laboratory of General Biochemistry and Physical Pharmacy, Department of Pharmaceutics,
Ghent University, Ghent, Belgium c University of Lille, School of Pharmacy, Lille, France
Journal of Controlled Release 110 (2005) 67-78
Chapter 4
54
Abstract
In this paper, the release of proteins from a novel self-gelling hydrogel based on biodegradable dextran microspheres is investigated. The protein-loaded macroscopic gels are obtained by
hydration of mixtures of oppositely charged hydroxyethyl methacrylate-derivatized dextran microspheres with a protein solution. In media of low ionic strength (100 mM Hepes pH 7.0) it was found that the release of the entrapped model proteins (lysozyme, BSA and IgG) was slower than
in saline (150 mM NaCl, 100 mM Hepes pH 7.0). The reason behind this observation is that substantial adsorption of the proteins onto the microspheres’ surface and/or absorption in the
microspheres takes place. Confocal images showed that independent of their crosslink density the microspheres are impermeable for BSA and IgG. BSA, bearing a negative charge at neutral pH, was
adsorbed onto the surface of positively charged microspheres. Lysozyme, which is positively charged at neutral pH, was able to penetrate into the negatively charged microspheres. In saline, the gels showed continuous release of the different proteins for 25 to 60 days. Importantly,
lysozyme was quantitatively and with full preservation of its enzymatic activity released in about 25 days. This emphasizes the protein friendly technology to prepare the protein-loaded gels.
Mathematical modeling revealed that protein release followed Fick’s second law, indicating that the systems are primarily diffusion controlled. These results show that these hydrogels are very
suitable as injectable matrix for diffusion-controlled delivery of pharmaceutically active proteins.
Keywords: dextran, injectable hydrogels, self-assembling, mathematical modeling, diffusion-
controlled protein delivery
Protein release from hydrogels composed of oppositely charged dextran microspheres
55
1. Introduction
Hydrogels are an attractive class of materials for the controlled release of pharmaceutical proteins
and for tissue engineering applications.[1-7] The release of the entrapped proteins can be tailored by varying the crosslink density, the water content and the polymer composition of the hydrogels.
Crosslinking can be established either with chemical or physical methods.[8] The latter is most favorable since the use of organic solvents or crosslinking agents, often toxic and potentially
destructive for the protein, is avoided. In physically crosslinked systems non-permanent bonds are created by physical interactions between the polymer chains.[9-15] In recent years there is a growing interest in physically crosslinked systems where gel formation is through self-assembly.[16-19]
Network formation can be obtained after mixing the hydrogel components (e.g. stereocomplex gels[20-23]) or after a certain trigger (e.g. temperature,[24-28] pH[29, 30] or biological stimuli[31]). These
systems can be administered by injection as liquid formulation and gelation occurs in situ. Recently we reported on a novel self-gelling hydrogel based on oppositely charged dextran
microspheres.[32] This system combines the injectability of polymeric microspheres with physical crosslinking through ionic interactions. A macroscopic hydrogel is formed by simply mixing aqueous dispersions of anionically and cationically charged microspheres. Importantly, it was
demonstrated that the ionic interactions creating the physical network could be broken when exposed to shear. Further, the gel is reformed when the shear is removed, indicating the reversible
character of the system. This study is focused on the release of proteins in these novel hydrogels, composed of dex-HEMA-
MAA and dex-HEMA-DMAEMA microspheres. Dex-HEMA microspheres are degradable at physiological pH and temperature. The network degradation is caused by OH- driven hydrolysis of
the carbonate-ester, linking the dextran backbone and the HEMA side chains. Both in vitro and in
vivo degradability of dex-HEMA microspheres has been extensively studied and described previously by our group.[33-36] The release of three model proteins varying in isoelectric point (pI)
and size was studied as function of the solid content of the gel and the crosslink density of the dextran microspheres that form the gel. Possible matrix-protein interactions were studied by
adsorption experiments and confocal laser scanning microscopy. Finally, fluorescence recovery after photobleaching was used to gain insight into the mobility of proteins in the complex
network of charged polymeric microspheres.
2. Materials and Methods
2.1. Materials
Dextran T40 (from Leuconostoc ssp.), N,N,N’,N’-tetramethylethylenediamine (TEMED), 2-
hydroxyethyl methacrylate (HEMA) and lysozyme (from hen egg white) were provided by Fluka (Buchs, Switzerland). Poly(ethylene glycol) (PEG) 10000 and potassium peroxodisulfate (KPS) were
purchased from Merck (Darmstadt, Germany). N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes) was obtained from Acros Chimica (Geel, Belgium). Methacrylic acid (MAA),
Chapter 4
56
dimethylaminoethyl methacrylate (DMAEMA), bovine serum albumin (BSA, fraction V, minimum
96 %), fluorescein isothiocyanate bovine serum albumin (FITC-BSA), fluorescein isothiocyanate (FITC, isomer I, 90) and Micrococcus lysodeikticus were provided by Sigma-Aldrich (Zwijndrecht, The Netherlands). Bovine immunoglobulin G (IgG, fraction II) was obtained from ICN Biomedicals BV
(Zoetermeer, The Netherlands). The bicinchoninic acid (BCA) protein assay kit was purchased from Interchim (Montluçon, France). Hydroxyethyl methacrylate-derivatized dextran (dex-HEMA) was
synthesized and characterized according to Van Dijk-Wolthuis et al.[35] Dextran with an Mn of 16000 Da was selected, ensuring renal excretion during in vivo applications.[37] The DS´s (i.e. the number of
HEMA groups per 100 glucopyranose units) used in this study were 5, 8 and 18.
2.2. Preparation of charged dex-HEMA microspheres
Negatively and positively charged microspheres were prepared as described previously.[32] In short, the dextran microspheres, with a water content of 70 %, were obtained through radical
polymerization of dex-HEMA, emulsified in an aqueous poly(ethylene glycol) solution. Addition of either methacrylic acid (MAA) or dimethylaminoethyl methacrylate (DMAEMA) to the
polymerization mixture resulted in respectively negatively (dex-HEMA-MAA) and positively (dex-HEMA-DMAEMA) charged microspheres. The crosslinked particles were purified by multiple washing and centrifugation steps and ultimately the microspheres were lyophilized. The particle
size and size distribution of the microspheres were determined with a laser blocking technique, using an Accusizer® (model 770, Particle Sizing Systems, Santa Barbara, CA, USA).
The mean volume diameters of the dex-HEMA-MAA and the dex-HEMA-DMAEMA microspheres
were 10 μm. The microsphere charge was confirmed by ξ-potential measurements using a
Malvern Zetasizer 2000 (Malvern Instruments, Worcestershire, UK).[38] ξ-potentials varied from -13 to
-16 mV for dex-HEMA-MAA microspheres and from +12 to +15 mV for dex-HEMA-DMAEMA microspheres.
2.3. Formation of physically crosslinked hydrogels using charged microspheres
Hydrogels consisting of equal amounts of positively (dex-HEMA-DMAEMA) and negatively (dex-
HEMA-MAA) charged microspheres were prepared by mixing the lyophilized microspheres. Next,
the microsphere mixture was hydrated in Hepes buffer for 1h (100 mM pH 7.0) at 4 °C. The solid
content of the gels was varied between 15 % and 30 %. Hydrogels containing 1 mg protein per 100 mg gel were prepared by hydration of the microsphere mixture in a protein solution (25 mg/mL) in 100 mM Hepes (100 mM pH 7.0).
2.4. Adsorption and absorption of proteins to the microspheres
Possible absorption and adsorption of proteins to the negatively or positively charged microspheres was studied by incubating 10 mg dry microspheres with 1 mL of protein solution
500 μg/ml in Hepes buffer (100 mM, pH 7.0). After 1 h of incubation at room temperature the
dispersions were centrifuged (1 min 10,000 rpm) and the protein concentration in the supernatant was determined with the BCA® Protein Assay (described in section 2.7). Subsequently, the
Protein release from hydrogels composed of oppositely charged dextran microspheres
57
microsphere pellet was redispersed in 1 ml 150 mM NaCl, Hepes buffer pH 7.0 and incubated for 1
day at room temperature. Thereafter, the dispersions were treated and analyzed as described above. Confocal images of FITC-labeled lysozyme and BSA in the gels were taken using a confocal
scanning laser microscope (model MRC1024 UV, Bio-Rad, Hemel Hempstead, UK).
2.5. Rheological experiments
Rheological experiments were performed on hydrogels (25 % solid) with and without protein
loading. The rheological measurements on the hydrogels were performed using a controlled stress rheometer (AR1000-N, TA Instruments, Etten-Leur, The Netherlands), equipped with an acrylic flat plate geometry (20 mm diameter) and a gap of 500 μm.[32] Hydrogels were prepared as described
in section 2.3 and thereupon introduced between the two plates. A solvent trap was used to prevent evaporation of the solvent. The viscoelastic properties of the gels were determined by
measuring the G’ (shear storage modulus) and G” (loss modulus) at 20 °C with a constant strain of
1 % and constant frequency of 1 Hz. Creep experiments were done to evaluate the extent of recovery of the material after deformation. Therefore a shear stress of 50 Pa was applied while the
strain was monitored. After 5 min the stress was removed and the recovery of the sample was monitored by measuring the strain during 10 min.
2.6. In vitro protein release
Hydrogels were prepared and loaded with the various proteins as described in section 2.3. Lysozyme (Mw 14000 g/mol), BSA (Mw 67000 g/mol) and IgG (Mw 150000 g/mol) were used as model proteins. Their diffusion coefficients in water (D0) are respectively: 1.04 x 10-6 cm²/s,[39] 0.59 x
10-6 cm²/s[39] and 0.40 x 10-6 cm²/s.[40] The hydrodynamic diameter of the proteins (d) was calculated using the Einstein-Stokes Equation (1):
πηd
kTD
30 = (1)
where D0 is the diffusion coefficient of the protein, k is the Boltzmann constant, T is the absolute
temperature and η is the viscosity of the solvent. The calculated hydrodynamic diameters for
lysozyme, BSA and IgG are respectively 4.1, 7.2 and 10.7 nm. Different protein-loaded hydrogels were prepared, altering the solid content of the gels (15 % -
30 %) and the DS of the microspheres (DS 5, 8, 18). Each formulation was made in duplo. Non-protein loaded hydrogels were also included in the study. For every formulation, 500 mg gel was
prepared in 2 mL eppendorf cups that were weighed in advance. After hydration the gels were transferred into the release device (described below), closed with a rubber stop and the empty
eppendorf cups were weighed again to determine the exact weight of the gel introduced in the device. The release device is made of polyoxymethylene and consists of a gel and a release
compartment, with a diameter of respectively 8.5 mm and 15 mm and length of 8.8 mm and 30 mm (Fig. 1). As a result, cylindrical gels of 8.5 x 8.8 mm (diameter x length) were obtained.
Chapter 4
58
0.5 mL gel
3 mL buffer
0.5 mL gel
3 mL buffer
Figure 1: Schematic presentation of the release device.
Release buffer (3 mL, 100 mM Hepes pH 7.0, 0.02 % NaN3, with and without 150 mM NaCl) was
added to each formulation and the device was incubated on a shaking plate at 37 °C. Samples of
0.5 mL were taken at regular time intervals and replaced by an equal volume of fresh buffer. The
release samples were analyzed for their protein concentration using the BCA® Protein assay described in the next section.
2.7. Determination of protein concentration and enzymatic activity of lysozyme in release samples
The protein concentration in the release samples was determined with the BCA® Protein Assay.[41]
Standard protein solutions (concentration range 0.010 - 1 mg/mL) were prepared to generate calibration curves. Release samples (25 μL) were pipetted into a 96-microwells plate and 200 μL of
working reagent (= BCA reagent A: BCA reagent B, 50:1 v/v) was added. The plates were incubated
for 30 minutes at 37 °C followed by cooling down to room temperature. Subsequently the
absorbance was measured at 550 nm with a Microplate Manager® (Bio-rad Laboratories, Hercules,
CA, USA). The enzymatic activity of lysozyme in some selected release samples was determined. The assay is
based on the hydrolysis of the outer cell membrane of Micrococcus lysodeikticus, resulting in solubilization of the affected bacteria and consequent decrease of light scattering.[42] The release samples were diluted to a concentration of 50 - 100 μg/mL. Next, 10 μL of sample was added to
1.3 mL of M. lysodeikticus suspension (0.2 mg/mL, 100 mM Hepes buffer pH 7.0) and the decrease in turbidity was measured for 3 minutes at 450 nm. The % remaining enzyme activity was obtained
by comparing the activity to that of a reference lysozyme solution (100 μg/mL). UV scans (Perkin-Elmer Lambda 2 UV/VIS spectrophotometer, Überlingen, Germany, 250 nm-
350 nm) of release samples containing lysozyme were taken and compared to a scan of a reference lysozyme solution (100 μg/mL) to verify the possible presence of protein aggregates.
Lysozyme denaturation was checked by fluorescence spectroscopy of release samples (Fluorolog fluorimeter, Jobin Yvon Horriba, 300 nm–450 nm) and also compared to a freshly made lysozyme solution (23 μg/mL).
2.8. Mathematical modeling of in vitro protein release
An analytical solution of Fick’s second law of diffusion was used to quantitatively describe protein release from the investigated hydrogels. The hydrogels are regarded as monolithic structures since
the proteins were molecularly dispersed in the macroscopic gels, although they might show some
Protein release from hydrogels composed of oppositely charged dextran microspheres
59
micro-heterogeneities due to variable packing of the microspheres. Furthermore, the model
considers that the edges of the cylindrical gels were not exposed to the release medium. Thus, the mathematical analysis could be restricted to one dimension and the release kinetics could be described as follows:[43]
2
2
x
cD
t
c
∂∂
∂∂ ⋅= (2)
where c denotes the concentration of the protein within the polymeric system, being a function of time t and position x; D represents the apparent diffusion coefficient of the protein. Considering perfect sink conditions throughout the experiment and the fact that only one circular
surface of the cylindrical gels was exposed to the release medium, the following solution of Fick’s second law of diffusion can be derived and used to describe protein release from the investigated
hydrogels:[43]
⎟⎟⎠
⎞⎜⎜⎝
⎛⋅⋅
⋅⋅+⋅−⋅
⋅+⋅−= ∑
∞
=∞
tDL
n
nM
M
n
t2
22
022 4
)12(exp
)12(
81
ππ
(3)
where Mt and M∞, represent the absolute cumulative amounts of protein released at time t, and infinite time, respectively; L denotes the height of the cylindrical hydrogel. If protein release leveled off below 100 %, the experimentally determined plateau value (amount
of mobile protein) was considered as 100 % reference value for protein diffusion.
2.9. Fluorescence recovery after photobleaching (FRAP)
FRAP was used to study the mobility of the proteins in the gels composed of oppositely charged
particles, as well as in the microspheres themselves.[44] The FRAP measurements were performed using a setup as described previously.[45, 46] In detail, a
confocal scanning laser microscope (model MRC1024 UV, Bio-Rad, Hemel Hempstead, UK) modified for bleaching arbitrary regions, was used. The 488-nm line of a 4 W Ar-ion laser (model Stabilite 2017; Spectra-Physics, Darmstadt, Germany) was used to bleach uniforms disks with a
typical diameter of 25 μm. It is assumed that the bleaching phase is very short (100 to 200 ms) so that the amount of fluorescence recovery that will take place during bleaching is negligible
compared to the characteristic recovery time. The microscope was equipped with a 10x objective lens (CFI Plan Apochromat; Nikon, Badhoevedorp, The Netherlands). Next, a highly attenuated laser
beam measured the fluorescence recovery in the bleached area, which is due to the diffusion of fluorescent probes from the surrounding unbleached area into the bleached spot. The diffusion coefficient can be calculated from the experimental recovery curve by fitting of the appropriate
FRAP model. The derivation of the FRAP model for a uniform disk bleached by a low numerical aperture lens has been described earlier.[47]
The gels were loaded with fluorescein isothiocyanate (FITC) labeled proteins. FITC-BSA was used as provided by the supplier whereas lysozyme was labeled as follows: 300 mg lysozyme and 12 mg
FITC were each dissolved in 60 mL borate buffer (100 mM, pH 8.5). While stirring, the FITC solution was added drop wise to the lysozyme solution and the resulting solution was stirred for 16 h. Next,
Chapter 4
60
the protein solution was extensively dialyzed against water (at 4 °C) and the FITC-lysozyme was
collected after freeze-drying.
Protein stock solutions were prepared by dissolving 180 mg lysozyme or BSA and 20 mg FITC-lysozyme, respectively, FITC-BSA in 10 mL buffer (Hepes 100 mM pH 7.0).
The microspheres were prepared as described in section 2.2. After the washing and centrifugation steps, 500 mg of both dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres were dispersed in
5 mL buffer (Hepes 100 mM pH 7) and subsequently vigorously mixed. The vials were rinsed with 5 mL buffer that was subsequently added to the particle dispersion. After centrifugation (5 min 2000 rpm) 200 mg gel was transferred into eppendorf vials (1.5 mL) and 100 μL fluorescently
labeled protein stock solution was added. These mixtures were again intensively vortexed en centrifuged (5 min 2000 rpm). Samples containing either positively or negatively charged
microspheres were prepared in the same way. Just after preparation, the supernatant in the vials containing protein-loaded gels composed of oppositely charged microspheres was colorless and
the gels were yellow, whereas in the vials containing only microspheres of the same charge, the supernatant had a bright yellow color. This indicates that the fluorescently labeled protein was fully entrapped in the physically crosslinked network.
To perform FRAP experiments a spatula tip of the protein-loaded hydrogel was placed on an objective glass, on which an adhesive spacer (Secure-Seal Spacer, Molecular Probes, Leiden, The
Netherlands) of 0.5 mm thickness (adhering at both sides) was fixed, and subsequently protected with a cover glass. In this way evaporation and convection in the sample was prevented.
As a control, the diffusion coefficients of the fluorescent probes were measured in a sucrose solution (50 % w/w). The viscosity of the sucrose-protein solutions was determined using a Lauda
MGW 540 SK viscosimeter (Lauda MGW, Germany).
3. Results and discussion
3.1. Adsorption and absorption of proteins to the microspheres
Confocal images were taken to visualize the distribution of the proteins in the gel matrices.
Figures 2A and B show respectively lysozyme and BSA in gels composed of dextran microspheres with DS 5 in a medium with low ionic strength (100 mM Hepes, pH 7.0, ionic strength 17 mM). This figure illustrates that lysozyme was able to penetrate into the microspheres, while BSA was only
visible between the microspheres. The same results (not shown) were found for the gels composed of microspheres with higher crosslink densities (DS 8 and 18), indicating that lysozyme
(dh = 4.1 nm) is small enough to diffuse into the microspheres even when the DS is 18, while BSA (dh = 7.2 nm) is too large to penetrate into the microspheres. Figure 2A also shows that lysozyme is
not able to penetrate into all microspheres. Control experiments (Fig. 2C and D) reveal that lysozyme is able to penetrate into negatively charged microspheres, whereas positively charged microspheres do not absorb this protein. This result can be explained by the positive charge that
lysozyme bears at neutral pH (pIlysozyme = 9.3[48]). BSA, which is negatively charged at pH 7 (pIBSA = 4.7[49]), is adsorbed onto the surface of positively charged microspheres (Fig. 2B). No
penetration into neither the negatively nor positively charged microspheres is observed.
Protein release from hydrogels composed of oppositely charged dextran microspheres
61
Adsorption and/or absorption of the proteins to the microspheres were quantified as described in
section 2.4. Table 1 lists the results. In the presence of negatively charged microspheres, the microspheres ad(b)sorbed 75 % of the added lysozyme in medium of low ionic strength (100 mM Hepes, pH 7.0), whereas no adsorption onto the positive microspheres was found. BSA adsorbed
onto positively charged microspheres (35 % of amount added) and showed no adsorption onto negatively charged particles. These results are in good agreement with the confocal images and
can be explained by the net charge that BSA and lysozyme have at neutral pH. When the protein solutions were added to mixtures of positively and negatively charged microspheres, the adsorbed
amount was considerably lower (50 and 20 % for lysozyme and BSA, respectively), indicating that the proteins are only able to interact with part of the microsphere population. IgG (polyclononal, pI 5-10) showed the least adsorption to the particles (5-20 %), independent of the charge of the
microspheres. When the microspheres, preadsorbed with protein, were incubated in Hepes buffered saline (150 mM NaCl, 100 mM Hepes, pH 7.0) almost quantitative desorption occurred,
again indicating that the adsorption is due to electrostatic interactions between the proteins and microspheres of opposite charge.
Figure 2: Confocal images of FITC-lysozyme (2A) and FITC-BSA (2B) entrapped in hydrogels prepared by mixing
oppositely charged dextran microspheres (DS 5). Figure 2C and D show the confocal images of FITC-lysozyme in dispersions of negatively and positively charged dextran microspheres.
A
DC
B
Chapter 4
62
Table 1: Influence of ionic strength and protein charge on the adsorption and/or absorption of lysozyme, BSA
and IgG to charged microspheres.
Protein (charge) microsphere
charge
% adsorbed/absorbed after
1 h without NaCl
% adsorbed/absorbed
after 1 day after addition of NaCl after 1 h
Lysozyme (+) - 75 0
+ 0 0
+ and - 50 0
BSA (-) - 0 0
+ 35 10 + and - 20 10
IgG (polyclonal) - 5 0
+ 10 5
+ and - 20 5
3.2. In vitro protein release
To study the in vitro protein release from the ionically crosslinked gels, three model proteins,
lysozyme, BSA and IgG, differing in hydrodynamic radius, molecular weight and isoelectric point were used. The influence of protein properties, solid content of the gels, DS of the microspheres
and the ionic strength of the release buffer was investigated. Since the proteins investigated bear a charge at neutral pH and show adsorption onto the microspheres, it cannot be excluded that by
this process the interaction between the microspheres is diminished. It was shown that the
rheological parameters (G’, G”, tan(δ)) and the recovery during creep of a protein-loaded and
control gel prepared in Hepes buffer (100 mM, pH 7.0) were equal, indicating that proteins do not
influence the network properties of the gel. A continuous release of the proteins from both the 15 % and 30 % gels composed of microspheres
with DS 5, 8 and 18 was observed in Hepes buffered saline (100 mM, pH 7.0, 150 mM NaCl). Figure 3 shows a representative example. This figure illustrates that 50 % of the entrapped lysozyme, BSA and IgG was released in 4, 5 and 60 days, respectively.
The structure of proteins might be changed after their release from polymeric matrices because of stress factors applied during the preparation of the protein-loaded materials.[50] Also, degradation
of the matrices might be associated with protein degradation.[51-55] Therefore, we studied the structure of the released lysozyme with spectroscopic techniques (fluorescence, UV) and a
bioactivity assay was done to quantify its enzymatic activity. Fluorescence spectroscopy revealed that there were no shifts in the maximum fluorescence intensity peak of the released protein when
compared to native lysozyme, indicating that no structural damage of the protein has occurred. Moreover, UV scans showed no shift in maximum absorbance and no extra peaks in the 310-350 nm region, which indicates that no protein aggregates were formed. Finally, the lysozyme
activity assay showed that the specific activity of released lysozyme was the same as that of native lysozyme. It can therefore be concluded that neither degradation nor aggregation of the lysozyme
during preparation of the gel and/or during release had occurred.
Protein release from hydrogels composed of oppositely charged dextran microspheres
63
0 10 20 30 40 50 600
102030405060708090
100
BSAIgG
lysozyme
time (days)
cum
ulat
ive
rele
ase
(%)
Figure 3: Cumulative release of lysozyme ( ), BSA ( ) and IgG ( ) from hydrogels (15 % solid, DS 8). Symbols:
experimental values; curves: fitted theory (Equation 3). Data are shown as average (n=2).
Good agreement between the experimentally measured and theoretically calculated (Equation 3)
protein release kinetics was obtained in all cases (e.g. Fig. 3 and 4). This indicates that the release of the model proteins from the hydrogels is primarily diffusion controlled during the entire release period. Dependent on their size and surface charge, all proteins, are hindered by the microsphere
network. Smaller proteins with a hydrodynamic diameter smaller than the pores in the dextran microspheres are able to diffuse in and out of the microspheres. In contrast, larger protein
molecules are not able to penetrate into the microspheres and have to search their way between the microspheres, leading to a longer journey before they reach the gel surface and diffuse into
the release medium. But for both extreme situations, and in agreement with observations, the release of both small and large proteins is governed by diffusion. Based on the mathematical
analysis, the apparent diffusion coefficients of the respective proteins in the gels were determined (Table 2). The observed differences correlate very well with the molecular weight and hydrodynamic radii of the proteins (D(lysozyme) > D(BSA) > D(IgG)). In all cases, except for lysozyme with
microspheres of DS 8 and 18 and IgG with microspheres of DS 18, doubling the % solid content did not lead to a significant decrease in D (Unpaired t test, p > 0.05). The insignificant differences
between the diffusion coefficients of the proteins in the gels of 15 and 30 % solid content can be explained as follows. In the initial situation, 30 % gels will be composed of 30 % dry microspheres
hydrated with 70 % water (the microspheres were prepared as such that their equilibrium water content amounted 70 %), leaving no water in the spaces between the microspheres. In a 15 % gel, obtained by mixing 150 mg dry dextran microspheres with 850 mg water, the dry microspheres
will absorb 350 mg water, meaning 500 mg (50 % of the total hydrogel mass) is present between the microspheres. During the release experiment the gels are brought into contact with an excess
of water, which makes that they swell. A higher % of microspheres results in a higher swelling ratio (1.3 for 15 %; 1.8 for 30 %) of the macroscopic gel, eventually leading to gels of about the same
equilibrium water content (85-90 %). The data given in Table 2 also show that the 15 % hydrogels composed of microspheres of DS 18 (highest crosslink density) exhibited the fastest release. We
have previously shown that dextran gels with DS above 10 are dimensionally stable.[56] It can be expected that due to swelling the microspheres with a low DS (5 and 8) are pressed onto each
Chapter 4
64
other, which will restrict diffusion of the protein. However, the high crosslink density (DS 18)
prevents strong swelling of the microspheres, leading to larger pores between the microspheres when compared to the other gels and resulting in the faster release. Table 2 shows that the diffusion coefficients of the different model proteins in the 30 % gels are
not dependent on the microsphere crosslink densities. Confocal images (Fig. 2A) showed that lysozyme is absorbed by positively charged microspheres, independent of their crosslink density.
As a result, lysozyme is distributed in and between the microspheres. Hence, the lysozyme release is not influenced by the crosslink density of the microspheres and will be comparable for DS 5, 8
and 18. Confocal images (Fig. 2B) also showed that BSA is unable to penetrate into the microspheres, independent of their crosslink density. Consequently, BSA is only present in the pores between the microspheres resulting in a release rate that is independent of the DS of the
microspheres.
Table 2: Diffusion coefficients of lysozyme, BSA and IgG determined by fitting Equation (3) to the
experimentally measured protein release kinetics: effects of the solid content of the hydrogel, type of protein,
degree of substitution of dextran. Unless indicated otherwise, the release medium was Hepes buffer (100 mM,
pH 7.0, 0.02 % NaN3, 150 mM NaCl).
solid content protein DS D, 10-7 cm²/s (± s.d.) R²
15 % lysozyme 5 5.4 ± 0.6 0.99-1.00
8 5.1 ± 0.1 0.99-1.00
8 * 3.4 ± 0.3 0.99-1.00
18 8.8 ± 0.1 0.98-0.99
BSA 5 4.3 ± 0.2 0.98-0.99
8 4.4 ± 0.0 0.99
8 * 1.5 ± 0.1 0.97-0.99
18 6.8 ± 0.8 0.99
IgG 5 2.1 ± 0.1 0.99
8 2.1 ± 0.2 1.00
18 3.6 ± 0.1 0.99
30 % lysozyme 5 4.1 ± 0.6 0.99-1.00
8 3.9 ± 0.2 0.99
18 3.7 ± 0.0 1.00
BSA 5 4.1 ± 0.2 0.97-0.98
8 4.1 ± 0.5 0.97
18 5.0 ± 0.2 0.98
IgG 5 2.1 ± 0.5 0.98-0.99
8 2.6 ± 0.2 0.99-1.00
18 2.4 ± 0.3 0.99
* release medium: Hepes buffer (100 mM, pH 7.0, 0.02 % NaN3)
Protein release from hydrogels composed of oppositely charged dextran microspheres
65
The release of lysozyme and BSA from 15 % gels composed of microspheres with DS 8 was studied
in media of low (100 mM Hepes, pH 7.0) and physiological ionic strength (100 mM Hepes, pH 7.0, 150 mM NaCl) (Fig. 4). Equation 3 was fitted to the experimentally determined protein release kinetics and the calculated diffusion coefficients are reported in Table 2. In media with
physiological ionic strength, the diffusion coefficients of lysozyme and BSA were, respectively, 1.5 and 3 times higher than in media with low ionic strength. In both media full release was observed,
indicating that the adsorption/absorption, occurring in media of low ionic strength, is reversible. Adsorption/absorption and subsequent desorption processes retards the mobility of the proteins
in the gels and will consequently lower their diffusion coefficients in media of low ionic strength.
0 5 10 15 20 25 300
102030405060708090
100
lysozyme 150 mM NaCllysozyme no NaCl
cum
ulat
ive
lyso
zym
e re
leas
e (%
)
BSA 150 mM NaCl
time (days)
BSA no NaCl
Figure 4: The release of lysozyme ( ) and BSA ( ) from a 15 % gel (DS 8) in 100 mM Hepes, 150 mM NaCl,
pH 7.0, 0.02 % NaN3 (full symbols) and 100 mM Hepes, pH 7.0, 0.02 % NaN3 (open symbols). Symbols:
experimental values; curves: fitted theory (Equation 3). Data are shown as average (n=2).
3.3. Fluorescence recovery after photobleaching (FRAP)
To study the mobility of lysozyme and BSA on a micro scale in the microsphere dispersions and gels, FRAP measurements were performed. As a control, the diffusion coefficients of the proteins
were determined in 50 % sucrose solutions. This resulted in diffusion coefficients in water of 0.85 x 10-6 cm²/s (lysozyme) and 0.31 x 10-6 cm²/s (BSA). Figure 5 shows some representative bleaching
and recovery curves. The diffusion coefficients of the proteins in gels composed of microspheres with DS 5, 8 and 18 and dispersions of either cationically or anionically charged microspheres with DS 8, as determined by FRAP, are listed in Table 3. Lysozyme and BSA present in dispersions
(100 mM Hepes, pH 7.0) containing non-interacting positively charged or negatively charged microspheres, respectively, have diffusion coefficients similar to those in water. Obviously, the
proteins are repelled by the equal charges of the microspheres and diffuse freely through the dispersions. Upon mixing of the positively charged lysozyme with interacting negatively charged
microspheres or negatively charged BSA with positively charged microspheres, the opposite phenomenon occurs. The proteins are ionically attracted to the spheres, resulting in 10-20 times lower diffusion coefficients as compared to their values in water. Diffusion coefficients of both
proteins in gels composed of equal amounts of positively and negatively charged microspheres (100 mM Hepes, pH 7.0) were about the same as those observed in microsphere dispersion having
Chapter 4
66
opposite charge as the proteins. To minimize the ionic interactions between the proteins and the
microspheres, FRAP was performed on gels (DS 8) containing 150 mM NaCl protein solution (Table 3). In these hydrogels, the diffusion of lysozyme and BSA was, respectively, 2 to 4 times faster than in low ionic strength media. These findings are in good agreement with the adsorption and
protein release data as described in sections 3.1 and 3.2. In general, the diffusion coefficients obtained with FRAP are 5 to 20 times smaller than the D’s calculated from the release experiments.
These discrepancies result from the differences in experimental setup. The diffusion coefficients of the proteins during their release are macroscopic D’s, whereas those measured with FRAP are on a
microscopic level. It might be possible that in gels two populations of protein molecules are present. One with a high mobility, present in the pores between the microspheres and a diffusion coefficient (almost equal) as that in water, and one fraction with restricted mobility (ad(b)sorbed
by the microspheres). The fraction of protein molecules with high mobility diffuses too rapidly for our confocal microscope to determine their D and consequently the D of the protein molecules
interacting with the microspheres is obtained using FRAP.
-10 0 10 20 30 40 50 60 70 800.80
0.85
0.90
0.95
1.00
time (s)
BSAlysozymefluor
esce
nce
reco
very
Figure 5: Fluorescence recovery after photobleaching of FITC-lysozyme ( ) and FITC-BSA ( ) in dex-HEMA-
MAA/dex-HEMA-DMAEMA hydrogels in Hepes buffer (100 mM, pH 7.0, 150 mM NaCl).
Table 3: Diffusion Coefficients (10-8 cm²/s) of the model proteins in gels (30 % solid) determined with FRAP
(data are shown as averages ± standard deviation (n=5-7)). Unless indicated otherwise, the medium was Hepes
buffer (100 mM, pH 7.0, 0.02 % NaN3).
Gels DS
5 8 8* 18
Lysozyme 6.9 ± 0.8 7.2 ± 1.0 14 ± 2.0 3.8 ± 1.2
DSA 3.1 ± 0.6 2.2 ± 0.6 8.5 ± 3.0 0.8 ± 0.2
Dispersions (DS 8) Negatively charged microspheres Positively charged microspheres
Lysozyme 4.6 ± 1.2 31 ± 2
BSA 82 ± 16 4.0 ± 1.0
* Hepes buffer (100 mM, pH 7.0, 0.02 % NaN3, 150 mM NaCl)
Protein release from hydrogels composed of oppositely charged dextran microspheres
67
4. Conclusions
Physically crosslinked hydrogels are an attractive class of protein delivery systems since, during
preparation and protein loading, harsh crosslinking conditions are avoided, thereby maintaining the biological activity of the proteins. This paper reports on the in vitro protein release from an
injectable self-assembling hydrogel based on oppositely charged dextran microspheres. The gels show a continuous release of the entrapped model proteins with full preservation of the
enzymatic activity of lysozyme. This emphasizes the protein-friendly nature of the hydrogel. In conclusion, these hydrogel systems are very suitable for the diffusion-controlled release of pharmaceutical proteins. At present we are studying the degradation behavior of the gels.
Chapter 4
68
References
1. Hoffman AS. Hydrogels for biomedical applications. Adv. Drug Deliver. Rev. 2002;43:3-12.
2. Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chem. Rev. 2001;101:1869-1879.
3. Drury JL, Mooney DJ. Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials 2003;24:4337-4351.
4. Peppas NA, Bures P, Leobandung W, Ichikawa H. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 2000;50:27-46.
5. Panyam J, Labhasetwar V. Biodegradable nanoparticles for drug and gene delivery to cells and tissue. Adv. Drug Deliver. Rev. 2002;55:329-347.
6. Freiberg S, Zhu XX. Polymer microspheres for controlled drug release. Int. J. Pharm. 2004;282:1-18.
7. Sinha VR, Trehan A. Biodegradable microspheres for protein delivery. J. Control. Release 2003;90:261-280.
8. Hennink WE, van Nostrum CF. Novel crosslinking methods to design hydrogels. Adv. Drug Deliver. Rev. 2002;54:13-36.
9. Nagahara S, Matsuda T. Hydrogel formation via hybridization of oligonucleotides derivatized in water-soluble vinyl polymers. Polym. Gels Netw. 1996;4:111-127.
10. Stenekes RJH, Talsma H, Hennink WE. Formation of dextran hydrogels by crystallization. Biomaterials 2001;22:1891-1898.
11. Gacesa P. Alginates. Carbohyd. Polym. 1988;8:161-182.
12. Rowley JA, Madlambayan G, Mooney DJ. Alginate hydrogels as synthetic extracellular matrix materials. Biomaterials 1999;20:45-53.
13. Berger J, Reist M, Mayer JM, Felt O, Peppas NA, Gurny R. Structure and interactions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications. Eur. J. Pharm. Biopharm. 2004;57:19-34.
14. Sreenivasachary N, Lehn J-M. Gelation-driven component selection in the generation of constitutional dynamic hydrogels based on guanine-quartet formation. P. Natl. Acad. Sci. USA 2005;102:5938-5943.
15. Gombotz WR, Wee SF. Protein release from alginate matrices. Adv. Drug Deliver. Rev. 1998;31:267-285.
16. Hamley IW, Ansari IA, Castelletto V, Nuhn H, Rosler A, Klok HA. Solution self-assembly of hybrid block copolymers containing poly(ethylene glycol) and amphiphilic ß-strand peptide sequences. Biomacromolecules 2005;6:1310-1315.
17. Anseth KS, Metters AT, Bryant SJ, Martens PJ, Elisseef JH, Bowman CN. In situ forming degradable networks and their application in tissue engineering and drug delivery. J. Control. Release 2002;78:199-209.
18. Balakrishnan B, Jayakrishnan A. Self-cross-linking biopolymers as injectable in situ forming biodegradable scaffolds. Biomaterials 2005;26:3941-3951.
19. Tae G, Kornfield JA, Hubbell JA. Sustained release of human growth hormone from in situ forming hydrogels using self-assembly of fluoroalkyl-ended poly(ethylene glycol). Biomaterials 2005;26:5259-5266.
20. De Jong SJ, De Smedt SC, Wahls MWC, Demeester J, Kettenes-van den Bosch JJ, Hennink WE. Novel self-assembled hydrogels by stereocomplex formation in aqueous solution of enantiomeric lactic acid oligomers grafted to dextran. Macromolecules 2000;33:3680-3686.
21. De Jong SJ, van Eerdenbrugh B, van Nostrum CF, Kettenes-van den Bosch JJ, Hennink WE. Physically crosslinked dextran hydrogels by stereocomplex formation of lactic acid oligomers: degradation and protein release behavior. J. Control. Release 2001;71:261-275.
22. Bos GW, Jacobs JJL, Koten JW, Van Tomme SR, Veldhuis TFJ, van Nostrum CF, Den Otter W, Hennink WE. In situ crosslinked biodegradable hydrogels loaded with IL-2 are effective tools for local IL-2 therapy. Eur. J. Pharm. Sci. 2004;21:561-567.
23. Li SM, Vert M. Synthesis, characterization and stereocomplex-induced gelation of block copolymers prepared by ring opening polymerization of L(D)-lactide in the presence of poly(ethylene glycol). Macromolecules 2003;36:8008-8014.
24. Pochan DJ, Schneider JP, Kretsinger J, Ozbas B. Thermally reversible hydrogels via intramolecular folding and consequent self-assembly of a de novo designed peptide. J. Am. Chem. Soc. 2003;125:11802-11803.
25. Ruel-Gariepy E, Leroux J-C. In situ-forming hydrogels--review of temperature-sensitive systems. Eur. J. Pharm. Biopharm. 2004;58:409-426.
26. Jeong B, Kim SW, Bae YH. Thermosensitive sol-gel reversible hydrogels. Adv. Drug Deliver. Rev. 2002;54:37-51.
27. Huang X, Lowe TL. Biodegradable thermoresponsive hydrogels for aqueous encapsulation and controlled release of hydrophilic model drugs. Biomacromolecules 2005;6:2131-2139.
28. Cohn D, Sosnik A, Garty S. Smart hydrogels for in situ generated implants. Biomacromolecules 2005;6:1168-1175.
29. Zhang R, Tang M, Bowyer A, Eisenthal R, Hubble J. A novel pH- and ionic-strength-sensitive carboxy methyl dextran hydrogel. Biomaterials 2005;26:4677-4683.
Protein release from hydrogels composed of oppositely charged dextran microspheres
69
30. Chiu HC, Lin YF, Hsu YH. Effects of acrylic acid on preparation and swelling properties of pH-sensitive dextran hydrogels. Biomaterials 2002;23:1103-1112.
31. Murakami Y, Maeda M. Hybrid hydrogels to which single-stranded (ss) DNA probe is incorporated can recognize specific ssDNA. Macromolecules 2005;38:1535-1537.
32. Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Self-gelling hydrogels based on oppositely charged dextran microspheres. Biomaterials 2005;26:2129-2135.
33. Franssen O, Vandervennet L, Roders P, Hennink WE. Degradable dextran hydrogels: controlled release of a model protein from cylinders and microspheres. J. Control. Release 1999;60:211-221.
34. Cadée JA, Brouwer LA, den Otter W, Hennink WE, van Luyn MJA. A comparative biocompatibility study of microspheres based on crosslinked dextran or poly(lactic-co-glycolic) acid after subcutaneous injection in rats. J. Biomed. Mater. Res. 2001;56:600-609.
35. Van Dijk-Wolthuis WNE, Tsang SKY, Kettenes-van den Bosch JJ, Hennink WE. A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer 1997;38:6235-6242.
36. Van Dijk-Wolthuis WNE, van Steenbergen MJ, Underberg WJM, Hennink WE. Degradation kinetics of methacrylated dextrans in aqueous solution. J. Pharm. Sci. 1997;86:413-417.
37. Arturson G, Wallenius G. The renal clearance of dextran of different molecular sizes in normal humans. Scan. J. Clin. Lab. Inv. 1964;1:81-86.
38. De Geest BG, Déjugnat C, Sukhorukov GB, Braeckmans K, De Smedt SC, Demeester J. Self-rupturing microcapsules Adv. Mater. 2005;17:2357-2361.
39. Merrill EW, Dennison KA, Sung C. Partitioning and diffusion of solutes in hydrogels of poly(ethylene oxide). Biomaterials 1993;14:1117-1126.
40. Burczak K, Fujisato T, Hatada M, Ikada Y. Protein permeation through poly(vinyl alcohol) hydrogel membranes. Biomaterials 1994;15:231-238.
41. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC. Measurement of protein using bicinchoninic acid. Anal. Biochem. 1985;150:76-85.
42. Shih P, Malcolm BA, Rosenberg S, Kirsch JF, Wilsow AC. Reconstruction and testing of ancestral proteins
In Methods in Enzymology, Academic Press: 1993; Vol. 224, pp 576-590.
43. Crank J. The mathematics of diffusion. Clarendon Press: Oxford, 1975.
44. Watkins AW, Anseth KS. Investigation of molecular transport and distributions in poly(ethylene glycol) hydrigels with confocal laser scanning microscopy. Macromolecules 2005;38:1326-1334.
45. De Smedt SC, Meyvis TKL, Demeester J. Diffusion of macromolecules in dextran methacrylate solutions and gels as studied by confocal scanning laser microscopy. Macromolecules 1997;30:4863-4870.
46. Meyvis TKL, De Smedt SC, Van Oostveldt P, Demeester J. Fluorescence recovery after photobleaching: a versatile tool for mobility and interaction measurements in pharmaceutical research. Pharm. Res. 1999;16:1153-1162.
47. Braeckmans K, Peeters L, Sanders NN, De Smedt SC, Demeester J. Three-dimensional fluorescence recovery after photobelaching with the confocal scanning laser microscope. Biophys. J. 2003;85:2240-2252.
48. Petersen SB, Jonson V, Fojan P, Wimmer R, Pedersen S. Sorbitol prevents the self-aggregation of unfolded lysozyme leading to an up to 13 degreesC stabilisation of the folded form. J. Biotechnol. 2004;114:269-278.
49. Hu J, Li S, Liu B. Adsorption of BSA onto sulfonated microspheres. Biochem. Eng. J. 2005;23:259-263.
50. Van de Weert M, Hennink WE, Jiskoot W. Protein instability in poly(lactic-co-glycolic acid) microparticles. Pharm. Res. 2000;17:1159-1167.
51. Schwendeman SP. Recent advances in the stabilization of proteins encapsulated in injectable PLGA delivery systems. Crit. Rev. Ther. Drug 2002;19:73-98.
52. Jiang W, Gupta RK, Deshpande MC, Schwendeman SP. Biodegradable poly(lactic-co-glycolic acid) microparticles for injectable delivery of vaccine antigens. Adv. Drug Deliver. Rev. 2005;57:391-410.
53. Tamber H, Johansen P, Merkle HP, Gander B. Formulation aspects of biodegradable polymeric microspheres for antigen delivery. Adv. Drug Deliver. Rev. 2005;57:357-376.
54. Li L, Schwendeman SP. Mapping neutral microclimate pH in PLGA microspheres. J. Control. Release 2005;101:163-173.
55. Lucke A, Kiermaier J, Göpferich A. Peptide acylation by poly(alpha-hydroxy esters). Pharm. Res. 2002;19:175-181.
56. Hennink WE, Talsma H, J.C.H B, Smedt SCD, Demeester J. Controlled release of proteins from dextran hydrogels. J. Control. Release 1996;39:47-55.
5 Degradation behavior of dextran hydrogels composed of
positively and negatively charged microspheres
Sophie R. Van Tommea, Cornelus F. van Nostruma, Stefaan C. de Smedtb, Wim E. Henninka
a Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS),
University Utrecht, Utrecht, The Netherlands bLaboratory of General Biochemistry and Physical Pharmacy, Department of Pharmaceutics,
Ghent University, Ghent, Belgium
Biomaterials 27 (2006) 4141-4148
Chapter 5
72
Abstract
This paper reports on the degradation behavior of in situ gelling hydrogel matrices composed of
positively and negatively charged dextran microspheres. Rheological analysis showed that, once the individual microspheres started to degrade, the hydrogel changed from a mainly elastic tot a
viscoelastic network. It was shown with gels composed of equal amounts of cationic and anionic microspheres that both a higher crosslink density of the particles and a decrease in water content of the hydrogels resulted in a slower degradation, ranging from 65 to 140 days. Dispersions
containing cationic, neutral or anionic microspheres completely degraded within 30, 55 or 120 days, respectively. The microspheres were loaded with rhodamine-B-dextran and degradation
was studied with confocal microscopy and fluorescence spectroscopy. After a lag time of 3 days rhodamine-B-dextran started to release from the positive microspheres with a 50 % release after
16 days. In contrast, release of rhodamine-B-dextran from the negative microspheres started after 10 days with a 50 % release after 36 days. The faster degradation of the positively charged microspheres as compared to the negatively
charged microspheres is attributed to stabilization of the transition state in the hydrolysis process by the protonated tertiary amine groups present in the cationic microspheres. On the other hand,
the presence of negatively charged groups causes repulsion of hydroxyl anions resulting in a slower degradation. Combining the oppositely charged microspheres in different ratios makes it
possible to tailor the network properties and the degradation behavior of these hydrogels, making them suitable for various applications in drug delivery and tissue engineering.
Keywords: dextran microspheres, self-gelling, injectable hydrogel, degradation behavior, release
Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
73
1. Introduction
During the past decades, hydrogels, composed of synthetic or natural polymers, have been widely
investigated as protein delivery systems.[1-4] Network formation can be established by either chemical or physical crosslinking methods, of which the latter has some distinct advantages for
pharmaceutical and biomedical applications.[5] Many physically crosslinked hydrogels have been developed, based on e.g. stereocomplex formation, hydrogen bonding, hydrophobic interactions
and ionic interactions.[6-11] Particularly attractive are those polymers that can be injected and form gels spontaneously at the site of injection.[6-8, 10, 12-14] Besides for drug delivery applications, (injectable) hydrogels are of great interest in tissue engineering due to their good hydrophilicity
and mechanical properties.[15-17] For the delivery of macromolecular drugs and for tissue engineering applications, biodegradability
of the hydrogels is an important issue. Degradable delivery systems do not require surgical removal and, importantly, release of the entrapped proteins can be tailored by the degradation
characteristics of the gel. For tissue engineering applications, scaffolds for the development of new tissue need to provide mechanical support until sufficient extra-cellular matrix has been formed. Tailoring of the degradation rate of hydrogels allows to specifically design systems for various
applications.[18] Previously, we reported on a novel self-gelling hydrogel, based on physical crosslinking between
water-swollen dextran microspheres of opposite charge[19]. The ionic interactions between the microspheres can be broken when exposed to stress and rebuilt upon removal of the stress,
making the system suitable for injection. The individual microspheres are composed of an interpenetrating network of dextran and poly(hydroxyethyl methacrylate) (pHEMA) that is
crosslinked by carbonate ester bonds between the two polymer chains. The carbonate ester bonds guarantee degradability of the microspheres and thus the hydrogel composed thereof under physiological conditions.[20-22] Recently, it was shown that proteins could be entrapped in
the hydrogel matrix by simply hydrating the microspheres in a protein solution. Modeling the release of model proteins revealed a diffusion-controlled delivery.[23]
This paper focuses on the degradation behavior of the charged dextran microspheres and macroscopic hydrogels composed of various amounts of positively and negatively charged
microspheres. Release of rhodamine-B-dextran from both microsphere types was studied. The influence of degradation of the microspheres on the network properties of the hydrogels was
studied rheologically. Swelling behavior of different types of dex-HEMA-MAA (= negatively charged microspheres)/dex-HEMA-DMAEMA (= positively charged microspheres) gels, varying in degree of substitution of the microspheres, solid content of the gel and ratio positively to
negatively charged microspheres, was investigated.
Chapter 5
74
2. Materials and methods
2.1. Materials
Dextran T40 (from Leuconostoc ssp.), N,N,N’,N’-tetramethylethylenediamine (TEMED) and 2-
hydroxyethyl methacrylate (HEMA) were provided by Fluka (Buchs, Switzerland). Poly(ethylene glycol) (PEG) 10 kDa and potassium peroxodisulfate (KPS) were purchased from Merck (Darmstadt,
Germany). N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes) was obtained from Acros Chimica (Geel, Belgium). Methacrylic acid (MAA) and dimethylaminoethyl methacrylate (DMAEMA) were provided by Sigma-Aldrich (Zwijndrecht, The Netherlands). Rhodamine-B-dextran (Mw 70
kDa) was purchased from Invitrogen B.V. (Breda, The Netherlands). Hydroxyethyl methacrylate-derivatized dextran (dex-HEMA) was synthesized and characterized according to Van Dijk-Wolthuis
et al.[24] The degrees of substitution (DS, i.e. the number of HEMA groups per 100 glucopyranose units) used in this study were 5, 8, 10 and 18.
2.2. Preparation of charged dex-HEMA microspheres
Negatively and positively charged microspheres were prepared as described previously.[19] In short,
aqueous solutions of PEG (40 % (w/w)) and dex-HEMA (20 % (w/w)) were prepared in Hepes buffer (100 mM, pH 7.0). The dextran microspheres, with a water content of 70 %, were obtained through
radical polymerization of dex-HEMA, emulsified in the aqueous poly(ethylene glycol) solution. Addition of either methacrylic acid (MAA) or dimethylaminoethyl methacrylate (DMAEMA) to the
polymerization mixture (final concentration 25 mM) resulted in negatively (dex-HEMA-MAA) and positively (dex-HEMA-DMAEMA) charged microspheres, respectively. The crosslinked particles were purified by multiple washing and centrifugation steps and finally the microspheres were
lyophilized. Microspheres were also prepared with 10 times less MAA (2.5 mM) and 10 times more DMAEMA (250 mM). Furthermore, neutral microspheres were prepared without the addition of
MAA or DMAEMA. The particle size and size distribution of the microspheres were determined with a laser blocking
technique, using an Accusizer® (model 770, Particle Sizing Systems, Santa Barbara, CA, USA). The
microsphere charge was confirmed by ξ-potential measurements using a Malvern Zetasizer 2000
(Malvern Instruments, Worcestershire, UK).[25]
Fluorescently labeled microspheres (DS 10) were obtained by adding a rhodamine-B-dextran solution to the dex-HEMA/PEG mixture, prior to polymerization (ratio rhodamine-B-dextran/dex-
HEMA: 1/183 (w/w)). Encapsulation efficiencies were determined by measuring the fluorescence in the PEG and wash fractions.
2.3. Formation of physically crosslinked hydrogels using charged microspheres
Hydrogels consisting of equal amounts of positively (dex-HEMA-DMAEMA) and negatively (dex-
HEMA-MAA) charged microspheres were prepared by mixing lyophilized microspheres. Next, the
microsphere mixture was hydrated in Hepes buffer (100 mM, pH 7.0) for 1 h at 4 °C. The solid
content of the gels was varied between 15 and 25 % (w/w). Hydrogels consisting of variable
Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
75
amounts of positively and negatively charged microspheres and dispersions of solely negatively,
positively or neutral microspheres were prepared in the same way as the hydrogels.
2.4. Monitoring the swelling behavior of the hydrogels during degradation
Gels (200 mg) were prepared as described in section 2.3 and transferred into 1 mL glass vials (diameter of 6.5 mm) and gently centrifuged (2 min 1000 rpm). Subsequently, the gels were
equilibrated at 4 °C for 16 h. Next, the length of the gels was measured (L0) and 600 μL Hepes
buffered saline (100 mM, pH 7.0, 150 mM NaCl) was added. The length of the gels was generally 5
to 10 mm. The vials were placed in a water bath at 37 °C and at regular time intervals the length of
the gels was measured (Lt) to calculate the swelling ratio (S = Lt/L0). The solid content of the gels, the DS of the microspheres and the ratio positively to negatively charged microspheres were varied (respectively 15 % and 25 %, DS 5, 8 and 18, ratios (+/-) 0/100, 25/75, 50/50, 75/25 and
100/0). For the degradation of neutral microspheres and micropsheres prepared with a higher and lower concentration of DMAEMA and MAA, respectively, dispersions in buffer were prepared
composed of 15 % microspheres (DS 8).
2.5. Rhodamine-B-dextran release
Rhodamine-B-dextran was entrapped in the dex-HEMA-MAA and dex-HEMA-DMAEMA
microspheres as described in section 2.2. Hydrogels (500 mg, 15 % solid, microspheres of DS 10) were prepared as described in section 2.3. The release of rhodamine-B-dextran from negatively and positively charged microspheres was studied, as well as from a gel composed of both
positively and negatively charged microspheres. The release experiments were done as described
previously[23]. In short, after hydration for 1 h at 4 °C, the gels were transferred into a release device
consisting of a gel and a release compartment (8.8 mm x 8.5 mm and 15 mm x 30 mm (diameter x length), respectively) and closed with a rubber stop. Release buffer (3 mL, 100 mM Hepes, pH 7.0, 150 mM NaCl, 0.02 % NaN3) was added to each gel and the device was incubated on a shaking
plate at 37 °C. Samples of 0.5 mL were taken at regular time intervals and replaced by an equal
volume of fresh buffer. The concentration of rhodamine-B-dextran in the release samples was
measured using a FLUOstar OPTIMA Microplate-based multi-detection reader (Isogen Life Science, Maarssen, The Netherlands) equipped with a 550 nm excitation and a 600 nm emission filter. Standard rhodamine-B-dextran solutions (concentration range 0.001-0.050 mg/mL) were prepared
to generate a calibration curve. Degradation products of the gels did not affect the calibration curve (linear, R² = 0.99, coincides with calibration curve in buffer). Gels were made in triple and
samples (100 μL) were analyzed thrice. At various time points during the degradation of microspheres loaded with rhodamine-B-dextran,
samples were taken and studied with confocal fluorescence microscopy. A Leica TCS-SP confocal laser scanning microscope equipped with a 568 nm Krypton laser (Leica Microsystems) was used (40x oil immersion objective).
Chapter 5
76
2.6. Rheological behavior of degrading gels
During the initial stages of degradation of the hydrogels (400 mg, 22 % initial solid content, microspheres of DS 8) their network properties were monitored. At various moments during
swelling experiments (section 2.4) rheological analysis on the gels was performed using a controlled stress rheometer (AR1000-N, TA Instruments, Etten-Leur, The Netherlands), equipped
with an acrylic flat plate geometry (20 mm diameter) and a gap of 500 μm.[19] A solvent trap was used to prevent evaporation of the solvent. The viscoelastic properties of the gels were
determined by measuring the G’ (shear storage modulus) and G” (loss modulus) at 20 °C with a
constant strain of 1 % and constant frequency of 1 Hz. Creep experiments on dex-HEMA-MAA/dex-HEMA-DMAEMA gels were done by application of a shear stress of 1 Pa while the strain was
monitored. After 1 minute the stress was removed and the recovery of the sample was monitored by measuring the strain during 2 minutes.
3. Results
3.1. Characterization of charged dex-HEMA microspheres
The mean volume diameters of the dex-HEMA-MAA and the dex-HEMA-DMAEMA microspheres were 9 μm. Dex-HEMA microspheres containing the standard amount of MAA or DMAEMA had a
ξ-potential of -13 to -16 mV and +12 to +15 mV, respectively. The microspheres prepared with 10
times less MAA and those with 10 times more DMAEMA had ξ-potentials of -8 and +22 mV,
respectively.
3.2. Swelling behavior of gels composed of oppositely charged dextran microspheres
The degradation of hydrogels composed of dex-HEMA-MAA and dex-HEMA-DMAEMA
microspheres was studied by swelling experiments at 37 °C and pH 7.0. Figure 1 shows that
hydrogels composed of equal amounts of cationic and anionic microspheres (DS 8) reached their maximum swelling (swelling ratio 1.25 to 1.75) after approximately 3 to 8 days. Thereafter the
hydrogels started to disintegrate. Gels composed of 25 % initial solid content degraded within 95 days, whereas gels of 15 % initial solid content degraded within 65 days.
The influence of the crosslink density of the microspheres on the degradation profile of the hydrogels is illustrated in Figure 2. A higher crosslink density of the microspheres resulted in longer degradation times. Complete degradation of the hydrogels (25 % solid) was found after 66, 95 and
140 days for microspheres of respectively DS 5, 8 and 18.
Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
77
0 10 20 30 40 50 60 70 80 90 1000.000.250.500.751.001.251.501.752.002.25 15%
25%
time (days)
swel
ling
ratio
Figure 1: Swelling of hydrogels (15 % ( ) and 25 % ( ) solid) composed of microspheres of DS 8. Data are
shown as average (n=2).
0 20 40 60 80 100 120 1400.00.20.40.60.81.01.21.41.61.82.02.2
DS 8
DS 18
DS 5
time (days)
swel
ling
ratio
Figure 2: Swelling of hydrogels (25 % solid) composed of microspheres with varying DS (5 ( ), 8 ( ), 18 ( )).
Data are shown as average (n=2).
Figure 3 shows that the charge of the microspheres and the ratio positively/negatively charged
microspheres in the gels substantially influenced the degradation time of the hydrogels. Complete degradation of the microspheres was found after 30, 55 and 125 days for positive, neutral and
negative microspheres, respectively. Changing the charge density of the negative and positive microspheres led to degradation times of 15 and 70 days for microspheres containing a higher
amount of DMAEMA or lower amount of MAA, respectively (results not shown). Decreasing the ratio positively/negatively charged microspheres resulted in increasing degradation times of 75, 85 and 105 days, for gels consisting of 75/25, 50/50 and 25/75 positively/negatively charged
microspheres, respectively (Fig. 3).
Chapter 5
78
0 15 30 45 60 75 90 105 120 1350.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5 100+/0-75+/25-
50+/50-
0+/100-
neutral
25+/75-
time (days)
swel
ling
ratio
Figure 3: Swelling of hydrogels (25 % solid, DS 8) composed of different ratios of positively charged/negatively
charged microspheres (100+/0- ( ), 75+/25- ( ), 50+/50- ( ), 25+/75- ( ), 0+/100- ( )). Data are shown as
average (n=2). Swelling of neutral microspheres (15 % solid, DS 8) (∗). Data are shown as average ± standard
deviation (n=3).
3.3. Rhodamine-B-dextran release
To gain more insight into the degradation characteristics of gels composed of oppositely charged microspheres, both positively and negatively charged microspheres loaded with rhodamine-B-
dextran were prepared and used for the hydrogel formation (Fig. 4).
Figure 4: Confocal fluorescence image (magnification 40x) of rhodamine-B-dextran loaded microspheres (DS 8).
Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
79
During degradation of the hydrogels, rhodamine-B-dextran was released into the surrounding
medium (Fig. 5). Formulations were prepared in which positively charged microspheres loaded with rhodamine-B-dextran were combined with non-loaded negatively charged microspheres and vice versa. Release of rhodamine-B-dextran from the positively charged microspheres in
dispersions containing these microspheres only, started after 2 days and reached a plateau after approximately 30 days. This is in accordance to the results of the swelling experiments that
showed that in these dispersions the microspheres were degraded in about 30 days. Hydrogels (50 % positively and 50 % negatively charged microspheres) in which the positive microspheres
were loaded with rhodamine-B-dextran, started to release rhodamine-B-dextran after approximately 4 days. In comparison to the positively charged microspheres, the negatively charged ones released rhodamine-B-dextran less rapidly (Fig. 5).
0 5 10 15 20 25 30 35 40 45 500
102030405060708090
100 50+ rho/50-100+ rho/0-50- rho/50+100- rho/0+50+ rho/50- rho
time (days)
rhod
amin
e-B-
dext
ran
rele
ase
(%)
Figure 5: Rhodamine-B-dextran release from hydrogels (15 % solid, DS 10) composed of different ratios
positive/negative microspheres: ( ) 100+/0-; ( ) 0+/100-; other symbols represent 50+/50- gels with the
rhodamine-B-dextran loaded in the cationic ( ), anionic ( ), or in both microspheres (∗). Data are shown as
average ± standard deviation (n=3).
In dispersions containing only negative microspheres, rhodamine-B-dextran started to release after a lag time of 10 days; after 33 days 50 % of the entrapped rhodamine-B-dextran was released.
Surprisingly, in hydrogels composed of equal amounts of positively and negatively charged microspheres, in which the negative microspheres were loaded with rhodamine-B-dextran, the
release of rhodamine-B-dextran was even more delayed: its release started after a lag time of 17 days and 50 % was released after 37 days. This is in contrast to the results of the swelling tests
that suggest a faster degradation when the ratio positive/negative microspheres was increased (Fig. 3).
Chapter 5
80
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8
-0.65
-0.55
-0.45
-0.35
-0.25
-0.15
-0.05
log t
log
(Mt/M
∞)
0 5 10 15 20 25 30 35 40 45 500
102030405060708090
100
time (days)
% re
leas
e
Figure 6A (left): Rhodamine-B-dextran release from hydrogels (15 % solid, DS 10) composed of 50 % positively
and 50 % negatively charged microspheres, both loaded with rhodamine-B-dextran, plotted on a logarithmic
scale
Figure 6B (right): Rhodamine-B-dextran release from hydrogels (15 % solid, DS 10) composed of 50 % positively
and 50 % negatively charged microspheres, both loaded with rhodamine-B-dextran (( ) calculated release, (∗)
experimental release)
The release of rhodamine-B-dextran from a gel composed of 50 % positively and 50 % negatively charged microspheres, both loaded with rhodamine-B-dextran, started after 4 days; 50 % was
released in 28 days. The double logarithmic plot of release versus time showed a linear relationship with slope = 1.02 ± 0.03 (Fig. 6A), indicating that the release from this formulation follows zero-
order kinetics. The release from the latter gel can be predicted by taking the average of the release from: (1) the 50+/50- gel in which the cationic microspheres are loaded with rhodamine-B-dextran; and (2) the corresponding gel in which the anionic microspheres are loaded with rhodamine-B-
dextran. Figure 6B shows that an excellent correlation exists between the calculated and the experimental release profile of this formulation.
3.4. Confocal microscopy
Degradation of the rhodamine-B-dextran loaded microspheres was monitored by confocal microscopy (Table 1). The cationic microspheres showed clear signs of degradation after 12 days of
incubation at pH 7.0 and at 37 °C. Fluorescence was present between the microspheres, indicating
that the rhodamine-B-dextran was released from the microspheres. At day 21 most of the fluorescent-labeled dextran had leaked out of the positive microspheres and after 43 days the
positive microspheres were fully degraded. This is in agreement with the swelling experiments that showed complete degradation after approximately 30 days (Fig. 3, 100+/0-). In contrast, the negatively charged microspheres showed no signs of degradation after 43 days, which is in
agreement with the swelling data (Fig. 3, 100-/0+). At day 54, the anionic microspheres were still visible but they were bathing in fluorescence, indicating that, due to degradation, fluorescently
labeled dextran was leaking out of the microspheres. Finally, after 90 days, no microspheres could be observed.
A B
Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
81
3.5. Rheological properties of degrading hydrogels
The rheological properties of the hydrogels (22 % solid, DS 8, ratio +/-: 50/50) incubated at pH 7.0
and at 37 °C were measured in time (Fig. 7). This Figure shows that the G’ dropped from 2050 ±
730 Pa (day zero) to 210 ± 100 Pa and 10 ± 0 after 8 days and 16 days of degradation, respectively,
while the tan(δ) increased from 0.09 ± 0.02 to 0.14 ± 0.02 and 1.40 ± 0.14 over the same time
period. Decrease in gel strength, evidenced by the drop in G’, is attributed to water uptake and
swelling of the microspheres. Once the network starts to disintegrate (day 16), the interactions become weaker, causing a change from a mainly elastic to a viscoelastic network as evidenced
from the increase in tan(δ). These results are in good agreement with the swelling behavior shown
in Figures 1 and 2.
0 days 0.3 days 3 days 8 days 16 days0
500
1000
1500
2000
2500
3000
0.00
0.25
0.50
0.75
1.00
1.25
1.50
1.75
swelling time
G' (
Pa)
tan(
δ )
Figure 7: Storage modulus (G’ (Pa), black bars) and tan (δ) (open bars) as a function of the swelling time of
hydrogels (22 % solid, DS 8) composed of equal amounts of cationic and anionic microspheres. Data are shown
as average ± standard deviation (n=4).
Chapter 5
82
Table 1: Confocal fluorescence microscopy images of positively and negatively charged microspheres loaded
with rhodamine-B-dextran (DS 8) as a function of the degradation time.
degradation time positively charged microspheres negatively charged microspheres
4 days
12 days
21 days
43 days
54 days
90 days
Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
83
4. Discussion
Degradation of macroscopic dex-HEMA gels has been studied by van Dijk-Wolthuis et al.[20, 21] and
Franssen et al.[22] Hydrolysis of the carbonate ester bond between the dextran backbone and the crosslinked HEMA side chains results in degradation of the dex-HEMA gels at physiological pH.
Degradation data of macroscopic dex-HEMA gels showed that a higher DS resulted in a longer degradation time since more crosslinks need to be hydrolyzed. Decreasing the initial water
content also caused slower degradation.[21] The same trends are now visible with regard to dex-HEMA microspheres. Swelling of hydrogels composed of dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres is due to water uptake by the microparticles and water absorption in the
pores of the gels composed of oppositely charged crosslinked microspheres. Furthermore, it was described previously that, as a consequence of the hydrolysis of the carbonate esters, leading to a
decreased crosslink density, the elasticity of the dex-HEMA network drops, allowing the microspheres to absorb more water resulting in additional swelling of the gels.[26] After the initial
swelling phase, further degradation causes weight loss and decrease in gel volume. Increasing the crosslink density (DS 5-18) of the microspheres and the solid content of the gels (15-25 %) led to significantly longer degradation times (65 to 140 days) (Fig. 1 and 2).
The degradation data demonstrated that faster degradation of positively charged microspheres and slower degradation of negatively charged microspheres occurred when compared to neutral
ones. It was found previously that the hydrolysis of the carbonate esters in dex-HEMA at physiological pH is catalyzed by OH- ions.[20] Taking this hydroxyl catalysis into account, it is
reasonable that the hydrolysis of the positively charged microspheres (containing NH+(CH3)2 groups) occurs faster than that of the negatively charged ones (containing COO- groups). Firstly,
the anionic COO- groups repel the OH- ions whilst the cationic NH+(CH3)2 groups attract OH- ions, resulting in a local environment of lower and higher pH, respectively. Secondly, it was shown by van de Wetering et al.[27] that in DMAEMA, at pH 4-8 the NH+(CH3)2
group can approach the
carbonyl oxygen, rendering the methacrylate ester more susceptible to hydrolysis. Such interaction might also affect the carbonate ester between the dextran backbone and
HEMA side chains. In this way the transition state of the hydrolysis of the carbonate ester is stabilized by the proton of the NH+(CH3)2
group, resulting in a decrease of activation energy for
hydrolysis and faster degradation. This theory is supported by an even more accelerated degradation of microspheres containing 10 times more DMAEMA groups (15 vs. 30 days).
Microspheres containing 10 times less MAA degraded significantly faster than the negative microspheres containing the higher amount of MAA (60 vs. 125 days). In Figure 8 the radical polymerization of dex-HEMA in the presence or absence of MAA or DMAEMA and the resulting
network is schematically presented for the three types of microspheres. The possible interactions influencing the degradation behavior of the microspheres are illustrated.
Chapter 5
84
OR
O
R
OR
O
O
O
OHO
HOOOO
O
O
O
O
OHO
HOOOO
O
OOR
O
ran
dex-HEMA-MAA microspheres (R=R2)
dex-HEMA-DMAEMA microspheres (R=R3)
O
O
OHO
HOOOO
O
OO
O
ran
O
O
OHO
HOOOO
O
OO
O
ran
HN
OH
O
O
OHO
HOOO
O
R2= H R3=R1= N
dex-HEMA microspheres (R=R1)
O
O
OHO
HOOOO
O
O
O
O
ran
O
O
OHO
HOOO
O
OH
Figure 8: Schematic presentation of the radical polymerization of dex-HEMA in the presence of MAA or
DMAEMA (R=R1: dex-HEMA microspheres, R=R2: dex-HEMA-MAA microspheres, R=R3: dex-HEMA-DMAEMA
microspheres). The effect of charge in the network on its degradation behavior is illustrated.
Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
85
Release of rhodamine-B-dextran entrapped in the microspheres was degradation controlled. Faster
degradation of positively charged microspheres led to an earlier release of the rhodamine-B-dextran from these microspheres when compared to rhodamine-B-dextran release from negatively charged microspheres. Addition of negatively charged microspheres to positively charged
microspheres loaded with rhodamine-B-dextran, or vice versa, in both cases resulted in an increased lag time for release when compared to release from dispersions containing only the
loaded microspheres. In a hydrogel network the rhodamine-B-dextran that is released from the degrading microspheres, can be entrapped in the pores between the ionically crosslinked
particles. Release from a gel composed of positive and negative microspheres, both loaded with rhodamine-B-dextran, calculated from the release data of gels composed of 50 % loaded microspheres and 50 % non-loaded microspheres, shows a good similarity with the experimental
release from these gels. The correspondence between the experimental and calculated release profiles indicates that the release profiles of various combinations of positive/negative
microspheres can be predicted. In this way specific release systems can be designed, such as a zero-order release system as demonstrated in this paper.
5. Conclusion
In this study the degradation behavior of a self-gelling injectable hydrogel composed of oppositely charged dextran microspheres was investigated. The swelling behavior and degradation rate of
the gels can be tailored by varying the crosslink density of the microspheres, the % solid content of the gels and the ratio positively to negatively charged microspheres. Increasing the crosslink
density, the % solid and the amount of negatively charged microspheres, results in significantly longer degradation times. Varying the amount of positive and negative microspheres makes it
possible to modulate the degradation behavior of the complete hydrogel. The possibility to tailor the network properties and degradation times of these hydrogels makes them attractive for various drug delivery and tissue engineering applications.
Acknowledgement
The authors would like to thank the Center for Cell Imaging of the Faculty of Veterinary Medicine (Utrecht University, The Netherlands) for use of the confocal microscope. Marjan Fretz (Department
of Pharmaceutics, Utrecht University, the Netherlands) is gratefully acknowledged for help with confocal microscopy. The research was supported by a grant of the Dutch Ministry of Economic
affairs.
Chapter 5
86
References
1. Peppas NA, Bures P, Leobandung W, Ichikawa H. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 2000;50:27-46.
2. Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chem. Rev. 2001;101:1869-1879.
3. Hoffman AS. Hydrogels for biomedical applications. Adv. Drug Deliver. Rev. 2002;43:3-12.
4. Gupta P, Vermani K, Garg S. Hydrogels: from controlled release to pH-responsive drug delivery. Drug Discov. Today 2002;7:569-579.
5. Hennink WE, van Nostrum CF. Novel crosslinking methods to design hydrogels. Adv. Drug Deliver. Rev. 2002;54:13-36.
6. De Jong SJ, De Smedt SC, Wahls MWC, Demeester J, Kettenes-van den Bosch JJ, Hennink WE. Novel self-assembled hydrogels by stereocomplex formation in aqueous solution of enantiomeric lactic acid oligomers grafted to dextran. Macromolecules 2000;33:3680-3686.
7. Stenekes RJH, Talsma H, Hennink WE. Formation of dextran hydrogels by crystallization. Biomaterials 2001;22:1891-1898.
8. Tae G, Kornfield JA, Hubbell JA. Sustained release of human growth hormone from in situ forming hydrogels using self-assembly of fluoroalkyl-ended poly(ethylene glycol). Biomaterials 2005;26:5259-5266.
9. Huang X, Lowe TL. Biodegradable thermoresponsive hydrogels for aqueous encapsulation and controlled release of hydrophilic model drugs. Biomacromolecules 2005;6:2131-2139.
10. Jeong B, Kim SW, Bae YH. Thermosensitive sol-gel reversible hydrogels. Adv. Drug Deliver. Rev. 2002;54:37-51.
11. Berger J, Reist M, Mayer JM, Felt O, Peppas NA, Gurny R. Structure and interactions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications. Eur. J. Pharm. Biopharm. 2004;57:19-34.
12. Hatefi A, Amsden B. Biodegradable injectable in situ forming drug delivery systems. J. Control. Release 2002;80:9-28.
13. Packhaeuser CB, Schnieders J, Oster CG, Kissel T. In situ forming parenteral drug delivery systems: an overview. Eur. J. Pharm. Biopharm. 2004;58:445-455.
14. Ruel-Gariepy E, Leroux J-C. In situ-forming hydrogels--review of temperature-sensitive systems. Eur. J. Pharm. Biopharm. 2004;58:409-426.
15. Kim B-S, Mooney DJ. Development of biocompatible synthetic extracellular matrices for tissue engineering. Trends Biotechnol. 1998;16:224-230.
16. Drury JL, Mooney DJ. Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials 2003;24:4337-4351.
17. Lévesque SG, Lim RM, Shoichet MS. Macroporous interconnected dextran scaffolds of controlled porosity for tissue-engineering applications. Biomaterials 2005;26:7436-7446.
18. Lee KY, Bouhadir KH, Mooney DJ. Controlled degradation of hydrogels using multi-functional cross-linking molecules. Biomaterials 2004;25:2461-2466.
19. Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Self-gelling hydrogels based on oppositely charged dextran microspheres. Biomaterials 2005;26:2129-2135.
20. Van Dijk-Wolthuis WNE, van Steenbergen MJ, Underberg WJM, Hennink WE. Degradation kinetics of methacrylated dextrans in aqueous solution. J. Pharm. Sci. 1997;86:413-417.
21. Van Dijk-Wolthuis WNE, Hoogeboom JAM, van Steenbergen MJ, Tsang SKY, Hennink WE. Degradation and release behavior of dextran-based hydrogels. Macromolecules 1997;30:4639-4645.
22. Franssen O, Vandervennet L, Roders P, Hennink WE. Degradable dextran hydrogels: controlled release of a model protein from cylinders and microspheres. J. Control. Release 1999;60:211-221.
23. Van Tomme SR, De Geest BG, Braeckmans K, De Smedt SC, Siepmann F, Siepmann J, van Nostrum CF, Hennink WE. Mobility of model proteins in hydrogels composed of oppositely charged dextran microspheres studied by protein release and fluorescence recovery after photobleaching. J. Control. Release 2005;110:67-78.
24. Van Dijk-Wolthuis WNE, Tsang SKY, Kettenes-van den Bosch JJ, Hennink WE. A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer 1997;38:6235-6242.
25. De Geest BG, Déjugnat C, Sukhorukov GB, Braeckmans K, De Smedt SC, Demeester J. Self-rupturing microcapsules Adv. Mater. 2005;17:2357-2361.
26. Stubbe BG, Braeckmans K, Horkay F, Hennink WE, De Smedt SC, Demeester J. Swelling pressure observations on degrading dex-HEMA hydrogels. Macromolecules 2002;35:2501-2505.
27. Van de Wetering P, Zuidam NJ, van Steenbergen MJ, van der Houwen OAGJ, Underberg WJM, Hennink WE. A mechanistic study of the hydrolytic stability of poly(2-(dimethylamino)ethyl methacrylate). Macromolecules 1998;31:8063-8068.
Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres
87
6 Effect of particle size and charge on the network properties
of microsphere-based hydrogels
Sophie R. Van Tommea, Cornelus F. van Nostruma, Marjolein Dijkstrab, Stefaan C. De Smedtc, Wim E. Henninka
aDepartment of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS),
University Utrecht, Utrecht, The Netherlands bSoft Condensed Matter group, Department of Physics and Astronomy, Debye Institute for
NanoMaterials science, University Utrecht, Utrecht, The Netherlands cLaboratory of General Biochemistry and Physical Pharmacy, Department of Pharmaceutics,
Ghent University, Ghent, Belgium
Submitted for publication
Chapter 6
90
Abstract
This work describes the effect of the particle size (distribution) and charge on the network
properties and injectability of self-assembling hydrogels, based on ionic crosslinking between dextran microspheres. Copolymerization of hydroxyethyl methacrylate-derivatized dextran (dex-
HEMA), emulsified in an aqueous poly(ethylene glycol) (PEG) solution, with methacrylic acid (MAA) or dimethylaminoethyl methacrylate (DMAEMA) resulted in negatively and positively charged
microspheres, respectively, at physiological pH. The monomer/HEMA ratio was ranged between 6
and 57, resulting in microspheres with zeta (ξ)-potentials from -6 to -34 mV and +3 to +23 mV, for
the monomers MAA and DMAEMA, respectively. Microspheres with various sizes and size
distributions were obtained by sieving of polydisperse batches or by altering the emulsification
procedure. Rheological analysis showed that an increasing ξ-potential of the microspheres led to
stronger hydrogels. Relatively small microspheres (7 μm) with a narrow size distribution (99 %
smaller than 14 μm) gave rise to stronger hydrogels when compared to microspheres of 20 μm with a broad distribution (99 % smaller than 50 μm). When small microspheres were combined
with large microspheres of opposite charge, it was found that the strongest gels were obtained with 75 % small and 25 % large microspheres, instead of the ‘standard’ equal amounts (50/50) of
positively and negatively charged microspheres. Computer modeling confirmed these findings and showed that the most favorable composition, related to the lowest potential energy, comprised of 75 % small microspheres. Taking both charge and size effects into account, the
storage moduli (G’) of the almost fully elastic hydrogels could be tailored from 400 to 30000 Pa. Injectability tests showed that hydrogels composed of equal amounts of oppositely charged
microspheres (-6 and +6 mV, average particle size 7 μm) could be injected through 25G needles using a static load of 15 N, an ISO accepted value. In conclusion, a variety of options to control the
network properties of macroscopic hydrogels is provided, related to the charge and particle size of the composing dextran microspheres.
Keywords: particle size distribution, zeta potential, dextran, hydrogels, injectable
Effect of particle size and charge on the network properties of microsphere-based hydrogels
91
1. Introduction
Controlled delivery of pharmaceutically active proteins is an important topic in the advanced drug delivery field. The bioavailability of proteins after oral administration is low due to chemical and
physical degradation in the gastrointestinal (GI) tract. Parenteral delivery avoids the harsh conditions in the GI tract, but is associated with a lower patient comfort due to repeated
administration.[1] Numerous approaches have been evaluated to obtain a prolonged circulation time of the protein (e.g. by conjugation to poly(ethylene glycol) (PEG))[2] and to assure the
bioactivity of the therapeutic is preserved. The latter can be accomplished by entrapment of the protein into polymeric matrices. Various delivery vehicles were designed for the controlled release
of bioactive proteins covering among others nanoparticles[3-5] microspheres[6-9] and macroscopic hydrogels.[10-12] Special attention should be given to the structural changes of the proteins that might be induced during encapsulation in the delivery system.[1] Damage of these fragile
molecules and as a result, loss of their therapeutic activity, e.g. as a result of exposure to organic solvent or crosslinking agents, can be avoided when making use of self-assembling systems.[13]
In situ formation of hydrogels, mainly based on physical crosslinking between polymer chains can be accomplished through, amongst others, hydrogen bonding,[14, 15] crystallization,[16] hydrophobic
interactions as a result of temperature changes,[17] ionic interactions[18-20] or stereocomplexes.[21, 22] Besides in the protein delivery field, hydrogels have been widely used in tissue engineering applications.[23-27] Hydrogels do not only act as scaffolds, embedding cells and providing support
for new formed tissue, but they can also deliver growth factors and other signaling proteins at the right site in a sustained manner.[28]
A variety of natural and synthetic polymers has been used for the design of hydrogel matrices, of which the latter provide more control over chemical and physical properties. Dextran is an
attractive polymer for hydrogel formulations, meeting most of the requirements regarding biocompatibility and biodegradability.[29] Recently, we described a novel injectable self-gelling
system based on physical crosslinking between dextran-based microspheres, creating macroscopic hydrogels with tailorable properties regarding protein release and network properties.[18, 30] It was found that the oppositely charged microspheres showed a different
degradation behavior, providing various delivery options in the case a bioactive substance would be entrapped between as well as inside the individual microspheres.[31]
In this paper the effect of the charge of the dextran microspheres as well as their size and size distribution on the network properties of the macroscopic hydrogels was studied. Additionally and
importantly for applications, the injectability of these hydrogel systems was investigated.
2. Materials and methods
2.1. Materials
Dextran T40 (from Leuconostoc ssp.), N,N,N’,N’-tetramethylethylenediamine (TEMED) and 2-
hydroxyethyl methacrylate (HEMA) were obtained from Fluka (Buchs, Switzerland). Poly(ethylene
Chapter 6
92
glycol) (PEG) 10000 and potassium peroxodisulfate (KPS) were provided by Merck (Darmstadt,
Germany). N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes) was purchased from Acros Chimica (Geel, Belgium). Methacrylic acid (MAA) and N,N-dimethyl aminoethyl methacrylate (DMAEMA) were provided by Sigma-Aldrich (Zwijndrecht, The Netherlands).
2.2. Preparation of charged dex-HEMA microspheres
2.2.1. Preparation of microspheres with various charge densities
Dextran was derivatized with hydroxyethyl methacrylate (dex-HEMA) according to van Dijk-
Wolthuis et al.[32] The degree of substitution (DS, i.e. the number of HEMA groups per 100
glucopyranose units) used in this study was 10. Charged dex-HEMA microspheres were prepared in an all-aqueous environment as described previously with some minor modifications.[33, 34] Most
batches were prepared on a 50 g scale (total mixture), in 50 ml plastic vials. The emulsion was obtained by vortexing for 3 minutes. Charged monomers, methacrylic acid (MAA) or N,N-dimethyl
aminoethyl methacrylate (DMAEMA) were added to dex-HEMA prior to emulsification in PEG solution. In the standard procedure, the concentration of monomer (either MAA or DMAEMA) was
25 mM, corresponding to a monomer/HEMA ratio of 11. The ratio was varied from 6 to 57, using a monomer concentration of 12.5 mM to 125 mM, respectively. Finally, after 3 washing and
centrifugation steps with reversed osmosis water, the microspheres were lyophilized.
2.2.2. Preparation of microspheres with various particle size distributions
To obtain particles with a relatively narrow particle size distribution, large batches (500 g scale) of
negatively and positively charged microspheres (standard monomer/HEMA ratio of 11) were prepared.[18] The microsphere dispersions were consecutively sieved through a series of metal
sieves with decreasing pore size (50, 20, 15, 10 and 5 μm) by means of ultrasonic vibrations (using a wet sieving system that comprised of an Electronic Sieve Vibrator (EMS 755) and an Ultrasonic Processor (UDS 751), purchased from Topaz GmbH, Dresden, Germany). The final dispersions were
centrifuged at 4000 rpm for 1 h after which the obtained pellets and supernatants were freeze-dried. The microspheres were obtained from the pellet and will be further on referred to as ‘small’.
Charged microspheres with a larger average particle size and a broad size distribution compared to the standard particles were obtained by emulsification of the dextran/PEG mixture with a 3-
blade propeller stirrer at 60 rpm for 1 h prior to polymerization with KPS and TEMED. The microspheres with the standard monomer/HEMA ratio of 11 were prepared on a 75 g scale in a 125 ml beaker. Ultimately, after 3 washing and centrifugation steps, the microspheres were
lyophilized. These microspheres will be further on referred to as ‘large’.
Effect of particle size and charge on the network properties of microsphere-based hydrogels
93
2.3. Characterization of the charged microspheres
2.3.1. Determination of the particle size and particle size distribution
Particle size and particle size distribution of the various microsphere batches were obtained using an Optical Particle Sizer (Accusizer Model 780, Particle Sizing Systems, Santa Barbara, USA).
Calibration of the instrument was performed with latex beads (1-100 μm) (Duke Scientific Corporation). The lyophilized microspheres were suspended in reversed osmosis (r.o.) water prior
to the particle size measurement. Light microscopy images of hydrated microspheres were obtained using a Nikon eclipse TE2000-U (Nikon instruments Europe B.V., Badhoevedorp, The Netherlands).
2.3.2. ξ (zeta)-potential measurements
The ξ-potential of the microspheres was measured by laser Doppler electrophoresis with a
Zetasizer Nano-Z (Malvern Instruments Ltd., Worcestershire, United Kingdom) using a folded capillary cell (DTS 1060). Calibration of the instrument was performed with DTS 1050 latex beads
(zeta potential transfer standard, Malvern). The lyophilized microspheres were suspended in buffer (Hepes 5 mM pH 7.0 or ammonium acetate 5 mM pH 6.0) and homogenized thoroughly before the measurement.
2.4. Preparation of microsphere-based hydrogels
Hydrogels were obtained after hydration of equal weights of freeze-dried positively and negatively charged microspheres. Prior to addition of buffer (Hepes 100 mM pH 7.0) the positively and
negatively charged lyophilized microspheres were intensively mixed. The microspheres were
allowed to hydrate for 1 h at 4 °C before rheological analysis of the formed hydrogels was
performed. For various experiments (specified in the text) different ratios of positively/negatively
charged microspheres were used. Unless stated otherwise, the solid content of the gels was 15 % (w/w).
2.5. Rheological analysis
Rheological analysis of hydrogels was done using a controlled stress rheometer (AR1000-N, TA
Instruments, Etten-Leur, The Netherlands), equipped with an acrylic flat plate geometry (20 mm diameter) and a gap of 500 μm. Hydrogels (200 mg) were prepared as described in section 2.4 and
subsequently introduced between the two plates. A solvent trap was used to prevent evaporation of the solvent. The viscoelastic properties of the sample were determined by measuring the G’
(shear storage modulus) and G” (loss modulus) at 20 °C with a constant strain of 1 % and constant
frequency of 1 Hz. Creep experiments were performed to evaluate the extent of recovery of the material after deformation. In the creep experiments, a shear stress of 10 Pa was applied while the
strain was monitored.
Chapter 6
94
2.6. Computer modeling of the particle interactions
Computer modeling of the hydrogels was done. It was assumed that the microspheres interact with the Derjaguin-Landau-Verwey-Overbeek (DLVO) screened-Coulomb pair potential Vij(r):[35, 36]
r
e
aa
eZZ
Tk
rV r
ji
aa
BjiB
ijji κκ
κκλ
−+
++=
)1)(1(
)( )(
where r is the distance between two microspheres with respective radii ai and aj, and charges Zie
and Zje. Here the Bjerrum length is λB = e2/(4πεε0kBT), with ε the relative dielectric constant of the
solvent, ε0 the dielectric permittivity of vacuum, e the elementary charge, kB Boltzmann’s constant
and T the absolute temperature. The same contact value βε= |ZiZj|λB /((1+κ ai ) (1+κ aj )) for all pair
interactions was assumed, as the ξ-potentials for the anionic and cationic microspheres are almost
equal. The solid fraction was set at 20 % and the ratio anionic/cationic microspheres was varied.
Furthermore, the negatively charged microspheres were 5 μm in size, while the positively charged ones were 9 μm. Monte-Carlo simulations in the canonical ensemble were performed, i.e. the
number of both species, the volume of the box, and the temperature were fixed. A cubic box with linear dimension of 45 μm was used and periodic boundary conditions were applied. For more technical details on the simulations we refer to Frenkel et al.[37] Equilibration was checked by
monitoring the potential energy of the system U*, which is the sum of all pair interactions:
1 ( )
1 1
* ijN N V r
i j i
U υ
−
= = +
=∑∑
When equilibrium was reached, a production run of 1x105 sweeps was performed (one displacement attempt per particle), while sampling was performed once every sweep.
2.7. Injectability of the hydrogels
The injectability of the hydrogels was investigated using a compression device (Lloyd LR5K-plus)
fitted with a 100 Newton Load cell (Lloyd Instruments Ltd, Hampshire, United Kingdom). Dispersions or hydrogels (500 mg) were introduced into glass syringes (2 ml, internal diameter =
8.65 ± 0.2 mm) equipped with a fixed needle (25 G 5/8 inch), by means of a spatula. The samples were subjected to a static load (5-30 N) during 30 sec while the displacement of the plunger was
monitored. The syringes were weighted before and after each experiment to measure the amount of sample ‘ejected’. The data were processed with Nexygen Ondio software (Lloyd Instruments Ltd.).
Effect of particle size and charge on the network properties of microsphere-based hydrogels
95
3. Results and discussion
3.1. Preparation of dex-HEMA microspheres with various charges and size distributions
Using the water-in-water emulsion preparation procedure, positively and negatively charged dex-HEMA microspheres were obtained by copolymerization with DMAEMA and MAA, respectively,
with an equilibrium water content of 70 %. The size (distribution) and charge of the microspheres was varied. Table 1 gives an overview of the different batches and their characteristics.
Table 1: Characteristics of the various microsphere (dex-HEMA-MAA and dex-HEMA-DMAEMA) batches.
negatively charged microspheres (dex-HEMA-MAA)
MAA/HEMA
ratio (mol/mol)
emulsification
method
referred to as vol-wt* mean
diameter (μm)
99 % <
than (μm) ξ-potential
(mV)
6 vortex standard 12 22 -6
11 vortex standard 14 28 -8
23 vortex standard 11 20 -13
34 vortex standard 11 20 -20
45 vortex standard 5 12 -27
57 vortex standard 9 28 -34
11 turrax/sieving small 7 14 -7 11 stirrer large 16 50 -8
positively charged microspheres (dex-HEMA-DMAEMA)
DMAEMA/HEMA
ratio (mol/mol)
emulsification
method
referred to as vol-wt* mean
diameter (μm)
99 % <
than (μm) ξ-potential
(mV)
6 vortex standard 13 29 +3
11 vortex standard 13 27 +6
23 vortex standard 10 19 +11
34 vortex standard 4 15 +16
45 vortex standard 4 14 +23
57 vortex standard 9 18 +23
11 turrax/sieving small 7 13 +6 11 stirrer large 25 50 +8
* volume-weighted
3.1.1. Effect of the preparation method on the particle size distribution of dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres
Representative size distributions of ‘small’, ‘large’ and ‘standard’ microspheres, obtained via sieving,
stirring or vortexing, respectively, are depicted in Figure 1. Emulsification of the dex-HEMA/PEG mixture using a vortex led to polydisperse particles with an average volume-weighted diameter of 10 μm. 99 % of the microspheres were smaller than 20 μm.
For 3 batches prepared by vortexing, a significantly smaller particle size was observed (4-5 μm).
Chapter 6
96
To narrow down the particle size distribution, sieving of polydisperse particles was applied. Before
sieving, the average particle size was 10 μm (99 % smaller than 20 μm). Remarkably, during the sieving process most microspheres were able to pass all sieves, even with the smallest pore size (5 μm). This is possibly due to the high water content of the particles (70 %), which makes them
quite flexible and allows deformation to penetrate through the pores of the membranes. The final microsphere dispersion was centrifuged, which resulted in a pellet of microspheres with an
average size of 7 μm and a relatively narrow size distribution (99 % smaller than 13-14 μm) (Fig. 2A). Light microscopy images revealed that the supernatant mostly contained fragments of
microspheres. The microsphere yield after the sieving and centrifugation steps was ~35 %. Microspheres with a slightly larger average diameter (16-25 μm) and a broad size distribution (99 % smaller than 50 μm) (Fig. 2B) when compared to the ‘standard’ particles obtained by vortexing,
were prepared by creating the dex-HEMA/PEG emulsion at a lower speed with a 3-blade propeller. A size distribution with 2 populations was obtained, both smaller than 50 μm (Fig. 1).
0 5 10 15 20 25 30 35 40 45 500.0
0.5
1.0
1.5
2.0
2.5stirrervortexultraturrax + sieving
diameter (µm)
Volu
me
%
Figure 1: Particle size distribution (dex-HEMA-MAA microspheres, MAA/HEMA ratio of 11) for different
preparation methods (ultraturrax and sieving: light gray line; vortex: gray line; stirrer: black line).
Figure 2: Light microscopy images of (A) small dex-HEMA-MAA microspheres obtained after sieving of a
polydisperse batch, (B) large dex-HEMA-MAA microspheres with a broad size distribution, prepared at a low
speed with a 3-blade propeller.
10 μm 10 μm
A B
Effect of particle size and charge on the network properties of microsphere-based hydrogels
97
3.1.2. Effect of the monomer/HEMA ratio on the ξ-potential of charged dex-HEMA
microspheres
Figure 3 shows that with increasing the monomer/HEMA ratio, the surface charge (reflected by the
ξ-potential) of the resulting particles increased. This figure also shows that the ξ-potential
(absolute values) of the dex-HEMA-DMAEMA microspheres was slightly lower than the ξ-potential
of the dex-HEMA-MAA microspheres prepared at the same monomer/HEMA ratio. Since the pKa of pDMAEMA is ~7.5,[38] the DMAEMA units in the network are not fully protonated at pH 7, while
MAA (pKa ~ 4.7[39]) is fully deprotonated under the same conditions. An alternative explanation is that DMAEMA is less reactive than MAA, leading to less incorporation in the microspheres.
0 10 20 30 40 50 60-40
-30
-20
-10
0
10
20
30
40
ratio monomer/HEMA
zeta
pot
entia
l (m
V)
Figure 3: Effect of the monomer/HEMA ratio on the ξ-potential of dex-HEMA-MAA ( pH 7) and dex-HEMA-
DMAEMA ( pH 7; pH 6) microspheres ( , dex-HEMA microspheres , pH 7).
Higher ξ-potentials for the DMAEMA particles were found when measuring the particle charge at
pH 6 (Fig. 3) confirming the partial protonation at pH 7.0. Dex-HEMA microspheres, without
additional charged monomer showed a slightly negative ξ-potential (~ -1.3 mV, Fig. 3), likely
caused by adsorption of anions.
3.2. Effect of the particle size on the gel strength
Rheology was used to study the effect of the particle size and size distribution on the network
properties of microsphere-based hydrogels. Figure 4 shows the G’ for hydrogels consisting of equal amounts (w/w) of negatively and positively charged microspheres, but with various size
(small vs. large). The zeta potential was -7 mV for dex-HEMA-MAA and +7 mV for dex-HEMA-DMAEMA microspheres), while the solid content of the hydrogels was varied (15-25 %).
Chapter 6
98
15% 20% 25%0
5000
10000
15000
20000
25000
30000
35000
40000 small/smalllarge/smalllarge/large
solid content (w/w)
G' (
Pa)
Figure 4: Storage modulus (G’) of hydrogels consisting of equal amounts (w/w) of dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres as a function of the solid content of the hydrogels, for particles with different
sizes (small/small ( ), large/small ( ), large/large ( )) (tan(δ) of the different gels was < 0.1) (n= 3).
When large particles were used, no gel formation was observed at a solid content of 15 %. At the
same concentration of small microspheres, on the other hand, a hydrogel with a G’ of 5000 Pa was formed. A higher % solid did lead to gel formation for the large microspheres, but the gels were
substantially weaker than those consisting of small particles (G’= 1100 and 13500 Pa, respectively, for 20 % solid). A combination of small particles with larger particles led to hydrogels with
intermediate strength (e.g. G’= 8300 Pa for 20 % solid). Various weight ratios of small negatively charged and large positively charged microspheres were
combined and the gel properties were monitored (15 % solid) (Fig. 5). As a comparison, the gel strength (15 % solid) was investigated for various weight ratios of negatively and positively charged small microspheres.
0-/100+20-/80+
40-/60+50-/50+
60-/40+80-/20+
100-/0+0
500100015002000250030003500400045005000
ratio -/+ microspheres
G' (
Pa)
Figure 5: Storage modulus (G’) of dispersions and hydrogels (15 % solid (w/w)) consisting of ( ) small
negatively charged and large positively charged microspheres or ( ) only small microspheres (positively and
negatively charged) in various ratios (w/w) (n= 2). The ξ-potential of the particles was +/- 7 mV.
Effect of particle size and charge on the network properties of microsphere-based hydrogels
99
As previously shown,[18] only negatively charged or positively charged microspheres did not lead to
the formation of a gel (tan(δ) > 1). When a mixture of large microspheres (80 %) with a small
fraction of small microspheres (20 %) was used, no gel formation occurred. The opposite
combination, 80 % small negatively charged and 20 % large positively charged microspheres, led
to the formation of an almost fully elastic gel (G’ ~1600 Pa, tan(δ)= 0.1). Using both small negatively
charged microspheres and small positively charged microspheres, comparable network properties
were found for the ratios 20/80 and 80/20 anionic/cationic microspheres (G’ ~2000 Pa, tan(δ)
~0.15).
For the system described above, the batches of large microspheres had a high polydispersity (average volume-weighted (vol-wt) diameter 25 μm, 99 % smaller than 50 μm) and thus contain a
fraction of small particles. Figure 6 shows the G’ of dispersions and hydrogels consisting of microspheres with a less broad size distribution (dex-HEMA-MAA microspheres: 5 μm, 99 % smaller than 12 μm, -27 mV; dex-HEMA-DMAEMA microspheres: 9 μm, 99 % smaller than 18 μm, +23 mV).
No network formation (tan(δ) > 1) was observed when an excess of large particles was present
(75 % w/w) and above this concentration gels were formed of which the gel strength (G’)
increased with increasing fraction of small particles reaching a maximum at 75 % small and 25 %
large microspheres (G’ ~19000 Pa, tan(δ) ~0.06 ). At higher percentage of small microspheres, no
gel formation was observed.
25-/75+40-/60+
50-/50+60-/40+
75-/25+90-/10+
0500
1000150020002000
7000
12000
17000
ratio -/+ microspheres
G' (
Pa)
Figure 6: Storage modulus (G’) of dispersions and hydrogels consisting of ( ) small negatively charged and
large positively charged or ( ) large negatively charged and small positively charged microspheres in various
weight ratios.
In the case described above, the small particles were negatively charged while the large particles were positively charged. The reversed situation was also investigated (dex-HEMA-MAA
microspheres: 11 μm, 99 % smaller than 20 μm, -20 mV; dex-HEMA-DMAEMA microspheres: 4 μm, 99 % smaller than 15 μm, +16 mV). Figure 6 shows again that a dramatic increase of the gel strength was observed when the amount of small microspheres was increased compared to the
fraction of large ones. No hydrogel formation could occur when an abundance of large particles
was present in the mixture (75 %) as evidenced by tan(δ) > 1.
Chapter 6
100
The results presented in Figure 6 can be explained by the fact that one large microsphere will be
able to interact with many small microsphere. To obtain as many interactions as possible, the number of small microspheres needs to be higher than the number or amount of large microspheres (Fig. 7). For the combination dex-HEMA-MAA microspheres of 5 μm with dex-HEMA-
DMAEMA microspheres of 9 μm (depicted in Fig. 6), for the ratio where the gel had the highest strength (75-/25+), it can be calculated that the total number of small particles was 17 times higher
than the number of large particles. Further, the total area of the small microspheres with an average size of 5 μm was 5 times higher than that of the large particles with an average size of
9 μm. It can be anticipated that the large microspheres are able to utilize their surface more efficiently than the small microspheres. Figure 7 illustrates that the large microspheres can almost use their full surface to interact with the small particles while the small ones only bind to the large
ones at a few connection points.
Figure 7: Schematic presentation of a hydrogel consisting of large and small microspheres (A) and merely small
microspheres (B). The dark-colored and light-colored microspheres represent particles of opposite charge.
To fully comprehend the interactions between the negatively and positively charged microspheres of different size, computer modeling was carried out. For various ratios negative/positive
microspheres, the potential energy of the system was calculated. The potential energy is a measure for the strength of the gel and is calculated by summing over all microsphere-
interactions. Equally charged microspheres repel, while oppositely charged microspheres attract. The lowest potential energy corresponds to the most favorable composition, i.e. the best stabilized
system. Figure 8 shows the potential energy (U*) as a function of the ratio negatively/positively charged microspheres of 5 and 9 μm, respectively. The highest (less negative) U* was found in
those systems with an excess (≥ 90 %) of negatively or positively charged particles. As shown in
Figure 6, no gel formation occurs in these cases. The most favorable composition, related to the lowest U* (most negative) showed to be the system containing ~75 % of negatively charged
microspheres. Keeping in mind that the anionic microspheres are ‘small’ (5 μm), compared to the cationic microspheres (9 μm), these findings fully correlate with the rheology data reported in
Figure 6.
A B
Effect of particle size and charge on the network properties of microsphere-based hydrogels
101
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0-7
-6
-5
-4
-3
-2
-1
0
fraction negatively charged microspheres
U*
Figure 8: Calculated potential energy (U*) as a function of the fraction of negatively charged small
microspheres.
3.3. Effect of the microspheres charge on the gel strength
Dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres were prepared with varying
monomer/HEMA ratios yielding particles with different ξ-potentials (Fig. 3). Increasing the ratio
monomer/HEMA in the initial dextran/PEG mixture from 6 to 23 led to a significant increase in G’ of
the corresponding hydrogels from 350 to 6000 Pa (Fig. 9). The tan(δ) was in all cases below 0.1,
confirming the almost fully elastic properties of the hydrogels. Strikingly, a further increase of the monomer/HEMA ratio, led to a decrease of the G’ to 1400 Pa. For the highest charge of the
particles a slight increase in gel strength to 4000 Pa was again observed.
0 10 20 30 40 50 600
100200300400500
10002000300040005000600070008000
0.00.10.20.30.40.50.60.70.80.91.0
monomer/HEMA ratio
G' (
Pa)
tan(
δ )
Figure 9: Storage modulus (G’) ( ) and tan(δ) ( ) of hydrogels (15 % solid) consisting of equal amounts (w/w)
of dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres (n=3). The microspheres differed in the
monomer/HEMA ratio used for the preparation.
Chapter 6
102
Table 1 shows that the ξ-potential (absolute value) of the dex-HEMA-DMAEMA microspheres is in
most cases less than that of the dex-HEMA-MAA microspheres prepared at the same
monomer/HEMA ratio. Therefore, hydrogels were also prepared by combining negatively charged
and positively charged microspheres with similar absolute ξ-potentials, instead of with equal
monomer/HEMA ratios. Figure 10 shows that the combination of dex-HEMA-MAA microspheres of
-6 mV with dex-HEMA-DMAEMA microspheres of +6 mV led to much stronger gels (G’= 2000 Pa,
tan(δ)= 0.04) than the combination with dex-HEMA-DMAEMA microspheres of +3 mV (G’= 350 Pa,
tan(δ)= 0.08). Although not as pronounced, this phenomenon was also seen with dex-HEMA-MAA
microspheres of -20 mV combined with dex-HEMA-DMAEMA microspheres of 23 mV (G’= 2250 Pa,
tan(δ)= 0.16) instead of dex-HEMA-DMAEMA microspheres of +16 mV (G’= 1700 Pa, tan(δ)= 0.13).
These observations however, do not explain the drop in G’ when the monomer/HEMA ratio increased from 23 to 34. As already mentioned in section 3.2, the difference in particle size between the positively and negatively charged microspheres was quite large for the microsphere
batches with monomer/HEMA ratios of 34 and 45, and therefore, as discussed above, the 50/50 weight ratio used is not optimal to obtain the highest gel strength in those cases. Instead of the
50/50 dex-HEMA-MAA/dex-HEMA-DMAEMA microsphere combinations (Fig. 9), the ‘optimal’ ratios are plotted (Fig. 11). The characteristics of the various combinations are listed in Table 2. Compared
to Figure 9, Figure 11 shows that in line with expectations the G’ increased with increasing ξ-
potential of the microspheres. Figure 11 illustrates that, when taking the particle size and ξ-
potential into account, hydrogels can be designed with tailorable strength for the aimed
application.
-6 mV +3 mV-6 mv + 6 mV
-20 mV + 16 mV
-20 mV + 23 mV0
500
1000
1500
2000
2500
3000
G´(
Pa)
Figure 10: Storage modulus (G’) of several hydrogels (15 % solid) consisting of equal amounts (w/w) of
positively and negatively charged microspheres. Various ξ-potential combinations were investigated (n=2).
Effect of particle size and charge on the network properties of microsphere-based hydrogels
103
Table 2: Various combinations of dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres and their specific characteristics.
monomer/HEMA ratio ξ-potential (mV) vol-wt mean diameter (μm) combination #
- + - + - + ratio -/+
1 6 11 -6 +6 12 13 50/50
2 11 11 -8 +6 14 13 50/50
3 23 23 -13 +11 11 10 50/50
4 34 34 -20 +16 11 4 25/75
5 45 57 -27 +23 5 9 75/25
1 2 3 4 50
2500
5000
7500
10000
12500
15000
17500
20000
combination #
G' (
Pa)
Figure 11: Storage modulus of hydrogels (15 % solid) consisting of various combinations of dex-HEMA-MAA
and dex-HEMA-DMAEMA microspheres (n=3, except for # 4 where n=1). The combination # refers to Table 2.
3.4. Injectability of the hydrogels
One important application that can be foreseen for the hydrogels composed of oppositely
charged dextran microspheres is the use as delivery system for pharmaceutical proteins. Injectability of the delivery systems is preferred, thereby avoiding surgery. The injectability of the
microsphere-based hydrogels was evaluated during compression tests in which the hydrogels were loaded into glass syringes (internal diameter = 8.65 ± 0.2 mm), equipped with a fixed 25G
(5/8 inch) needle. Such needles are typically used for subcutaneous injection, which is the aimed route of administration for these hydrogel systems. It was shown that the rubber plunger had a low friction (1 N) with the syringe wall.
Next, the injection force required for a dispersion of non-charged dex-HEMA microspheres (15 % solid w/w) was studied. A static load of 5 N was applied during 30 sec while the displacement of
the plunger (= extension) was monitored (Fig. 12A). The dispersion showed to be easily and reproducibly ‘ejectable’ at a flow of ~ 135 mg/min, under these conditions, as expected. The
Chapter 6
104
behavior of hydrogels (15 % solid w/w) composed of equal amounts of dex-HEMA-MAA and dex-
HEMA-DMAEMA microspheres (‘standard’ monomer/HEMA ratio of 11) was investigated at static load of 30 N. Figure 12B shows that approximately 15 N was needed to obtain a constant flow, which is regarded acceptable.[40] Compared to the dex-HEMA microsphere-dispersion, the flow
rate was smaller, ~70 mg/min, for a higher static load (15 N vs. 5 N for dex-HEMA microspheres). It was shown in section 3.3 that stronger hydrogels were obtained when the monomer/HEMA
ratio of the microspheres was increased from 11 to 23. Figure 13 shows the differences in displacement of the plunger during 3 consecutive measurements for hydrogels composed of
microspheres with a monomer/HEMA ratio of 11 (corresponding to ξ-potentials of -7 and +6 mV)
(A) and a monomer/HEMA ratio of 23 (corresponding to ξ-potentials of -13 and +11 mV) (B).
A static load of 15 N was enough to introduce flow in the systems independent of the microsphere
charge, but the injectability of the hydrogels composed of particles with the highest charge was hampered as evidenced by the decrease in flow speed from ~60 mg/min to ~10 mg/min during 3 consecutive experiments.
0 5 10 15 20 25 30 35 400
1
2
3
4
5
6
0.0
0.5
1.0
1.5
time (s)
load
(N)
exte
nsio
n (m
m)
0 5 10 15 20 25 3005
101520253035
0.0
0.2
0.4
0.6
0.8
1.0
time (s)
load
(N)
exte
nsio
n (m
m)
Figure 12: Extension ( ) of (A) a dex-HEMA microsphere-dispersion (15 % solid (w/w)) and (B) a hydrogel
composed of dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres (15 % solid (w/w), monomer/HEMA ratio
11) during an injectability test (static load (—)).
0 5 10 15 20 25 300.0
0.2
0.4
0.6
0.8
1.0
123
time (s)
exte
nsio
n (m
m)
0 10 20 30 400.00.20.40.60.81.01.21.4 1
23
time (s)
exte
nsio
n (m
m)
Figure 13: Displacement of the plunger (extension) during 3 consecutive compression test of (A) dex-HEMA-MAA/dex-HEMA-DMAEMA hydrogel (monomer/HEMA ratio 11, 15 % solid (w/w)) and (B) dex-HEMA-MAA/dex-
HEMA-DMAEMA hydrogel (monomer/HEMA ratio 23, 15 % solid (w/w)).
A B
A B
Effect of particle size and charge on the network properties of microsphere-based hydrogels
105
A difference in flow behavior was observed between the two differently charged hydrogels
described above. While the ‘standard’ hydrogel (ξ-potentials ~6 mV, absolute value) showed the
formation of homogenous gel-threads during ejection, the hydrogel with microspheres with a
higher ξ-potential (~12 mV, absolute value) was ejected as a discontinuous fiber with water
droplets daggling along. The injectability of hydrogels composed of microspheres with an even
higher ξ-potential (~25 mV) was not possible under the tested conditions. Initially, water was
pushed out of the hydrogel, leading to a denser packing of the microspheres inside the syringe. Eventually, a cake of microspheres remained. Apparently, a higher charge density on the
microspheres results in expulsion of water from the hydrogel upon compression and can be explained by the increased inter-particle interactions. These experiments show that it is possible to
administer hydrogels based on dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres, using needles suitable for subcutaneous injection, provided that the particle interactions are not too strong.
4. Conclusions
In this work, the network properties of self-assembling hydrogels based on ionic interactions
between dextran microspheres were investigated. Dextran hydrogels have shown to possess favorable properties such as biocompatibility and biodegradability. The potential of the hydrogels described in this paper as protein delivery matrices has been studied previously. In the current
study a number of options to control their network properties is provided. Both the surface charge of the microspheres as well as their size and size distribution was varied by adjusting the charged
monomer/HEMA ratio and several parameters in the microsphere preparation process. The gel strength showed to be tailorable from 400 to 30000 Pa, making them suitable for a range of drug
delivery and tissue engineering applications in which diverse mechanical properties are required. The hydrogels showed to be injectable using a force of 15 N, which is accepted by ISO standards.
Acknowledgement
Jan Rikken and Jean-Paul Kleeven from N.V. Organon (Oss, The Netherlands) are gratefully acknowledged for help with the injectability tests. Wouter Bult (Department of Nuclear Medicine,
University Medical Center, Utrecht, The Netherlands) is gratefully acknowledged for assistance with the sieving procedure. This research was financially supported by SenterNovem (IS042016).
Chapter 6
106
References
1. Jorgensen L, Moeller EH, van de Weert M, Nielsen HM, Frokjaer S. Preparing and evaluating delivery systems for proteins. Eur. J. Pharm. Sci. 2006;29:174-182.
2. Bailon P, Berthold W. Polyethylene glycol-conjugated pharmaceutical proteins. Pharm. Sci. Technol. To. 1998;1:352-356.
3. Panyam J, Labhasetwar V. Biodegradable nanoparticles for drug and gene delivery to cells and tissue. Adv. Drug Deliver. Rev. 2002;55:329-347.
4. Bilati U, Allémann E, Doelker E. Strategic approaches for overcoming peptide and protein instability within biodegradable nano-and microparticles. Eur. J. Pharm. Biopharm. 2005;59:375-388.
5. Chung Y-I, Tae G, Yuk SH. A facile method to prepare heparin-functionalized nanoparticles for controlled release of growth factors. Biomaterials 2006;27:2621-2626.
6. Cadée JA, Brouwer LA, den Otter W, Hennink WE, van Luyn MJA. A comparative biocompatibility study of microspheres based on crosslinked dextran or poly(lactic-co-glycolic) acid after subcutaneous injection in rats. J. Biomed. Mater. Res. 2001;56:600-609.
7. Sinha VR, Trehan A. Biodegradable microspheres for protein delivery. J. Control. Release 2003;90:261-280.
8. Jiang W, Gupta RK, Deshpande MC, Schwendeman SP. Biodegradable poly(lactic-co-glycolic acid) microparticles for injectable delivery of vaccine antigens. Adv. Drug Deliver. Rev. 2005;57:391-410.
9. Kim HK, Chung HJ, Park TG. Biodegradable polymeric microspheres with open/closed pores for sustained release of human growth hormone. J. Control. Release 2006;112:167-174.
10. Peppas NA, Bures P, Leobandung W, Ichikawa H. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 2000;50:27-46.
11. Gupta P, Vermani K, Garg S. Hydrogels: from controlled release to pH-responsive drug delivery. Drug Discov. Today 2002;7:569-579.
12. Peppas NA, Hilt JZ, Khademhosseini A, Langer R. Hydrogels In biology and medicine: from molecular principles to biotechnology. Adv. Mater. 2006;18:1345-1360.
13. Hennink WE, van Nostrum CF. Novel crosslinking methods to design hydrogels. Adv. Drug Deliver. Rev. 2002;54:13-36.
14. Khutoryanskiy VV. Hydrogen-bonded interpolymer complexes as materials for pharmaceutical applications. Int. J. Pharm. 2007;334:15-26.
15. Mahler A, Reches M, Rechter M, Cohen S, Gazit E. Rigid, self-assembled hydrogel composed of a modified aromatic dipeptide. Adv. Mater. 2006;18:1365-1370.
16. Stenekes RJH, Talsma H, Hennink WE. Formation of dextran hydrogels by crystallization. Biomaterials 2001;22:1891-1898.
17. Ruel-Gariepy E, Leroux J-C. In situ-forming hydrogels--review of temperature-sensitive systems. Eur. J. Pharm. Biopharm. 2004;58:409-426.
18. Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Self-gelling hydrogels based on oppositely charged dextran microspheres. Biomaterials 2005;26:2129-2135.
19. Berger J, Reist M, Mayer JM, Felt O, Peppas NA, Gurny R. Structure and interactions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications. Eur. J. Pharm. Biopharm. 2004;57:19-34.
20. Kuo CK, Ma PX. Ionically crosslinked alginate hydrogels as scaffolds for tissue engineering: Part 1. Structure, gelation rate and mechanical properties. Biomaterials 2001;22:511-521.
21. Hiemstra C, Zhong ZY, Van Tomme SR, Hennink WE, Dijkstra PJ, Feijen J. Protein release from injectable stereocomplexed hydrogels based on PEG-PDLA and PEG-PLLA star block copolymers. J. Control. Release 2006;116:e19-e21.
22. De Jong SJ, De Smedt SC, Wahls MWC, Demeester J, Kettenes-van den Bosch JJ, Hennink WE. Novel self-assembled hydrogels by stereocomplex formation in aqueous solution of enantiomeric lactic acid oligomers grafted to dextran. Macromolecules 2000;33:3680-3686.
23. Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chem. Rev. 2001;101:1869-1879.
24. Drury JL, Mooney DJ. Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials 2003;24:4337-4351.
25. Mahoney MJ, Anseth KS. Three-dimensional growth and function of neural tissue in degradable polyethylene glycol hydrogels. Biomaterials 2006;27:2265-2274.
26. Lévesque SG, Shoichet MS. Synthesis of cell-adhesive dextran hydrogels and macroporous scaffolds. Biomaterials 2006;27:5277-5285.
27. Brandl F, Sommer F, Goepferich A. Rational design of hydrogels for tissue engineering: Impact of physical factors on cell behavior. Biomaterials 2007;28:134-146.
Effect of particle size and charge on the network properties of microsphere-based hydrogels
107
28. Baldwin SP, Mark Saltzman W. Materials for protein delivery in tissue engineering. Adv. Drug Deliver. Rev. 1998;33:71-86.
29. Van Tomme SR, Hennink WE. Biodegradable dextran hydrogels for protein delivery applications. Expert Rev. Med. Devices 2007;4:147-164.
30. Van Tomme SR, De Geest BG, Braeckmans K, De Smedt SC, Siepmann F, Siepmann J, van Nostrum CF, Hennink WE. Mobility of model proteins in hydrogels composed of oppositely charged dextran microspheres studied by protein release and fluorescence recovery after photobleaching. J. Control. Release 2005;110:67-78.
31. Van Tomme SR, van Nostrum CF, De Smedt SC, Hennink WE. Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres. Biomaterials 2006;27:4141-4148.
32. Van Dijk-Wolthuis WNE, Tsang SKY, Kettenes-van den Bosch JJ, Hennink WE. A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer 1997;38:6235-6242.
33. Stenekes RJH, Franssen O, van Bommel EMG, Crommelin DJA, Hennink WE. The use of aqueous PEG/dextran phase separation for the preparation of dextran microspheres. Int. J. Pharm. 1999;183:29-32.
34. Franssen O, Hennink WE. A novel preparation method for polymeric microparticles without the use of organic solvents. Int. J. Pharm. 1998;168:1-7.
35. Verwey EJW, Overbeek JTG. Theory of the stability of lyotropic colloids. Elsevier: New York, 1948.
36. Derjaguin BV, Landau L. Theory of the stability of strongly charged lyophobic sols and of the adhesion of strongly charged particles in solution of electrolytes. Acta Physicochimica URSS 1941;14:633-662.
37. Frenkel D, Smit B. Understanding molecular simulation : from algorithms to applications 2nd ed.; Academic Press: London, 2002.
38. Van de Wetering P, Zuidam NJ, van Steenbergen MJ, van der Houwen OAGJ, Underberg WJM, Hennink WE. A mechanistic study of the hydrolytic stability of poly(2-(dimethylamino)ethyl methacrylate). Macromolecules 1998;31:8063-8068.
39. Grant DH, McPhee VA. Determination of methacrylic acid by coulometric titration Anal. Chem. 1976;48:1820-1820.
40. Sterile single-use syringes, with or without needle, for insulin, ISO 8537 (E), 1991-05-01.
7 Tuning the degradation behavior of ionically crosslinked
dextran hydrogels by variation of the cationic groups
Sophie R. Van Tomme, Najim El Morabit, Cornelus. F. van Nostrum, Wim. E. Hennink
Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS),
University Utrecht, Utrecht, The Netherlands
Chapter 7
110
Abstract
In this work, the degradation behavior of cationic microspheres that act as building blocks of
macroscopic hydrogels is reported. In the previous chapters of this thesis, dimethylaminoethyl methacrylate (DMAEMA) was copolymerized with hydroxyethyl methacrylate (HEMA)-derivatized
dextran, resulting in positively charged microspheres, which, in combination with anionic particles (containing methacrylic acid (MAA)), form macroscopic gels via ionic interactions. In the current study, hydroxypropyl methacrylamide (HPMA)-based cationic monomers, i.e. HPMA-DMAE
(dimethylaminoethanol) and HPMA-DMAPr (dimethylaminopropanol) were used to prepare positively charged microspheres. Both monomers contain a hydrolysis-sensitive carbonate ester
leading to a gradual charge loss in time. It was shown that upon incubation at pH 7.0 at 37 °C the
ξ-potential of microspheres containing DMAEMA remained constant for over 1 week, while the
microspheres containing HPMA-DMAE and HPMA-DMAPr lost their positive ξ-potential in 1 and
2 days, respectively. Rheological analysis of hydrogels consisting of equal amounts of positively
and negatively charged microspheres revealed that due to charge loss of the cationic microspheres an accelerated disintegration of the macroscopic hydrogels occurred. The hydrogels
containing DMAEMA or either HPMA-DMAE or HPMA-DMAPr degraded in 10 and 30 days, respectively, as evidenced by a complete loss of gel strength. This introductory study shows that
the degradation behavior of the ionically crosslinked microsphere-based hydrogels can be tailored by alteration of the cationic groups of the positively charged dextran microspheres.
Keywords: dextran microspheres, cationic monomers, degradation behavior
Tuning the degradation behavior of ionically crosslinked dextran hydrogels by variation of the cationic groups
111
1. Introduction
One of the major assets of hydrogels, especially in pharmaceutical and biomedical applications, is
the possibility to tailor the network properties and degradation behavior.[1-4] In Chapters 3 to 6 we reported on a novel self-assembling hydrogel system based on oppositely charged
microspheres.[5-8] Modified dextran was copolymerized with dimethylaminoethyl methacrylate (DMAEMA) or methacrylic acid (MAA) to yield positively and negatively charged microspheres,
respectively. Preliminary in vitro cytotoxicity tests revealed that the positively charged particles induced a decreased cell viability when compared to neutral, negatively charged or a combination of negatively and positively charged microspheres (Chapter 9 of this thesis). Whether this is due to
the positive charge of the particles, which can interact with negatively charged cell membranes and could cause membrane disruption and cell death, or due to the presence of soluble
pDMAEMA fragments, that have shown to be slightly cytotoxic,[9] is unclear, thus far. In the current study, positively charged microspheres were prepared using two hydroxypropyl methacrylamide
(HPMA)-based cationic monomers, i.e. HPMA-DMAE (dimethylaminoethanol) and HPMA-DMAPr (dimethylaminopropanol). A reduced toxicity of the microspheres can be foreseen, since the monomers contain a hydrolysis-sensitive carbonate ester with different degradation rates. The
influence of the monomer type and degradation rate on the ξ-potential of the cationic
microspheres was investigated. Further, the network properties of the hydrogels were monitored
in time, to assess the influence of charge loss of the positively charged microspheres on the macroscopic network.
2. Materials and Methods
2.1. Materials
Dextran T40 (from Leuconostoc ssp.), N,N,N’,N’-tetramethylethylenediamine (TEMED) and 2-hydroxyethyl methacrylate (HEMA) were obtained from Fluka (Buchs, Switzerland). Poly(ethylene
glycol) (PEG) 10,000 and potassium peroxodisulfate (KPS) were provided by Merck (Darmstadt, Germany). N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid (Hepes) was purchased from Acros Chimica (Geel, Belgium). Methacrylic acid (MAA), dimethylaminoethyl methacrylate
(DMAEMA), N,N-dimethylaminoethanol (DMAE), N,N-dimethylaminopropanol (DMAPr) and hydroquinone monomethyl ether were provided by Sigma-Aldrich (Zwijndrecht, The Netherlands).
2.2. Synthesis of cationic monomers
N-(2-hydroxypropyl)- methacrylamide (HPMA) was synthesized based on the method described by Oupicky et al.[10] HPMA-DMAE and HPMA-DMAPr were synthesized as previously reported by Luten et al.[11]
Chapter 7
112
2.3. Preparation of charged dex-HEMA microspheres
Hydroxyethyl methacrylate-derivatized dextran was obtained as described by van Dijk-Wolthuis et al.[12] The degree of substitution (DS, i.e. number of HEMA groups per 100 glucopyranose units)
used in this study was 10. Dex-HEMA microspheres were prepared in an all-aqueous environment according to Franssen et al.[13] with some minor modifications. The microspheres were prepared on
a 50 g scale (total amount of mixture) and the emulsion was obtained by vortexing for 3 min. The preparation of charged dex-HEMA microspheres has also been previously reported.[5] As ‘standard’
monomers, MAA and DMAEMA were used to obtain negatively and positively charged particles, respectively. In this study, additionally, HPMA-DMAE and HPMA-DMAPr were used as cationic monomers. The various monomers were added to the dextran phase prior to emulsification in PEG
after which radical polymerization resulted in charged dex-HEMA microspheres. Ammonium acetate buffer pH 5.0 (100 mM) was used instead of Hepes buffer pH 7.0 (100 mM) to inhibit
degradation of the hydrolysis-sensitive HPMA-based monomers. The monomer/HEMA ratio was
varied (from 1 to 16) to obtain particles with a broad range of zeta (ξ)-potentials. After extensive
washing, the microspheres were lyophilized.
2.4. Characterization of the charged microspheres
Particle size and particle size distribution of the microspheres were obtained using an Optical Particle Sizer (Accusizer Model 780, Particle Sizing Systems, Santa Barbara, USA). Calibration of the
instrument was performed with latex beads (1-100 μm) (Duke Scientific Corporation). Microspheres, collected during the final centrifugation step, were suspended in reversed osmosis water prior to the particle size measurement.
The ξ-potential of the microspheres was measured by laser Doppler electrophoresis with a
Zetasizer Nano-Z (Malvern Instruments Ltd, Worcestershire, United Kingdom) using a folded
capillary cell (DTS 1060). Calibration of the instrument was performed with DTS 1050 latex beads (zeta potential transfer standard, Malvern). Microspheres, collected during the final centrifugation step, were suspended in buffer (Hepes 5 mM pH 7.0 or ammonium acetate 5 mM pH 5.0) and
homogenized thoroughly before the measurement.
2.5. ξ-potential measurements of degrading microspheres
Microspheres (50 mg) were dispersed in 1 ml buffer (Hepes, 100 mM, pH 7.0 or ammonium
acetate, 100 mM, pH 5.0) and incubated at 37 °C. At regular time intervals, buffer was added to a
total volume of 10 ml and the vials were centrifuged (5 min 3500 rpm). The supernatant was
discarded and the washing procedure was repeated thrice. Finally, 1 ml buffer was added and the
vials were incubated at 37 °C. One drop of the microsphere suspension was added to 2 ml Hepes 5
mM pH 7.0 or ammonium acetate 5 mM pH 5.0 buffer after which ξ-potential measurements were
performed as described in section 2.4.
Tuning the degradation behavior of ionically crosslinked dextran hydrogels by variation of the cationic groups
113
2.6. Preparation of self-assembling hydrogels
Hydrogels (300 mg, solid content 20 % (w/w)) were obtained after hydration of equal amounts of positively and negatively charged microspheres. Prior to the addition of buffer (Hepes, 100 mM, pH
7.0) the oppositely charged lyophilized microspheres were intensively mixed. The hydrogels were
allowed to hydrate for 1 h at 4 °C. To monitor their degradation behavior, the hydrogels were
incubated at 37 °C and at regular time intervals rheological analysis was performed. For each time
point, 3 identical gels were prepared.
2.7. Rheological analysis
Rheology measurements on hydrogels were performed using a controlled stress rheometer (AR1000-N, TA Instruments, Etten-Leur, The Netherlands), equipped with an acrylic flat plate
geometry (20 mm diameter) and a gap of 500 μm. Hydrogels were prepared as described in section 2.6 and thereupon introduced between the two plates. A solvent trap was used to prevent
evaporation of the solvent. The viscoelastic properties of the sample were determined by
measuring the G’ (shear storage modulus) and G” (loss modulus) at 20 °C during 20 min with a
constant strain of 1 % and constant frequency of 1 Hz.
3. Results and Discussion
3.1. Preparation of positively charged microspheres using different cationic monomers
Charged dex-HEMA microspheres with an equilibrium water content of 70 % were obtained making use of the PEG/dextran phase separation. It has been shown in Chapter 6 of this thesis that
by ranging the monomer/HEMA ratio between 6 and 57, the ξ-potential could be tailored
between -34 (with MAA) to +23 mV (with DMAEMA).[8] In this study, positively charged microspheres (average volume-weighted mean particle diameter of 10 μm) were also prepared by
addition of 2 other cationic monomers, HPMA-DMAE (pKa ∼ 9.9[11]) and HPMA-DMAPr (pKa ∼ 8.1[11])
(Fig. 1). Table 1 gives an overview of some representative microsphere types and their
corresponding ξ-potentials.
Table 1: Overview of the ξ-potentials of positive microspheres containing HPMA-DMAE or HPMA-DMAPr.
monomer monomer/HEMA ratio (mol/mol) ξ-potential (mV)
HPMA-DMAE 1 + 5
4 + 8
16 + 18
HPMA-DMAPr 1 + 5
4 + 8 16 + 16
Chapter 7
114
The HPMA-based monomers seem to have a higher affinity for the dextran phase than DMAEMA,
since for the latter a monomer/HEMA ratio of 34 was needed to obtain a ξ-potential of + 16 mV,[8]
whereas a (HPMA-DMAE or HPMA-DMAPr)/HEMA ratio of 16 was sufficient. Additionally, since the
pKa of both HPMA-based monomers is higher than the pKa of DMAEMA (pKa ∼ 7.5[14]) more
groups will be protonated at pH 7.0.
NO
O
OHN
O N O
O
OHN
O
Figure 1: Chemical structures of HPMA-DMAE (A) and HPMA-DMAPr (B).
3.2. Degradation behavior of cationic microspheres
In Chapter 5, it has been reported that cationic dex-HEMA microspheres containing DMAEMA
degraded four times faster than their negatively charged counterparts (dex-HEMA-MAA).[7] This was attributed to the stabilization of the transition state in the hydrolysis process by the
protonated tertiary amine groups present in the cationic microspheres. On the other hand, the presence of negatively charged groups causes repulsion of hydroxyl anions resulting in a slower
degradation of the dex-HEMA-MAA microspheres. Degradation of dex-HEMA microspheres is caused by hydrolysis of the carbonate esters between the dextran backbone and HEMA side
chains. Using HPMA-DMAE or HPMA-DMAPr as cationic monomers, additional carbonate esters were introduced. The degradation behavior of these monomers has been reported by Luten
et al.[11] At 37 °C and pH 7.4 the half-life of the monomers is 11 and 79 h, for HPMA-DMAE and
HPMA-DMAPr, respectively. The influence of the hydrolysis of the carbonate ester of the
monomers, when copolymerized in microspheres, on the ξ-potential of the particles was
investigated. During 20 days of incubation at pH 7.0 the ξ-potentials of the dex-HEMA and dex-
HEMA-MAA microspheres remainded constant (-1.5 mV and -10 mV, respectively), which indicates that no significant degradation occurred. This is in agreement with the results of the swelling experiments reported in Chapter 5 that showed that the neutral and anionic microspheres (with
comparable ξ-potentials as in the current study) degraded in 55 and 125 days, respectively.
At pH 5 no degradation of dex-HEMA-DMAEMA microspheres was observed, since carbonate
esters are most stable at this pH (Fig. 2A). Contrary, at pH 7, the ξ-potential of dex-HEMA-DMAEMA
microspheres gradually dropped in time and at day 14 the positive charge had completely disappeared. This gradual charge loss is due to hydrolysis of the particles, as mentioned above
(Fig. 2B).[7]
A B
Tuning the degradation behavior of ionically crosslinked dextran hydrogels by variation of the cationic groups
115
0 2 4 6 8 10 12 14-5
0
5
10
15
20
25
time (days)
ξ−po
tent
ial (
mV
)
0 5 10 15 20 25 300
5
10
15
20
25
time (days)
ξ−po
tent
ial (
mV
)
Figure 2: ξ-potential of dex-HEMA-DMAEMA microspheres in ammonium acetate buffer 5 mM pH 5.0 (A) or
Hepes buffer 5 mM pH 7.0 (B) in time during incubation at 37 °C (n=3).
0.5 1.0 1.5 2.0 2.50
4
8
12
16
20
24
28
time (days)
ξ -po
tent
ial (
mV)
Figure 3: ξ-potential of dex-HEMA-HPMA-DMAE ( ) and dex-HEMA-HPMA-DMAPr (—) microspheres in
Hepes buffer 5 mM pH 7.0 in time during incubation at 37 °C (n=3).
Figure 3 shows the ξ-potentials of dex-HEMA-HPMA-DMAE and dex-HEMA-HPMA-DMAPr
microspheres in time at pH 7 and 37 °C. The microspheres containing HPMA-DMAE lost their
charge completely in 1 day, while it took 2 days for the microspheres containing HPMA-DMAPr to
become neutral. This is in line with previous findings that HPMA-DMAE is hydrolyzed faster than HPMA-DMAPr.[11] However, the difference in degradation times, once copolymerized with dex-
HEMA was less pronounced.
3.3. Influence of the degradation of the cationic microspheres on the network properties of the macroscopic hydrogels
The network properties of the microsphere-based hydrogels were studied during rheology experiments. Figure 4 shows the storage modulus (G’) of hydrogels (20 % solid) consisting of equal amounts of positively and negatively charged microspheres, differing in cationic monomer and
charge density. For the microspheres containing DMAEMA as positively charged group, the gel strength and elasticity were comparable to previously reported results (Chapter 6).[8] A clear
increase of the G’ (from 3100 to 8800 Pa) was observed when the microsphere charge was
A B
Chapter 7
116
doubled. Hydrogels in which in the cationic particles HPMA-DMAE or HPMA-DMAPr were
introduced, showed the same profile, although the G’ values were slightly lower, with comparable
elasticity (tan(δ) ~0.15).
DMAEMA 8 mV
DMAEMA 16 mV
HPMA-DMAE 8 mV
HPMA-DMAE 16 mV
HPMA-DMAPr 8 mV
HPMA-DMAPr 16 mV0
100020003000400050006000700080009000
1000011000
G' (
Pa)
Figure 4: Storage modulus (G’) of hydrogels consisting of equal amounts of positively and negatively charged
microspheres, differing in cationic monomer and charge density (n=3).
It was shown in section 3.2 that the cationic microspheres containing HPMA-DMAE or HPMA-
DMAPr lost their charge after 1 and 2 days, respectively, at pH 7.0 at 37 °C, whereas those
containing DMAEMA started to loose their charge after 10 days. The influence of the decrease in ξ-
potential of the cationic microspheres on the network properties of the self-assembling macroscopic hydrogels in time was investigated by means of rheological analysis. Figure 5 shows
the G’ of hydrogels composed of equal amounts (w/w) of positively and negatively charged microspheres. The positively charged microspheres contained DMAEMA, HPMA-DMAE or HPMA-
DMAPr. The G’ of hydrogels with DMAEMA did not change significantly during the first 10 days
(G’= 3100 Pa), which is in agreement with the ξ-potential measurements that showed no charge
loss during this time period (Fig. 2A). After 10 days, the gel strength gradually decreased as a result
of reduced interaction possibilities between the charged microspheres. Finally, after 30 days the hydrogel was almost completely degraded (G’= 90 Pa). Hydrogels with the same charge density,
but consisting of equal amounts of dex-HEMA-MAA and dex-HEMA-HPMA-DMAE or dex-HEMA-
HPMA-DMAPr microspheres, showed a rapid decrease in G’ in time after incubation at 37 °C
(Fig. 5). The G’ reached half its initial value after 3 and 5 days, for HPMA-DMAE and HPMA-DMAPr
containing hydrogels, respectively. After 10 days the material did not show any gel-like properties
anymore, with a tan(δ) > 1. These findings can be explained by the ongoing hydrolysis of the
carbonate esters in the functional monomers leading to a loss of positive charges. Once the
particles loose their cationic character, the ionic interactions between the microspheres vanish and
the macroscopic network falls apart. The ξ-potential measurements reported in section 3.2
showed that the cationic microspheres containing HPMA-DMAE or HPMA-DMAPr had lost their charge after 1 and 2 days, respectively (Fig. 3), which is faster than would be expected from the
Tuning the degradation behavior of ionically crosslinked dextran hydrogels by variation of the cationic groups
117
rheological data of the macroscopic hydrogels. Presumably, in a hydrogel network, containing
both negatively and positively charged microspheres, hydrolysis of the carbonate ester in the
monomers is retarded by the presence of anionic groups. Although the ξ-potential measurements
showed a slightly slower charge loss of the HPMA-DMAPr containing microspheres when
compared to those with HPMA-DMAE, no significant differences in degradation behavior during rheology experiments was observed between the 2 types of hydrogels.
0 5 10 15 20 25 300
500
1000
1500
2000
2500
3000
3500DMAEMA
HPMA-DMAEHPMA-DMAPr
time (days)
G' (
Pa)
Figure 5: Storage modulus (G’) of hydrogels consisting of equal amounts of cationic (containing DMAEMA ( ),
HPMA-DMAE ( ) or HPMA-DMAPr ( )) and methacylic acid (MAA) containing microspheres in time, upon
incubation at 37 °C (n=3). The ξ-potential of the microspheres was 8 mV.
4. Conclusions
In conclusion, the results presented in this chapter demonstrate that it is possible to copolymerize
HPMA-DMAE or HPMA-DMAPr with dex-HEMA, leading to positively charged microspheres. The use of these hydrolysis-sensitive HPMA-based monomers in the self-assembling microsphere-
based dextran hydrogels, offers the opportunity to tailor the degradation rate of the positively charged microspheres, but also to tune the mechanical properties of the hydrogels in time. Additionally, a better cell compatibility is expected because of the degradability of the HPMA-
based groups and their reduced cytotoxicity as compared to pDMAEMA.
Chapter 7
118
References
1. Hoffman AS. Hydrogels for biomedical applications. Adv. Drug Deliver. Rev. 2002;43:3-12.
2. Drury JL, Mooney DJ. Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials 2003;24:4337-4351.
3. Kashyap N, Kumar N, Ravikumar MNV. Hydrogels for pharmaceutical and biomedical applications. Crit. Rev. Ther. Drug 2005;22:107-150.
4. Van Tomme SR, Hennink WE. Biodegradable dextran hydrogels for protein delivery applications. Expert Rev. Med. Devices 2007;4:147-164.
5. Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Self-gelling hydrogels based on oppositely charged dextran microspheres. Biomaterials 2005;26:2129-2135.
6. Van Tomme SR, De Geest BG, Braeckmans K, De Smedt SC, Siepmann F, Siepmann J, van Nostrum CF, Hennink WE. Mobility of model proteins in hydrogels composed of oppositely charged dextran microspheres studied by protein release and fluorescence recovery after photobleaching. J. Control. Release 2005;110:67-78.
7. Van Tomme SR, van Nostrum CF, De Smedt SC, Hennink WE. Degradation behavior of dextran hydrogels composed of positively and negatively charged microspheres. Biomaterials 2006;27:4141-4148.
8. Van Tomme SR, Van Nostrum CF, Dijkstra M, De Smedt SC, Hennink WE. Effect of particle size and charge on the network properties of microsphere-based hydrogels. 2007 submitted.
9. Van de Wetering P, Cherng J-Y, Talsma H, Hennink WE. Relation between transfection efficiency and cytotoxicity of poly(2-dimethylamino)ethyl methacrylate)/plasmid complexes. J. Control. Release 1997;49:59-69.
10. Oupicky D, Konak C, Ulbrich K. DNA complexes with block and graft copolymers of N-(2-hydroxypropyl)methacrylamide and 2-(trimethylammonio)ethyl methacrylate. J Biomater Sci Polym Ed 1999;10:573-590.
11. Luten J, Akeroyd N, Funhoff A, Lok MC, Talsma H, Hennink WE. Methacrylamide polymers with hydrolysis-sensitive cationic side groups as degradable gene carriers. Bioconjugate Chem. 2006;17:1077-1084.
12. Van Dijk-Wolthuis WNE, Tsang SKY, Kettenes-van den Bosch JJ, Hennink WE. A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer 1997;38:6235-6242.
13. Franssen O, Hennink WE. A novel preparation method for polymeric microparticles without the use of organic solvents. Int. J. Pharm. 1998;168:1-7.
14. Van de Wetering P, Zuidam NJ, van Steenbergen MJ, van der Houwen OAGJ, Underberg WJM, Hennink WE. A mechanistic study of the hydrolytic stability of poly(2-(dimethylamino)ethyl methacrylate). Macromolecules 1998;31:8063-8068.
Tuning the degradation behavior of ionically crosslinked dextran hydrogels by variation of the cationic groups
119
8 Macroscopic hydrogels by self-assembly of oligolactate-
grafted dextran microspheres
Sophie R. Van Tommea, Ad Mensb, Cornelus F. van Nostruma, Wim E. Henninka
aDepartment of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS),
University Utrecht, Utrecht, The Netherlands bInorganic Chemistry and Catalysis group, Department of Chemistry, Faculty of Science,
Utrecht University, Utrecht, The Netherlands
Submitted for publication
Chapter 8
122
Abstract
A novel approach is presented to create self-assembling hydrogels. Microspheres based on crosslinked dextran were chemically modified with L- or D-oligolactate chains. Both the degree of
substitution (DS, i.e. number of lactate grafts per 100 glucopyranose units) and the degree of polymerization (DP) of the oligomers were varied. Oligolactate-grafting of the particles was
confirmed by Fourier Transform-InfraRed (FT-IR) spectroscopy, Raman spectroscopy and X-ray photoelectron spectroscopy (XPS). Rheological analysis of aqueous dispersions of oligolactate-
grafted microspheres demonstrated that hydrophobic interactions between oligolactate chains on the surface of various microspheres resulted in the formation of an almost fully elastic gel. A mixture of microspheres substituted with L- or D-oligolactates of opposite chirality resulted in gels
with highest strength, likely due to stereocomplexation between the enantiomers. The network properties could be modulated (storage modulus (G’) from 500 to 12500 Pa) by varying the solid
content of the gel, the DS and the DP of the oligolactate grafts. This study shows that because of the biocompatible nature of the material and the possibility to tailor the gel properties, this
hydrogel system is an attractive candidate for pharmaceutical and biomedical applications.
Keywords: dextran microspheres, oligolactate, self-assembling hydrogel, hydrophobic
interactions, stereocomplexes
Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
123
1. Introduction
There is a high need for materials and technologies for the development of suitable pharmaceutical formulations of the modern generation of biotherapeutics, such as peptides,
proteins and plasmid DNA.[1] In this way, the active compounds are protected against premature degradation and released in a controlled manner. Both natural and synthetic polymers, as well as
lipids have been used to create delivery vehicles in all shapes and sizes, e.g. hydrogels, microspheres, liposomes, micelles, etc.[2] Hydrogels are hydrophilic polymeric networks capable of
imbibing substantial amounts of water. Dissolution of the polymer chains is prevented by chemical or physical crosslinking.[3] Their high water content makes them attractive candidates for
protein delivery and tissue engineering applications.[4-6] Various hydrogels have been developed, mainly based on chemical crosslinking of e.g. modified dextrans[7-10] and poly(ethylene glycol)s (PEG).[11] Despite the favorable characteristics of these
matrices regarding biodegradability and biocompatibility, the crosslinking conditions might adversely affect the structure and activity of the entrapped bioactive molecules. Therefore,
physically crosslinked hydrogels have gained increasing interest in recent years as an alternative for the chemically crosslinked gels as delivery system for among others pharmaceutically active
proteins. Various strategies have been developed to create in situ forming delivery systems. Chitosan,[12, 13] hyaluronic acid[14, 15] and PEG-based polymers[16] have been successfully used as
photopolymerizable systems in tissue engineering and drug delivery. Alginates spontaneously gellify in the presence of divalent cations like Ca2+ and have among others been used for
ophthalmic delivery[17] and as synthetic extracellular matrices.[18] In situ gelation can also occur in response to a certain trigger, e.g. temperature,[19-21] pH[22, 23] or after some time after mixing the
components. The latter can be the result of e.g. hydrophobic interactions,[24] ionic interactions[25] or stereocomplexation.[26, 27] Stereocomplexation between oligolactate grafts of opposite chirality on
dextran resulted in hydrogels after mixing aqueous solutions of dex-L-lactate and dex-D-lactate. Extensive in vitro and in vivo studies have shown that this system is promising as protein delivery system.[26, 28-30] Although the stereocomplexed dex-lactate hydrogels have successfully been used
as injectable matrices,[29] the high viscosity of the two components used to prepare the gels limits their practical applicability.
In this work, a novel hydrogel system is reported combining injectability of dextran microspheres and physical crosslinking between the microspheres. For this purpose, we have made use of the
favorable properties of lactic acid oligomers to connect the microspheres, creating a macroscopic gel. Lactic acid contains a chiral center, which enables the polymer chains to form stereocomplex
crystallites between the L- and D-enantiomers. The presence of stereocomplexes in polymeric matrices results in materials with higher mechanical strength, higher melting point and lower hydrolysis rate.[31, 32] Thus, oligolactate chains of opposite chirality were covalently grafted onto
microspheres that are formed by polymerization of (hydroxyethyl methacrylate)-substituted dextran (referred to as ‘dex-HEMA-lactate microspheres’).
Chapter 8
124
Figure 1: Schematic representation of the concept of the self-assembling hydrogel (grafts are not on scale).
Figure 1 illustrates the possible interactions that may occur upon hydration of the resulting dex-HEMA-lactate microspheres. The individual microspheres grafted with L- or D-oligolactate could
assemble due to hydrophobic interactions between the lactate-enriched domains on their surface. However, when microspheres, derivatized with oligolactate of opposite chirality, are mixed, stereocomplexes can be formed, resulting in relatively stronger interactions between the particles.
The characterization of the particles and their ability to create macroscopic gels are presented in this work.
2. Materials and methods
2.1. Materials
Dextran T40 (from Leuconostoc ssp.), N,N,N’,N’-tetramethyl ethylenediamine (TEMED) and 2-hydroxyethyl methacrylate (HEMA, 95 %) were provided by Fluka (Buchs, Switzerland).
Poly(ethylene glycol) (PEG) 10 kDa and potassium peroxodisulfate (KPS, >99 %) were purchased from Merck (Darmstadt, Germany). L-lactide ((3S-cis)-3,6-dimethyl-1,4-dioxane-2,5-dione, > 99.5 %)
and D-lactide ((3R-cis)-3,6-dimethyl-1,4-dioxane-2,5-dione, > 99.5 %), poly(L-lactic acid) (PLLA, intrinsic viscosity 0.59 dl/g) and poly(D-lactic acid) (PDLA, intrinsic viscosity 2.39 dl/g) were obtained from Purac Biochem BV (Gorinchem, The Netherlands) and used without further
purification. N-2-hydroxyethyl piperazine-N'-2-ethanesulfonic acid (Hepes), 1,1’-carbonyl diimidazole (CDI, 98 %), 4-N,N-dimethyl aminopyridine (DMAP, 99 %) and poly(vinyl alcohol) (PVA,
Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
125
88 % hydrolyzed) were purchased from Acros Chimica (Geel, Belgium). Stannous octoate (tin(II) bis(2-ethyl hexanoate) SnOct2, 95 %) and S-ethyl lactate (98 %) were provided by Sigma-Aldrich
(Zwijndrecht, The Netherlands).
2.2. Preparation of dex-HEMA microspheres
Hydroxyethyl methacrylate-derivatized dextran (dex-HEMA) with a substitution degree of 10 (DS, i.e. the number of HEMA groups per 100 glucopyranose units) was synthesized and characterized
according to van Dijk-Wolthuis et al.[33] Microspheres were prepared through radical polymerization of dex-HEMA in an all-aqueous environment, as previously described.[34, 35]
2.3. Synthesis of polydisperse enantiomeric oligolactate
L- and D-lactic acid oligomers were synthesized by a ring-opening polymerization reaction of
lactide with ethyl lactate as initiator, according to de Jong et al.[36] The average degree of polymerization (DPav) of the formed oligolactates was controlled by the ethyl lactate/lactide ratio.
The DPav used in this study were 5 and 13.
2.4. Grafting lactic acid oligomers to dex-HEMA microspheres
To couple the lactic acid oligomers to dex-HEMA microspheres, the hydroxyl groups of the
oligomers were activated using N,N’-carbonyldiimidazole (CDI) as described by de Jong et al.[26] Next, the activated oligomers (ethyl lactate-CI) were grafted to dex-HEMA microspheres, essentially as described for grafting of lactic acid oligomers to soluble dextran with some minor modifications.
In short, lyophilized dex-HEMA microspheres (500 mg) were dispersed in dry dimethyl sulfoxide (DMSO) (10 mL). Next, DMAP (100 mg) was dissolved in the mixture and subsequently, ethyl
lactate-CI (e.g. 230 mg to obtain a theoretical maximum DS of 8 (defined as the number of oligolactate grafts per 100 glucopyranose units and further referred to as ‘DStheor’)) was added and
the vials were placed on a roller bench for 10 days. Finally, the microspheres were washed with DMSO (3 times) and dichloromethane (DCM) (3 times) and subsequently dried overnight by evaporation of the DCM. Microspheres with different theoretical oligolactate substitution degrees
were prepared (DStheor 2, 4 and 8). As a control, the dex-HEMA microspheres were treated with non-CDI-activated oligolactate under the same conditions.
2.5. Preparation of PLLA and PDLA microspheres
PLLA en PDLA microspheres were prepared using an oil-in-water (o/w) solvent evaporation technique.[37] Briefly, 3 g PLLA or PDLA was dissolved in 13 ml DCM and subsequently poured into 180 ml PVA solution (1 % in water (w/v)). The o/w emulsion was stirred for 4 h (mechanical stirrer,
1200 rpm) during which the DCM was allowed to evaporate. After three washing and centrifugation steps (10 min, 3000 rpm), the microspheres were lyophilized.
Chapter 8
126
2.6. Fourier Transform-InfraRed spectroscopy (FT-IR)
Modification of the dex-HEMA microspheres with oligolactates was studied with FT-IR (Biorad FTS 6000 spectrometer (Varian Inc., Palo Alto, CA, USA)). Microsphere dispersions were prepared (5 %
w/w in Hepes buffer, 100 mM, pH 7.0) of which 10 μl was brought into a liquid sample cell with a CaF2 window and a path length of 7 μm. For a single spectrum, 1024 scans were accumulated and
subsequently the spectra were corrected for water vapor. Before and after calculation of the second derivative, the spectra were smoothed using a second order, seven points Savitzky-Golay
smoothing.[38]
2.7. Raman spectroscopy
Raman spectra of dry dex-HEMA microspheres, dex-HEMA-lactate microspheres and oligolactate were recorded at room temperature with a Kaiser RXN dispersive Raman spectrometer. A 532 nm
(60 mW) laser was used for excitation and a Peltier element-cooled Andor CCD camera was applied for detection.
2.8. X-ray Photoelectron Spectroscopy (XPS)
XPS was performed to analyze the surface composition of the oligolactate-grafted microspheres.
The XPS data were obtained with a Vacuum Generators XPS system using a CLAM-2 hemispherical
analyzer for electron detection. Non-monochromatic Al (Kα) X-Ray radiation was used for
generating the photo-electrons at an anode current of 20 mA at 10 KeV. The pass energy of the analyzer was set at 50 eV. The survey scan was taken with a pass energy of 100 eV. Dry microspheres were attached to double-sided tape before mounting in the analysis chamber. Both
C(1s) and O(1s) spectra were recorded.
2.9. Lactic acid determination
To investigate the extent of grafting of oligolactates to the dex-HEMA microspheres, the lactic acid
content of degraded microspheres was measured. In detail, dex-HEMA-lactate microspheres (10 mg) were hydrolyzed in alkaline conditions at room temperature overnight (1 ml 0.1 N NaOH) and subsequently the solution was neutralized (100 μl 1 N HCl). Analysis of the amount of lactic acid in
the samples was carried out on a Waters HPLC system (Waters associates Inc., Milford, MA, USA) consisting of a quaternary gradient pump (model 600E), an autosampler (model 717) and a
variable wavelength absorbance detector (model 486). An Alltech Prevail Organic acid column (5 μm, 250 x 4.6 mm) was used with KH2PO3 buffer (25 mM, pH 2.5) and acetonitrile/water
(95/5 w/w) as eluent A and B, respectively, at a flow rate of 1 ml/min. A gradient was run from 0 to 60 % B in 10 min, followed by an elution with 60 % B for 10 min, with a total run time of 30 min.
The injection volume was 50 μl and the detection wavelength was 210 nm. The calibration curve was linear between 0.01 and 4 μmol lactic acid.
Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
127
2.10. Rheology of microsphere-based hydrogels
Rheological analysis of dex-HEMA-lactate microsphere dispersions was performed using a controlled stress rheometer (AR1000-N, TA Instruments, Etten-Leur, The Netherlands), equipped with an acrylic flat plate geometry (20 mm diameter) and a gap of 500 μm. A solvent trap was used
to prevent evaporation of the solvent. Gels (200 mg) were prepared by dispersing the dry microspheres in buffer (Hepes, 100 mM, pH 7.0). When both oligo-L-lactate and oligo-D-lactate
substituted microspheres were used, the particles were thoroughly mixed in the dry state, using a
vortex, before hydration with buffer. The microspheres were allowed to hydrate overnight at 4 °C.
The viscoelastic properties of the samples were determined by measuring the G’ (shear storage
modulus) and G” (loss modulus) at 20 °C with a constant strain of 1 % and constant frequency of
1 Hz. The extent of recovery of the material after deformation was evaluated during creep
experiments. A shear stress (varying from 1-100 Pa) was applied while the strain was monitored. After 1 min the stress was removed and the recovery of the sample was monitored by measuring the strain during 2 min. To mimic the behavior of the system during injection, the oscillatory stress
was gradually increased until the gel started to flow. Subsequently, a constant stress was applied during a time sweep experiment monitoring the recovery of the network after flow. The influence
of the solid content (10-20 %), the chain length of the oligolactate grafts (DPav 5 and 13) and the degree of oligolactate substitution (DStheor 2-8) on the network properties was investigated.
For comparison, PLLA and PDLA microspheres were mixed, hydrated (Hepes buffer 100 mM pH 7.0) (20-60 % solid content (w/w)) and investigated for possible network formation.
3. Results and discussion
3.1. Preparation and oligolactate-substitution of dex-HEMA microspheres
Internally crosslinked dex-HEMA microgels with an equilibrium water content of 70 % and a mean volume diameter of 9 μm were prepared making use of an aqueous PEG/dextran two-phase
system.[34, 35] Next, the microsphere particles were dried, swollen with DMSO and then grafted with oligolactate chains (Fig. 2). In this way, structures are created, consisting of hydrophilic, dextran-based particles, substituted
with hydrophobic oligolactate. Derivatization of the microspheres did not affect the size, but the wetability visibly decreased, as evidenced from their poor dispersability in water. These findings
indicate that water penetration into the microspheres was retarded due to an increased hydrophobicity of the microsphere surface.
Chapter 8
128
O
O
OHO
HOOOO
O
O
O
O
ran
O
O
OH
HOOO
O
O
OH
HOOO
O OO
OO
O
O
O
n
O
O
OHO
HOHO
O
O
OHO
HOOOO
O
O
O
O
ran
O
O
OH
HOOO
O O
OOO
O
O
O
O
n O
NN
+DMSO
DMAP
dex-HEMA microspheres activated oligolactate
dex-HEMA-lactate microspheres
Figure 2: Schematic representation of oligolactate grafting on dex-HEMA microspheres, resulting in dex-HEMA-
lactate microspheres.
3.2. FT-IR and Raman spectroscopy
The presence of the oligolactate chains in dex-HEMA microspheres was confirmed by FT-IR and Raman spectroscopy (Fig. 3-5). The 2nd derivative FT-IR spectra of dex-HEMA microspheres showed a peak around 1750 cm-1, attributed to vibrations of carbonyl groups in the HEMA
chains.[39] The grafted microspheres showed an increase in peak height due to an increased intensity of the carbonyl vibration, dependent on the degree of lactate substitution (DStheor 8 >
DStheor 4 > DStheor 2) (DPav 13) (Fig. 3). The control samples, consisting of microspheres that were treated with non-activated oligolactate, showed a slight increase of the carbonyl peak, indicating
that some aspecific binding to the microparticles had occurred. Figure 4 shows the region 2700-3100 cm-1 of the Raman spectrum of unmodified dex-HEMA
microspheres. The observed peaks correspond to CH and CH2 stretching vibrations.[39] In Figure 5 the region 2800-3100 cm-1 of the Raman spectra of oligolactate and dex-HEMA-L-lactate microspheres (DStheor 8) are depicted. The Raman spectrum of oligolactate showed 3 dominant
peaks at 2885, 2950 and 3000 cm-1, corresponding to CH and CH3 stretching vibrations,[39] which were also detected in the spectrum of the oligolactate-substituted dex-HEMA microspheres. They
indicate the presence of oligolactate chains in the microspheres, most likely as a result of the grafting reaction.
Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
129
170017201740176017801800-0.0008
-0.0004
0.0000
0.0004
0.0008
0.0012
0.0016
wavenumber (cm-1)
abso
rban
ce
Figure 3: 2nd derivative FT-IR spectra of dispersions consisting of dex-HEMA microspheres ( ) and dex-HEMA-
D-lactate microspheres of different oligolactate substitution degrees (DStheor 2 ( ), DStheor 4(—), DStheor 8 ( );
DPav 13) in buffer. Dex-HEMA microspheres treated with non-activated oligolactate were measured as a
control ( ).
2700275028002850290029503000305031000
100
200
300
400
500
600
wavenumber (cm-1)
Ram
an in
tens
ity(a
rbitr
ary
units
)
Figure 4: Raman spectrum (region 2700-3100 cm-1) of unmodified dex-HEMA microspheres.
28002850290029503000305031000
2000
4000
6000
80008500
13500
18500
23500
wavenumber (cm-1)
Ram
an in
tens
ity(a
rbitr
ary
units
)
Figure 5: Raman spectra (region 2800-3100 cm-1) of oligolactate ( ) and dex-HEMA-L-lactate microspheres
(DStheor 8, DPav 13) (—).
Chapter 8
130
3.3. Lactic acid determination of degraded microspheres
To quantify the extent of grafting, the lactic acid content of the microspheres was determined. Therefore the dex-HEMA-lactate microspheres were incubated for 16 h under alkaline conditions
to yield dextran, poly(HEMA)[40] and lactic acid. Degradation of the dex-HEMA-lactate microspheres and subsequent lactic acid quantification revealed that 10-13 % and 16-20 % (for DPav 13 and 5,
respectively) of the amount of oligolactate added, was grafted to the microspheres. The relatively low grafting efficiency might indicate that the OH groups present at the outer layer of the
microspheres, are preferentially modified, by which reaction of the OH groups buried inside the particles with oligolactate is prevented.
3.4. XPS of oligolactate-grafted dex-HEMA microspheres
XPS provides information on chemical bonding and elemental composition of the outer layer (a
few nm) of a sample. Figure 6 shows representative C(1s) spectra of oligolactate, dex-HEMA microspheres and modified dex-HEMA microspheres. In the spectrum of oligolactate, 3 peaks at
291, 293 and 295 eV are visible, originating from C-H, C-O and C=O bonds, respectively. The dex-HEMA microspheres (without and with oligolactate grafts) only showed one peak, at 293 eV, originating from C-O bonds.[41, 42] C-O groups are the most abundant ones in dex-HEMA
microspheres. Grafting of the particles with oligolactate led to an increase in intensity of other groups (C=O and C-H), but due to the excessive amount of C-O groups, only one broad peak,
ascribed to C-O, was detected. For all microspheres, the O(1s)-C(1s) was ~246.5 eV indicating that the shifts in peak maxima of both O(1s) and C(1s)[41] were not due to specific properties of the
samples, but due to the charging of the products.
285 290 295 300 30515000
25000
35000
45000
55000
65000
binding energy (eV)
inte
nsity
(arb
itrar
y un
its)
Figure 6: XPS C(1s) spectra of oligolactate ( ), dex-HEMA microspheres ( ) and dex-HEMA-L-lactate
microspheres (DStheor 8, DPav 13) (—).
The Full-Width-at-Half-Height (FWHH) of the peaks of the different microsphere samples was
calculated. Grafting of the microspheres with oligolactate resulted in an increased FWHH from 2.00 eV (± 0.05 eV) (dex-HEMA microspheres) to 2.43 eV (± 0.05 eV) (dex-HEMA-L-lactate microspheres, average of DStheor 2-8). This is presumably due to shouldering of the peaks towards a higher
Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
131
binding energy as a result of an increasing amount of C=O groups present at the surface. The oxygen/carbon ratio of the oligolactate-substituted dex-HEMA microspheres was calculated by
determining the area under the curve of both the O(1s) (not shown) and C(1s) peaks. The dex-HEMA microspheres substituted with oligolactate (DStheor 2-8) showed a higher O/C ratio (1.62 ± 0.04) when compared to dex-HEMA microspheres (1.2) and confirms that substitution of dex-
HEMA microspheres with oligolactate has led to an enrichment of their surface with oxygen as a result of an increased number of carbonyl groups. Keeping the detection limit of peak shifts and
peak separations in mind (± 0.2 eV)[43] the differences between the substituted microspheres of various DS were too small to detect. Presumably, preferential grafting of the outer surface layer has
taken place leading to microspheres with comparable surface composition.
3.5. Rheological characterization of the self-assembling networks
Rheological studies were done to evaluate the ability of the oligolactate substituted microspheres to interact and form macroscopic hydrogels. For comparison, the possibility of PLLA and PDLA
microspheres to self-assemble through stereocomplexation was also investigated. Rheological experiments on the latter system, did not show, even in high concentrations (solid contents up to
60 % (w/w)), the formation of a network. Rheological analysis of dispersions of dex-HEMA microspheres (15 % solid w/w) as well as
microspheres treated with non-activated oligo-L or D-lactate behaved like a Newtonian liquid, showing that the particles did not interact with each other to form a network (data not shown). Although FT-IR analysis (Fig. 3) and Raman spectroscopy (Fig. 5) suggested that some oligolactate
chains might be bound non-covalently to the microspheres, it appeared to be insufficient to lead to network formation with neighboring microspheres.
Figure 7 depicts the effect of the solid content of the gels (defined as the % dry microspheres w/w) on the hydrogel properties of mixtures of dex-HEMA-L-lactate and dex-HEMA-D-lactate
microspheres (DStheor 8, DPav 13). In these hydrogel systems, water forms a continuous phase in which the microparticles, the solid components, are dispersed. A water content higher than 90 %
prevented the microspheres from interacting and no gel formation could be observed. A strong increase of the storage modulus from about 850 Pa to 12700 Pa was observed when the solid
content was increased from 10 to 20 %. Within this range, the tan(δ) (= G”/G’) was lower than 0.1,
indicating that almost fully elastic gels were formed. The discontinuous semisolid phase imparts the gel strength since a higher concentration of microspheres leads to a denser packing of the
particles and more opportunities to interact. Interestingly, dispersions (15 % solid (w/w), corresponding to 50 % (w/w) of free water) of the individual components dex-HEMA-L-lactate or dex-HEMA-D-lactate microspheres (DStheor 2,
DPav 13) also showed gel-like behavior with tan(δ) < 0.1 and storage moduli (G’) of 800 and 600 Pa,
respectively (Fig. 8). The G’ increased to 1900 Pa and 2400 Pa when the DStheor was increased from
2 to 8, for dex-HEMA-L-lactate and dex-HEMA-D-lactate microspheres, respectively. Obviously, a higher number of oligolactates present on the microspheres led to more physical crosslinks between the microspheres and thus stronger hydrogels. No significant differences in gel
properties were observed between oligo-L- or oligo-D-lactate grafted microspheres for the different DStheor. Apparently, hydrophobic interactions between the oligolactate chains on different
microspheres are sufficient to interconnect the microspheres and create an almost fully elastic
Chapter 8
132
network (tan(δ) < 0.1). This behavior was different from that of non-crosslinked, water soluble,
oligo-L- or oligo-D-lactate substituted dextran with a high DP and DS, which behaved as a typical
viscoelastic material due to association of longer oligolactates with oligomers of the same chirality.[26]
Importantly, for each DStheor, mixing equal amounts of microspheres substituted with oligolactates of opposite chirality led to stronger hydrogels, corresponding with G’s of 1400 and 4400 Pa for
DStheor 2 and 8 respectively (DPav 13) (Fig. 8). This is attributable to stereocomplex formation between L- and D-lactate (Fig. 8). This has been described previously for self-assembling hydrogels based on oligolactate-grafted dextran, which formed an almost elastic network as well.[26]
10% 12.5% 15% 17.5% 20 %0
2500
5000
7500
10000
12500
15000
17500
solid content
G' (
Pa)
Figure 7: Storage modulus (G’) at 20 °C of dispersions consisting of a mixture of equal amounts of dex-HEMA-L-
lactate and dex-HEMA-D-lactate microspheres (oligolactate DPav 13 DStheor 8) as a function of the solid content
(w/w) (n=3).
L D L+D0
5001000150020002500300035004000450050005500
DStheor 4DStheor 8
DStheor 2
G' (
Pa)
Figure 8: Storage modulus (G’) of hydrogels (15 % solid (w/w)) consisting of dex-HEMA-L-lactate microspheres
(L), dex-HEMA-D-lactate microspheres (D) or a mixture of both (L+D) at 20 °C, as a function of the DS (DStheor 2:
gray bars, DStheor 4: black bars, DStheor 8: white bars) (DPav 13) (n=3).
Microspheres were grafted (DStheor 2 and 8) with oligolactate of either DPav 5 or 13 and the influence of the oligolactate chain length on the rheological behavior was monitored (Fig. 9). The
Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
133
microspheres grafted with the shortest oligolactate chains (DPav 5) (15 % solid (w/w)) were able to interact and form hydrogels, but the gel strength could not be altered by varying the DStheor (2 vs.
8). It was previously shown that enantiomeric monodisperse lactic acid oligomers of DP 7 are amorphous, whereas blends of the corresponding L- and D-forms show crystallinity.[36] When grafted to dextran however, a DP lower than 11 did not result in hydrogel formation.[26] Taking this
into account, it is clear that when using oligolactate chains with a DPav 5 only a fraction of the oligomers will be sufficiently long to be able to interact with lactic acid oligomers on other
microspheres to yield stereocomplexes, so the possibilities to tailor the network properties diminishe. For hydrogels (15 % solid (w/w)) composed of dex-HEMA-L- and dex-HEMA-D-lactate
microspheres with a DStheor 8, an increase of DPav from 5 to 13 led to a strong increase in G’ from 1000 Pa to 4500 Pa (Fig. 9), most likely due to the presence of stereocomplexes acting as physical
crosslinks between the microspheres.
5 13
5001000150020002500300035004000450050005500
DStheor 2
DStheor 8
DPav(oligolactate)
G' (
Pa)
Figure 9: Storage modulus (G’) of dispersions containing equal amounts of dex-HEMA-L-lactate and dex-HEMA-
D-lactate microspheres with varying oligolactate substitution degrees (DStheor 2: white bars, DStheor 8: black bars),
as a function of the DPav of the oligolactate chains (15 % solid (w/w), 20 °C) (n=3).
0.5 1.0 1.5 2.0 2.5 3.00.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
time (min)
% s
train
dex
-HEM
A-la
ctat
eM
S di
sper
sion
Figure 10: Creep experiment on a hydrogel consisting of a mixture of equal amounts of dex-HEMA-L-lactate
and dex-HEMA-D-lactate microspheres (15 % solid (w/w), applied stress 50 Pa, 20 °C).
Chapter 8
134
Creep experiments were performed to investigate the extent of recovery of the hydrogels after
deformation (Fig. 10). Dispersions (15 % solid) containing dex-HEMA microspheres showed Newtonian behavior with a % strain of 8000 during the retardation phase and no recovery after withdrawal of the shear stress (results not shown). In contrast, under the same experimental
conditions, hydrogels (15 % solid (w/w)) composed of dex-HEMA-L-lactate and dex-HEMA-D-lactate microspheres (DStheor 8, DPav 13) reached a plateau deformation value of 1.2 % and
recovered almost completely when the stress was removed (Fig. 10). Dispersions containing solely oligo-L- or oligo-D-lactate grafted microspheres displayed the same deformation profile (not
shown). The reversibility of gelation was studied by measuring the gel properties at a constant strain (1 %, corresponding to a shear stress of 6 Pa) during 20 min and by subsequently increasing the
oscillatory stress until the network was broken (tan(δ) >1). Immediately hereafter the network
properties were monitored during a second time sweep. When the stress on the gels (equal
amounts of dex-HEMA-L-lactate and dex-HEMA-D-lactate microspheres, 15 % solid (w/w), DPav 13, DStheor 8) was increased, the gel was broken and flow started to occur at approximately 350 Pa (Fig. 11).
1 10 100 10000
50010001500200025003000350040004500500055006000
0.01
0.1
1
10
100
oscillatory stress (Pa)
G' (
Pa)
tan(
δ )
Figure 11: Storage modulus (G’) (—) and tan(δ)( ) of a dispersion containing equal amounts of dex-HEMA-L-
lactate and dex-HEMA-D-lactate microspheres (DStheor 8, DPav 13) (15 % solid (w/w), 20 °C) as a function of the
oscillatory stress.
Upon removal of the stress the network rebuilt almost instantaneously as indicated by the
recovery of the initially observed G’ (~ 5000 Pa) and low tan(δ) (< 0.1). These results indicate that
upon shear, the interactions between the microspheres can be temporarily and reversibly broken.
This is an important feature for application as injectable system since the gels need to exhibit a reversible yield point, i.e. they have to flow when a certain stress is applied after which the network
needs to be rebuilt. The yield point is dependent on the initial gel strength and can be tailored by varying the solid content, the number of lactate grafts and their DPav. Some representative examples are listed in Table 1.
Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
135
Table 1: Yield point of hydrogels differing in solid content consisting of equal amounts of dex-HEMA-L-lactate
and dex-HEMA-D-lactate microspheres with varying properties.
solid content (%) DPav DStheor yield point (Pa)
15 5 8 100
15 13 2 130
10 13 8 250 15 13 8 350
20 13 8 1000
4. Conclusions
This paper reports on the possibility to create macroscopic hydrogels based on self-assembling
dextran microspheres. Hydrophilic dex-HEMA microspheres were grafted with enantiomeric oligolactate chains creating surface hydrophobized microgels. Hydrophobic interactions between
oligolactate chains led to the formation of macroscopic networks. An even stronger type of physical crosslinks, stereocomplexes, was introduced by addition of oligolactate-substituted microspheres of opposite chirality. The network properties of these self-assembling hydrogels
showed to be tailorable by modifying the grafting density of the microspheres, the degree of polymerization of the oligolactate grafts, and the solid content of the gels. Interestingly for
application of these systems as injectable matrix, it was found that upon exposure to shear forces, the interactions between the microspheres were temporarily broken, allowing the gel to be
injected, after which the network was rebuilt. The biocompatible nature of the material as well as the possibility to tailor the gel properties makes this hydrogel attractive in a variety of
pharmaceutical applications.
Acknowledgement
The authors would like to thank Bert Weckhuysen (Inorganic Chemistry and Catalysis group,
Department of Chemistry, Faculty of Science, Utrecht University) for valuable discussions and critically reading this manuscript. Tom Visser and Fouad Soulimani (Inorganic Chemistry and Catalysis group, Department of Chemistry, Faculty of Science, Utrecht University) are gratefully
acknowledged for performing Raman spectroscopy and related scientific discussions. This work was financially supported by SenterNovem (IS042016).
Chapter 8
136
References
1. Orive G, Hernandez RM, Rodriguez Gascon A, Dominguez-Gil A, Pedraz JL. Drug delivery in biotechnology: present and future. Curr. Opin. Biotechnol. 2003;14:659-664.
2. Crommelin DJ, Storm G, Jiskoot W, Stenekes R, Mastrobattista E, Hennink WE. Nanotechnological approaches for the delivery of macromolecules. J. Control. Release 2003;87:81-88.
3. Hennink WE, van Nostrum CF. Novel crosslinking methods to design hydrogels. Adv. Drug Deliver. Rev. 2002;54:13-36.
4. Peppas NA, Hilt JZ, Khademhosseini A, Langer R. Hydrogels In biology and medicine: from molecular principles to biotechnology. Adv. Mater. 2006;18:1345-1360.
5. Hoffman AS. Hydrogels for biomedical applications. Adv. Drug Deliver. Rev. 2002;43:3-12.
6. Van Tomme SR, Hennink WE. Biodegradable dextran hydrogels for protein delivery applications. Expert Rev. Med. Devices 2007;4:147-164.
7. Van Dijk-Wolthuis WNE, Hoogeboom JAM, van Steenbergen MJ, Tsang SKY, Hennink WE. Degradation and release behavior of dextran-based hydrogels. Macromolecules 1997;30:4639-4645.
8. Franssen O, Vandervennet L, Roders P, Hennink WE. Degradable dextran hydrogels: controlled release of a model protein from cylinders and microspheres. J. Control. Release 1999;60:211-221.
9. Cadée JA, Brouwer LA, den Otter W, Hennink WE, van Luyn MJA. A comparative biocompatibility study of microspheres based on crosslinked dextran or poly(lactic-co-glycolic) acid after subcutaneous injection in rats. J. Biomed. Mater. Res. 2001;56:600-609.
10. Hiemstra C, Aa LJ, Zhong Z, Dijkstra PJ, Feijen J. Rapidly in Situ-Forming Degradable Hydrogels from Dextran Thiols through Michael Addition. Biomacromolecules 2007 in press.
11. Van de Wetering P, Metters AT, Schoenmakers RG, Hubbell JA. Poly(ethylene glycol) hydrogels formed by conjugate addition with controllable swelling, degradation, and release of pharmaceutically active proteins. J. Control. Release 2005;101:619-627.
12. Fujita M, Ishihara M, Morimoto Y, Simizu M, Saito Y, Yura H, Matsui T, Takase B, Hattori H, Kanatani Y. Efficacy of photocrosslinkable chitosan hydrogel containing fibroblast growth factor-2 in a rabbit model of chronic myocardial infarction. J. Surg. Res. 2005;126:27-33.
13. Berger J, Reist M, Mayer JM, Felt O, Peppas NA, Gurny R. Structure and interactions in covalently and ionically crosslinked chitosan hydrogels for biomedical applications. Eur. J. Pharm. Biopharm. 2004;57:19-34.
14. Park YD, Tirelli N, Hubbell JA. Photopolymerized hyaluronic acid-based hydrogels and interpenetrating networks. Biomaterials 2003;24:893-900.
15. Masters KS, Shah DN, Leinwand LA, Anseth KS. Crosslinked hyaluronan scaffolds as a biologically active carrier for valvular interstitial cells. Biomaterials 2005;26:2517-2525.
16. Anseth KS, Metters AT, Bryant SJ, Martens PJ, Elisseef JH, Bowman CN. In situ forming degradable networks and their application in tissue engineering and drug delivery. J. Control. Release 2002;78:199-209.
17. Cohen S, Lobel E, Trevgoda A, Peled Y. A novel in situ-forming ophtalmic drug delivery system for alginates undergoing gelation in the eye. J. Control. Release 1997;44:201-208.
18. Rowley JA, Madlambayan G, Mooney DJ. Alginate hydrogels as synthetic extracellular matrix materials. Biomaterials 1999;20:45-53.
19. Ruel-Gariepy E, Leroux J-C. In situ-forming hydrogels--review of temperature-sensitive systems. Eur. J. Pharm. Biopharm. 2004;58:409-426.
20. Cohn D, Sosnik A, Garty S. Smart hydrogels for in situ generated implants. Biomacromolecules 2005;6:1168-1175.
21. Lee J, Bae YH, Sohn YS, Jeong B. Thermogelling aqueous solutions of alternating multiblock copolymers of poly(l-lactic acid) and poly(ethylene glycol). Biomacromolecules 2006;7:1729-1734.
22. Ganguly S, Dash AK. A novel in situ gel for sustained drug delivery and targeting. Int. J. Pharm. 2004;276:83-92.
23. Ismail FA, Napaporn J, Hughes JA, Brazeau GA. In situ gel formulations for gene delivery: release and myotoxicity studies. Pharm. Dev. Technol. 2000;5:391-397.
24. Li J, Li X, Ni X, Wang X, Li H, Leong KW. Self-assembled supramolecular hydrogels formed by biodegradable PEO-PHB-PEO triblock copolymers and [alpha]-cyclodextrin for controlled drug delivery. Biomaterials 2006;27:4132-4140.
25. Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Self-gelling hydrogels based on oppositely charged dextran microspheres. Biomaterials 2005;26:2129-2135.
26. De Jong SJ, De Smedt SC, Wahls MWC, Demeester J, Kettenes-van den Bosch JJ, Hennink WE. Novel self-assembled hydrogels by stereocomplex formation in aqueous solution of enantiomeric lactic acid oligomers grafted to dextran. Macromolecules 2000;33:3680-3686.
Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres
137
27. Hiemstra C, Zhong ZY, Van Tomme SR, Hennink WE, Dijkstra PJ, Feijen J. Protein release from injectable stereocomplexed hydrogels based on PEG-PDLA and PEG-PLLA star block copolymers. J. Control. Release 2006;116:e19-e21.
28. De Jong SJ, van Eerdenbrugh B, van Nostrum CF, Kettenes-van den Bosch JJ, Hennink WE. Physically crosslinked dextran hydrogels by stereocomplex formation of lactic acid oligomers: degradation and protein release behavior. J. Control. Release 2001;71:261-275.
29. Bos GW, Jacobs JJL, Koten JW, Van Tomme SR, Veldhuis TFJ, van Nostrum CF, Den Otter W, Hennink WE. In situ crosslinked biodegradable hydrogels loaded with IL-2 are effective tools for local IL-2 therapy. Eur. J. Pharm. Sci. 2004;21:561-567.
30. Bos GW, Hennink WE, Brouwer LA, den Otter W, Veldhuis TFJ, van Nostrum CF, van Luyn MJA. Tissue reactions of in situ formed dextran hydrogels crosslinked by stereocomplex formation after subcutaneous implantation in rats. Biomaterials 2005;26:3901-3909.
31. Slager J, Domb AJ. Biopolymer stereocomplexes. Adv. Drug Deliver. Rev. 2003;55:549-583.
32. Tsuji H. Poly(lactide) stereocomplexes: formation, structure, properties, degradation and applications. Macromol. Biosci. 2005;5:569-597.
33. van Dijk-Wolthuis WNE, Tsang SKY, Kettenes-van den Bosch JJ, Hennink WE. A new class of polymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer 1997;38:6235-6242.
34. Franssen O, Hennink WE. A novel preparation method for polymeric microparticles without the use of organic solvents. Int. J. Pharm. 1998;168:1-7.
35. Stenekes RJH, Franssen O, van Bommel EMG, Crommelin DJA, Hennink WE. The preparation of dextran microspheres in an all-aqueous system: effects of the formulation parameters on particle characteristics. Pharm. Res. 1998;15:557-561.
36. De Jong S, van Dijk-Wolthuis WNE, Kettenes-van den Bosch JJ, Schuyl PJW, Hennink WE. Monodisperse enantiomeric lactic acid oligomers: preparation, characterization, and stereocomplex formation. Macromolecules 1998;31:6397-6402.
37. Cohen S, Yoshioka T, Lucarelli M, Hwang LH, Langer R. Controlled delivery systems for proteins based on poly(lactic/glycolic) acid micropsheres. Pharm. Res. 1991;8:713-720.
38. Savitzky A, Golay MJE. Smoothing and differentiation of data by simplified least squares procedures. Anal. Chem. 1964;36:1627-1639.
39. Socrates G. Infrared and Raman characteristic group frequencies. 3rd ed.; John Wiley & Sons: New York, 2004.
40. Vlugt-Wensink KDF, Jiang X, Schotman G, Kruijtzer G, Vredenberg A, Chung JT, Zhang Z, Versluis C, Ramos D, Verrijk R, Jiskoot W, Crommelin DJA, Hennink WE. In Vitro Degradation Behavior of Microspheres Based on Cross-Linked Dextran. Biomacromolecules 2006;7:2983-2990.
41. Briggs DJ, Grant JT. Surface analysis by Auger and X-ray photoelectron spectroscopy. Chicester : IM Publications 2003.
42. Wagner CD, Riggs WM, Davis LE, Moulder JF. Handbook of X-Ray photoelectron spectroscopy. 1st ed.; Perkin-Elmer Corporation (Physical Electronics): Eden Prairie, 1979.
43. Gijzeman OLJ, Mens AJM, van Lenthe JH, Mortier WJ, Weckhuysen BM. The effect of chemical composition and structure on XPS binding energies in zeolites. J. Phys. Chem. B 2003;107:678-684.
9 Possible applications and future perspectives
Chapter 9
140
Possible applications and future perspectives
141
Hydrogels have been used for a variety of purposes ranging from contact lenses, diapers and
wound dressings to transdermal patches and other controlled drug delivery devices. Their inherent hydrophilicity makes them especially attractive for controlled protein release and tissue engineering applications. In this thesis, novel self-assembling hydrogels are described, which are
based on physical interactions between dextran microspheres. Proteins or cells can be entrapped in the matrix by simple mixing, thereby avoiding the use of any harsh crosslinking conditions.
Various options were provided in this thesis to control and tailor the network properties, release profile and degradation behavior of the hydrogels.
The effect of the particle size on the strength of the microsphere-based hydrogels was described in Chapter 6. From the results presented in this chapter it was concluded that smaller
microspheres with a relatively narrow size distribution gave rise to stronger hydrogels when compared to larger microspheres. When positively and negatively charged microspheres with
comparable sizes were used, the highest gel strength was obtained with equal amounts of both microsphere types. However, when small particles of 5 μm with large particles of 9 μm were
combined, the strongest hydrogels were obtained when the ratio small/large microspheres was 3. Since these particles still had a rather broad size distribution, it can be anticipated that a better
control over the particle size and size distribution will provide more options to further fine-tune and tailor the network properties. A number of procedures for the preparation of monodisperse particles have been developed. Berkland et al. reported on the production of poly(D,L-lactide-co-
glycolide) (PLGA) microspheres with precisely controlled size and monodisperse size distribution.[1] PLGA, dissolved in an organic solvent was sprayed through a nozzle to form a cylindrical jet.
Acoustic excitations were subsequently used to break the jet into uniform droplets that were collected in an aqueous poly(vinyl alcohol) (PVA) solution. Microspheres of 30 μm were hardened
by extraction and evaporation of the solvent. By combining the acoustic vibrations with an annular carrier stream of PVA solution around the polymer jet, the particle size could be tailored from 5 to 500 μm. Additionally, the size distributions could be predefined by varying the experimental
parameters (e.g. carrier stream flow, nozzle size). All microspheres had a very narrow size distribution with > 95 % within 1-2 μm of the average size. In a subsequent study it was shown
that precise control of the microsphere size provides better control over the release kinetics of encapsulated drugs.[2] Böhmer et al. reported on the use of an ink-jet technology to produce
uniformly sized PLGA particles.[3] A piezo-driven printing head was submerged into an aqueous phase creating well-defined polymer particles, and even hollow capsules were prepared using this
technology. The particle size could be controlled by the nozzle diameter. Both systems described above are very suitable for the production of PLGA particles with precisely controlled size and monodisperse size distribution. With some adjustments they might also be applicable for the
preparation of microgels, such as dextran-based microspheres. Recently, De Geest et al. described the preparation of monodisperse hydroxyethyl methacrylate-
modified dextran (dex-HEMA) microspheres in a microfluidic device.[4] A dex-HEMA solution, to which a photoinitiator was added, was emulsified in a mineral oil via separate microchannels that
met in a T-junction, thereby creating droplets via tip-streaming. The droplets were cured by UV-irradiation while still in the oil-phase, resulting in almost monodisperse microspheres with a size of
9.9 ± 0.3 μm. At present, however, there are 2 major drawbacks associated with this technology. Firstly, the use of a poly(ethylene glycol) (PEG) solution as continuous phase appeared to be
Chapter 9
142
impossible because of a lack of sufficient shear between the 2 phases upon mixing. Therefore, De
Geest et al. had to extensively wash the microspheres with organic solvents to remove the mineral oil, which can be detrimental when proteins are encapsulated in the particles. Secondly, the production yield is very low as only ~0.15 mg of microgel particles was collected per hour.
However, microfluidics is a very promising method for the development of monodisperse dex-HEMA microspheres and it would definitely be interesting to invest in optimization and scaling up
of this technique.
For the further development of the above described self-assembling microsphere-based hydrogels, described in this thesis, into the clinic, some general issues need to be tackled, i.e. upscaling, sterility and safety. Large scale preparation of clinical grade dex-HEMA microspheres has
been described previously.[5] The microspheres were prepared at aseptic conditions and lyophilized. Preparation of charged dex-HEMA microspheres does not require any dramatical
changes in the setup. Additional cleaning steps would be needed if the same reactor is used for neutral, positively and negatively charged microspheres.
Aseptic preparation of the hydrogel components circumvents several safety concerns regarding bacterial contamination and endotoxin load. Importantly, the biocompatibility and possible
cytotoxicity of the hydrogel and its degradation products need to be studied. Certain cell viability tests have already been performed and are reported in section 2. The major applications that can be foreseen for the self-gelling systems described in this thesis are outlined and briefly discussed in
the next sections.
1. Controlled release of pharmaceutically active proteins
The tremendous progress of biotechnology the past 30 years has resulted in a whole new generation of bioactive proteins.[6] Consequently, there is an increasing need for suitable delivery methods of these protein pharmaceuticals.[7] The main advantages of controlled delivery systems
in general include: (1) protection of the fragile drug molecules against enzymatic or chemical degradation, (2) prolonged and constant therapeutic blood levels, (3) reduction of drug toxicity, (4)
reduced administration frequency, contributing to improved patient compliance. On the other hand, it has to be taken into account that the release devices (or their degradation products) can
provoke immune reactions or local discomfort for the patient (e.g. implants). Additionally, release of high drug doses, as a result of a burst effect, could induce severe toxicity.[8] During the
development process of each delivery system, not only the above mentioned concerns need to be taken into consideration, a number of critical issues such as production costs and scalability will ultimately determine whether a delivery system will enter the clinic.[7] Several advanced drug
delivery systems have reached the market, of which most are liposome-based.[9] Other suitable delivery systems include polymeric micelles, nanospheres, microspheres and hydrogels. However,
focusing on protein delivery in particular, of all the delivery methods described in literature, only a select group has been tested in vivo and only a few have entered human clinical trials.[7]
The soft and rubbery nature of hydrogels makes them especially attractive candidates for protein delivery application. It is beneficial if the hydrogel is formed in situ by means of physical crosslinking, avoiding conditions that are possibly toxic to cells or cause denaturation of the
Possible applications and future perspectives
143
entrapped proteins. The novel self-assembling hydrogels described in this thesis have shown to be
suitable for the diffusion- and degradation-controlled release of model compounds. Importantly, the activity of lysozyme was preserved during encapsulation and release. Next, in vivo release experiments with therapeutic relevant proteins need to be carried out. One could think of
cytokines (e.g. interleukins, interferons), growth factors (e.g. transforming growth factor, bone morphogenic proteins), hormones (e.g. human growth hormone (hGh), insulin) etc. that need to
be released in a controlled or sustained fashion. Recently, a good in vitro - in vivo correlation was described for the release of hGh from dex-HEMA microspheres,[10] whcih is important to predict the
clinical performance. Studying the in vitro - in vivo correlation of the microsphere-based dextran hydrogels would be of tremendous importance for the future of these systems in protein delivery.
2. Tissue engineering
A rising problem in health care is tissue loss or organ failure due to injury or disease (e.g. cancer, burns, ischemia). The standard therapy to treat such conditions is tissue or organ transplantation.
However, there is a severe donor shortage and waiting lists are increasing dramatically.[11, 12] An emerging strategy to overcome these obstacles is tissue engineering. Tissue engineering is, as defined by Langer et al., ‘an interdisciplinary field that applies the principles of engineering and life
sciences toward the development of biological substitutes that restore, maintain, or improve tissue function’.[13] It involves the culturing of cells in polymeric scaffolds that serve as a synthetic
extracellular matrix that organizes the cells in a three-dimensional architecture. Most used polymers for the production of scaffolds are based on PLGA and poly(lactic acid) (PLA). These
polymers are FDA approved for some pharmaceutical applications and provide strong mechanical support to the seeded cells. Their hydrophobicity and relatively harsh processing conditions,
however, limit the possibilities to entrap viable cells.[14] Highly hydrated hydrogels are attractive candidates for the design of scaffolds because of their biocompatibility and controllable degradability.[11, 14, 15] The hydrogels described in this thesis might possess the appropriate
mechanical properties for soft-tissue engineering (e.g. cartilage repair, partial organ replacement). To assess their potential as cell supporting matrices, introductory in vitro cytotoxicity tests were
performed.
2.1. Cytotoxicity of self-assembling microsphere-based dextran hydrogels
Microspheres were prepared under aseptic conditions. The cytotoxicity of dex-HEMA, dex-HEMA-MAA and dex-HEMA-DMAEMA microspheres as well as of a combination of the latter two was
assessed by means of a 5-(3-carboxymethoxyphenyl)-2-(4,5-dimethylthiazolyl)-3-(4-sulfophenyl)tetrazolium (inner salt) (MTS) extraction assay by Maike Wubben in the lab of Prof.
M.J.A. van Luyn (Medical Biology: Stem Cell & Tissue Engineering, Department of Pathology and Laboratory Medicine, University Medical Center Groningen, The Netherlands) (unpublished
observations). The MTS assay is a colorimetric approach to measure dehydrogenase enzyme activity, found in metabolically active cells.[16] The microspheres were incubated in RPMI medium
for 24 h at 37 °C under shaking conditions after which an extract was added to PK 84 cells (human
skin fibroblasts) and cytotoxicity was determined after 48 h. The percentage of proliferating cells is
Chapter 9
144
presented in Figure 1. The highest cellular activity was obtained after exposure of the cells to
extracts of ‘neutral’ dex-HEMA microspheres, which is in agreement with a previous study.[17] The cell proliferation was decreased after incubation of the cells with extracts of positively charged dex-HEMA-DMAEMA microspheres. Contrary, extracts of negatively charged dex-HEMA-MAA
microspheres alone, or in combination with dex-HEMA-DMAEMA microspheres, allowed 60 % cell proliferation. The cytotoxicity of the dex-HEMA-DMAEMA microspheres might result from the
presence of pDMAEMA fragments that possibly leaked out of the microspheres during incubation with medium and that have shown some cytotoxicity.[18]
PU latex + - +/- neutral0
102030405060708090
100110
prol
ifera
tion
(%)
Figure 1: Cell proliferation (%) after 48 h incubation of the cells with extracts of positively (+), negatively (-), a
combination of both (+/-) or neutral (= dex-HEMA) microspheres. As a positive and negative control,
polyurethane (PU) and latex were used, respectively (n=6).
Although these experiments indicate possible toxicity of the charged hydrogels, it is further important to study the viability of cells that are embedded in a hydrogel matrix. The following tests
were performed in collaboration with Dr. W. Dhert (Department of Orthopaedics, University Medical Center Utrecht, The Netherlands).
Dex-HEMA-MAA/dex-HEMA-DMAEMA microspheres (prepared at aseptic conditions) and dex-HEMA-L-lactate/dex-HEMA-D-lactate microspheres were sterilized by UV-irradiation for 2 h. Hydrogels (20 % solid (w/v) for dex-HEMA-MAA/dex-HEMA-DMAEMA and 15 % solid (w/v) for dex-
HEMA-L-lactate/dex-HEMA-D-lactate microsphere-gels) were prepared with osteogenic medium and subsequently mixed with goat bone marrow stromal cells (gBMSC) (440000 cells/ml gel).
Subsequently, 50 μl of each of these mixes were transferred into 96-wells plates and incubated with osteogenic medium that was refreshed every 48 h. After 1, 3 and 7 days, the medium was
removed and the gels were incubated with LIVE/DEAD® Viability/Cytotoxicity kit (Molecular Probes), which is based on the intracellular esterase activity and the plasma membrane integrity, fluorescently staining live cells green and dead cells red. Next, the hydrogels were analyzed with
fluorescence microscopy. As a control, the cells were cultured on tissue culture polystyrene.
Possible applications and future perspectives
145
control -/+ gel L/D gel0
102030405060708090
100
day 3day 7
day 1
cell
viab
ility
(%)
Figure 2: Cell viability of gBMSC after 1, 3 and 7 days of incubation with tissue culture polystyrene (left), dex-
HEMA-MAA/dex-HEMA-DMAEMA microspheres (middle) and dex-HEMA-L-lactate/dex-HEMA-D-lactate
microspheres (right).
Figure 2 shows the viable cells as percentage of total cells after 1, 3 and 7 days of encapsulation in the hydrogels. Compared to the control material, a slight decrease in cell viability was observed
after one day of incubation with the gels, for both microsphere-based types. After 3 and 7 days, a gradual decrease in cell survival occurred, which was more pronounced for the charged than for
the stereocomplexed hydrogels. Although these results might look discouraging, the following needs to be taken into account. It was reported by Nuttelman et al. that cell viability in PEG hydrogels, generally considered as fully biocompatible, was very low. Less than 10 % survival of
human mesenchymal stem cells (hMSC) was observed after embedment in the hydrogels.[19] This low survival was ascribed to the fact that cells have a low tendency to adhere to the highly
hydrated hydrogel surfaces. This absence of cell-matrix interactions can have a striking effect on the cell viability, differentiation, etc.[20] It was postulated that binding of cell-surface integrins to
integrin receptors on the extracellular matrix induces apoptosis-suppressing signals.[21] As a result, in hydrogels, providing an environment where no strong matrix-cell interactions occur, cell survival
will be low. Cell viability could be greatly enhanced by addition of specific functionalities, such as arginine-glycine-aspartic acid (RGD) peptide sequences, to the hydrogel matrix. Binding of the RGD sequence to integrins on cell surfaces, is known to support cell adhesion.[21] Nuttelman et al.
modified PEG hydrogels with pendant RGD sequences and found a significant increase in cell viability from 10 to 75 %.[20] It is therefore anticipated that modification of the microsphere-based
dextran hydrogels with cell-adhesion promoting ligands might greatly enhance cell survival and, as a result, their applicability as cell-supporting scaffold.
In Chapter 7 of this thesis, it was found that DMAEMA can be replaced by hydrolysis-sensitive HPMA-based monomers, resulting in cationic microspheres with different degradation rates. The HPMA-based polymers have shown to have a lower cytotoxicity than pDMAEMA.[22, 23] It would
therefore be very interesting to investigate the cytotoxicity of hydrogels in which the positive microspheres contain HPMA-based polymers instead of pDMAEMA.
Chapter 9
146
2.2. Combining self-assembling microsphere-based dextran hydrogels with bioactive calcium phosphate (CaP) ceramics
Besides for soft-tissue engineering, bone replacement therapy is a very interesting research area where hydrogels might find their application. Bone replacement can be achieved with biotolerant
(e.g. stainless steel), bioinert (e.g. titanium oxide) or bioactive materials (e.g. calcium phosphate (CaP) ceramics).[24] The latter can induce so called ‘bone-bonding’ or interaction between the implant and the bone tissue.[25] More specifically, an ion-exchange between the bioactive implant
and surrounding body fluids results in the deposition of a hydroxyapatite layer on the implant surface, chemically and crystallographically equivalent to the mineral phase in bone. Eventually,
complete fusion between the implant and bone tissue is achieved via a mechanically solid interface, in the absence of a fibrous tissue layer.[26]
CaP bulk ceramics have been used as bone filler in orthopedics and dentistry.[27] A drawback of CaP ceramics is their relative mechanical weakness and brittleness. To exploit their favorable bioactivity,
they are used as coatings on mechanically strong implant materials.[26] CaP nanocrystals in suspension can also be used as injectable filling of bony voids to aid in bone regeneration but dislocation of crystals into the surrounding tissue should be excluded. Presumably, this can be
achieved by entrapping the CaP nanocrystals in a biodegradable matrix. For this purpose, in collaboration with Prof. J. Jansen and Dr. S. Leeuwenburgh (Department of Periodontology and
Biomaterials, Radboud University Nijmegen Medical Center, The Netherlands) a CaP suspension was mixed with positively and negatively charged dextran microspheres and rheological analysis
of the composites was performed (Fig. 3A). The amount of CaP nanoparticles in the hydrogels was varied between 0 and 12%. Figure 3A shows a steep increase of the gel strength when CaP was
added to the microspheres. Furthermore, the composites showed to be mainly elastic with tan(δ)
< 0.1. Interestingly, cationic or anionic microspheres alone, mixed with CaP nanocrystals also showed gel formation with comparable strength and elasticity as a mixture of oppositely charged
microspheres with CaP. This can probably be ascribed to the cementing effect of the CaP suspension (Fig. 3B). This pilot study shows the potential of the microsphere-based dextran
hydrogels combined with bioactive CaP nanocrystals as biodegradable putty, moldable to fill specific defects in bone replacement therapy.
0 2 4 6 8 10 12 140
1020304050607080
% CaP
G' (
kPa)
Figure 3: (A) Storage modulus (G’) of hydrogels composed of equal amounts of dex-HEMA-MAA and dex-
HEMA-DMAEMA (20 % w/w) to which various amounts of a CaP nanocrystal suspension were added (n=3). (B)
SEM image of oppositely charged dex-HEMA microspheres mixed with CaP nanocrystals.
B A
Possible applications and future perspectives
147
3. Protein releasing hydrogel scaffolds
It has been shown that hydrogels are very suitable as protein delivery devices and as tissue
engineering scaffolds. Recently, there has been an increasing interest in combining these features in a multi-functional matrix in which cells are entrapped.[28-30] Ferreira et al. reported on a
photopolymerized dextran-based hydrogel, functionalized with RGD sequences, in which human embryonic stem cells (hESC) were embedded.[31] Additionally, vascular endothelial growth factor
(VEGF)-loaded PLGA microspheres were entrapped in the hydrogel matrix. In line with other publications, it was found that modification of the hydrogel with RGD peptides promoted hESC attachment and spreading on the hydrogel surface. Furthermore, an enhanced vascular
differentiation of hESC was observed, confirming the bioactivity of the scaffolds. Park et al. reported on injectable oligo(poly(ethylene glycol) fumarate) (OPF) hydrogels.[32] Both
rabbit marrow mesenchymal stem cells (rMSC) and gelatin microspheres, loaded with
transforming growth-factor-β1 (TGF-β1), were entrapped in the hydrogels. An upregulation of
cartilage-relevant genes was observed when rMSC were encapsulated together with TGF-β1
containing microspheres, confirming their differentiation into chondrocyte-like cells. This effect was not seen in ‘blank’ OPF hydrogels.
The systems described above have obvious advantages compared to plain hydrogel scaffolds. As mentioned in section 2.1, modification of hydrogels with specific cell-binding peptide sequences,
strongly enhances the cell viability and proliferation and can promote differentiation. Furthermore, differentiation of stem cells, embedded in the hydrogel matrix can be favored by the presence of
particular growth factors. So far, various polymeric microspheres have been embedded in hydrophilic cell-encapsulating matrices to obtain these effects. The self-assembling microsphere-
based dextran hydrogels, reported in this thesis, offer the possibility to create an ‘all-in-one’ protein releasing scaffold: growth factors can be entrapped inside the dex-HEMA microspheres, while cells
are embedded in the macroscopic network. Release of proteins from biodegradable dex-HEMA microspheres has been extensively studied.[10, 33] Various options were provided to tailor the release kinetics (e.g. crosslink density, water content). Additionally, in this thesis it was found that the
charge of the microspheres also affects the release profile of entrapped drugs. The properties of the macroscopic hydrogel can be influenced to a great extent by altering for instance the solid
content, charge density, particle size, charged monomer type, etc. Given the wide-ranging tailorability of these novel hydrogels, it can be foreseen that they will
conquer their place in this exciting and rapidly-growing research field that combines protein delivery with tissue engineering.
Chapter 9
148
References
1. Berkland C, Kim K, Pack DW. Fabrication of PLG microspheres with precisely controlled and monodisperse size distributions. J. Control. Release 2001;73:59-74.
2. Berkland C, King M, Cox A, Kim KK, Pack DW. Precise control of PLG microsphere size provides enhanced control of drug release rate. J. Control. Release 2002;82:137-147.
3. Böhmer MR, Schroeders R, Steenbakkers JAM, de Winter SHPM, Duineveld PA, Lub J, Nijssen WPM, Pikkemaat JA, Stapert HR. Preparation of monodisperse polymer particles and capsules by ink-jet printing. Colloid Surface A 2006;289:96-104.
4. De Geest BG, Urbanski JP, Thorsen T, Demeester J, DeSmedt SC. Synthesis of monodisperse biodegradable microgels in microfluidic devices. Langmuir 2005;21:10275-10279.
5. Vlugt-Wensink K. Dextran-based microspheres as controlled delivery systems for proteins. 2007. Utrecht University. Department of Pharmaceutics.
6. Pavlou A, Reichert J. Recombinant protein therapeutics - success rates, market trends and values to 2010. Nat. Biotechnol. 2004;22:1513-1519.
7. Cleland JL, Daugherty A, Mrsny R. Emerging protein delivery methods. Curr. Opin. Biotechnol. 2001;12:212-219.
8. Langer R. Drug delivery and targeting. Nature 1998;392:5-10.
9. Allen TM, Cullis PR. Drug delivery systems: entering the mainstream. Science 2004;303:1818-1822.
10. Vlugt-Wensink KDF, de Vrueh R, Gresnigt MG, Hoogerbrugge CM, van Buul-Offers SC, de Leede LGJ, Sterkman LGW, Crommelin DJ, Hennink WE, Verrijk R. Preclinical and clinical in vitro in vivo correlation of an hGH dextran microsphere formulation. 2007 submitted.
11. Lee KY, Mooney DJ. Hydrogels for tissue engineering. Chemical Reviews 2001;101:1869-1879.
12. Chen G, Ushida T, Tateishi T. Scaffold design for tissue engineering. Macromol. Biosci. 2002;2:67-77.
13. Langer R, Vacanti JP. Tissue engineering. Science 1993;260:920-926.
14. Drury JL, Mooney DJ. Hydrogels for tissue engineering: scaffold design variables and applications. Biomaterials 2003;24:4337-4351.
15. Fedorovich NE, Alblas J, De Wijn JR, Hennink WE, Verbout AJ, Dhert WJA. Hydrogels as extracellular matrices for skeletal tissue engineering: state of the art and novel application in organ printing - a review. Tissue Eng. 2007 in press.
16. Barltrop JA, Owen TC, Cory AH, Cory JG. 5-(3-carboxymethoxyphenyl)-2-(4,5-dimethylthiazolyl)-3-(4-sulfophenyl)tetrazolium, inner salt (MTS) and related analogs of 3-(4,5-dimethylthiazolyl)-2,5-diphenyltetrazolium bromide (MTT) reducing to purple water-soluble formazans As cell-viability indicators. Bioorg. Med. Chem. Lett. 1991;1:611-614.
17. De Groot CJ, Van Luyn MJA, Van Dijk-Wolthuis WNE, Cadee JA, Plantinga JA, Den Otter W, Hennink WE. In vitro biocompatibility of biodegradable dextran-based hydrogels tested with human fibroblasts. Biomaterials 2001;22:1197-1203.
18. Van de Wetering P, Moret EE, Schuurmans-Nieuwenbroek NME, van Steenbergen MJ, Hennink WE. Structure-activity relationships of water-soluble cationic methacrylate/methacrylamide polymers for nonviral gene delivery. Bioconjugate Chem. 1999;10:589-597.
19. Nuttelman CR, Tripodi MC, Anseth KS. In vitro osteogenic differentiation of human mesenchymal stem cells photoencapsulated in PEG hydrogels. J. Biomed. Mater. Res. A 2004;68A:773-782.
20. Nuttelman CR, Tripodi MC, Anseth KS. Synthetic hydrogel niches that promote hMSC viability. Matrix Biol. 2005;24:208-218.
21. Ruoslahti E, Reed J. Anchorage dependence, integrins, and apopthosis. Cell 1994;77:477-478.
22. Funhoff AM, van Nostrum CF, Lok MC, Kruijtzer JAW, Crommelin DJA, Hennink WE. Cationic polymethacrylates with covalently linked membrane destabilizing peptides as gene delivery vectors. J. Control. Release 2005;101:233-246.
23. Luten J, Akeroyd N, Funhoff A, Lok MC, Talsma H, Hennink WE. Methacrylamide polymers with hydrolysis-sensitive cationic side groups as degradable gene carriers. Bioconjugate Chem. 2006;17:1077-1084.
24. Osborn JF, Newesely H. Dynamic aspects of implant/bone interface. In Dental implants, G. H, Ed. Carl Hansen Verlag: Munich, 1980; pp 111-123.
25. Ducheyne P, Bianco P, Radin S, Schepers E. Bioactive materials: mechanisms and bioengineering considerations. In Bone-bonding biomaterials, Ducheyne P, van Blitterswijk CA and Kokubo T, Eds. Reed Healthcare Communications: Leiderdorp, 1993; pp 1-12.
26. Leeuwenburgh S. Electrosprayed calcium phosphate coatings for biomedical purposes. 2006. Radboud University Nijmegen Medical Center. Department of Biomaterials.
27. Schmitz JP, Hollinger JO, Milam SB. Reconstruction of bone using calcium phosphate bone cements: a critical review. J. Oral Maxillofac. Surg. 1999;57:1122-1126.
Possible applications and future perspectives
149
28. Kretlow JD, Klouda L, Mikos AG. Injectable matrices and scaffolds for drug delivery in tissue engineering. Adv. Drug Deliver. Rev. 2007;59:263-273.
29. Sokolsky-Papkov M, Agashi K, Olaye A, Shakesheff K, Domb AJ. Polymer carriers for drug delivery in tissue engineering. Adv. Drug Deliver. Rev. 2007;59:187-206.
30. Tessmar JK, Gopferich AM. Matrices and scaffolds for protein delivery in tissue engineering. Adv. Drug Deliver. Rev. 2007;59:274-291.
31. Ferreira LS, Gerecht S, Fuller J, Shieh HF, Vunjak-Novakovic G, Langer R. Bioactive hydrogel scaffolds for controllable vascular differentiation of human embryonic stem cells. Biomaterials 2007;28:2706-2717.
32. Park H, Temenoff JS, Tabata Y, Caplan AI, Mikos AG. Injectable biodegradable hydrogel composites for rabbit marrow mesenchymal stem cell and growth factor delivery for cartilage tissue engineering. Biomaterials 2007;28:3217-3227.
33. Franssen O, Vandervennet L, Roders P, Hennink WE. Degradable dextran hydrogels: controlled release of a model protein from cylinders and microspheres. J. Control. Release 1999;60:211-221.
10
Summary
Chapter 10
152
Summary
153
Wichterle et al. in 1960 were the first to describe the use of hydrophilic polymeric networks for the
design of soft contact lenses.[1] Later on, hydrogels have been used in a variety of applications, ranging from food additives to surgical sutures, wound dressing materials, tissue engineering scaffolds and drug delivery devices.[2-6] Their hydrophilic, soft and rubbery nature ensures minimal
tissue irritation and a low tendency of cells and proteins to adhere to the hydrogel surface. Chapter 1 provides a general introduction to hydrogels, their interesting features and the possible
applications in the biopharmaceutical world. Additionally, the aim of this thesis is outlined.
A variety of polymers, both natural and synthetic, has been used for the development of hydrogel systems.[7] Synthetic polymers generally lack an inherent bioactivity that supports cellular activities,
but they are well defined as it concerns degradability and mechanical properties. Additionally, the network properties can be tailored to a great extent by the building blocks. Natural polymers have the advantage of containing biologically recognizable moieties but depending on their origin,
they might be contaminated with pathogens, possibly eliciting an immune response. Furthermore, natural polymers may exhibit batch to batch variability and the corresponding hydrogels may not
provide acceptable mechanical properties for many applications. Of the numerous polymers proposed and designed for hydrogel formulations, polysaccharides are of particular interest
because of their biocompatibility and versatile physico-chemical properties.[8] In this thesis, research is focused on derivatives of the glucose homopolysaccharide dextran. Chapter 2 reviews
the current state of the art of dextran-based hydrogels for protein delivery applications. A summary is given of dextran derivatives and the corresponding crosslinking methods used or
developed in the past 5 to 10 years. Both chemically as well as physically crosslinked systems are discussed, focusing on network properties, protein delivery characteristics, degradation behavior and biocompatibility. Special attention was given to injectable, in situ forming hydrogels, since this
is the subject of this thesis.
In Chapter 3 a novel self-gelling hydrogel is presented based on physical interactions between
hydroxyethyl methacrylate-derivatized dextran (dex-HEMA) microspheres. The microspheres were
prepared via a water-in-water emulsion of dex-HEMA in poly(ethylene glycol) (PEG) and subsequent polymerization of the inner phase. Macroscopic gels were obtained by mixing
dispersions of oppositely charged microspheres, which in turn were prepared by addition of methacrylic acid (MAA) or dimethylaminoethyl methacrylate (DMAEMA) to the polymerization mixture. Rheological analysis showed that instantaneous gelation occurred when equal volumes of
negatively charged (dex-HEMA-MAA) and positively charged (dex-HEMA-DMAEMA) microspheres, dispersed in buffer solutions of pH 7, were mixed. The shear modulus of the mainly elastic
hydrogels could be tailored by varying the water content of the system. It was shown that the network structure could be broken by either a low pH or a high ionic strength of the medium,
confirming the charge interactions between the microspheres at pH 7, creating the macroscopic network. Important for possible applications, e.g. as injectable matrix for protein delivery or tissue
engineering, it was shown that the hydrogel system had a reversible yield point, meaning that above a certain applied stress, the system starts to flow, whereas when the stress is removed, the network is rebuilt.
Chapter 10
154
The feasibility of the ionically crosslinked hydrogels, described in Chapter 3 to deliver proteins in a
controlled fashion, with full preservation of their activity, is investigated in Chapter 4. Protein-
loaded macroscopic gels were prepared by hydration of mixtures of oppositely charged
microspheres with a protein solution. In this way, the proteins are entrapped in the pores between the microspheres. The release of 3 model proteins (lysozyme, BSA and IgG) was studied, both in saline (Hepes buffer 100 mM pH 7.0) as well as in media of low ionic strength (Hepes buffer
100 mM pH 7.0 150 mM NaCl). Confocal images showed that BSA, bearing a negative charge at neutral pH and low ionic strength, adsorbed onto the surface of dex-HEMA-DMAEMA
microspheres, whereas lysozyme, which is positively charged at neutral pH, was able to penetrate into the dex-HEMA-MAA microspheres. These findings explained the slower release rate of the
entrapped proteins in media of low ionic strength (50 % lysozyme and BSA release in 7 and 17 days, respectively), compared to saline (50 % lysozyme and BSA release in 4 and 6 days, respectively). In saline, the gels showed continuous release of the different proteins for 25 to
60 days. Importantly, lysozyme was quantitatively and with full preservation of its enzymatic activity released in about 25 days. This emphasizes the protein friendly technology to prepare the
protein-loaded gels. Mathematical modeling revealed that protein release followed Fick’s second law, indicative of a primarily diffusion-controlled release process. Finally, fluorescence recovery
after photobleaching (FRAP) was used to gain insight into the mobility of proteins in the complex network of charged polymeric microspheres. Lysozyme and BSA, dissolved in dispersions
containing either non-interacting positively charged (lysozyme) or negatively charged microspheres (BSA), showed diffusion coefficients similar to those in water, demonstrating that the protein does not interact with the dextran microspheres. Mixing of the proteins with hydrogels
composed of both types of microspheres resulted in diffusion coefficients that were significantly lower than those in water. In general, the diffusion coefficients obtained with FRAP were 5 to 20
times smaller than those calculated from the release experiments. These discrepancies likely resulted from the differences in experimental setup: release experiment on a macroscopic level,
FRAP are on a microscopic level. Overall, the results presented in Chapter 4 showed that these ionically crosslinked microsphere-based hydrogels are very suitable as matrix for diffusion-controlled delivery of proteins.
For most applications and especially for the delivery of macromolecular drugs and in tissue
engineering, biodegradability of the hydrogels is an important issue. Scaffolds for the development of new tissue need to provide mechanical support until sufficient extra-cellular
matrix has been formed. Tailoring of the degradation rate of hydrogels allows to specifically design systems for various applications.[9] In Chapter 5 the degradation behavior of the hydrogel matrices
composed of positively and negatively charged dextran microspheres is studied. Rheological analysis showed that, once the microspheres started to degrade, the hydrogel changed from a
mainly elastic tot a viscoelastic network. The degradation time of gels composed of equal amounts of cationic and anionic microspheres ranged from 65 to 140 days, depending on the crosslink density of the particles and the water content of the hydrogels. Positively charged, neutral and
negatively charged microspheres, dispersed in buffer (Hepes 100 mM pH 7.0) completely degraded within 30, 55 or 120 days, respectively. Confocal microscopy and fluorescence
spectroscopy of degrading microspheres, loaded with rhodamine-B-dextran revealed that after a lag time of 3 days the fluorescently labeled dextran started to release from the dex-HEMA-
Summary
155
DMAEMA microspheres with a 50 % release after 16 days. In contrast, release of rhodamine-B-
dextran from the dex-HEMA-MAA started after 10 days with a 50 % release after 36 days. The faster degradation of the cationic microspheres as compared to the anionic microspheres was attributed to 2 factors. Firstly, stabilization of the transition state in the hydrolysis process occurred by the
protonated tertiary amine groups present in the cationic microspheres. Secondly, the presence of negatively charged groups in the dex-HEMA-MAA microspheres caused repulsion of hydroxyl
anions resulting in a slower degradation. In conclusion, Chapter 5 shows that combining the oppositely charged microspheres in different ratios makes it possible to tailor the network
properties and the degradation behavior of these hydrogels. In Chapter 6 the research was focused on the influence of the size (distribution) and charge of the
ionic dextran microspheres on the network properties of the macroscopic gels. Additionally, the
injectability of the self-assembling hydrogels was studied. Microspheres with zeta (ξ)-potentials
from -6 to -34 mV and from +3 to +23 mV were prepared by varying the MAA/HEMA or DMAEMA/HEMA ratio, respectively, between 6 and 57 during polymerization of the microspheres.
Particles with various sizes and size distributions were obtained by sieving of polydisperse batches
or by altering the emulsifications parameters. Rheological analysis showed that an increasing ξ-
potential (absolute values) of the microspheres led to stronger hydrogels. The smallest
microspheres (7 μm) gave rise to stronger hydrogels when compared to microspheres of 20 μm
with a broad distribution. Taking both charge and size influence into account, the storage moduli (G’) of the almost fully elastic hydrogels could be tailored from 400 to 30000 Pa. Injectability tests
showed that hydrogels composed of equal amounts of oppositely charged microspheres (-6 and
+6 mV, average particle size 7 μm) could be injected through 25G needles using a static load of
15 N, an ISO accepted value. The results discussed in this chapter provide a variety of options to control the network properties of these microsphere-based hydrogels.
The positively charged dex-HEMA microspheres, used for the preparation of the self-assembling hydrogels described in Chapters 3 to 6 were prepared with DMAEMA as cationic monomer. In
Chapter 7, hydroxypropyl methacrylamide (HPMA)-based cationic monomers, i.e. HPMA-DMAE
(dimethylaminoethanol) and HPMA-DMAPr (dimethylaminopropanol) were used for the
preparation of the positively charged microspheres. Both monomers contain a hydrolysis-sensitive
carbonate ester leading to a gradual charge loss in time. The ξ-potential of microspheres
containing DMAEMA remained constant for over 1 week, while the microspheres containing
HPMA-DMAE and HPMA-DMAPr lost their positive ξ-potential after 1 and 2 days, respectively. As a
result, the macroscopic networks comprising equal amounts of positively and negatively charged
microspheres, disintegrated in time, due to a loss of the physical interactions between the microspheres. By altering the cationic monomer in the positive microspheres, the time span during which the gels lost their mechanical properties could be varied between 10 and 30 days.
A novel approach to create physically crosslinked microsphere-based hydrogels is described in
Chapter 8. Dex-HEMA microspheres were chemically modified with L- or D-oligolactate chains.
The actual grafting of the particles was confirmed by Fourier Transform-InfraRed (FT-IR)
spectroscopy, Raman spectroscopy and X-ray photoelectron spectroscopy (XPS). Rheological analysis of aqueous dispersions of oligolactate-grafted microspheres demonstrated that
Chapter 10
156
hydrophobic interactions between oligolactate chains grafted on the surface of various
microspheres resulted in the formation of an almost fully elastic gel. A mixture of microspheres substituted with L- or D-oligolactates of opposite chirality resulted in gels with highest strength, likely due to stereocomplexation between the enantiomers. Both the degree of substitution (DS,
i.e. number of lactate grafts per 100 glucopyranose units) and the degree of polymerization (DP) of the oligomers were varied and showed to influence the network properties to a great extent. The
results reported in this chapter show that besides ionic interactions, other physical crosslinking methods such as hydrophobic interactions or stereocomplexation, are good options for the design
of self-assembling hydrogels based on dextran microspheres. In Chapter 9 an overview is given of the possible applications and future perspectives of these
novel self-gelling systems. Because of its protein-friendly nature, the hydrogels are potentially very suitable for the delivery of pharmaceutically active proteins. In this thesis, only the release of model
proteins has been studied. Monitoring the release behavior of a pharmaceutically relevant protein is the next step to take. Eventually, in vivo studies need to be conducted to further develop clinical
applications of these injectable hydrogels. To assess the possibilities of the self-assembling microsphere-based dextran hydrogels, studied in
this thesis, in soft-tissue engineering applications, a number of preliminary cell-viability tests have been performed. It was found that the viability of goat bone marrow stromal cells (gBMSC) was
decreased by 85 and 50 % after 7 days for ionically crosslinked and stereocomplexed hydrogels, respectively, when compared to the control (tissue culture polystyrene). Further studies are needed to gain insight into the specific causes of cell death provoked by the gels. It is known from
literature that the generally good biocompatibility of hydrogels is mostly insufficient to guarantee cell viability, as many hydrogels do not provide an ideal environment for cell-adhesion.[10] Various
research groups described that this hurdle can be overcome by modification of the matrix with cell-adhesive peptides.[11, 12]
Additionally, some pilot experiments were performed to estimate the potential of the ionically crosslinked hydrogels for bone replacement. Promising results have been obtained using charged dextran microspheres in combination with negatively charged calcium phosphate (CaP)
nanocrystals (12 wt%), for which a twenty-fold increase of gel strength was observed, caused by the cementing effect of the CaP nanoparticles. In this way, dislocation of the CaP crystals within
the surrounding tissue is prevented by their entrapment in the hydrogel and this moldable ‘putty’ could be directly injected into bone defects.
In conclusion, this thesis describes a novel approach for the design of self-assembling hydrogels,
based on physical crosslinking between dextran microspheres. They provide an ‘all-in-one’ protein releasing hydrogel that can act as cell-supporting scaffold. The inherent biocompatibility of dextran-based hydrogels could likely be even more improved by specific modification of the
polymers. Numerous options to tailor the gel properties and release behavior have been outlined, making these systems attractive candidates in a variety of pharmaceutical applications. It can be
foreseen that these injectable matrices will find their use in a combination of cell entrapment and delivery of signaling proteins to promote new tissue growth.
Summary
157
References
1. Wichterle O, Lim D. Hydrophilic gels for biological use. Nature 1960;185:117-118.
2. Peppas NA. Hydrogels and drug delivery. Curr. Opin. Colloid In. 1997;2:531-537.
3. Peppas NA, Bures P, Leobandung W, Ichikawa H. Hydrogels in pharmaceutical formulations. Eur. J. Pharm. Biopharm. 2000;50:27-46.
4. Peppas NA, Hilt JZ, Khademhosseini A, Langer R. Hydrogels In biology and medicine: from molecular principles to biotechnology. Adv. Mater. 2006;18:1345-1360.
5. Clark AH, Ross-Murphy SB. Structural and mechanical properties of biopolymer gels. In Advances in Polymer Sciences, Springer: Berlin, 1987; Vol. 83, pp 57-192.
6. Kashyap N, Kumar N, Ravikumar MNV. Hydrogels for pharmaceutical and biomedical applications. Crit. Rev. Ther. Drug 2005;22:107-150.
7. Davis KA, Anseth KS. Controlled release from crosslinked degradable networks. Crit. Rev. Ther. Drug 2002;19:385-423.
8. Coviello T, Matricardi P, Marianecci C, Alhaique F. Polysaccharide hydrogels for modified release formulations. J. Control. Release 2007;119:5-24.
9. Brandl F, Sommer F, Goepferich A. Rational design of hydrogels for tissue engineering: Impact of physical factors on cell behavior. Biomaterials 2007;28:134-146.
10. Fedorovich NE, Alblas J, De Wijn JR, Hennink WE, Verbout AJ, Dhert WJA. Hydrogels as extracellular matrices for skeletal tissue engineering: state of the art and novel application in organ printing - a review. Tissue Eng. 2007 in press.
11. Ferreira LS, Gerecht S, Fuller J, Shieh HF, Vunjak-Novakovic G, Langer R. Bioactive hydrogel scaffolds for controllable vascular differentiation of human embryonic stem cells. Biomaterials 2007;28:2706-2717.
12. Park H, Temenoff JS, Tabata Y, Caplan AI, Mikos AG. Injectable biodegradable hydrogel composites for rabbit marrow mesenchymal stem cell and growth factor delivery for cartilage tissue engineering. Biomaterials 2007;28:3217-3227.
Appendix
Nederlandse samenvatting
List of abbreviations List of publications
Curriculum Vitae Dankwoord/Acknowledgement
Appendix
160
Nederlandse samenvatting
161
1. Hydrogelen
Hydrogelen zijn 3-dimensionale netwerken die opgebouwd zijn uit hydrofiele polymeren. Ze zijn in staat grote hoeveelheden water te absorberen, wat ze bijzonder aantrekkelijk maakt voor tal van farmaceutische toepassingen. Het gebruik van hydrogelen werd voor het eerst beschreven in 1960
voor de ontwikkeling van zachte contactlenzen. Daarnaast zijn ze onder andere gebruikt als materiaal voor wondverzorging en chirurgische hechtingen. Door hun zacht oppervlak en hoog
watergehalte veroorzaken hydrogelen een minimale weefselirritatie en zijn ze uiterst geschikt als celdrager voor de aanmaak van nieuw weefsel of als afgiftesysteem voor therapeutische eiwitten.
Voor de bereiding van hydrogelen kunnen polymeren van zowel natuurlijke als synthetische oorsprong gebruikt worden. Alginaat, chitosan, collageen en hyaluronzuur zijn enkele voorbeelden van frequent gebruikte natuurlijke polymeren. Het belangrijkste voordeel van deze
materialen is hun biocompatibiliteit. Daartegenover staan, over het algemeen, zwakke mechanische eigenschappen en variabiliteit, die inherent is aan de grondstoffen van natuurlijke
oorsprong. Het gebruik van synthetische polymeren, waarvan poly(ethyleen glycol) (PEG) en poly(acrylaten) de bekendste zijn, leidt tot hydrogelen met sterke mechanische eigenschappen en
biedt bovendien de mogelijkheid om de netwerkeigenschappen op een reproduceerbare manier te sturen.
2. Crosslinkingsmethoden
De polymeerketens die de bouwstenen vormen van een hydrogel zijn ‘gecrosslinkt’ om het oplossen van de polymeren in een waterige omgeving te voorkomen. Deze crosslinks kunnen tot
stand gebracht worden via chemische of fysische bindingen. Chemische crosslinks zijn permanent en wanneer gevormd, met behulp van bijvoorbeeld radicalen en enzymen, kunnen ze enkel door
afbraak van het materiaal verbroken worden. Hydrogelen die op deze manier vervaardigd zijn, beschikken over sterke mechanische eigenschappen maar kunnen niet waarborgen dat de ingesloten eiwitten of cellen intact blijven tijdens de vernettingsprocedure. Fysische crosslinks
kunnen tot stand gebracht worden door onder andere waterstofbruggen alsmede door ionische en hydrofobe interacties, wat geen nadelige effecten heeft voor de ingesloten eiwitten. Bovendien
zijn deze fysische interacties reversibel en kunnen onder bepaalde condities verbroken worden waardoor de hydrogel injecteerbaar is. Figuur 1 toont een schematische weergave van een
gecrosslinkt netwerk.
hydrofielepolymeerketensknooppunten
hydrofielepolymeerketensknooppunten
Figuur 1: Schematische weergave van een hydrogel.
Appendix
162
3. In situ gelvorming
Hydrogels kunnen ontworpen worden in verschillende geometrieën, zoals gelbolletjes in de grootteorde van nano- en micrometer, of macroscopische gelimplantaten. De kleine gelbolletjes kunnen door middel van injectie toegediend worden, terwijl de implantaten een chirurgische
ingreep vereisen. Tegenwoordig wordt veel onderzoek gedaan naar hydrogelen die gevormd worden op de plaats van toediening. In deze gevallen wordt het materiaal in vloeibare vorm
geïnjecteerd waarna het netwerk zich in situ vormt, na een bepaalde tijd of vanwege een specifieke impuls zoals temperatuursstijging. Bijzonder aantrekkelijk zijn bovendien die gelen die
hun inhoud afgeven na een bepaalde stimulus, zoals glucoseconcentratie, zuurtegraad, temperatuur of antigenbinding.
4. Dextraan hydrogelen
Van de polymeren die tot nu toe onderzocht zijn voor de bereiding van hydrogels, is dextraan een van de meest veelbelovende. Dextraan is een polysaccharide dat zowel synthetisch als bacterieel
geproduceerd kan worden en zeer geschikt is voor chemische modificatie. Hoofdstuk 2 van dit
proefschrift geeft een literatuuroverzicht van het belangrijkste onderzoek verricht op het gebied
van dextraanhydrogelen over de afgelopen 5 à 10 jaar. In onze onderzoeksgroep zijn verschillende dextraanhydrogelen ontwikkeld, die gebaseerd zijn op
zowel fysische als chemische vernetting. Bovendien werd een methode ontwikkeld om dextraanmicrosferen (gelbolletjes ~ 1-100 μm) te vervaardigen in een waterige oplossing waardoor het gebruik van organische oplosmiddelen wordt vermeden. Hierbij wordt gebruik
gemaakt van de onmengbaarheid van dextraan- en PEG-oplossingen in water. Emulsificatie van een oplossing van gemodificeerd dextraan (dex-HEMA) in een PEG-oplossing, gevolgd door
radicaalpolymerisatie van het dex-HEMA, resulteert in dex-HEMA microsferen.
Het doel van dit proefschrift was de ontwikkeling en karakterisering van macroscopische hydrogelen die opgebouwd zijn uit dex-HEMA microsferen die op hun beurt via fysische crosslinks met elkaar vernet zijn. Hoofdstuk 3 beschrijft een in situ gelerend gelsysteem, dat gebaseerd is op
tegengesteld geladen dex-HEMA microsferen, die via ionische interacties een macroscopisch netwerk vormen. Door middel van reologiemetingen werden de netwerkeigenschappen
onderzocht. Er werd aangetoond dat door aanbrengen van afschuifkrachten, de viscoelastische gel gaat vloeien. Er werd verder aangetoond dat het netwerk opnieuw wordt gevormd wanneer de
afschuifkracht werd opgeheven. In Hoofdstuk 6 wordt ingegaan op de invloed van de lading en
grootte van de microsferen op de netwerkeigenschappen, waaronder de gelsterkte. Zo werd
onder andere gevonden dat kleine microsferen sterkere gelen vormen dan de grotere varianten. Verder werd aangetoond dat het mogelijk is om de gelsterkte te variëren door de lading van de
microsferen te veranderen. Fysische crosslinking in hydrogelen kan ook verkregen worden door hydrofobe interacties. Deze benadering is onderzocht in Hoofdstuk 8, waarin dex-HEMA
microsferen gederivatiseerd werden met korte melkzuurketens die via hydrofobe domeinen de microsferen met elkaar verbinden. Reologie werd toegepast om de geleigenschappen te bepalen.
Er werd aangetoond dat onder meer de ketenlengte en de hoeveelheid lactaatketens een cruciale rol spelen in de netwerkvorming.
Nederlandse samenvatting
163
5. Hydrogelen als gecontroleerd afgiftesysteem voor eiwitten
Eén van de belangrijkste toepassingen voor hydrogelen is de gecontroleerde afgifte van therapeutisch relevante eiwitten. De vooruitgang van de biotechnologie de afgelopen 30 jaar heeft gezorgd voor een nieuwe generatie bioactieve eiwitten. Parallel hiermee is er een urgente
behoefte aan geschikte afgiftesystemen voor deze farmaceutische eiwitten ontstaan. Het insluiten van eiwitten in een matrix biedt de volgende voordelen: (1) bescherming van de fragiele
geneesmiddelen tegen chemische of enzymatische degradatie, (2) constante en verlengde therapeutische bloedconcentraties, (3) lagere geneesmiddel-gerelateerde toxiciteit, (4)
verminderde toedieningsfrequentie en daaraan gekoppeld een verhoogde compliantie voor de patiënt. Maar men moet er ook rekening mee houden dat de afgiftematrix zelf, m.a.w. de hydrogel, ongewenste immuunreacties kan induceren en dat een zogenaamde ‘burst-afgifte’, of een
plotselinge afgifte van een grote hoeveelheid eiwit, tot sterke bijwerkingen kan leiden. Daarnaast moeten tijdens het ontwikkelingsproces van afgiftesystemen ook de productiekosten en
opschalingsmogelijkheden in acht genomen worden. Dusver zijn slechts een gering aantal van de systemen, ontwikkeld voor de gecontroleerde afgifte van eiwitten, in vivo getest en een nog
geringer aantal is in een klinische studie geëvalueerd. Reeds eerder is aangeven dat fysisch gecrosslinkte hydrogelen te prefereren zijn boven chemische
gecrosslinkte varianten vanwege hun eiwitvriendelijke bereidingsprocedure. In de spontaan-vormende hydrogels, die beschreven zijn in dit proefschrift kan een therapeutisch eiwit ingesloten worden door eenvoudige menging met de microsferen. In Hoofdstuk 4 wordt de afgifte
beschreven van 3 modeleiwitten uit gelen die opgebouwd zijn uit tegengesteld geladen dex-HEMA microsferen. Er werd aangetoond dat diffusie de drijvende kracht was in het afgifteproces.
De enzymatische activiteit van het afgegeven lysozyme bleek volledig intact te zijn gebleven, wat een belangrijk gegeven is voor de verdere toepasbaarheid van het hydrogelsysteem. Aanbevolen
wordt om in een volgende onderzoeksfase de in vivo afgifte van therapeutisch relevante eiwitten te evalueren.
6. Hydrogelen als celdragers in weefseltechnologie
Het verlies van gezond weefsel door bijvoorbeeld ziekte of slecht functioneren van organen is een belangrijk probleem in de gezondheidszorg. Orgaan- of weefseltransplantatie is de
standaardtherapie voor deze patiënten, maar een nijpend donortekort creëert lange wachtlijsten. De toepassing van weefseltechnologie of ‘tissue engineering’ zou in veel gevallen een oplossing
kunnen bieden. Hierbij worden cellen opgenomen in een matrix die functioneert als synthetische extracellulaire matrix om de cellen te organiseren in een 3-dimensionale structuur. Als de goede
condities gewaarborgd kunnen worden (toevoer van voedingsstoffen, mechanische steun, etc.) groeien de cellen uit tot nieuw weefsel terwijl de synthetsiche matrix geleidelijk afgebroken wordt. De ingesloten cellen kunnen specifieke celtypes zijn, bijvoorbeeld hepatocyten wanneer een deel
van de lever moet vervangen worden, maar ook stamcellen kunnen gebruikt worden. Deze cellen kunnen zich, afhankelijk van de externe factoren differentiëren tot specifieke celtypes. De zachte,
hydrofiele aard van hydrogelen en de mogelijkheid om de mechanische eigenschappen, en de afbraaksnelheid te sturen, maakt hen bijzonder aantrekkelijk als celdragers in ‘tissue engineering’.
Het degradatiegedrag van de ionische gecrosslinkte hydrogelen, beschreven in dit proefschrift, is
Appendix
164
bestudeerd in Hoofdstuk 5. Het is gebleken dat de positief geladen microsferen 4 keer sneller
degraderen dan de microsferen met een negatieve lading. Gevonden werd dat door variatie in de verhouding positieve/negatieve microsferen het degradatieprofiel van de hydrogelen gestuurd
kan worden. Verder werd aangetoond dat, wanneer er bijvoorbeeld een eiwit ingesloten wordt in de microsferen, het afgiftepatroon afhangt van de lading van de microsferen, wat het mogelijk
maakt de afgifte te sturen door de verhouding van positieve/negatieve microsferen te variëren. In Hoofdstuk 7 wordt verder ingegaan op het degradatiegedrag van deze hydrogelen en in het
bijzonder de afbraak van de positief geladen microsferen. De groep verantwoordelijk voor de positieve lading, werd vervangen door een andere chemische verbinding die een afbreekbare binding bezit. Op deze manier verliezen de microsferen na verloop van tijd hun positieve lading,
waarna, door een gebrek aan interactiemogelijkheden met de negatieve microsferen, het gelnetwerk uit elkaar valt. Deze inleidende experimenten toonden aan dat het mogelijk is om op
deze manier het degradatiegedrag van de in situ-gelerende hydrogelen verder te moduleren om ze geschikt te maken voor een scala aan toepassingen.
De toepasbaarheid van de spontaan-vormende fysisch gecrosslinkte hydrogelen beschreven in dit
proefschrift is verder onderzocht in Hoofdstuk 9. Speciale aandacht is besteed aan eiwitafgifte en
weefseltechnologie, zoals hierboven al werd aangestipt, de twee meest veelbelovende
toepassingen voor deze systemen. Verder worden de resultaten van enkele inleidende studies, waarin de in vitro toxiciteit van de hydrogelen werd geëvalueerd, kort besproken.
In de verschillende hoofdstukken van dit proefschrift is aangetoond dat zowel het afgifteprofiel, het degradatiegedrag als de netwerkeigenschappen van de op microsferen gebaseerde
hydrogelen kunnen gestuurd worden naar gelang de gewenste toepassing. Het kan voorzien worden dat deze hydrogelen hun toepassing zullen vinden in de combinatie van eiwitafgifte en
weefseltechnologie. Hierbij kan men denken aan een ‘all-in-one’ matrix waarin bijvoorbeeld groeifactoren in de microsferen zijn ingesloten, terwijl cellen zijn ingebed in de poriën tussen de gelbolletjes, zoals schematisch weergegeven in Figuur 2.
Figuur 2: Schematische weergave van een uit microsferen bestaande hydrogel als celdrager in tissue
engineering.
Nederlandse samenvatting
165
List of abbreviations
167
List of abbreviations
BCA bicinchoninic acid BSA bovine serum albumin
CaP calcium phosphate CDI N,N’-carbonyldiimidazole DCM dichloromethane
DMAE N,N-dimethylaminoethanol DMAEMA N,N-dimethylaminoethyl methacrylate
DMAP 4-N,N-dimethylaminopyridine DMAPr N,N-dimethylaminopropanol
DMSO dimethylsulfoxide DP degree of polymerization
DPav average DP DS degree of substitution DStheor theoretical DS
FITC fluorescein isothiocyanate FRAP fluorescence recovery after photobleaching
FTIR fourier transform infrared spectroscopy G´ storage modulus
G” loss modulus HEMA hydroxyethyl methacrylate Hepes N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid
HPLC high pressure liquid chromatography HPMA N-2-hydroxypropyl-methacrylamide
IgG immunoglobulin G KPS potassium peroxodisulfate
MA methacrylate MAA methacrylic acid
PDLA poly(D-lactic acid) PEG poly(ethylene glycol) PLGA poly(D,L-lactice-co-glycolide)
PLLA poly(L-lactic acid) PVA poly(vinyl alcohol)
RGD arginine-glycine-aspartic acid rh recombinant human
tan(δ) tangens(delta)
TEMED N,N,N’,N’-tetramethylethylenediamine XPS X-ray photoelectron spectroscopy
List of publications
169
List of publications
Papers
Van Tomme SR, van Nostrum CF, Dijkstra M, De Smedt SC, Hennink WE. Effect of particle size and charge on the network properties of microsphere-based hydrogels. Submitted for publication. Van Tomme SR, Mens A, van Nostrum CF, Hennink WE. Macroscopic hydrogels by self-assembly of oligolactate-grafted dextran microspheres. Submitted for publication. Hiemstra C, Zhong ZY, Van Tomme SR, van Steenbergen MJ, Jacobs JJL, Den Otter W, Hennink WE,
Feijen J. In vitro and in vivo protein delivery from in situ forming poly(ethylene glycol)–poly(lactide) hydrogels. J. Control. Rel. 2007;119:320-327.
Van Tomme SR, Hennink WE. Biodegradable dextran hydrogels for protein delivery applications. Expert Rev. Med. Devices 2007;4:147-164.
Hiemstra C, Zhong ZY, Van Tomme SR, Hennink WE, Dijkstra PJ, Feijen J. Protein release from injectable stereocomplexed hydrogels based on PEG–PDLA and PEG–PLLA star block copolymers.
J. Control. Rel. 2006;116:e19-e21. Van Tomme SR, van Nostrum CF, De Smedt SC, Hennink WE. Degradation behavior of dextran
hydrogels composed of positively and negatively charged microspheres. Biomaterials 2006;27:4141-4148.
Van Tomme SR, De Geest BG, Braeckmans K, De Smedt SC, Siepmann F, Siepmann J, van Nostrum CF, Hennink WE. Mobility of model proteins in hydrogels composed of oppositely charged dextran microspheres studied by protein release and fluorescence recovery after photobleaching. J.
Control. Rel. 2005;110:67-78. Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Self-gelling
hydrogels based on oppositely charged dextran microspheres. Biomaterials 2005;26:2129-2135. Bos GW, Jacobs JJL, Koten JW, Van Tomme SR, Veldhuis TFJ, van Nostrum CF, den Otter W, Hennink
WE. In situ crosslinked biodegradable hydrogels loaded with IL-2 are effective tools for local IL-2 therapy. Eur. J. Pharm. Sci. 2004;21:561-567.
Patents
Van Tomme SR, van Steenbergen MJ, van Nostrum CF, Hennink WE. Gel composition comprising
charged polymers. European Patent Application EPC 04076424.3 Van Tomme SR, Hennink WE. Aqueous gels comprising microspheres. European Patent Application
EPC 06076563.3
Appendix
170
Abstracts
Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Self-gelling hydrogels based on oppositely charged dextran microspheres.
Oral presentation during the 13th Conference of the Dutch Society for Biomaterials and Tissue Engineering (Lunteren, The Netherlands, 2004).
Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Oppositely charged dextran microspheres for the design of reversible macrogels. Poster presentation during the 32nd Controlled Release Society Annual Meeting (Miami Beach, USA,
2005).
Van Tomme SR, van Nostrum CF, Hennink WE. Degradation behavior of in situ gelling hydrogels
composed of oppositely charged dextran microspheres. Oral presentation during the 14th Conference of the Dutch Society for Biomaterials and Tissue
Engineering (Lunteren, The Netherlands, 2005). Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Controlled
release of proteins from self-assembling hydrogels based on oppositely charged dextran microspheres.
Oral presentation during the Globalization of Pharmaceutics Network Meeting (Lawrence, USA, 2006).
Van Tomme SR, van Steenbergen MJ, De Smedt SC, van Nostrum CF, Hennink WE. Reversible macrogels based on oppositely charged dextran microspheres. Poster presentation during the American Association of Pharmaceutical Scientists Annual Meeting
(San Antonio, USA, 2006). Van Tomme SR, Hennink WE. Oligo-lactate grafted dextran microspheres self-assemble into
macroscopic hydrogels. Oral presentation during the pre-satellite meeting of the Pharmaceutical Sciences World Congress
(Amsterdam, The Netherlands, 2007)
Poster presentation during the Pharmaceutical Sciences World Congress (Amsterdam, The
Netherlands, 2007) Van Tomme SR, Hennink WE. In situ gelling hydrogels based on oligolactate-grafted dextran
microspheres. Poster presentation during the 34nd Controlled Release Society Annual Meeting (Long Beach, USA,
2007)
List of publications
171
Curriculum Vitae
173
Curriculum Vitae Date of birth
Place of Birth Nationality
Residence
25th October 1980
Bruges, Belgium Belgian
Utrecht, The Netherlands
Education
1998-2003 Pharmacist Faculty of Pharmaceutical Sciences, Ghent University
1994-1998 Latin-Mathematics State High School, Bruges
Working experience and internships
2003 - 2007 PhD student at the Department of Pharmaceutics, Utrecht Institute for Pharmaceutical Sciences (UIPS), Faculty of Science, Utrecht
University Supervisors: Prof. Dr. Wim Hennink, Prof. Dr. Stefaan De Smedt
Project: Hydrophilic polymeric systems for drug delivery purposes The results of this work are described in this thesis
Spring-summer 2003 Internship (6 months) at public pharmacy Waelkens in Bruges
Spring 2003 Internship (1 month) at University Hospital Pharmacy in Ghent
Spring 2002 Erasmus exchange student (3 months) at Department of Pharmaceutics, Utrecht University
Project: Physically crosslinked dextran hydrogels for the controlled delivery of Interleukin-2 in cancer therapy
Supervisors: Dr. Gert Bos, Prof. Dr. Wim Hennink
Achievements
Best paper award 2004 of the European Journal of Pharmaceutical Sciences
Best oral presenter award during 13th conference of the Dutch Society for Biomaterials and Tissue Engineering (2004)
Third prize oral presentations during Globalization of Pharmaceutics Education Network meeting (University of Kansas, 2006)
Dankwoord
175
Dankwoord/Acknowledgement “…wisest is she who knows she does not know…”
(Socrates)
Hier zit ik dan, lekker een dagje thuis (heerlijk dat systeem!) om ‘op mijn gemak’ mijn dankwoord te kunnen schrijven… Muziekje erbij, kat op schoot, potje koffie in de hand, aantal letters op papier: nul. In de loop der 4 jaar denk je af en toe “die mag ik niet vergeten in mijn dankwoord” en
iedereen mag dan wel denken dat ik een uitermate georganiseerd persoontje ben, ik ben toch net niet georganiseerd genoeg om dat soort dingen bij te houden. Bijgevolg zijn we al 2 tango-cd’s
verder, heeft Merlijn al hele melodieën bij elkaar gespind, sta ik stijf van de koffie, maar staat er nog niets zinnigs op papier…
Welaan, Professor Hennink, Wim, het lijkt me gepast bij jou te beginnen. Samen met Gert Bos heb
je er 5 jaar geleden voor gezorgd dat ik als student de smaak van het onderzoek te pakken kreeg. Dat heeft ertoe geleid dat ik een jaar later als AIO bij Biofarmacie aan de slag kon. Ik had me geen betere promotor kunnen bedenken. Af en toe heb ik je vervloekt, dikwijls heb ik veel plezier met je
gehad, maar wat het belangrijkste is, je hebt me heel veel geleerd. Ik kon altijd bij je terecht voor grote en kleine ‘uitdagingen’. Je gaf me de kans me op vele gebieden te ontplooien en te
bewijzen. We hebben best wel moeilijke tijden doorstaan samen, bijvoorbeeld toen je Nissan het begaf en we strandden in Groningen en IK gelukkig wel een mobieltje had… Ondertussen heeft
de moderne technologie ook jou veroverd en rij je nu met een Bijzonder Mooie Wagen, voorzien van een GPS die je tot ‘SophieSophie’ hebt omgedoopt omwille van het ‘kalm blijven’ van de dame (ironie o ironie!). Ik blijf in de toekomst nog even rondhangen in de vakgroep dus ik denk dat
je me nog geregeld ‘op de kast’ krijgt ☺.
De afgelopen maanden waren ook voor jou zwaar, zeker toen je in juni “3 dames op hetzelfde moment moest afwerken” (dixit WH). Bedankt voor alles Wim! En voor de laatste keer… opposites
attract! Professor De Smedt, beste Stefaan, bedankt dat je mijn tweede promotor wilde zijn. Ondanks het
feit dat we niet op regelmatige basis werkbesprekingen hadden, hebben onze discussies een zeer waardevolle bijdrage geleverd aan mijn onderzoek en aan het tot stand komen van dit proefschrift.
Je kritische vragen hebben me dikwijls geprikkeld dieper te graven dan ik initieel van plan was, met leuke resultaten tot gevolg. Bovendien was het ook aangenaam eens gewoon Vlaams te
kunnen ‘klappen’!
Mijn co-promotor, Dr. van Nostrum, beste Rene, bedankt voor de begeleiding en de inspirerende werkbesprekingen.
Voorlopig zeg ik de academische wereld nog geen vaarwel want in september ga ik als Post Doc aan de slag bij Professor Storm. Beste Gert, je was niet bij mijn AIO-project betrokken, dus ik ken je
voorlopig enkel als goedlachse, praatgrage tegenpool van Wim, die een spreekuur heeft ingesteld om mensen te ontvangen… Ik verheug me op onze samenwerking en heb er alle vertrouwen in
dat het een uitdagende tijd zal worden. Mevrouw Hennink omschrijft je niet voor niets als de
‘Latin-versie van Wim’… ☺
Appendix
176
OctoPlus NV was nauw betrokken bij mijn onderzoeksproject. Leo de Leede, Ruud Verrijk, Delphine
Ramos, Karin Wensink, Bas Kremer, Marie-Andree Yessine, bedankt voor de nuttige discussies en
feedback. Hopelijk zien we op termijn een formulering met de ‘plusmin-bolletjes’ op de markt ☺.
Gedurende de afgelopen 4 jaar hebben een aantal mensen op de een of andere manier een
professioneel steen(tje) bijgedragen tot mijn project. Een aantal keer was de samenwerking zo vruchtbaar dat er een copublicatie is uitgevloeid, een andere keer bleef het bij een goeie discussie
en leerzame ervaring. Graag bedank ik dan ook: Wouter Bult, Bruno De Geest, Joseph Demeester, Wouter Dhert, Marjolein Dijkstra, Natalja Fedorovich, Marco Harmsen, Jean-Paul Kleeven, Govert Kruijtzer, Sander Leeuwenburgh, Enrico Mastrobattista, Ad Mens, Frank Nijsen, Jan Rikken, Juergen
en Florence Siepmann, Fouad Soulimani, Marc Sutter, Marja van Luyn, Tom Visser, Herman Vromans, Bert Weckhuysen, Maike Wubben, Sander Zielhuis. Ik ben vast iemand vergeten, bij deze
mijn excuses, je krijgt een drankje van me!
Een van de taken van een AIO is het begeleiden van studenten, niet altijd een gemakkelijke taak voor iemand die achter in de rij stond toen ‘geduld’ uitgedeeld werd… Dieter, Astrid, Eveline, Mila en Bénédicte, bedankt voor jullie inzet! Astrid en Eveline, jullie project was erg kort (1 maand), maar
jullie ijver was formidabel. Ik ben blij dat je ook het AIO-pad bent ingeslaan Eveline! Mila, je enthousiasme was aanstekelijk. Ondanks de tegenvallende resultaten (waar jij voor niets tussen zat,
that’s research) bleef je altijd goedgezind en gemotiveerd. Succes verder allemaal.
Promotieonderzoek is als de Tour de France (ik ben helemaal niet into wielrennen, maar de Tour komt dit jaar zoveel in het nieuws dat het niet te negeren valt): je hebt vlakke stukken, bergetappes
en tijdritten. Gelukkig dat ik mijn ‘bolletjes’ onderwerp niet te letterlijk heb genomen, want om een promotie tot een goed einde te brengen moet je zowel de gele, groene en bolletjestrui aangehad
hebben! (ja jongens, ik weet zelfs waarvoor ze staan! Dank je papa ☺) En af en toe, moet je
gewoon een rustdag nemen of je erbij neerleggen dat je wel eens moet opgeven, ‘relax, take it
easy, for there is nothing that we can do’ (Mika), en daar heb je onder andere collega’s voor. Collega’s, ze komen en gaan, zeker bij Biofarmacie, een groep met een halfwaardetijd korter dan
die van een geëxciteerde fluorescente probe, maar overall zijn ze een constante factor, dikwijls mede-lotgenoten, in het leven van een AIO. Weinig vakgroepen kunnen tippen aan die van ons als het om feestjes en gezelligheid gaat. Labuitjes, kerstdiners, proeverijen, promoties, verjaardagen,
BBQ’s, trouwerijen en niet te vergeten de 90’s NOW fuiven (Tivoli gaat bijna automatisch kaartjes voor ons reserveren denk ik). Het moge duidelijk zijn, de afgelopen 4 jaar waren zeker niet saai op
dat gebied. Biofarmacie-collega’s bedankt! Ik kan niet iedereen opnoemen, maar uiteraard verdienen enkelen een speciale vermelding…
Marion, ik was blij toen jij me na een jaar kwam vergezellen op het lab. Samen (met Natasa en Suzanne) brachten we het nodige vrouwelijke leven in de brouwerij op de door mannen
overheerste 6de verdieping (ondertussen is de emancipatiegolf gelukkig ook daar doorgedrongen). We hebben veel plezier gehad samen! Veel succes met het afronden van je AIO-project (nog een jaar en jij bent aan de beurt!).
In de loop der jaren werden we vergezeld door Albert, Niels en Frank. Jongens, het werd er alleszins niet saaier op met jullie! Niels bedankt voor de grappige noot en om me te laten
kennismaken met de wondere wereld van poker…
Dankwoord
177
Martin, ik ben blij dat jij er was. Je hebt een goeie dosis relativeringsvermogen waaraan het een
(beginnende) AIO soms ontbreekt. Bedankt voor alle leuke momenten en voor de knuffels en schouderklopjes als ik het nodig had!
Professor van Steenbergen ☺, Mies, mijn steun en toeverlaat op het gebied van procedures en
apparaten, bedankt voor alle hulp en onuitputbare (soms gekke) ideeën. Zonder jou waren de
afgelopen 4 jaar zeker nog zwaarder geweest. Najim, de laatste 6 maanden van mijn project was jij analist ‘voor mij’. Bedankt voor al je werk. Er
was niet genoeg tijd om voldoende data te verzamelen voor een publicatie, maar wel voor een hoofdstuk in dit proefschrift! De laatste maanden heb ik het lableven grotendeels achter me gelaten en me teruggetrokken in
de AIO-kamer (alias ‘typkamer’), speciaal bedacht voor ‘terminale’ AIO’s. Je mist alle nieuwtjes en roddels van het lab, maar je krijgt er een boel gezelligheid voor terug. Myrra, Cris en Birgit, het
waren leuke tijden. Birgit, jammer dat we zo laat onze gezamenlijke passie ontdekten. Het was erg leuk om ongegeneerd over pirouettes en arabesques te kunnen kletsen. We zien elkaar in de
toekomst zeker nog bij jazzdance! Barbara, Marjan, Maryam en Tina, mede dankzij jullie voel ik me nu als een vis in het water op de 7de! Ellen, wij vormen samen (en een occasionele student) de Vlaamse enclave bij Biofarmacie. Je blijft
tussen al die Hollanders toch de enige die me helemaal verstaat als ik er een typisch Vlaams woord tussenlap. Misschien slaag jij er nog in Wim ervan te overtuigen dat ‘welaan’ echt niet te pas en te
ompas gebruikt wordt door alle Belgen… Samen hebben we Utrecht al ontelbare keren onveilig gemaakt en er komen zeker nog meer plezante fuifavonden aan!
En dan zijn er de AIO’s die wel bij onze vakgroep behoren maar die we enkel bij presentaties en uitjes zien: Joost en Wouter. Ik kan erg genieten van jullie humor en jullie gezelschap op feestjes!
Ook de andere reguliere feestbeesten, Sabrina, Niels (succes met jouw promotie-onderzoek in het mooie Zuid-Afrika!), Marcel, Karlijn, Enrico, Helma (jammer dat je weg bent bij Biofarmacie!), Ethlinn, Holger, Rolf,… bedankt voor alle leuke avonden de afgelopen 4 jaar!!
Een gezonde geest in een gezond lichaam… Vreselijk cliché, maar wel waar. Ik ben blij dat ik het
een jaar geleden aandurfde om auditie te doen bij een dansgroep in Utrecht. La Vinia danseressen, bedankt voor de inspannende, pijnlijke (anders train je effe in een week voor een spagaat!),
ontspannende en soms echt hilarische momenten tijdens de voorbije repetities en uitvoeringen! And there’s more to come…!
Een van de voordelen van het AIO-bestaan is dat je veel mensen leert kennen. During the GPEN conference in Lawrence (Kansas, USA), I met Allyn, the chair of the session during which I had to
present my work. Allyn, you introduced me to SoCo&Lime ☺ and taught me that “Life isn’t always
a fairytale, but it can still be a good story”. As you recently said, our friendship is still young so we have plenty of time to be friends. Good luck with finishing your PhD next year!
De verhuis van het mooie Brugge naar het even mooie maar toch heel andere Utrecht heeft op persoonlijk gebied een en ander teweeggebracht. Weinig vriendschappen zijn bestand tegen de
eroderende werking van ‘afstand’. Maar de vriendschappen die blijven bestaan, zijn voor het leven. Tina, ik verhuisde naar Utrecht, jij naar Valencia. Maar er is meer nodig dan afstand om ‘kleuterklas-
vriendinnen’ uit elkaar te krijgen hé! Lang leve msn ☺.
Appendix
178
Julie, mijn lieve paranimf, samen begonnen we aan een Utrecht-avontuur als farmaciestudentes
uit Gent. Those were the days! Sindsdien kan onze vriendschap niet meer kapot. Je hebt met me meegeleefd toen het leven hier wat lastiger was en me met warme knuffels uit sommige dipjes
geholpen (jaja, dat kan telefonisch ☺). Ik kijk alweer uit naar onze volgende jaarlijkse ‘onder-
vriendinnen-sauna-verwendag’. Bedankt voor je vriendschap en om mijn paranimf te willen zijn!
Met een nieuwe stad en nieuwe job komen natuurlijk ook nieuwe vrienden… De belangrijkste
nieuwe persoon in mijn leven sinds ik in Utrecht woon ben jij Cristianne (alias Cris, chica, guapa,…), mijn paranimf. Al heel snel na het begin van ons promotie-avontuur werden we vriendinnen. We zijn samen als AIO begonnen, zullen samen als AIO eindigen en zijn, als kers op de
taart, elkaars paranimf! Wat moeten de mensen daarvan denken ☺. Als Thelma & Louise hebben
we Amerika in een convertible onveilig gemaakt, 2 keer zelfs! Onvergetelijk: Cruisen met Cris door California in een Chrysler Convertible! Ik ben benieuwd wat we nog allemaal zullen uitspoken! Het
maakt niet uit (er zullen ongetwijfeld cocktails mee gemoeid zijn ☺), plezier zullen we zeker hebben. Chica, bedankt voor alles! Als ik van 1 ding zeker ben in het leven, is het wel van onze
vriendschap.
Collega’s, vrienden,… daarna beland ik vanzelf bij (zo goed als) familie. Familie Fonteyne, Christine en Noël, het was altijd leuk om tijdens de ontelbare weekends waarin we naar Brugge gingen bij
jullie aan tafel aan te schuiven. Bedankt voor jullie steun en interesse de afgelopen 4 jaar. Sorry dat
ik jullie zoon naar het verre noorden heb ‘ontvoerd’ ☺.
Patrick, lieve Nonkel P, je kent me al van voor ik geboren was. Langs de zijlijn heb je me gevolgd in
de verschillende fasen van mijn leven. We hebben dikwijls samen gefilosofeerd over het leven en je hebt me veel geleerd. Je stimuleert me om met volle teugen te genieten van het leven, “take a
walk on the wild side”… Bedankt voor de gezellige babbels, lekkere etentjes en sponsoring van dit promotiegebeuren. Maar vooral bedankt omdat je er voor me bent!
Mama en papa, zonder jullie was ik dit dankwoord niet aan het schrijven. Jullie hebben me altijd
gestimuleerd om te doen wat ik wou (qua studies dan hé, laten we niet overdrijven ☺) en jullie
hebben me de beste kansen gegeven die iemand zich kan wensen, op alle gebied. Ik heb zoveel
van jullie geleerd en dankzij jullie al veel van de wereld gezien. Jullie hebben me mezelf laten ontwikkelen tot de persoon die ik ben vandaag. Ik zie jullie graag! Papa, “il n’appartient qu’aux grands hommes d’avoir de grands défauts”… dat ik nog veel “à nos
faiblesses” mag antwoorden als jij zegt “à nos amours”!
Mama, ‘compagnon in goeie en slechte dagen’ ☺, mijn grote voorbeeld in het leven, de sterke en
vooral lieve vrouw die er altijd voor me is, zonder jou zat ik hier zeker niet. Het is dankzij jouw 2 uur
durende peptalk tijdens een rit van Gent naar Utrecht dat ik niet mijn boeltje heb gepakt en gestopt ben met promoveren in oktober 2003. Wie kent me beter dan jij, mijn hartsvriendin? We
maken al (bijna!) 27 jaar plezier samen, gaan samen op reis, delen kommer en kwel, zagen tegen elkaar, zijn soms kortaf tegen elkaar (vooral ik, ik weet het, sorry daarvoor!),... Ik besef dat het niet altijd makkelijk is dat ik zo ver van huis woon. Gelukkig ben jij een superhippe mama die sms, email
en msn vlot de baas kan! Jouw liefde en knuffels overbruggen iedere afstand! You’re simply the best, mama!
Dankwoord
179
“The greatest thing you’ll ever learn is just to love and be loved in return…” (Moulin Rouge)
Alleen als je het zelf meegemaakt hebt, besef je hoeveel tijd het layouten van een proefschrift kost.
Maar dankzij mijn persoonlijke, professionele IT-consultant kent Word geen geheimen meer voor
mij! ☺ Matthias, mijn liefje, uiteraard heb je de afgelopen 4 jaar nog veel meer voor me gedaan. Je
hebt me zonder gemor mijn droom laten najagen en dat zal ik nooit vergeten. Ik bewonder je
uithoudingsvermogen om 2 jaar lang te pendelen tussen Gent en Utrecht, voor mij… Je hebt er altijd voor gezorgd dat ik met beide voeten op de grond bleef staan en liet het niet toe dat ik de AIO zou worden die 7 dagen per week op het lab was. Je leefde mee met mijn uitdagingen, was
geïnteresseerd in mijn experimenten en toonde hoe trots je was op mijn verwezenlijkingen. Het leven met een AIO is niet altijd makkelijk, zeggen ze. Als het dan nog gaat om iemand die door
haar vader omschreven wordt als “lief maar kattig” en een sticker met “Caution, I can go from 0 to bitch in 2.5 seconds” op haar bureaustoel heeft, dan mag het duidelijk zijn dat jij wel heel bijzonder
bent! Je brengt me aan het lachen, verwent me, gelooft in me en laat me volledig mezelf zijn. Je bent mijn beste vriend, ideale reispartner, soulmate. Mijn liefje, bedankt voor je onvoorwaardelijke
liefde en steun. Ik hou van je! Ik ben benieuwd wat het leven nog voor ons in petto heeft… Dat was het, de eindstreep is bereikt. Iedereen, BEDANKT!
Op naar een volgende uitdaging…!
Maar eerst… tijd voor champagne!
“Aim at a high mark and you’ll hit it.
No, not the first time, nor the second and maybe not the third.
But keep on aiming and keep on shooting for only practice will make you perfect. Finally, you’ll hit the Bull’s Eye of Success”
(Miss Annie Oakley)