Supramolecular biomaterials : introducing a modular approach

Supramolecular biomaterials : introducing a modularapproachDankers, P.Y.W.

DOI:10.6100/IR608696

Published: 01/01/2006

Document VersionPublishers PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

A submitted manuscript is the author's version of the article upon submission and before peer-review. There can be important differencesbetween the submitted version and the official published version of record. People interested in the research are advised to contact theauthor for the final version of the publication, or visit the DOI to the publisher's website. The final author version and the galley proof are versions of the publication after peer review. The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

Citation for published version (APA):Dankers, P. Y. W. (2006). Supramolecular biomaterials : introducing a modular approach Eindhoven:Technische Universiteit Eindhoven DOI: 10.6100/IR608696

General rightsCopyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright ownersand it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

Users may download and print one copy of any publication from the public portal for the purpose of private study or research. You may not further distribute the material or use it for any profit-making activity or commercial gain You may freely distribute the URL identifying the publication in the public portal ?

Take down policyIf you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediatelyand investigate your claim.

Download date: 17. Mar. 2018

http://dx.doi.org/10.6100/IR608696https://research.tue.nl/en/publications/supramolecular-biomaterials--introducing-a-modular-approach(ac7c7249-33d5-4060-9719-518b15cde8cf).html

Uitnodigingtot het bijwonen van

de openbare verdediging van mijn proefschrift

SupramolecularBiomaterials

Introducing a Modular Approach

op dinsdag 16 mei 2006 om 16.00 uur.

De promotie vindt plaats in het auditorium van de Technische Universiteit

Eindhoven.

Na afloop van de plechtigheid vindt er een receptie plaats waarvoor u ook van harte

bent uitgenodigd.

Patricia Y.W. DankersFazantstraat 14

5702 NG Helmond06 111 28 193

[email protected]

PMS293Zwart

Omslag proefschrift Patricia DankersRugdikte: 9,2mm o.b.v. 172 paglaminaat: Glans Druk: PMS 293 / zw

Patricia Dankers, Markt 75, 5701 RJ HelmondTU/e: Macromolecular and Organic ChemistryHeO 4.43Eindhoven University of TechnologyP.O. Box 513, 5600 MB EindhovenPhone: +31 40 2472187E-mail: [email protected]

Supramolecular Biomaterials

Introducing a Modular Approach

Patricia Y.W. DankersMacromolecular and Organic Chemistry

Su

pra

mo

lecu

lar B

iom

ate

rials P

atricia

Y.W

. Dan

kers

Supramolecular Biomaterials Introducing a Modular Approach

Supramolecular Biomaterials Introducing a Modular Approach

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de Rector Magnificus, prof.dr.ir. C.J. van Duijn, voor een

commissie aangewezen door het College voor Promoties in het openbaar te verdedigen op

dinsdag 16 mei 2006 om 16.00 uur

door

Patricia Yvonne Wilhelmina Dankers

geboren te Helmond

Dit proefschrift is goedgekeurd door de promotor: prof.dr. E.W. Meijer Copromotor: dr. R.P. Sijbesma This research has been financially supported by the Council for Chemical Sciences of the Netherlands Organization for Scientific Research (NWO-CW). Omslagontwerp: Patricia Y.W. Dankers, Jan-Willem Luiten, Koen Pieterse Druk: JWL Producties A catalogue record is available from the Library Eindhoven University of Technology ISBN-10: 90-386-2638-X ISBN-13: 978-90-386-2638-3

Manuscript commissie: prof.dr. E.W. Meijer (Technische Universiteit Eindhoven) dr. R.P. Sijbesma (Technische Universiteit Eindhoven) prof.dr. S.I. Stupp (Northwestern University, USA) prof.dr. M.J.A. van Luyn (Rijksuniversiteit Groningen) prof.dr.ir. F.P.T. Baaijens (Technische Universiteit Eindhoven)

Contents Chapter 1. The biomaterials trinity mechanical properties, biodegradability and bioactivity 1.1 Introduction 2 1.2 Mechanical properties and biodegradability 3 1.3 Bioactivity 4 1.4 Self-assembly approaches of peptide conjugates 7 1.5 Self-assembly approaches in tissue engineering 8 1.6 Aim of the thesis 9 1.7 Outline of the thesis 10 1.8 References 11 Chapter 2. Supramolecular biomaterials 2.1 Introduction 14 2.2 Modular design 15 2.3 Syntheses 16 2.4 Polymer morphologies 19 2.5 Thermal properties 22 2.6 Infrared spectroscopy 23 2.7 Mechanical properties 25 2.8 Processability 27 2.9 Cell proliferation of 3D-scaffolds 27 2.10 Biocompatibility 28 2.11 Degradation behaviour 30 2.12 Modular blends 30 2.13 Conclusions 33 2.14 References 34 Chapter 3. Oligo(trimethylene carbonate) based supramolecular materials 3.1 Introduction 36 3.2 Synthesis and characterization 37 3.3 Mechanical properties 38 3.4 Thermal properties 40 3.5 Material morphology 42 3.6 Materials processing 43 3.7 Degradation behaviour 44 3.8 Biocompatibility 46 3.9 Conclusions 47 3.10 References 48 Chapter 4. Chemical and biological properties of supramolecular polymer systems

based on oligocaprolactones 4.1 Introduction 50 4.2 Synthesis of UPy-synthons and UPy-polymers 51 4.3 Solution properties 51 4.4 Material properties 55 4.5 Thermal properties 56 4.6 Tissue response of end-modified and chain-extended UPy-polymers 58 4.7 In-vivo implantation of mixtures 60 4.8 Mass loss of implanted mixtures 60 4.9 Morphology and crystallinity of implanted mixtures 60 4.10 Tissue response of mixtures 63 4.11 Conclusions and discussion 64 4.12 References 65

Chapter 5. Supramolecular bioactive compounds UPy-peptides and UPy-proteins

5.1 Introduction 68 5.2 UPy-peptides 70 5.3 UPy-proteins 74 5.4 Conclusions and discussion 77 5.5 References 78 Chapter 6. A modular and supramolecular approach to bioactive scaffolds 6.1 Introduction 80 6.2 Modular design 81 6.3 Bioactive films 82 6.4 Cell adhesion and spreading 82 6.5 Cell binding strength 87 6.6 In-vivo behaviour 88 6.7 Conclusions 89 6.8 References 90 Chapter 7. Tunable ligand availability at the surface of supramolecular

polymer films 7.1 Introduction 92 7.2 Modular design 93 7.3 Film preparation 95 7.4 Extraction behaviour 95 7.5 Surface availability 97 7.6 Cell adhesion and spreading 99 7.7 In-vivo behaviour 100 7.8 Redesigning the modular systems 102 7.9 Conclusions and discussion 105 7.10 References and notes 107 Chapter 8. Towards the engineering of bioactive surfaces 8.1 Introduction 110 8.2 Reactive micro-contact printing 110 8.3 Supramolecular micro-contact printing 112 8.4 Biotin containing supramolecular films 116 8.5 UPy-modified hydrogels 120 8.6 UPy-dimerization in water 123 8.7 Epilogue: conclusions and outlook 125 8.8 References 127 Experimental section 129 Colour figures 159 Summary 161 Samenvatting 163 Curriculum Vitae 167 Woord van dank 169

1 The biomaterials trinity - mechanical properties,

biodegradability and bioactivity

Several biomaterials have been designed as application in the biomedical field. They are investigated for their use in regenerative medicine as drug delivery vehicles or as scaffolds for tissue engineering. These scaffolds have to meet several factors to ultimately form the right tissue; important factors in designing such scaffolds are their biocompatibility, ease of processing, mechanical properties, degradability of the materials and bioactivity. This chapter focuses mainly on examples of incorporation of bioactive molecules such as peptides and proteins into scaffold materials. First, non-covalent modification via simply mixing of bioactives with polymers and covalent modification of materials with bioactive molecules is discussed. Then, a self-assembly approach is described in which systems are produced via non-covalent interactions.

Chapter 1

2

1.1 Introduction Biomaterials are used in biomedical applications such as drug delivery and regenerative

medicine.1,2 Initially, biomaterials were used as inert prostheses to restore the malfunctioning of organs in the human body.3 Then, when knowledge about the human body developed, they were designed to be either biodegradable or bioactive in order to be able to react to signals at the site of implantation and to be integrated with the human body. At that point tissue engineering started. Tissue engineering (TE) is used to replace or cure tissues and organs by an autologous implant (Fig. 1.1).4 In the tissue engineering concept, a biopsy is taken from a patient from which cells are isolated.5 The cells are expanded in vitro and cultured on a polymer scaffold (biomaterial), to ultimately form a new tissue. This newly formed tissue or organ will be transplanted back into the patient.

biodegradable

polymer scaffoldautologous cells cells on scaffold new tissue

Figure 1.1. The tissue engineering concept. A biopsy is taken from a patient. Cells are isolated out of this tissue and are expanded. Then the cells are cultured on the ideal polymer scaffold. Ultimately, a new tissue is formed which can be transplanted into the patient.

Many organs and tissues are subject of investigations to be transplanted using TE.6 To this end, it is important that the scaffolds, which have to guide and regulate the formation of the novel tissue, match with the tissue required. Here, it is proposed that the ideal scaffold, besides being biocompatible and processable, also has to have the right mechanical properties, has to be biodegradable and has to be bioactive (Fig. 1.2). Difficulties arise in the precise design of these so-called third generation materials3 caused by synthetic reasons and lack of knowledge about what the exact properties have to be. Many examples have been shown in which at least two of the three standards are met (Fig. 1.2). Here, a number of examples are shown in which mechanical properties, biodegradability and the incorporation of bioactivity are investigated.

mechanicalproperties

degradability

bioactivity

idealscaffold

Figure 1.2. The biomaterials trinity. The ideal scaffold has to meet three properties. It has to have the right mechanical properties, it has to be degradable and incorporation of bioactive molecules has to be possible.

The biomaterials trinity mechanical properties, biodegradability and bioactivity

3

1.2 Mechanical properties and biodegradability The mechanical properties of the scaffold are important because every tissue and organ in the

human body shows different mechanical behaviour.7 Mechanical properties of biomaterials can, for instance, be regulated by varying molecular weight or using co-polymers. As an example, the material properties of poly(trimethylene carbonate) (PTMC) are strongly dependent on the molecular weight of the polymers.8 It has been shown that very high molecular weight PTMC with Mn above 200 kg/mol has excellent mechanical properties8 (E = 6 MPa, break = 12 MPa; break = 830%). This PTMC displayed strain-induced crystallization in contrast to low molecular weight counterparts. In another example it has been shown that photopolymerization of hyaluronic acid modified with methacrylic anhydride into networks resulted in materials with different compressive moduli, ranging from 2 to over 100 kPa depending on the molecular weight (from 50 to 1100 kDa).9 Furthermore, co-polymers of TMC and D,L-lactide (DLLA) display different mechanical properties than the homopolymers10.

Often, the mechanical properties and rate of degradation are connected. Control over the rate of degradation is of major importance, because the formation of new tissue has to occur simultaneously (Fig. 1.3). If the polymer matrix is already degraded before the tissue is formed, the construct will fall apart. It has been shown that different polymers have different degradation rates7: whereas degradation of polycaprolactone takes more than two years11, poly(glycolic acid) is degraded within weeks. To get control over degradation rates, many co-polymers have been investigated.7 Also, the molecular weight of the polymers and morphology of the scaffolds12 play an important role in the rate of degradation.7 Besides that, degradation rates can differ between in-vitro and in-vivo degradation studies. As an example, PTMC hardly degrades in vitro, but shows complete resorption after one year of subcutaneous implantation.13 In addition, it is important to regulate the mechanism of degradation. Whereas surface erosion is accompanied with gradual mass loss, bulk erosion shows the accumulation of acid in the interior of the material which ultimately leads to disintegration accompanied by a burst release of acid.

It is beyond this chapter to give an overview of all polymer systems that have been investigated concerning their mechanical properties and degradation behaviour. However, it is important to keep in mind that both properties can be regulated by varying molecular weight, by applying different scaffold morphologies or by using several co-polymers.

0

100

time

mec

hani

cal p

rope

rtie

s (%

)

scaffold

tissue

Figure 1.3. Degradation of scaffolds has to be accompanied by simultaneous formation of new tissue.

Chapter 1

4

1.3 Bioactivity Besides control over mechanical properties and degradability, the incorporation of certain

bioactive factors is important, because in that way the cells can be guided to form the right tissue. Examples of bioactive molecules are peptide sequences and proteins.14-16

A B

C D

Figure 1.4. Bioactive materials can be produced by two methods: A. non-covalent modification or, B. covalent coupling of bioactive molecules. C. Non-covalent modified materials are dynamic but rather unstable, which leads to delivery of the bioactives that can exert their function in the environment. D. Covalent functionalized materials are stable but not dynamic, which results in local functioning of the activity at the surface of the material.

In general, two methods are used to modify materials with bioactives. This can be done by

simply mixing, which leads to dynamic systems which are rather unstable (Fig. 1.4A and 1.4C). However, this approach might be very useful in drug delivery systems, provided the rate of release can be controlled. In addition, covalent modification of polymers shows great promise (Fig. 1.4B and 1.4D). The latter results in stable materials that are not dynamic. This lack of dynamics can be beneficial, because in this way the bioactive molecule has to exert its function localized at the surface of the material. However, it can also be a major drawback because the system cannot adapt to the environment. Additionally, the synthetic versatility remains limited and the polymers require rather high processing temperatures which mostly results in deactivation of the bioactives.

Many examples have been shown in which materials were made bioactive according to both methods. The first example shows non-covalent modification of poly(lactic-co-glycolic) (PLGA) matrices with vascular endothelial growth vector (VEGF) and platelet derived growth factor (PDGF) (Fig. 1.5).17 VEGF was mixed with the PLGA polymer and PDGF was incorporated into microspheres of PLGA and alginate before processing. A porous scaffold was formed which showed dual growth factor release, in which each growth factor showed distinct kinetics (Fig. 1.5B). Rapid formation if a mature vascular network was observed after subcutaneous implantation.


5

A B C

PDGF

VEGF

PLGA

PLGA, alginateand PDGF

pore

PLGA

Figure 1.5. Schematic representation of bioactive porous polymer scaffolds for growth factor release. The different scaffolds show distinct release kinetics.17 A. VEGF was mixed with PLGA. B. PDGF was incorporated in PLGA-alginate microspheres before processing with PLGA into scaffolds. C. PDGF containing PLGA-alginate microspheres were processed with PLGA and VEGF.

Secondly, a smart material was designed, that could spontaneously assemble into a growth factor bearing hydrogel network at physiological pH (Fig. 1.6). Vinylsulfone-poly(ethylene glycol) (PEG) macromers were incubated with thiol-containing RGD (Arg-Gly-Asp) peptides, transforming growth factor 1 (TGF-1) and VEGF121 with a C-terminal cysteine. After that, dithiol-peptides containing a cleavage site for matrix metalloproteinases (MMPs) were used as cross-linking moieties, resulting in the formation of the hydrogel network.18 VEGF121 was covalently cross-linked whereas TGF-1 was non-covalently incorporated. This network could be proteolytically degraded in vitro by cell-derived MMPs.18 Endothelial cells could adhere and spread out after 3 days on these PEG-peptide hydrogels containing VEGF, whereas on the reference gels without VEGF hardly any adhesion and no spreading could be observed.

TGF-1

VEGF121,CysPEG

icross-linking peptide

adhesion peptide

Ac-GCGYGRGDSPG-NH2

Ac-GCRDGPQGIWGQDRCG-NH2

Figure 1.6. Vinylsulfone-poly(ethylene glycol) (PEG) macromers were incubated with thiol-containing RGD peptides, TGF-1, VEGF121 with a C-terminal cysteine and dithiol-peptides containing a cleavage site for MMPs as cross-linking peptides. This resulted in the formation of a hydrogel network at (i =) physiological temperature and pH.18

The modification of materials with peptides is synthetically more accessible. The most extensively studied peptide is probably the cell adhesion RGD19 peptide. Many polymers, materials and surface have been modified with this sequence to study cell adhesion or as tissue engineering application.20 An elegant example is shown in which peptide graft co-polymers were synthesized using metathesis ring opening polymerization (ROMP) of peptide norbornene

Chapter 1

6

monomers.21,22 The peptide norbornene monomers were synthesized on the solid support and consisted of the cell adhesion GRGDS (Gly-Arg-Gly-Asp-Ser)19 peptide sequence and its synergistic PHSRN (Pro-His-Ser-Arg-Asn) sequence (Fig. 1.7)23,24 It has been shown that cell adhesion of fibroblasts to fibronectin was inhibited.

GRGDS

NH

NH

NH

O

ONH

NH

OH

O

O

O

O

NH

NHNH

tBu

Pmc

COOtBu

O

N

N

N

NH

NH

O

ONH

NH

OH

O

O

O

NH

NH NH

ONH

O

O

Pmc

Trt

Trt

TrtPHSRN

Figure 1.7. Norbornene peptide monomers used for ring opening metathesis polymerization (ROMP) consisting of the GRGDS peptide sequence or of the PHSRN peptide sequence. Co-polymerization results in the formation of a graft co-polymer containing both peptides.21, 22 The used protection groups are 2,2,4,6,7-pentamethyldihydrobenzofurane-5-sulfonyl (Pmc), t-butyl, t-butoxy (OtBu) and trityl (Trt).

Another example in which surfaces have been modified with RGD peptides is shown in figure 1.8. Star-shaped PEG polymers containing isocyanate end-groups were reacted with amine containing glass surfaces after which GRGDSC peptides were reacted with the remaining isocyanate groups.25 Gradients of RGD, varying from 1 RGD to less than 0.1 RGD per star, were produced. The amount of human osteosarcoma cells adhered and spread on the film increased with increasing concentration of RGD.

In-vivo evaluation of poly(methylmethacrylate) (PMMA) beads covalently functionalized with acrylamide containing cyclic-RGD peptides after implantation in bones of rabbits showed enhanced bone ingrowth in the presence of the peptides.26 This indicates that the RGD-peptides are also active in vivo.

PEG with isocyanate groups

cell adhesion peptideH2N-GRGDS-CONH2

PEG with amine and ureagroups

Figure 1.8. Star-shaped poly(ethylene glycol) polymers on glass were modified with RGD peptides.26


7

1.4 Self-assembly approaches of peptide conjugates It is proposed that high control over both stability and dynamics of bioactive materials can be

accomplished by using supramolecular chemistry. Supramolecular materials are systems in which the separate building blocks are held together via non-covalent interactions. In recent years the importance of peptides as building blocks in supramolecular architectures has been demonstrated. Many examples of oligo-peptide based self-assembled aggregates have been disclosed in which hybrid conjugates are synthesized by the combination of peptide sequences with for example all kinds of polymers27,28, long alkyl-chains or phospholipids29.These conjugates have great potential in the biomedical field. However, none of these systems have been used for real tissue engineering applications, except for one (Paragraph 1.5). Nevertheless, these supramolecular architectures are discussed as examples of possible supramolecular biomaterials because they are able to operate in water or at the polymer-water interface. Therefore, they might find their application as TE or drug delivery scaffolds. Another advantage of the use of supramolecular building blocks is their controlled way of synthesis; the synthesis method is also mentioned.

As an example, the self-assembly of various amphiphilic peptides has been studied by varying the length of the N-terminal alkyl tail to the GANPNAAG (Gly-Ala-Asn-Pro-Asn-Ala-Ala-Gly) sequence30, known for its preferred -hairpin conformation. This leads to several types of aggregates with different peptide conformations varying from random-coil to -sheet architectures.30 Also, these GANPNAAG peptides have been non-covalently incorporated in liposomes using peptide sequences that were functionalized with alkyl chains on both the N-terminus and C-terminus31. The folding of the peptide was stabilized into a -hairpin conformation owing to the two alkyl-tails (Fig. 1.9A). In contrast, functionalization of the peptide with only one alkyl-tail resulted in the formation of a random-coil instead of a -hairpin (Fig. 1.9B). These GANPNAAG-alkyl conjugates were entirely synthesized using solid-phase techniques. A polymer-peptide conjugate that forms spherical aggregates was also completely synthesized using solid-phase chemistry32 (Fig. 1.9C). First an amine functionalized polystyrene polymer was coupled to a resin, after which the GANPNAAG peptide was built up.32

liposome

A

OO

NH

O

NH

ONH

NH2

O

NH

NH

O

O

NH

NH2

O

ONH

NH

NH

O

O

ON

O

O

n

n

O

NH

ONH

NH2

O

NH

NH

O

O

NH

NH2

O

ONH

NH

NH

O

O

ON

O

10

10

C

O

NH

ONH

NH2

O

NH

NH

O

O

NH

NH2

O

ONH

NH

NH2

O

O

ON

O

10B

liposome

Figure 1.9. GANPNAAG peptide conjugates. A. The GANPNAAG sequence modified with two alkyl chains assembles into a -hairpin while B. the GANPNAAG peptide modified with one alkyl tail shows a random-coil conformation when incorporated in liposomes.31 C. GANPNAAG-modified polystyrene was entirely synthesized on the solid support.32

Chapter 1

8

Micellar structures have been created from block copolymers that were synthesized on solid supports loaded with peptides.33 Living free radical polymerization (LFRP) initiators were coupled to the antimicrobial tritrpticin (Val-Arg-Arg-Phe-Pro-Trp-Trp-Trp-Pro-Phe-Leu-Arg-Arg) peptide on the resin. Subsequently, nitroxide mediated or atom transfer radical polymerizations were performed to form the peptide-polymer aggregates.

The self-assembly of PEG-based peptide hybrids containing -helical coiled-coil peptide sequences can be regulated by varying the amino acid residues.34 Organized nanostructures based on these PEG-peptide hybrids have been found both in solution and in solid state35. Also, the biological activity can be correlated to this self-assembly behaviour.34 Functionalization of the peptide sequence with PEG was performed on the resin.

Other supramolecular architectures have been made based on small liquid-crystalline oligo-peptide derivatives. Oligo(glutamic acid) derivatives, synthesized in solution, have been studied with respect to their liquid-crystalline behaviour.36 Furthermore, low molecular weight hydrogelators consisting of 1,3,5-cyclohexyl-tricarboxamide-phenylalanine have been investigated.37 The stability of the hydrogelators have been studied using various amino acid based substituents connected to the core.38 In this case, solution phase chemistry was used to synthesize these hydrogelators. 1.5 Self-assembly approaches in tissue engineering

A beautiful supramolecular system which can be used in tissue engineering has been developed by Stupp et al. This system consists of peptide-based amphiphilic molecules that form threedimensional nanofibres (Fig. 1.10). These peptide-amphiphiles (PA) are constructed of at least three important regions; a long alkyl tail, a (flexible) linker region consisting of amino acids and the bioactive part39,40. Cysteine residues could be built in between the alkyl tail and bioactive part, which can be oxidized to form disulfide bonds resulting in polymerization of the supramolecular structure39. These peptide amphiphiles (PA) were entirely synthesized on the resin.

It has been shown that two bioactive PA molecules are able to co-assemble into nanofibres by electrostatic interactions (Fig. 1.10A). While the negative charged PA 1 and 3 self-assemble in acidic pH, PA 2 and 4 with a positive charge self-assemble at a basic pH (Fig. 1.10A). The molecule pairs 1/2 and 3/4 co-assemble at neutral pH. Many applications of PA nanofibres have been investigated, varying from mineralization of hydroxyapatite crystals on cross-linked PA nanofibres39, oligonucleotide binding via introduction of oligonucleotide moieties41, magnetic resonance imaging using attached contrast agent molecules (like DOTA derivatives)42, templated assembly of lipophilic inorganic nanoparticles on the PA nanofibres via base-pairing43, to avidin binding to biotin presenting PA fibres44.


9

A B

NH

NH

NH

NH

NH

NH

NH

NH

NH

O

O

O

O

O

O

O

O

ONH

NH

NH

NH

O

O

O

NH2

OHO

O

COOH

NH

NH

NH

NH

NH

NH

NH

NH

NH

O

O

O

O

O

O

O

O

ONH

NH

NH

NH

O

O

O

NH2O

O

NH2

OH

OH

NH

NH2NH

NH

NH

NH

NH

NH

NH

NH

NH

NH

O

O

O

O

O

O

O

O

ONH

NH

NH

OHO

O

O

O

O

SH SH

SH SHP OHOHO

NH

NHNH2

OH

COOH

NH

NH

NH

NH

NH

NH

NH

NH

NH

O

O

O

O

O

O

O

O

ONH

NH

NH

NH

O

O

O

OSH SH

SH SH

NH2 NH2

NH2

O

1. C16H31O-NH-AAAAGGGEIKVAV-COOH

2. C16H31O-NH-AAAAGGGKYIGSR-CONH2

3. C16H31O-NH-CCCCGGGS(P)RGDS-COOH

4. C16H31O-NH-CCCCGGGKIKVAV-CONH2 Figure 1.10. Self-assembly of peptide amphiphiles (PA) into nanofibres. A. Chemical structures of 4 PAs. PA 1 and 3 self-assemble at acidic pH and PA 2 and 4 at basic pH. PA-couples of 1/2 or 3/4 co-assemble at neutral pH. B. Representation of an IKVAV-containing PA. Figure taken from Stupp et al.40

Finally, PA nanofibres containing the laminin IKVAV (Ile-Lys-Val-Ala-Val) can possibly be used for tissue engineering of nerves. Selective differentiation of neural progenitor cells was accomplished via incorporation of this IKVAV peptide40 (Fig. 1.10B). The PA nanofibre gels with this bioactive sequence induced very rapid differentiation of the cells into neurons while discouraging the development of astrocytes40. This well-designed hydrogel system has high potential as application for tissue engineering. In addition, it would be nice to design a system which is also based on supramolecular interactions, but displays strong, elastomeric material properties.

1.6 Aim of the thesis

Inspired by the view of an ideal scaffold and the self-assembly approaches shown in this chapter, we propose a novel, supramolecular approach to biomaterials. This approach is employed in order to bridge the gap between covalent modification and simply (non-covalently) mixing of molecules and polymers, and to simultaneously get control over mechanical properties and degradation behaviour. For this purpose, low molecular weight prepolymers and biomolecules are modified with quadruple hydrogen bonding ureido-pyrimidinone (UPy) units (Fig. 1.11A). These UPy-moieties dimerize strongly via quadruple hydrogen bonding and display high association constants (Kass = 106107 L mol-1) in organic solvents46-49. Using a modular approach, it is proposed that these UPy-functionalized molecules can assemble non-covalently via directional, specific interactions into co-polymeric and/or bioactive supramolecular materials (Fig. 1.11B). Materials can be produced without tedious synthetic procedures. Besides that, it becomes easy to vary the amount and nature of the bioactive molecules and the nature of the polymers. In this way, we try to regulate the

Chapter 1

10

mechanical properties, the degradability and the bioactivity of the supramolecular biomaterials. Ultimately, the supramolecular (bioactive) biomaterials will be applied as ideal scaffolds for regenerative medicine.

N

N

O N

HO

H

NH

N

N

ON

HO

H

NH

A B

Figure 1.11. The modular approach to bioactive supramolecular biomaterials. A. The self-complementary hydrogen bonding ureido-pyrimidinone (UPy) moiety in a supramolecular polymer. B. The modular approach to constructing bioactive materials with various properties via simply mixing different UPy-functionalized biomolecules (green and red moieties) with UPy-polymers.50 1.7 Outline of the thesis

The feasibility of the modular approach is investigated with a number of studies. Chapter 2 shows three different methods to functionalize oligocaprolactones with UPy-moieties. The UPy-polymers show several morphologies that can be used to get control over the dynamics and stability of the material when bioactives are incorporated. Additionally, the UPy-modified oligocaprolactones are studied with respect to their processability into scaffolds and their biological behaviour. The degradation behaviour is investigated when used in a co-polymeric system with UPy-functionalized oligo(ethylene glycol).

Besides UPy-oligocaprolactones, also UPy-modified oligo(trimethylene carbonates) are studied as described in chapter 3. Bifunctional and trifunctional UPy-oligo(trimethylene carbonates) are explored as biomaterial, especially their processability into 3D-scaffolds. Co-polymeric systems of bifunctional and chain-extended oligocaprolactones are shown in chapter 4. Different mixtures of both polymers are produced in order to regulate the material properties such as thermal behaviour and mechanical properties. In addition, the differences in biological behaviour are studied when subcutaneously implanted in rats.

To introduce bioactivity into the UPy-modified polymeric materials, several peptide sequences and two model proteins are functionalized with UPy-moieties. This is described in chapter 5. Two coupling methods are used to produce UPy-peptides by solid-phase synthesis, while the UPy-proteins are synthesized via expressed protein ligation. The UPy-functionalized peptides are assembled into UPy-modified oligocaprolactone films in order to proof the modular approach to bioactive materials. Cell adhesion experiments in vitro as well as in-vivo implantation studies are shown in chapter 6. The research in chapter 7 takes this approach a step further by introducing different binding motifs in the UPy-peptides and UPy-polymers in order to regulate the stability and dynamics of the bioactive films. Several film preparation methods are investigated to tune ligand availability at the surface of the films.

Finally, several examples towards engineering of supramolecular surfaces are shown in chapter 8 to investigate the scope and limitations of the supramolecular system presented in this


11

thesis. The UPy-UPy dimerization at the polymer-water interface is studied. Different examples are explored including supramolecular micro-contact printing, avidin binding to UPy-modified biotin containing surfaces, UPy-modified hydrogels and UPy-UPy dimerization in water between UPy-modified proteins. 1.8 References 1. Langer, R.; Peppas, N. A. AIChE J. 2003, 49, (12), 2990-3006. 2. Langer, R.; Tirrell, D. A. Nature 2004, 428, (6982), 487-492. 3. Hench, L. L.; Polak, J. M. Science 2002, 295, (5557), 1014, 1016-1017. 4. Griffith, L. G.; Naughton, G. Science 2002, 295, (5557), 1009-1010, 1012-1014. 5. Kim, B.-S.; Mooney, D. J. Trends in Biotech. 1998, 16, (5), 224-230. 6. Vacanti, J. P.; Langer, R. Lancet 1999, 354 Suppl 1, SI32-4. 7. Ratner, B. D. H., A.S.; Schoen, F.J., Lemons, J.E. Biomater. Sci. 2004. 8. Pego, A. P.; Grijpma, D. W.; Feijen, J. Polymer 2003, 44, (21), 6495-6504. 9. Burdick, J. A.; Chung, C.; Jia, X.; Randolph, M. A.; Langer, R. Biomacromolecules 2005, 6, (1), 386-391. 10. Pego, A. P.; Poot, A. A.; Grijpma, D. W.; Feijen, J. J. Biomed. Mater. Sci.: Mater. Med. 2003, 14, (9), 767-773. 11. Pitt, C. G.; Gratzl, M. M.; Kimmel, G. L.; Surles, J.; Schindler, A. Biomaterials 1981, 2, (4), 215-20. 12. Lu, L.; Peter, S. J.; Lyman, M. D.; Lai, H. L.; Leite, S. M.; Tamada, J. A.; Uyama, S.; Vacanti, J. P.; Langer, R.;

Mikos, A. G. Biomaterials 2000, 21, (18), 1837-1845. 13. Pego, A. P.; Van Luyn, M. J. A.; Brouwer, L. A.; van Wachem, P. B.; Poot, A. A.; Grijpma, D. W.; Feijen, J. J.

Biomed. Mater. Res., Part A 2003, 67A, (3), 1044-1054. 14. Boontheekul, T.; Mooney, D. J. Curr. Opin. Biotechn. 2003, 14, (5), 559-565. 15. Maskarinec, S. A.; Tirrell, D. A. Curr. Opin. Biotechn. 2005, 16, (4), 422-426. 16. Hirano, Y.; Mooney, D. J. Adv. Mater. 2004, 16, (1), 17-25. 17. Richardson, T. P.; Peters, M. C.; Ennett, A. B.; Mooney, D. J. Nature Biotechn. 2001, 19, (11), 1029-1034. 18. Seliktar, D.; Zisch, A. H.; Lutolf, M. P.; Wrana, J. L.; Hubbell, J. A. J. Biomed. Mater. Res., Part A 2004, 68A, (4),

704-716. 19. Ruoslahti, E. Annu. Rev. Cell Develop. Biol. 1996, 12, 697-715. 20. Hersel, U.; Dahmen, C.; Kessler, H. Biomaterials 2003, 24, (24), 4385-4415. 21. Maynard, H. D.; Okada, S. Y.; Grubbs, R. H. Macromolecules 2000, 33, (17), 6239-6248. 22. Maynard, H. D.; Okada, S. Y.; Grubbs, R. H. J. Am. Chem. Soc. 2001, 123, (7), 1275-1279. 23. Aota, S.; Nomizu, M.; Yamada, K. M. J. Biol. Chem. 1994, 269, (40), 24756-61. 24. Leahy, D. J.; Aukhil, I.; Erickson, H. P. Cell 1996, 84, (1), 155-64. 25. Groll, J.; Fiedler, J.; Engelhard, E.; Ameringer, T.; Tugulu, S.; Klok, H.-A.; Brenner, R. E.; Moeller, M. J.

Biomed. Mater. Res., Part A 2005, 74A, (4), 607-617. 26. Kantlehner, M.; Schaffner, P.; Finsinger, D.; Meyer, J.; Jonczyk, A.; Diefenbach, B.; Nies, B.; Holzemann, G.;

Goodman, S. L.; Kessler, H. ChemBioChem 2000, 1, (2), 107-114. 27. Klok, H.-A. J. Polym. Sci., Part A: Polym. Chem. 2005, 43, (1), 1-17. 28. Hawker Craig, J.; Wooley Karen, L. Science 2005, 309, (5738), 1200-5. 29. Lwik, D. W. P. M.; van Hest, J. C. M. Chem. Soc. Rev. 2004, 33, (4), 234-245. 30. Ayres, L.; Hans, P.; Adams, J.; Loewik, D. W. P. M.; van Hest, J. C. M. J. Polym. Sci., Part A: Polym. Chem.

2005, 43, (24), 6355-6366. 31. Lwik, D. W. P. M.; Linhardt, J. G.; Adams, P. J. H. M.; van Hest, J. C. M. Organic & Biomolecular Chemistry

2003, 1, (11), 1827-1829. 32. Reynhout, I. C.; Lwik, D. W. P. M.; van Hest, J. C. M.; Cornelissen, J. J. L. M.; Nolte, R. J. M. Chem. Commun.

2005, (5), 602-604. 33. Becker, M. L.; Liu, J.; Wooley, K. L. Biomacromolecules 2005, 6, (1), 220-228. 34. Vandermeulen, G. W. M.; Hinderberger, D.; Xu, H.; Sheiko, S. S.; Jeschke, G.; Klok, H.-A. ChemPhysChem

2004, 5, (4), 488-494. 35. Klok, H.-A.; Vandermeulen Guido, W. M.; Nuhn, H.; Rosler, A.; Hamley Ian, W.; Castelletto, V.; Xu, H.;

Sheiko Sergei, S. Farad. Disc. 2005, 128, 29-41. 36. Nishii, M.; Matsuoka, T.; Kamikawa, Y.; Kato, T. Org. Biomolec. Chem. 2005, 3, (5), 875-880.

Chapter 1

12

37. Friggeri, A.; van der Pol, C.; van Bommel, K. J. C.; Heeres, A.; Stuart, M. C. A.; Feringa, B. L.; van Esch, J. Chem. Eur. J. 2005, 11, (18), 5353-5361.

38. van Bommel, K. J. C.; van der Pol, C.; Muizebelt, I.; Friggeri, A.; Heeres, A.; Meetsma, A.; Feringa, B. L.; van Esch, J. Angew. Chem., Int. Ed. 2004, 43, (13), 1663-1667.

39. Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Science 2001, 294, (5547), 1684-1688. 40. Silva, G. A.; Czeisler, C.; Niece, K. L.; Beniash, E.; Harrington, D. A.; Kessler, J. A.; Stupp, S. I. Science 2004,

303, (5662), 1352-1355. 41. Guler, M. O.; Pokorski, J. K.; Appella, D. H.; Stupp, S. I. Bioconj. Chem. 2005, 16, (3), 501-503. 42. Bull, S. R.; Guler, M. O.; Bras, R. E.; Venkatasubramanian, P. N.; Stupp, S. I.; Meade, T. J. Bioconj. Chem. 2005,

16, (6), 1343-1348. 43. Li, L.-S.; Stupp, S. I. Angew. Chem., Int. Ed. 2005, 44, (12), 1833-1836. 44. Guler, M. O.; Soukasene, S.; Hulvat, J. F.; Stupp, S. I. Nano Lett. 2005, 5, (2), 249-252. 45. Niece, K. L.; Hartgerink, J. D.; Donners, J. J. J. M.; Stupp, S. I. J. Am. Chem. Soc. 2003, 125, (24), 7146-7147. 46. Beijer, F. H.; Sijbesma, R. P.; Kooijman, H.; Spek, A. L.; Meijer, E. W. J. Am. Chem. Soc. 1998, 120, (27), 6761-

6769. 47. Folmer, B. J. B.; Sijbesma, R. P.; Versteegen, R. M.; van der Rijt, J. A. J.; Meijer, E. W. Adv. Mater. 2000, 12, (12),

874-878. 48. Sijbesma, R. P.; Beijer, F. H.; Brunsveld, L.; Folmer, B. J. B.; Hirschberg, J. H. K. K.; Lange, R. F. M.; Lowe, J.

K. L.; Meijer, E. W. Science 1997, 278, (5343), 1601-1604. 49. Sntjens, S. H. M.; Sijbesma, R. P.; van Genderen, M. H. P.; Meijer, E. W. J. Am. Chem. Soc. 2000, 122, (31),

7487-7493. 50. Dankers, P. Y. W.; Harmsen, M. C.; Brouwer, L. A.; Van Luyn, M. J. A.; Meijer, E. W. Nature Mater. 2005, 4,

(7), 568-574.

2 Supramolecular biomaterials

In this chapter, it is shown that supramolecular polymers consisting of oligocaprolactones end-functionalized with quadruple hydrogen bonding ureido-pyrimidinone (UPy) moieties are eminently suitable as biomaterial. Different supramolecular structures were made using three coupling methodologies of a UPy-unit to an oligocaprolactone; the UPy-group was directly attached, was coupled via a urethane functionality, or via a urea group. The first, directly coupled UPy-polymer is assumed to only yield chain extension which results in the formation of higher virtual molecular weight polymers. The other two, in which the UPy-unit is coupled via a urethane or urea linkage, will form besides chain-extended polymers also UPy-UPy dimer stacks via urethane or urea hydrogen bonding in the lateral direction. Differences between the urethane and urea coupled UPy-polymers are supposed to be caused by the length and dynamics of the stacks owing to weaker versus stronger hydrogen bonding, respectively. The urethane coupled UPy-oligocaprolactone was studied extensively with respect to its processability, biocompatibility and degradability. This polymer was processed into several morphologies, varying from meshes, to films and grids, on which fibroblast cells were able to proliferate. Next to that, this UPy-polymer and other water-soluble UPy-moieties were shown to be biocompatible using several direct and indirect in-vitro toxicity studies. The polymer did not degrade in-vitro during a period of more than 100 days, however, the degradation was accelerated when lipase enzymes were used. Several blends were produced via intimate mixing of end-functionalized UPy-oligocaprolactone with UPy-modified poly(ethylene glycol) to investigate whether the in-vitro degradability could be tuned. However, the UPy-functionalized poly(ethylene glycol) dissolved within hours and no additional mass loss was observed for the UPy-oligocaprolactone. Moreover, it is proposed that supramolecular polymers are promising biomaterials owing to their low-temperature processability, biocompatibility and easy tunability by mixing and matching the right components into the desired scaffold.

Part of this work has been published: Patricia Y.W. Dankers, D.J.M. van Beek, A. Tessa ten Cate, Rint P. Sijbesma, E.W. Meijer, Polym. Mater. Sci. Eng. 2003, 88, 52; Patricia Y.W. Dankers, Martin C. Harmsen, Linda A. Brouwer, Marja J.A. van Luyn, E.W. Meijer, Nature Mat. 2005, 4 (7), 568

Chapter 2

14

2.1 Introduction During the last few decades the field of supramolecular chemistry has emerged as a leading

discipline. Complex structures have been designed and studied, varying from protein assemblies to responsive supramolecular polymers. The monomeric units in supramolecular polymers are held together via reversible and highly directional non-covalent interactions such as metal-ligand coordination, - stacking or hydrogen bonding.1 Because the non-covalent interactions are reversible, these materials are highly dynamic and are able to react to external stimuli, such as changes in solvent and temperature.

This chapter focuses on the use of hydrogen bonding in supramolecular polymer systems. Hydrogen bonds are formed between atoms such as oxygen or nitrogen that have a larger electronegativity than hydrogen. The atom to which the hydrogen is connected, is referred to as the hydrogen bond donor (D) and the other atom is called the hydrogen bond acceptor (A). The typical bonding energy of a hydrogen bond is 10-80 kJ/mol in the gas phase. Because single hydrogen bonds are too weak to form stable aggregates, arrays of multiple hydrogen bonds are often used. These arrays can be complementary or self-complementary.

N

N

O N

HO

H

NH

N

N

ON

HO

H

NH

A B

N

N

N NR'

O

O

R

H H

H

N

N

O

R

NH

O

NR'H

HN

N

NO

R

HNO

H

R'

H

N

N

N NR'

O

O

R

H H

H

N

N

NNR'

O

O

R

HH

H

N

N

O

R

NH

O

NR'H

H

N

N

O

R

NH

O

NR' H

H

keto enol keto-2

D

D

A

A

D

D

A

A

A

D

A

D

D

A

D

A

Figure 2.1. The 2-ureido-pyrimidinone (UPy) unit. A. The different tautomers in which the UPy can exist; the self-complementary forms that are able to dimerize via quadruple hydrogen bonding (keto and enol), and the tautomer that cannot dimerize (keto-2). The hydrogen bonding arrays are indicated showing next to the primary hydrogen bonds (solid line), also the attractive (black arrow) and repulsive (grey arrow) secondary electrostatic interactions. B. Schematic representation of UPy-functionalized prepolymers. End-modification of telechelic prepolymers results in chain extension via UPy-UPy dimerization.

One of the first supramolecular complexes developed, is one between a host and its complementary barbiturate guest2, with an association constant of 2.1 104 M-1 in chloroform. An example of a self-complementary unit is the ureido-triazine moiety with a dimerization constant of 2.0 104 M-1 in chloroform3. Most interesting, the 2-ureido-pyrimidinone (UPy) group is a self-complementary quadruple hydrogen bonding unit3,4 with an even higher dimerization constant of 6 107 M-1 in chloroform, 1 107 M-1 in chloroform saturated with water and, 6 108 M-1 in

Supramolecular biomaterials

15

toluene.5 The UPy-moiety exists as a mixture of three tautomers of which two can dimerize, the keto and enol tautomer (Fig. 2.1A). Both dimerizing tautomers have different dimerization constants as a result of diagonal secondary electrostatic interactions.6,7 The keto tautomer displays an AADD array, whereas the enol form consists of an DADA array (Fig. 2.1A). The electrostatic effects have been quantified8 and it has been shown that each primary hydrogen bonding interaction showed a contribution of ~8 kJ/mol to the free energy of complexation. Each attractive or repulsive secondary interaction increases or decreases the free energy with 2.9 kJ/mol, respectively. This implies that the keto UPy-tautomer has less repulsive secondary interactions and therefore an higher dimerization constant then the enol UPy-tautomer.

Supramolecular polymers are formed when hydrogen bonding units are applied as associating end-groups of bifunctional molecules. The association constants must be sufficiently high to get a high degree of polymerization, which results in real polymer properties. A nice example of supramolecular polymer aggregates based on complementary units is shown by Craig et al., in which several oligonucleotides are used to produce self-assembled A-B type polymers via base pairing.9,10 An example of self-complementary units as associating moieties in supramolecular polymers is the UPy-unit. Short bifunctional molecules consisting of two UPy-units connected to a hexyl spacer have been studied4, showing that they display high degrees of polymerization in solution and in the bulk.

Next to small molecules as spacers between the associating units, also prepolymers have been investigated as spacing moieties. Polytetrahydrofuran (PTHF) has been modified with adenine and cytosine groups which resulted in a change of the materials properties11. Whereas the PTHF prepolymer is a waxy solid, the functionalized materials can be processed from the melt. The UPy-unit was also applied to yield chain-extended prepolymers via hydrogen bonding (Fig. 2.1B). Many prepolymers have already been modified with UPy-groups, varying from polydimethylsiloxanes, poly(ethylenebutylenes) (PEB) to polycarbonates, polyethers and polyesters.12-18 These functionalized polymers showed a dramatic change in material properties. For instance PEB is a highly viscous liquid, but becomes strong and flexible owing to UPy-functionalization13,16. An important issue is the purity of the functionalized material and its degree of functionalization.16 Mono-functionalized chains can act as chain stoppers and thereby decreasing the virtual molecular weight.

The UPy-modified materials have excellent properties in the bulk and are easy to process because of their low melt viscosities. Therefore, they might be applied in fields of cosmetics, printing, adhesives, coatings and personal care19. Here, we will investigate the applicability of supramolecular polymers based on UPy-groups in biomaterials research as an application in tissue engineering (TE). Therefore, FDA approved oligocaprolactone prepolymers were modified with UPy-moieties using different modification methods. The UPy-polymers were studied with respect to both their supramolecular material properties and biomedical relevant behaviour. Supramolecular blends were obtained via mixing UPy-oligocaprolactone with UPy-oligo(ethylene glycol) in order to possibly regulate in-vitro degradability. 2.2 Modular design

Several UPy-modified polycaprolactone (PCL) prepolymers were designed in order to investigate the coupling method of the UPy-group to the polymer with respect to the materials properties and morphology (Fig. 2.2). Therefore, oligocaprolactone (Mn = 2.0 kg/mol) was end-functionalized with a UPy-unit via direct coupling yielding PCL2000UPy2 (1), via a urethane group resulting in PCL2000UPy2 (2) or a urea functionality yielding PCL2000UPy2 (3). The

Chapter 2

16

dependence of the length of the PCL chain on the material properties was studied with an UPy-oligocaprolactone (Mn = 4.5 kg/mol) modified via a urea group (4). As control polymers, PCL prepolymers were functionalized with a benzyl ester-urea (5) or acid-urea (6) group. Finally, UPy-oligo(ethylene glycol) (Mn = 2.0 kg/mol) functionalized via a urethane group (7) was designed in order to make blends with UPy-oligocaprolactones to study possible degradability.

A

D

2

N

NH

NH

NH

O

ONH

O

O O

O

On

1

2

3

5

6

E

2N

NH

NH

NH

O

ONH

O

NH

O

O

On

2N

NH

NH

NH

O

OO

OO

n

7

C

2N

NH

NH

NH

O

ONH

O

NH

O

O

On

N

NH

NH

NH

O

ONH

O

OO n

2

NH

NH

ONH

O

O O

O

On O

O 2

NH

NH

ONH

O

O O

O

On OH

O2

B

41 2 3 4

Figure 2.2. Modular design of the modified prepolymers. A. Oligocaprolactone (Mn = 2.0 kg/mol) was modified with UPy-units, via directly coupling (1), via coupling of a urethane moiety (2) or a urea group (3). B. Oligocaprolactone (Mn = 4.5 kg/mol) was modified with UPy-groups via a urea functionality (4). C. As reference, oligocaprolactone (Mn = 2.0 kg/mol) was functionalized with a benzyl ester-urea (5) or acid-urea moiety (6). D. Oligo(ethylene glycol) (Mn = 2.0 kg/mol; 7) was end-modified with UPy-groups in order to produce blends with UPy-oligocaprolactone. E. Schematic representation of the UPy-modified oligocaprolactone polymers (1-4). 2.3 Syntheses

The syntheses of the compounds used are described in Scheme 2.1. First, two UPy-synthons13,20 1b and 2b, and a benzyl ester-urea-hexyl-isocyanate synthon 3b were synthesized (Scheme 2.1A). Methyl-isocytosine 1a was reacted with 1,1-carbonyldiimidazole (CDI) in dimethylsulfoxide (DMSO), resulting in 1b as a white solid with a yield of 98%. The UPy-hexyl-isocyanate synthon 2b was synthesized from methyl-isocytosine 2a in an excess of 1,6-hexanediisocyanate, yielding 98% 2b as a white powder. Synthon 3b was produced via first stirring glycine benzyl ester toluene-4-sulfonate in dichloromethane with sodium hydroxide in water and then reaction of the obtained glycine benzyl ester 3a with an excess of 1,6-hexanediisocyanate in toluene, which resulted in a white product with a yield of 67%.


17

N

NH

NH

NH

O

ONCO

OHO

OO

OHO O

n

n

OHO

OO

OHO O

n

n

A

ONH

NH

O

ONCO

ONH2

O

BO N

H

OOH

O

NH

O

O

OO

O

OOn

NH2 OO

O

OOn

2

2

C

D

EOH

OH

n

2

N

NH

NH2O

2a 2b

N

NH

NH2O

1a 1b

3a 3b

4a 4b

4c

NH2 OO

O

OOn

24c 1

2

N

NH

NH

NH

O

ONH

O

O O

O

On

2

NH2 OO

O

OOn

24c3 2

N

NH

NH

NH

O

ONH

O

NH

O

O

On

5

6

NH

NH

ONH

O

O O

O

On O

O 2

NH

NH

ONH

O

O O

O

On OH

O2

7

N

NH

NH

NH

O

ONH

O

OO n

2

N

NH

NH

NH

O

OO

OO

n

NN

N

NHO

O

NH

i ii

iii

iv

v

vi

vii

viii

x

ix

xi

Scheme 2.1. Synthesis of the modified prepolymers. A. Synthesis of the UPy-synthons (1b, = 98% and 2b, = 98%) and benzylester-synthon (3b, = 67%). i = CDI, DMSO, 80 C; ii = 1,6-hexanediisocyanate, 100 C; iii = 1,6-hexanediisocyanate, toluene, 21 C. B. Synthesis of an amine-terminated PCL prepolymer (Mn = 2.0 kg/mol; 4c, = 95%) from 4b, = 73%. iv = PCLdiol, DPTS, DCC, chloroform, 21 C; v = H2, EtOAc:MeOH, Pd/C. C. Synthesis of the three different UPy-PCL (Mn = 2.0 kg/mol) polymers; PCL2000UPy2 directly coupled (1, = 95%), PCL2000UPy2 (2, = 73%), PCL2000UPy2 urea coupled (3, = 79%). vi = 1b, chloroform, 75 C; vii = 2b, chloroform, DBTDL, 75 C; viii = 2b, chloroform, 75 C. D. Synthesis of the two control PCL (Mn = 2.0 kg/mol) prepolymers; PCL2000benzylester2 (5, = 95%) and PCL2000acid2 (6, = 60%). ix = 3b, chloroform, DBTDL, 75 C; x = H2, Pd/C, dioxane, t-BuOH, H2O. E. Synthesis of PEG2000UPy2 (Mn = 2.0 kg/mol; 7, = 58%). xi = 2b, chloroform, DBTDL, 75 C.

Chapter 2

18

Secondly, next to a commercially available hydroxy-terminated oligocaprolactone, an amine-terminated oligocaprolactone was needed to synthesize UPy-polymers 1, 3 and 4. To this end, commercially available hydroxy-terminated oligocaprolactone (Mn = 2.0 kg/mol) was reacted with 6-(benzyloxycarbonyl)amino-hexanoic acid 4a in chloroform in the presence of dicyclohexylcarbodiimide (DCC) and 4-(dimethylamino)pyridinium-4-toluenesulfonate (DPTS) (Scheme 2.1B). Protected polymer 4b was obtained as a white material in a yield of 73%. The amines were deprotected by hydrogenation using a Pd/C catalyst in ethyl acetate:methanol (4:1) (Scheme 2.1B), which resulted in 4c as a white material in a yield of 95%. The amine-terminated PCL 4c was immediately used for the subsequent reactions (to synthesize 1 and 3), because storage of 4c gives rise to amidation reactions of the amine with the ester functionality in the main chain.

The different UPy-oligocaprolactones 1-4 were synthesized according to scheme 2.1C. UPy-polymers 1 and 3 were obtained via reaction of the amine-terminated PCL 4c in chloroform with the UPy-synthons 1b or 2b, respectively. They were obtained as white fibrous materials in yields of 95% for 1 and 79% for 3. Especially, polymer 3 was obtained as flexible fibrous material. UPy-polymer 4 was produced in a similar manner as 3. However, -caprolactone had to be polymerized first, yielding a hydroxy-terminated oligocaprolactone with a Mn of 4.5 kg/mol. Subsequently, this polymer was functionalized with UPy-moieties as described for 3. UPy-polymer 4 was obtained as a flexible white fibrous material with a yield of 60%. The bifunctional UPy-oligocaprolactone with a urethane functionality (PCL2000UPy2 2) was obtained via reaction of hydroxy-terminated PCL with 2b in chloroform with dibutyltindilaurate (DBTDL) as catalyst12,13, which resulted in UPy-polymer 2 as a white fibrous flexible material with a yield of 73%. Noticeably, all reaction mixtures in which UPy-units were coupled to oligocaprolactones became more viscous in the course of the reaction, indicating the formation of high virtual molecular weight polymers due to dimerization of the UPy-moieties.

As controls, polymers 5 and 6 were synthesized, starting from hydroxy-terminated oligocaprolactone which was reacted with 3b in chloroform with DBTDL as catalyst (Scheme 2.1D). Polymer 5 was obtained as a white powder in a 95% yield. Subsequently, polymer 5 was hydrogenated with Pd/C as catalyst in a mixture of t-butanol, dioxane and water. This resulted in 6 as a slightly flexible, white material with a yield of 60%. Finally, commercially available oligo(ethylene glycol) was reacted with UPy-synthon 2b in chloroform using DBTDL as catalyst (Scheme 2.1E). The UPy-oligo(ethylene glycol) 7 was obtained as a white, brittle material in a yield of 58%.

All polymers were characterized with NMR and IR spectroscopy. The characteristic peaks of the hydrogen-bonded protons are clearly visible around 13.1, 11.9, and 10.1 ppm for all UPy-modified materials (1-4 and 7) as measured in deuterated chloroform. IR showed that the UPy-moieties are present in the UPy-modified materials as keto-tautomer with characteristic bands around 1700, 1669, 1587 and 1526 cm-1.21 For the UPy-oligocaprolactones modified via a urea functionality (3 and 4) an additional IR vibration could be seen around 1621 cm-1 which is assigned to the carbonyl-stretch of the urea-group.

Remarkably, the waxy hydroxy-terminated and amine-terminated oligocaprolactones became flexible materials owing to functionalization, especially for polymers 2, 3 and 4. Polymers 1 and 6 became slightly flexible. Additionally, waxy oligo(ethylene glycol) also showed a morphology change, as it became brittle owing to UPy-functionalization resulting in UPy-polymer 7. Whereas UPy-oligocaprolactones 1 and 2 were molecular dissolved at a concentration of 16 mM


19

in tetrahydrofuran (THF), the urea-coupled UPy-polymer 3 formed a physical gel at the same concentration (Fig. 2.3). This indicates the formation of strong aggregates of 3 (see below).

1 2 3 Figure 2.3. Different UPy-oligocaprolactones in THF (16 mM); PCL2000UPy2 directly coupled (1), PCL2000UPy2 urethane coupled (2) and PCL2000UPy2 urea coupled (3). 2.4 Polymer morphologies

Atomic force microscopy (AFM) was used to study the morphology of polymers 1-6. UPy-polymers 1 and 2 are single-phase materials at room temperature (Fig. 2.4), assuming that the morphologies shown are caused by crystallization of the PCL-part. The black regions in the phase plot are artefacts due to large height fluctuations in the sample. UPy-polymer 3 shows a completely different picture both in the height and the phase image (Fig. 2.4). Fibrous structures are found similar to those reported for bisurea containing thermoplastic elastomers22,23. The presence of these fibres is a strong indication that UPy-UPy dimers stack in the lateral direction as a result of hydrogen bonding between the urea groups. The apparent width is about 6 nm, which is mainly limited to the curvature of the probe tip and has the same order of magnitude as found for the bisurea stacks (10 nm).22,23 Although polymer 4 has the same UPy-urea moiety, it does not show fibrous structures at room temperature (Fig. 2.5). It displays much thicker and shorter fibrous crystals. These are probably caused by crystallization of the longer PCL-chains, which are slightly forced into a fibrous morphology.

As control polymers, the PCL2000benzylester2 (5) and PCL2000acid2 (6) were investigated with AFM (Fig. 2.5). These polymers show a similar trend as the UPy-modified polymers (1 versus 2 or 3, respectively). Polymer 5 shows a more crystalline morphology, probably caused by crystallization of PCL and the absence of hydrogen bonding between the benzyl ester-moieties. In contrast, polymer 6 shows fibre-like structures, similar as found for UPy-polymer 3. These fibres are proposed to be caused by lateral stacking of the carboxylic acid dimers as a result of hydrogen bonding between the urea and probably the urethane groups. Apparently, the benzyl-groups in polymer 5 prohibit the formation of dimer stacks. These findings are reflected in the macroscopic morphologies of 5 and 6; films of 5 turned out to be opaque and brittle, whereas films of 6 are clear and slightly flexible.

Chapter 2

20

2 31

height0 1 m height height0 1 m 0 1 m

phase phase phase15 nm 10 nm 10 nm

20o 25o 25o

Figure 2.4. Atomic force microscopy measurements on the UPy-polymers, PCL2000UPy2 directly coupled (1), PCL2000UPy2 urethane coupled (2) and PCL2000UPy2 urea coupled (3) with a PCL-part of Mn = 2.0 kg/mol. The images were recorded at 18 C. The enlargements show 0.2 m.

5 64



60o 40o 30o

Figure 2.5. Atomic force microscopy measurements on the urea coupled PCL4500UPy2 (4; Mn = 4.5 kg/mol), and on the reference polymers, PCL2000benzylester2 (5) and PCL2000acid2 (6). The images were recorded at 18 C


21

Temperature dependent AFM measurements were performed to investigate UPy-polymers 2 and 4 into more detail. Although UPy-polymer 2 also has the possibility of UPy-UPy dimer stacking in the lateral direction through hydrogen bonding between the urethanes, fibrous structures were not found at temperatures up to 40 C (Fig. 2.6A). This is possibly the result of the small difference between the melting points of the PCL and the presumed UPy-urethane melting point (see below).

UPy-polymer 4 behaves differently at elevated temperatures (Fig. 2.6B). At 40 C already some crystallites disappeared and some new, much thinner, fibres were formed. At 50 C almost all PCL crystallites have disappeared and the thickness of the fibres is reduced to 6 nm, which was also found for 3. This is the temperature at which the PCL part of the polymer is molten. Investigation of the thermogram of 3 (see below) shows, that the PCL part of UPy-polymer 3 is already molten at room temperature. This explains the observation of fibres of 3 at room temperature.

A B25 oC 40 oC 50 oC4 42



10o 25o 15o

Figure 2.6. Atomic force microscopy measurements on A. the UPy-polymer, PCL2000UPy2 urethane coupled (2) at 25 C and on B. the UPy-polymer, PCL4500UPy2 urea coupled (4) at 40 C and 50 C.

Thus, from this study it is assumed that hydrogen bonding in the lateral direction is necessary for the stacking of UPy-UPy dimers (Fig. 2.7). Therefore, the directly coupled UPy-polymer is proposed to be only chain-extended. UPy-polymers 3 and 4 are assumed to form hydrogen bonds between the urea-moieties. However, for UPy-polymer 4 it is proposed that, particularly at room temperature, there is a competition between crystallization of the (longer) PCL-part and the formation of lateral UPy-UPy dimer stacks, or that the crystalline PCL-parts mask the UPy-UPy dimer stacks. UPy-polymer 2 possibly forms stacks caused by urethane hydrogen bonding.

Chapter 2

22

However, also for 2 holds that the crystallization of the PCL-part is more prone to happen than the formation of the stacks or that the crystalline PCL-parts mask the UPy-UPy dimer stacks.

1 2 3 4 Figure 2.7. Schematic representations of the UPy-modified oligocaprolactones with a directly coupled UPy-unit (1), urethane coupled UPy-moiety (2) and urea coupled UPy-group (3 and 4). 2.5 Thermal properties

Differential scanning calorimetry (DSC) was used to confirm the proposal made above by studying the intrinsic thermal behaviour of polymers 1-6 at a scanning rate of 20 C/min (Fig. 2.8; Table 2.1). The thermograms of polymers 1 and 2 are similar. The glass transition temperature (Tg) and heat capacity (Cp) values are of comparable magnitude. Besides that, a crystallization peak (Tc) is observed in the heating run. However, the crystallization enthalpy (Hc) is lower for 2 than for 1. Also, the heat of fusion (Hm) is lower for 2, which means that 2 is less crystalline than 1. This is in accordance with visual observations as a film of 1 is more opaque than a film of 2. The melting (Tm) of the PCL-part is observed around 44 C for both 1 and 2. An additional melting peak is observed at 60 C for polymer 2. This is most probably due to melting of the lateral UPy-UPy dimer stacks caused by hydrogen bonding between the urethane groups.

A B

1

2

3

4

5

6

Figure 2.8. Thermal properties of A. the oligocaprolactones (Mn = 2.0 kg/mol) modified with UPy-units, via direct coupling (1), a urethane moiety (2) or a urea group (3) and of an oligocaprolactone (Mn = 4.5 kg/mol) modified with UPy-groups via a urea functionality (4) and B. as a control, of oligocaprolactones (Mn = 2.0 kg/mol) functionalized with a benzyl ester-urea (5) or acid-urea moiety (6). The second heating runs (20 C/min) are shown.


23

Table 2.1. Thermal properties of the modified oligocaprolactones 1-6 and hydroxy-terminated oligocaprolactone (Mn = 2.0 kg/mol; diol).

2nd run Tg Cp Hc Tc,top Hm,1 Tm,top1 Hm,2 Tm,top2 1 -59 0.58 -21 -27 57 44 - - 2 -58 0.43 -8 -23 35* 43 -* 60 3 -62 0.27 - - 15 22 14 132 4 -61 0.23 - - 32 39 9 122 5 -59 0.29 -29 3 32 48 - - 6 -57 0.11 - - 24 44 19 98

diol - - - - 62* 43 -* 49 Values of the second heating runs (20 C/min) are shown. The following variables are shown: the glass transition temperature (Tg),

the heat capacity Cp, the crystallization enthalpy (Hc), the crystallization peak (Tc,top), the heat of fusions (Hm,1 and Hm,2), the melting peaks (Tm,top1 and Tm,top2). The temperatures, T are given in C, the heat capacity Cp in J/gC and the enthalpies, H in J/g. * the total heat of fusion is depicted as Hm,1

A small Tg is measured for the urea-coupled polymers 3 and 4, while the Tm of the PCL-part is

found at 22 C and 39 C, respectively. An additional melting transition was observed at 132 C and 122 C, for 3 and 4, respectively. This transition is assumed to be caused by stacking of UPy-UPy dimers in the lateral direction owing to urea hydrogen bonding. Similar results were found for the two control polymers 5 and 6. Whereas, 5 shows similar behaviour as 1 and thus no lateral stacks are formed, polymer 6 behaves more like 2 or 3. Polymer 6 displays an additional melting peak at 98 C. However, this peak is very broad and found at much lower temperatures, which might indicate that the lateral stacks are not as strong and highly ordered as is proposed for polymers 3 and 4. Thus, the DSC measurements confirm the results obtained with AFM; UPy-UPy dimer stacks are formed in polymers 3, 4 and 6, whereas, polymer 2 is able to form weaker lateral aggregates. 2.6 Infrared spectroscopy

Polymers 1-4 were investigated with infrared spectroscopy (Fig. 2.9). In first instance, the polymers 1-3 were studied at 25 C and 60 C. All polymers show a strong vibration at 1725 cm-1, which is attributed to the carbonyl stretch vibration of the PCL-part24. The exact position of this peak differs between the various polymers at 25 C and shifts upon heating to 1730 cm-1 at 60 C. The higher wavenumber of the carbonyl vibration for polymer 3 at 25 C can be ascribed to the fact that this polymer already starts to melt at 25 C, whereas the other polymers are still in the solid state. UPy-polymer 3 shows a small vibration at 3333 cm-1, which can be ascribed to the hydrogen bonded N-H stretching vibrations of the urea. This peak is not observed for polymer 2, indicating that this polymer does not form strong hydrogen bond based stacks like 3 does, neither at 25 C nor at 60 C. However, it does not exclude this possibility as N-H stretching vibrations are often observed as very broad peaks and could therefore well be present but by other bonds.

The UPy-moieties of polymers 1-3 are in the keto-tautomeric form, because characteristic vibrations are found around 1700, 1669, 1587 and 1526 cm-1.21 An additional vibration is visible for polymer 3 at 1621 cm-1, which is assigned to the carbonyl stretch vibration of the urea group.

Chapter 2

24

A

2

3

(

1

B

2

3

1

* * **

* * **

Figure 2.9 Infrared spectra of UPy-modified oligocaprolactones (Mn = 2.0 kg/mol) to which the UPy-unit is directly coupled (1), coupled via a urethane moiety (2) or a urea group (3), recorded at 25 C (A) or at 60 C (B). The characteristic UPy-vibrations are depicted with *, the carbonyl stretch vibration of the PCL part is shown with and the carbonyl stretch vibration of the urea-group is depicted with .

IR spectra were measured at a range of elevated temperatures for UPy-polymers 1-4 (Fig. 2.10). For polymer 1 and 2, the only major change at rising temperature is found between 50-60 C or 60-70 C, respectively. A shift of the vibration at 1669 to 1660 cm-1 is observed, which correlates to the melting peaks found with DSC measurements for 1 and 2. The urea-coupled UPy-polymers 3 and 4 show a shift and decrease of the attributed urea vibration at 1621 cm-1 upon heating, until the peak vanishes around 130 C. The latter is assigned to the melting of the UPy-UPy dimer stacks caused by urea hydrogen bonding in the lateral direction.


25

B

C

2

3

1

A

D

4

* * **

* * * *

* * *

*

* * **

Figure 2.10 Infrared spectra of the UPy-polymers recorded at different temperatures. A. The directly coupled UPy- oligocaprolactone (1), B. coupled via a urethane moiety (2), C. via a urea group (3), or D. via a urea group with a longer PCL-part (4). The characteristic UPy-vibrations are depicted with *, the carbonyl stretch vibration of the PCL part is shown with and the carbonyl vibration of the urea-group is depicted with . 2.7 Mechanical properties

Possible differences in mechanical properties between polymers 2-4 were studied with uniaxial tensile testing (Fig. 2.11 and Table 2.2). Modified polymers 1, 5 and 6 turned out to be too brittle to perform tensile tests. Both solvent cast and compression-moulded samples were measured. A significant difference in mechanical properties is seen between the two methods of preparation except for polymer 3. This is possibly caused by the assumed differences in crystallinity. Polymers 2 and 3 seem to behave similar. Polymer 4 is the stiffest and strongest polymer, especially when prepared by solvent casting. Cyclic testing of polymer 4 turned out to influence the tensile properties (data not shown). Polymer 4 was stretched for more than 200 times in a strain-driven experiment. Repeatedly stretching of the polymer had no significant influence on the Youngs modulus. However, stress and especially strain at break decreased, which indicates fatigue. Approximately 5% strain deformation was shown.

Chapter 2

26

A B

2 3

4

4

23

Figure 2.11. Tensile testing graphs of the UPy-functionalized PCL prepolymers; the UPy-oligocaprolactone coupled via a urethane moiety (2; Mn = 2.0 kg/mol) or via a urea group (3; Mn = 2.0 kg/mol or 4; Mn = 4.5 kg/mol). Films were made via A. drop casting from chloroform or via B. compression moulding. Table 2.2. Tensile test data of films of the UPy-functionalized PCL prepolymers (2-4) made via drop casting or compression moulding.

cast E (MPa) yield (MPa) yield (%) break (MPa) break (%) max (MPa) 2 49 2 3.1 0.2 7 1 3.3 0.1 27 2 3.8 0.2 3 34 5 3.0 0.4 11 1 3.8 0.2 38 8 3.9 0.2 4 213 5 12 1 6 2 8.2 0.1 32 5 12.0 0.1

moulded E (MPa) yield (MPa) yield (%) break (MPa) break (%) max (MPa) 2 37 1 4.1 0.3 11 2 4.0 0.2 52 6 4.5 0.2 3 22 2 3.5 0.1 22 2 4.2 0.1 42 6 4.8 0.1 4 124 9 7.1 0.2 6 1 4.2 0.2 45 3 8.1 0.1

The Youngs modulus (E), yield-stress (yield), yield-strain (yield), stress at break (break), elongation at break (break) and the maximum stress (max) are shown.

We have noticed that the stress-strain behaviour depends on the synthesized batch of

polymer16 and on the film preparation method (see above). For example, films made via solvent casting from chloroform (this chapter) behave dramatically different than films produced by drop casting from tetrahydrofuran (THF) (Chapter 4). Furthermore, it is of major importance to have high control over the degree of UPy end-modification, because mono-functionalized chains can act as chain stopper, which results in a change of the mechanical properties. Polymers made via an optimized synthesis route have better mechanical properties16. In addition to that, the crystallinity of the PCL-part of the UPy-oligocaprolactones plays an important role. PCL can crystallize in time, by which the mechanical properties change. As an example, tensile testing on a different sample of polymer 4, showed strains at break between 90-180%. This illustrates the importance of a high degree of UPy-functionalization and a high control over crystallinity.


27

2.8 Processability Owing to the reversible nature of the hydrogen bonds, the PCL2000UPy2 2 material could

easily be processed via different techniques into several scaffolds (Fig. 2.12A) varying from films and fibres to meshes and grids. Films were made by solvent casting or compression moulding. Melt spinning and electrospinning were used to make fibres and meshes. Grids consisting of filaments with a width down to approximately 220 m were produced via Fused Deposition Modeling (FDM) which is a rapid prototyping technology that is used to build up 3D structures that can be used in tissue engineering25, 26. The dynamic nature of the supramolecular materials gives rise to low melt viscosity, which makes processing of these UPy-polymers at temperatures just below 80 C possible.

The mechanical properties of the different scaffolds were investigated using tensile testing (Fig. 2.12B). The Youngs moduli of the scaffolds are smaller for the FDM grid and electrospun mesh as compared to the Youngs modulus of the film; indicating the difference in porosity. Furthermore, the elongation at break differs enormously; the grids, fibres and meshes can be strained less far. These stress-strain measurements show that the mechanical properties strongly depend on the scaffold morphology.

film

fibres

electrospun scaffold

FDM

10 mm

10 mm

0.1 mm

10 mm

film

fibre

FDM grid

electrospun mesh

A B

Figure 2.12. Processability of the supramolecular UPy-materials. A. The PCL2000UPy2 polymer was processed into several scaffolds; films, fibres, meshes (SEM picture), and grids via Fused Deposition Modeling (FDM). B. The different scaffolds were tested with respect to their mechanical properties. 2.9 Cell proliferation on 3D-scaffolds

Mouse 3T3 fibroblast cells were cultured on the electrospun and FDM scaffolds in the presence of fetal bovine serum (FBS) (Fig. 2.13). Confocal laser scanning microscopy (CLSM) was used to visualize the cells with cell tracker green (colouring living cells green) and propidium iodide (colouring dead cells red). It is shown that the cells proliferate and become more spread out after 6 days in the case of the electrospun scaffold (Fig. 2.13.A). In the case of the FDM scaffold, cells also adhere and proliferate. The different cross-sections show that the cells adhere and spread on the edges and on top of the FDM fibres (Fig. 2.13B). Hardly any red cells were visible on both scaffolds, indicating that the cells are viable. Thus, 3D-scaffolds for TE applications could be produced with the UPy-polymers.

Chapter 2

28

electrospun scaffold day 1 day 6A

B FDM scaffold I. day 3 II. day 3

Figure 2.13. Mouse fibroblasts (7.2104 cells/well) cultured on scaffolds produced from PCL2000UPy2. The cells were visualized with confocal laser scanning microscopy with a cell tracker green (living cells) and propidium iodide (dead cells) staining after 1, 3 and 6 days of culturing. The scale bars represent 100 m, except for the optical microscopy picture of the FDM scaffold, that bar represents 400 m. A. The electrospun scaffold with cells visualized after two time points (day 1 and 6) and B. different cross-sections (I and II) of the FDM scaffold with cells visualized after 3 days of culturing. A colour version of this figure is available on page 159. 2.10 Biocompatibility

The scaffold materials must not be toxic and harmful to the surrounding tissue. Even their degradation products must not cause processes like cell death and inflammation. Particular standards for toxicity are the viability and proliferation rate of cells. Therefore, first the biocompatibility of the UPy-unit was tested in vitro with two water-soluble model compounds; tris(hydroxymethyl)aminomethane (Tris) and poly(ethylene glycol) (PEG; Mn = 5.6 kg/mol) modified with UPy-moieties (Fig. 2.14). These model compounds have been chosen because of the different synthetic methods applied; using a CDI-activated ethylpentyl isocytosine20 for the UPy-Tris and the UPy-hexyl-isocyanate synthon13 for the UPy-PEG.

The toxicity of the UPy-unit was tested with two different viability tests, the MTT toxicity27 and the LDH leakage test28, using rat macrophages. The tetrazolium-based colorimetric assay, or MTT test, is based on the conversion of the yellow 3-[4,5-dimethylthiazol-2-yl]-2,4-diphenyl tetrazolium bromide (MTT) to purple formazan crystals by metabolically active cells.27 The LDH (lactate dehydrogenase) leakage test is based on the leakage of NADH (nicotinamide adenine dinucleotide) and LDH oxidases out of cells when their cell membranes have become permeable.28 The toxicity of different concentrations of the molecules (1, 101, 102, 103 and 104 mM) was investigated. The tetrazolium-based colorimetric assay showed that the viability of the cells stayed above 80% for every concentration of UPy-Tris and stayed at least above 65% for every concentration of UPy-PEG (Fig. 2.14). The LDH leakage test displayed similar results, although the viability of the cells incubated with 1 mM of the two water-soluble UPy-molecules was somewhat lower when determined with the LDH leakage test. These tests strongly indicate that the UPy-moiety is not toxic and thus biocompatible.


29

N

NH

O NH

O

NH OH

OH

OH

N

NH

O NH

O

NH

NH

OO

O

On

A B

concentration (mM) concentration (mM)1 10-1 10-2 10-3 10-4 0 1 10-1 10-2 10-3 10-4 0

Figure 2.14. Toxicity studies on water-soluble UPy-moieties with NR8383 rat macrophages. MTT and LDH viability tests on A. UPy-Tris and B. UPy-PEG (Mn = 5.6 kg/mol).

Secondly, the toxicity of UPy-functionalized oligocaprolactone (PCL2000UPy2) was tested in vitro with an indirect LDH viability assay28 using 3T3 mouse fibroblasts. Medium was incubated with compression-moulded films of the selected materials (Fig. 2.15A); PCL2000UPy2 (P), latex (C1) as negative control (indicating low cell viability) and UHMWPE (C2) as positive control (indicating high cell viability). The viability of the cells in culture medium (C3) was set at 100% viability. The LDH test showed that the viability of the cells stayed above 80% for the UPy-modified oligocaprolactone, which again shows that the synthesized UPy-polymer is not toxic.

day 1 day 2 day 4

B

C

P C1 C2 C3

A

Figure 2.15. Biocompatibility of bifunctional UPy-modified oligocaprolactone (PCL2000UPy2). A. The indirect LDH viability test on PCL2000UPy2 (P), on UHMWPE (C2) as a positive control and on latex (C1) as negative control. The viability of the cells in culture medium (C3) was set at 100%. B and C. Proliferation of mouse fibroblasts on drop cast films of PCL2000UPy2 from B. THF or from C. chloroform. The scale bars indicate 100 m.

The proliferation of fibroblasts was investigated on similar films but then drop cast from tetrahydrofuran (THF) or chloroform. The fibroblasts were seeded with two cell densities (5.7 103 cells/cm2 and 2.3 104 cells/cm2) both giving similar results (Fig. 2.15). The cells were

Chapter 2

30

able to proliferate in time to ultimately form a confluent layer. No differences could be seen in proliferation rate of the cells when cultured on films drop cast from either THF or chloroform. Therefore, fibroblast cells proliferate well on the UPy-functionalized PCL, indicating the biocompatibility of the material.

Finally, subcutaneous implantation of PCL2000UPy2 films in rats confirmed these findings by showing no priming of the immune system (Chapters 4 and 6). The UPy-moieties and UPy-oligocaprolactones are assumed to be non-toxic and could be safely used as biomaterial with respect to its biocompatibility. 2.11 Degradation behaviour

The supramolecular PCL2000UPy2 material was compared with high molecular weight PCL with respect to its biodegradability in vitro. Polymeric films of PCL2000UPy2 and high molecular weight PCL were shaken in a phosphate buffered saline solution. The high molecular weight PCL samples showed even after 130 days almost no mass loss; only 0.5% of weight loss was observed. Moreover, no visible macroscopic changes occurred and the materials remained as plastic films. The PCL2000UPy2 films showed somewhat different degradation behaviour compared to PCL. Although after 130 days only 2% of the material was degraded, the films became less clear and more brittle, indicating an increase of crystallinity in time (see also Chapter 4). This was confirmed by differential scanning calorimetry (DSC) measurements: the heat of fusion became higher and the melting peaks slightly shifted to higher temperatures. The rearrangement is supposed to be a result of the reversible binding of the repeating units in the supramolecular materials. Tensile testing showed that the material became stiffer: an increase in the Youngs modulus and a decrease in strain at break upon degradation were observed.

Enzymatic degradation studies were performed on PCL2000UPy2 using different lipases: Amano lipase AYS, a lipase from Rhizopus niveus and a lipase from Thermomyces lanuginosus. No mass loss was observed up to 30 days using the Amano lipase AYS. However, the use of the lipase from Rhizopus niveus resulted in a mass loss of 12% after 30 days. However, no significant change in molecular weight was observed measuring gel permeation chromatography (GPC). In contrast, during enzymatic degradation of PCL2000UPy2 with the lipase from Thermomyces lanuginosus chain scission was demonstrated with GPC techniques. After already 15 days 90% mass loss was found. Thus, PCL2000UPy2 does hardly degrade in buffer but becomes more crystalline in time. The degradation of PCL2000UPy2 can be enzymatically tuned with lipases in vitro. 2.12 Modular blends

Supramolecular blends were produced via intimate mixing of UPy-modified oligocaprolactone 2 (PCL2000UPy2) and UPy-functionalized poly(ethylene glycol) 7 (PEG2000UPy2) with the intention to regulate the degradability of the produced material. It might even be possible that the presence of 7 could accelerate the dissolution or hydrolysis of 2.

First, different mixtures of 2 and 7 (100:0, 80:20, 60:40, 50:50, 40:60, 20:80 and 0:100 PCL2000UPy2:PEG2000UPy2) were studied with respect to their material properties. Thick films (~0.3 mm), were created, which resulted in materials with the properties summarized in Fig. 2.16. AFM measurements showed the presence of 7 at the surface of the mixed films (Fig. 2.16B and 2.16C) organized in platelets. Furthermore, IR-ATR spectroscopy confirmed this finding; upon addition of more 7 the C-O vibration of the ether (at 1100 cm-1) becomes larger, while the C-O vibration of the ester (at 1160 cm-1) decreases (Fig. 2.16E).


31

A B

C D

E F

G H

100:080:2060:4050:5040:60

20:800:100

100:080:20

60:4050:5040:60

20:80

0:100

100% 2 80:20 2:7

60:40 2:7 100% 7

72

2:7 2:72nd heating run 1st heating run

() 100:0 73 5 80:20 56 4 60:40 55 4 50:50 41 5 40:60 49 5 20:80 39 4 0:100 48 5

2:7

Figure 2.16. Mixtures of PCL2000UPy2:PEG2000UPy2 (2:7). AFM measurements at 18 C (left = height and right = phase image) on thick (~0.3 mm) films of A. 100% PCL2000UPy2 (1 m images; z-range height = 200 nm; z-range phase = 50), B. 80:20 PCL2000UPy2:PEG2000UPy2 (2 m images; z-range height = 250 nm; z-range phase = 50), C. 60:40 PCL2000UPy2:PEG2000UPy2 (2 m images; z-range height = 100 nm; z-range phase = 50) and, D. 100% PEG2000UPy2 (2 m images; z-range height = 200 nm; z-range phase = 40). Thick films of the following ratios of PCL2000UPy2:PEG2000UPy2, 100:0, 80:20, 60:40, 50:50, 40:60, 20:80 and 0:100 were investigated using, E. IR-ATR spectroscopy, F. Water contact angle () measurements and, G. DSC measurements (second heating run 20 C/min) and H. DSC measurements (first heating run 20 C/min).

Chapter 2

32

Differences between the bottom and top of the films could be seen, showing stronger vibrations of C-O of the ether at the bottom than at the top of the film (data not shown). Water contact angle measurements showed a decrease of the contact angle owing to addition of 7 to 2, resulting in the formation of more hydrophilic surfaces (Fig. 2.16F). Noticeable is that the water contact angles are different at the top than at the bottom of the drop cast films. For 100% 2 no difference was found, but for the films containing 7 the contact angles at the bottom became around 30. DSC shows an increase of the heat of fusion upon addition of 2 to 7 (Fig. 2.16G and 2.16H). The PCL-part has crystallized to a large extent in the blends, because the heat of fusion of the PCL-part is larger in the first than in the second heating run. Furthermore, the mixed films became more brittle owing to a higher concentration of more crystalline 7.

A C

B D

t0 - 1st

t4 - 1st

t0 - 2nd

t4 - 2nd

t0 - 1st

t4 - 1st

t0 - 2nd

t4 - 2nd

80:20 2:760:40 2:7

60:40 2:7 80:20 2:7

Figure 2.17. Mixtures of PCL2000UPy2:PEG2000UPy2 (2:7) after incubation in PBS buffer at 37 C. A. AFM measurements at 18 C (left = height and right = phase image) on the incubated thick (~0.3 mm) films of 80:20 PCL2000UPy2:PEG2000UPy2 after 4 months (2 m images; z-range height = 300 nm; z-range phase = 90) and B. DSC measurements on the corresponding mixture (80:20) before (t0) and after incubation for 4 months (t4); the second (2nd) and first (1st) heating runs 20 C/min are shown. C. AFM measurements at 18 C (left = height and right = phase image) on the incubated thick (~0.3 mm) films of 60:40 PCL2000UPy2:PEG2000UPy2 after 4 months (2 m images; z-range height = 250 nm; z-range phase = 40) and D. DSC measurements on the corresponding mixture (60:40) before (t0) and after incubation for 4 months (t4); the second (2nd) and first (1st) heating runs 20 C/min are shown.

Secondly, the

Supramolecular biomaterials : introducing a modular approach

Documents

Transcript of Supramolecular biomaterials : introducing a modular approach