Optimization of Gelatin-Based Cellular Coating of MSC for ...
Transcript of Optimization of Gelatin-Based Cellular Coating of MSC for ...
University of Kentucky University of Kentucky
UKnowledge UKnowledge
Theses and Dissertations--Chemical and Materials Engineering Chemical and Materials Engineering
2021
Optimization of Gelatin-Based Cellular Coating of MSC for Optimization of Gelatin-Based Cellular Coating of MSC for
Myocardial Infarction Therapy Myocardial Infarction Therapy
Kara Amelle Davis University of Kentucky, [email protected] Author ORCID Identifier: https://orcid.org/0000-0002-9059-947X Digital Object Identifier: https://doi.org/10.13023/etd.2021.247
Right click to open a feedback form in a new tab to let us know how this document benefits you. Right click to open a feedback form in a new tab to let us know how this document benefits you.
Recommended Citation Recommended Citation Davis, Kara Amelle, "Optimization of Gelatin-Based Cellular Coating of MSC for Myocardial Infarction Therapy" (2021). Theses and Dissertations--Chemical and Materials Engineering. 132. https://uknowledge.uky.edu/cme_etds/132
This Doctoral Dissertation is brought to you for free and open access by the Chemical and Materials Engineering at UKnowledge. It has been accepted for inclusion in Theses and Dissertations--Chemical and Materials Engineering by an authorized administrator of UKnowledge. For more information, please contact [email protected].
STUDENT AGREEMENT: STUDENT AGREEMENT:
I represent that my thesis or dissertation and abstract are my original work. Proper attribution
has been given to all outside sources. I understand that I am solely responsible for obtaining
any needed copyright permissions. I have obtained needed written permission statement(s)
from the owner(s) of each third-party copyrighted matter to be included in my work, allowing
electronic distribution (if such use is not permitted by the fair use doctrine) which will be
submitted to UKnowledge as Additional File.
I hereby grant to The University of Kentucky and its agents the irrevocable, non-exclusive, and
royalty-free license to archive and make accessible my work in whole or in part in all forms of
media, now or hereafter known. I agree that the document mentioned above may be made
available immediately for worldwide access unless an embargo applies.
I retain all other ownership rights to the copyright of my work. I also retain the right to use in
future works (such as articles or books) all or part of my work. I understand that I am free to
register the copyright to my work.
REVIEW, APPROVAL AND ACCEPTANCE REVIEW, APPROVAL AND ACCEPTANCE
The document mentioned above has been reviewed and accepted by the student’s advisor, on
behalf of the advisory committee, and by the Director of Graduate Studies (DGS), on behalf of
the program; we verify that this is the final, approved version of the student’s thesis including all
changes required by the advisory committee. The undersigned agree to abide by the statements
above.
Kara Amelle Davis, Student
Dr. Bradley J. Berron, Major Professor
Dr. Stephen Rankin, Director of Graduate Studies
Optimization of Gelatin-Based Cellular Coating of MSC for Myocardial Infarction Therapy
________________________________________
DISSERTATION ___________________________________
A dissertation submitted in partial fulfillment of the
requirements for the degree of Doctor of Philosophy in the College of Engineering
at the University of Kentucky
By Kara Davis
Lexington, Kentucky Director: Dr. Bradley J Berron, Professor of Chemical Engineering
Lexington, Kentucky 2021
Copyright © Kara Davis, 2021 https://orcid.org/0000-0002-9059-947X
ABSTRACT OF DISSERTATION
Optimization of Gelatin-Based Cellular Coating of MSC for Myocardial Infarction Therapy
Cardiovascular disease remains the number one threat to American lives. During
an acute myocardial infarction (AMI), blood flow is blocked and results in the formation
of scar tissue. As the body’s immune system responds, inflammatory signaling causes an
increase in both scar tissue size and the patient’s risk for further chronic heart failure. In
order to reduce the risk of continued heart disease inflammatory signaling must be
reduced. Stem cell therapies have the ability to alter the immune system’s pro-
inflammatory signal. However, stem cell retention is limited due to blood flow shear.
Gelatin methacrylate (GelMA) based coatings have been shown to increase the retention
of these therapeutic cells inside the infarcted myocardium. This dissertation evaluates the
GelMA coating process and aims to optimize this therapy for future trials.
Biotin-streptavidin (SA) affinity is the foundation for our GelMA coating process.
GelMA coatings are grown from eosin bound to the cell surface through a biotin-SA
complex. In this work, we demonstrate the high binding affinity through a cardiac flow
model. We show that biotinylated cardiac cells can be immobilized onto a SA
functionalized glass substrate and withstand large shear. While biotin-SA affinity allows
for glass immobilization, we have yet to study what drives the adhesion of our biotin-SA
grown GelMA coatings in vivo. We examine two different binding sites for GelMA
coated cells: the extracellular matrix (ECM) or resident cells that make up the
myocardium. GelMA is created from the degradation of collagen, however, our studies
show collagen-binding integrins on the surface of cells do not promote adhesion of
GelMA. Through decellularization of heart ECM, we were able to show GelMA-ECM is
a strong binding mode that is driven by collagen-gelatin binding. As our coatings drive
adhesion through collagen-collagen binding, I hypothesize that increasing the amount of
GelMA on the cell surface will result in increased retention in vivo. In our previous work,
cells were biotinylated through covalent attachment to free cell surface amines. In an
effort to increase the density of biotin on a cell surface, we utilize a lipid insertion
approach. The use of cholesterol anchors increased streptavidin density on the surface of
cells further driving polymerization and allowing for an increased fraction of cells coated
with gelatin (83%) when compared to covalent methods (52%). Altogether, through the
use of cholesterol anchors and our in vivo understanding of coating specificity, we have
created a path towards improving the retention of MSC in vivo for post-MI therapy.
KEYWORDS: myocardial infarction, stem cell therapy, cellular coatings, biotin-
streptavidin affinity, cell surface engineering, lipid insertion
Kara Davis
[07/13/2021]
Optimization of Gelatin-Based Cellular Coating of MSC for Myocardial Infarction Therapy
By Kara Davis
Dr. Bradley Berron Director of Dissertation
Dr. Stephen Rankin
Director of Graduate Studies
07/13/2021 Date
DEDICATION
to me
-An inside family joke where we raise a glass and make a toast to ourselves. A realdedication to a family full of love and support fueled by laughter. I owe all my life
iii
ACKNOWLEDGMENTS
I would like to first thank my PhD Advisor, Dr. Brad Berron, for his
encouragement, direction and motivation over these last four years. While I admire his
passion for research I most admire his dedication to research topics that interest him and
make him happy. I would also like to give a special thank you to Dr. Ahmed Abdel-Latif
who has served as an extraordinary, supportive co-advisor. I am honored to have worked
alongside someone with such a wide knowledge in cardiovascular research. Also thank
you to my other PhD committee members Dr. Dziubla and Dr. Knutson and outside
examiner Dr. Chapelin.
Beyond the support of their excellent mentorship, I would also like to thank Dr.
Victoria King and the members of the Center for Clinical and Translational Science
training fellowship. I have learned tremendous knowledge through this opportunity. I am
also grateful for the support of my other lab members.
Overall, I owe the deepest thank you to my loving family. I have been blessed
with amazing parents who support me and have built me to be the successful woman I
am. I will never be able to repay you for the good you have done in my life. To my
brother who pushed me through my last few years of my bachelors, and my Aunt and
Uncle who opened their home to me when my college experience first began. Lastly, to
my loving fiancé who always puts my goals and my happiness above everything else.
iv
TABLE OF CONTENTS
LIST OF TABLES ............................................................................................................ VI
LIST OF FIGURES ......................................................................................................... VII
LIST OF ABBREVIATIONS ........................................................................................ XIII
CHAPTER 1. INTRODUCTION ................................................................................... 1
1.1 INTRODUCTION .................................................................................................................. 1
CHAPTER 2 BACKGROUND ..................................................................................... 6
2.1 MYOCARDIAL INFARCTION AND STEM CELL TREATMENT ............................................................. 6
2.2 STREPTAVIDIN AND BIOTIN BINDING ....................................................................................... 8
2.3 CELL SURFACE MODIFICATION .............................................................................................. 9
2.4 THE MYOCARDIUM EXTRACELLULAR MATRIX ......................................................................... 14
2.5 INTEGRINS ....................................................................................................................... 15
CHAPTER 3 INCREASED RETENTION OF CARDIAC CELLS TO GLASS
SUBSTRATES THROUGH STREPTAVIDIN-BIOTIN AFFINITY .............................. 18
3.1 INTRODUCTION ................................................................................................................ 18
3.2 MATERIALS AND METHODS ................................................................................................ 21
3.3 RESULTS AND DISCUSSION .................................................................................................. 26
3.4 CONCLUSION. .................................................................................................................. 36
CHAPTER 4 GELATIN COATING ENHANCES THERAPEUTIC CELL
ADHESION TO THE INFARCTED MYOCARDIUM VIA ECM BINDING ............... 37
4.1 INTRODUCTION ................................................................................................................ 37
v
4.2 MATERIALS AND METHODS ................................................................................................ 39
4.3 RESULTS AND DISCUSSION .................................................................................................. 46
4.4 CONCLUSION ................................................................................................................... 56
4.5 .............................................................................................................................................. 57
CHAPTER 5 INCREASED YIELD OF GELATIN COATED THERAPEUTIC
CELLS THROUGH CHOLESTEROL INSERTION ....................................................... 58
5.1 INTRODUCTION ................................................................................................................ 58
5.2 MATERIALS AND METHODS ................................................................................................ 60
5.3 RESULTS ......................................................................................................................... 65
5.4 DISCUSSION ..................................................................................................................... 74
5.5 CONCLUSION. .................................................................................................................. 78
CHAPTER 6 CONCLUSIONS AND FUTURE DIRECTION .................................. 79
6.1 CONCLUSIONS .................................................................................................................. 79
6.2 FUTURE DIRECTION ........................................................................................................... 81
APPENDIX ....................................................................................................................... 84
APPENDIX 1. CHAPTER 3 STATISTICAL VALUES .................................................................................. 84
APPENDIX 2. CALIBRATION OF CY3 ON EPOXY SLIDE .......................................................................... 85
APPENDIX 3. STREPTAVIDIN-PHYCOERYTHRIN CALIBRATION ................................................................ 87
REFRENCES ..................................................................................................................... 88
KARA DAVIS VITA ....................................................................................................... 98
vi
LIST OF TABLES
TABLE 2.1 ADVANTAGES AND DISADVANTAGES OF VARIOUS CELL SURFACE MODIFICATION
TECHNIQUES. .............................................................................................................. 14
vii
LIST OF FIGURES
FIGURE 2.1: BIOTIN AND STREPTAVIDIN BINDING DIAGRAM ............................................... 8
FIGURE 2.2: COMMON METHODS OF ANCHORING A COATING TO THE MAMMALIAN CELL
SURFACE. .................................................................................................................... 10
FIGURE 2.3: INTEGRIN CONFIGURATION AND HUMAN HETERODIMERS. ............................... 16
FIGURE 3.1: FUNCTIONALIZATION OF A GLASS SUBSTRATE TO INCREASE CELL RETENTION.
A) EPOXY COATING OF A GLASS SUBSTRATE FOLLOWED BY STREPTAVIDIN
CONJUGATION AND BINDING OF BIOTINYLATED CELLS. B) FLUORESCENT IMAGINING
OF SA-CY3 COATING ON EPOXY SLIDES AND NEGATIVE CONTROL / UNMODIFIED
SLIDES. STREPTAVIDIN DENSITY DETERMINED USING CY3 CALIBRATION SLIDE (SCALE
BAR = 1MM). ............................................................................................................... 20
FIGURE 3.2: VIABILITY OF CARDIOMYOCYTES DETERMINED BY TRYPAN BLUE STAINING
DURING THE FIRST 5 HOURS FOLLOWING RETRIEVAL. BLACK ARROWS INDICATE CM
WITH INTACT MEMBRANE. .......................................................................................... 27
FIGURE 3.3: VIABILITY OF UNMODIFIED H9C2 VERSUS NHS-BIOTIN MODIFIED H9C2
THROUGH CALCEIN, ETHIDIUM AND CASPASE ASSAYS. ............................................... 29
FIGURE 3.4: RETENTION OF H9C2 FOLLOWING PBS RINSING WHEN INCUBATED WITH
LAMININ AND POLY-L-LYSINE COVER GLASS COMPARED TO BIOTINYLATED H9C2 ON
AN ESA FUNCTIONALIZED EPOXY COVER. BSA BLOCKED GLASS SUBSTRATE WAS USED
AS A CONTROL. A) MICROSCOPIC IMAGES OF CELL RETENTION AT INCUBATION TIMES
OF 5, 10, 20 AND 40 MINUTES (SCALE BAR = 50 µM). B) SUMMARY PLOT OF CELLS
RETENTION VERSUS INCUBATION TIME. RETENTION RESULTS ARE AN AVERAGE OF 3
REPLICATES WITH 3 AREAS COUNTED PER COVERSLIP. ................................................ 30
viii
FIGURE 3.5: A) RETENTION OF CARDIOMYOCYTES FOLLOWING THE INTRODUCTION OF PBS
FOR UNMODIFIED CELLS ON BSA BLOCKED SUBSTRATE VERSUS BIOTINYLATED CELLS
ON ESA FUNCTIONALIZED SUBSTRATE (SCALE BAR = 50 µM). B) VIABILITY OF
UNMODIFIED CM VERSUS NHS-BIO MODIFIED CM DETERMINED BY TRYPAN BLUE
AND CALCEIN ASSAY. C) CONTRACTION OF ADHERED CARDIOMYOCYTES FOLLOWING
FLUIDIC STIMULI. ARROW INDICATES THE POSITION OF THE DEFORMATION WAVE
DURING CELL CONTRACTION. ...................................................................................... 31
FIGURE 3.6: TIME FOR DYE TO REACH FULL CONCENTRATION THROUGH INJECTION WITH
HAND PIPETTING VERSUS PERFUSION WITH 160 ML/MIN PARALLEL-PLATE FLOW
CHAMBER. RETENTION OF UNMODIFIED H9C2 ON PLAIN GLASS SUBSTRATE VERSUS
NHS-BIOTIN MODIFIED H9C2 ON ESA. MICROSCOPIC IMAGES OF CELL RETENTION AT
T=1, 4, 6 AND 9 SECONDS. ........................................................................................... 34
FIGURE 3.7: CALCIUM TRANSIENTS OBTAINED FROM SINGLE VENTRICULAR
CARDIOMYOCYTES. 1 HZ ELECTRICAL FIELD STIMULATED TO STEADY STATE
FOLLOWED BY CESSATION OF PACING FOLLOWED BY A 10MM CAFFEINE PUFF (BLUE
VERTICAL ARROW). FOR UNMODIFIED CM/COVER GLASS SYSTEM (N=5) VERSUS ESA-
BIOTIN SYSTEM (N=5). SYSTEMS ARE COMPARED BASED ON CM MOVEMENT AND
OBSERVED CA2+ TRANSIENT. (Y-AXIS=F340/F380, X-AXIS=TIME(S)) ......................... 35
FIGURE 4.1: INTEGRINS AND ECM COMPONENTS SERVE AS POTENTIAL BINDING MODES FOR
GELMA COATED BMC IN VIVO. A) SCHEMATIC OF POTENTIAL BINDING MODES FOR
GELMA COATED BMC WITHIN THE MYOCARDIUM. B) IMMUNOFLUORESCENT
IMAGINING OF INFARCT HEAR TISSUE STAINED WITH PRIMARY b1 ANTIBODY
ix
FOLLOWED BY DONKEY ANTI-RABBIT ALEXA FLUOR 568 NO PRIMARY ANTIBODY WAS
USED AS A NEGATIVE CONTROL. ................................................................................. 39
FIGURE 4.2:FLOW CYTOMETRY CONFIRMS EFFECTIVE GELMA COATING OF BMC. A)
SCHEMATIC OF GELMA COATING PROCESS THROUGH MODIFICATION OF CELL
SURFACE AMINES. B) FLOW CYTOMETRY ANALYSIS OF COATED VERSUS UNCOATED
BMCS STAINED WITH PRIMARY COLLAGEN ANTIBODY FOLLOWED BY SECONDARY AF-
647 (EITC, EOSIN ISOTHIOCYANATE). ........................................................................ 47
FIGURE 4.3: INCREASED RETENTION WAS OBSERVED FOR COATED VERSUS UNCOATED
CELLS. A) FLOW CYTOMETRY ANALYSIS OF DIGESTED HEART TISSUE DEMONSTRATES
HIGHER PERCENTAGE OF GFP+ CELLS IN MICE TREATED WITH COATED CELLS
COMPARTED TO MICE TRANSPLANTED WITH UNCOATED CELLS. PANEL A SHOWS THE
EXPRESSION OF CD45 AND GFP IN DIGESTED HEART TISSUE WHERE THE MAJORITY OF
GFP CELLS WERE CD45+. B) QUANTITATIVE ANALYSIS OF GPF+BMC IN DIGESTED
HEART TISSUE FOR COATED VERSUS UNCOATED CELLS BY FLOW CYTOMETRY (SIGNAL
IN THE PBS ARM REPRESENT BACKGROUND). C) IMMUNOHISTOCHEMICAL ANALYSIS
DEMONSTRATED HIGHER NUMBERS OF RETAINED GFP+ BMC IN MICE TREATED WITH
COATED CELLS. D) QUANTITATIVE ANALYSIS OF GPF+BMC IN DIGESTED HEART
TISSUE FOR COATED VERSUS UNCOATED CELLS BASED ON IMMUNOHISTOCHEMICAL
ANALYSIS (SIGNAL IN THE PBS ARM REPRESENT BACKGROUND). (N = 3 MICE/GROUP;
*P<0.05, **P<0.01). .................................................................................................. 49
FIGURE 4.4:EXPRESSION OF b1 INTEGRINS DOES NOT PROMOTE GELMA ADHESION. A)
EXPERIMENTAL PROTOCOL FOR EXPERIMENTS EXAMINING THE ROLE OF INTEGRINS IN
CELL ADHESION OF GELMA COATED CELLS. B) HUVEC AND H9C2 CELLS ANALYZED
x
BY FLOW CYTOMETRY FOR b1 INTEGRINS BY PRIMARY ANTI-b1 FOLLOWED BY
SECONDARY ANTI-AF647. C) MICROSCOPIC IMAGINING OF BMC INCUBATED WITH
CULTURED CELLS (HUVEC OR H9C2) THEN SUBJECTED TO RINSING BY
CENTRIFUGATION AT 800XG FOR 5 MINUTES. (N=5 SLIDES PER TEST GROUP). ............. 51
FIGURE 4.5: A GREATER NUMBER OF GELMA COATED BMCS ARE RETAINED ON DECELL
HEART TISSUE THAN UNCOATED BMC. A) QUANTITATIVE ANALYSIS OF THE NUMBER
OF BMC PER AREA DETERMINED BY IMAGE J BEFORE AND AFTER WASHING FOR
COATED VERSUS UNCOATED BMC. B) MICROSCOPIC IMAGES OF COATED VERSUS
UNCOATED GFP+BMC ON DECELLULARIZED TISSUE BEFORE AND AFTER RINSING.
(N=5 DECELL TISSUE PER TEST GROUP; P**<0.01) ...................................................... 53
FIGURE 4.6:COLLAGEN SURFACES SIGNIFICANTLY CONTRIBUTE TO THE RETENTION OF
GELMA COATED CELLS WHEN COMPARE TO OTHER ECM COMPONENTS. A) MEAN
NUMBER OF BMC PER AREA DETERMINED BY IMAGE J BEFORE AND AFTER WASHING
FOR COATED VERSUS UNCOATED BMC. B) MICROSCOPIC IMAGES OF GELMA COATED
GFP+BMC ON VARIOUS ECM SUBSTRATES BEFORE AND AFTER RINSING. (N=5 SLIDES
PER TEST GROUP; P*<0.05, P**<0.01) ........................................................................ 55
FIGURE 4.7: COATING IS ESSENTIAL FOR BMC ADHESION. A) AVERAGE AMOUNT OF BMCS
PER AREA DETERMINED BY IMAGEJ BEFORE AND AFTER WASHING FOR COATED VERSUS
UNCOATED BMC. B) MICROSCOPIC IMAGES OF UNCOATED GFP+BMC ON VARIOUS
ECM SUBSTRATES BEFORE AND AFTER CENTRIFUGAL RINSING. (N=5 SLIDES PER TEST
GROUP; P*<0.05, P**<0.01) ....................................................................................... 56
FIGURE 5.1: CELL SURFACE ENGINEERING METHODS TO INTRODUCE BIOTIN ONTO CELL
SURFACE. A) CELL SURFACE AMINES COVALENTLY BOUND TO SULFO-NHS-BIOTIN
xi
(NHS-BIO). B) CHOLESTEROL-PEG-BIOTIN (CHOLBIO) INSERTED DIRECTLY INTO
PHOSPHOLIPID BILAYER. ............................................................................................. 60
FIGURE 5.2: HIGH CONCENTRATIONS OF CHOLBIO ALLOW FOR A SIGNIFICANT INCREASE IN
CELL SURFACE SA OVER NHS-BIO. A) DEPENDENCY OF BIOTIN SURFACE DENSITY ON
SOLUTION CONCENTRATION OF CHOLBIO1K. B) IMPACT OF INCREASED
CONCENTRATION OF CHOLBIO1K ON MSC VIABILITY. ............................................... 67
FIGURE 5.3: MOLECULAR WEIGHT OF CHOLBIO PEG CHAIN DID NOT SIGNIFICANTLY
IMPACT SA LOADING. A) CHOLBIO AT 10µM RESULTED IN A SIGNIFICANT INCREASE IN
CELL SURFACE BIOTIN DENSITY WHEN COMPARED TO NHS-BIO (P*<0.05, P**<0.01,
N=5). DATA OBTAINED THROUGH PE QUANTIBRITE CALIBRATION BEADS. B)
VIABILITY OF CHOLBIO(X) AT 10 µM VERSUS NHS-BIO BY CALCEIN, ETHIDIUM AND
CASPASE ASSAYS. HIGHER MOLECULAR WEIGHT OF CHOLBIO RESULTED IN
SIGNIFICANTLY LOWER VIABILITY THAN NHS-BIO (P*<0.05, N=5). C) CHOLBIO AT 1
µM RESULTED IN SIGNIFICANTLY LESS BIOTIN ON THE CELL SURFACE WHEN COMPARED
TO NHS-BIO (P*<0.05, P**<0.01, N=5). DATA OBTAINED THROUGH PE
QUANTIBRITE CALIBRATION BEADS. D) VIABILITY OF CHOLBIO(X) AT 1 µM
VERSUS NHS-BIO BY CALCEIN, ETHIDIUM AND CASPASE ASSAYS. ............................ 69
FIGURE 5.4: CELL SURFACE MODIFICATION THROUGH LIPID INSERTION INCREASED THE
AMOUNT OF MSC EFFECTIVELY COATED WITH GELMA. A) FLOW CYTOMETRY
IMAGES OF GELMA PROCESSED GFP+MSC STAINED WITH ANTIBODY AGAINST
COLLAGEN FOLLOWED BY SECONDARY AF-647 ANTIBODY. B) CHOLBIO1K-Y GELMA
PROCESSED GFP+MSC RESULTED IN SIGNIFICANTLY HIGHER GELATIN COATED CELLS
THAN NHS-BIO PROCESSING (N=5, P*<0.01). ............................................................ 71
xii
FIGURE 5.5: CELL SURFACE MODIFICATION THROUGH LIPID INSERTION RESULTED IN MORE
UNIFORM COATINGS ON THE MSC SURFACE. A) REPRESENTATIVE MICROSCOPIC
IMAGES OF CHOLBIO(1K) ANCHORED GELMA COATING RESULTING IN MORE GELATIN
POSITIVE MSC, BY SECONDARY AF-647 ANTIBODY, THAN THE NHS-BIO ANCHORED
GFP+MSC. ZOOMED IN IMAGE OF INDIVIDUAL CELL SHOWS MORPHOLOGY OF CELL
COATING. (SCALE BAR = 50 µM) .................................................................................. 72
FIGURE 5.6: INCREASED COATING EFFICIENCY DID NOT IMPACT MSC ABILITY TO MIGRATE
AND PROLIFERATE. A) MICROSCOPIC AND QUANTITATIVE REPRESENTATION OF
CHOLBIO1K-10 COATED MSC VERSUS UNCOATED MSC MIGRATION ACROSS AN
INTRODUCED GAP. B) MICROSCOPIC AND QUANTITATIVE COMPARISON OF COLONIES
FORMED BY CRYSTAL-VIOLET STAINED COATED MSCS VERSUS UNCOATED MSC
AFTER 8 DAYS IN CULTURE. ........................................................................................ 73
FIGURE 5.7: 6 HOUR LPS STIMULATION. A) MACROPHAGE TNFA SIGNAL FOLLOWING
INCUBATION WITH CULTURED MEDIA (CM) FROM CHOLBIOAK-10 COATED MSC AND
UNCOATED MSC. B) MACROPHAGE IL-10 SIGNAL FOLLOWING INCUBATION WITH
CULTURED MEDIA (CM) FROM CHOLBIOAK-10 COATED MSC AND UNCOATED MSC.
................................................................................................................................... 74
xiii
LIST OF ABBREVIATIONS
AMI Acute Myocardial Infarction
AF Alexa Fluorescent
BM Bone Marrow
BMC Bone Marrow Cell
CFU Colony Forming Unit
CholBio Cholesterol-PEG-Biotin
CM Cardiomyocyte
CM* Conditioned Media (Chapter 5 only)
DMSO Dimethyl Sulfoxide
ECM Extracellular Matrix
EITC Eosin Isothiocyanate
eSA Epoxy Streptavidin Coated Surface
FBS Fetal Bovine Serum
FITC Fluorescein isothiocyanate
GelMA Gelatin Methacrylate
GFP Green Fluorescent Positive
H9C2 Rat Cardiac Myoblast
HUVEC Human umbilical vein endothelial cell
LAD Left Anterior Descending Coronary Artery
LV Left Ventricle
xiv
LVEF Left Ventricle Ejection Fraction
MSC Mesenchymal Stem Cell
NHS-Bio Sulfo-NHS-Biotin
PBS Phosphate Buffer Solution
PE Phycoerythrin
SA Streptavidin
1
CHAPTER 1. INTRODUCTION
1.1 Introduction
Acute myocardial infarction (AMI) causes nearly two American deaths every
minute [1]. Critically, AMI causes severe damage to the heart and leaves patients at
increased risk for chronic heart disease. MI is caused by a blockage of blood flow in the
patient’s coronary artery, starving the left ventricle of oxygen and causing localized cell
death. The immune system then responds in a pro-inflammatory state causing irreversible
scar tissue expansion and a weakening of the heart.
As the majority of damage done to the heart is caused by the body’s own immune
response, many have explored the use of immunomodulating and immunosuppressive
mesenchymal stem cells (MSCs). However, even when injected straight into the heart
wall, results have showed less than 1% of MSC are retained within the heart [2]. Without
the persistence of cells in the area of therapeutic need, the cell’s effect is diluted across
the body, and therapeutic efficacy is reduced [3, 4]. One way to overcome this
insufficient MSC retention is through cell surface modification [5]. Specifically, cells can
be engineered with gelatin based cellular coatings that increase their binding potential
withing the peri-infarct region [6].
Cell surface engineering is a technique commonly used throughout many
biomedical research areas including cancer [7, 8], drug delivery [9, 10] and,
cardiovascular protection [6, 11]. The field of cell surface engineering first achieved
significant attention in the 1990s when the Hubbell lab effectively encapsulated islets of
Langerhans and various cells with poly (ethylene glycol) [12-14]. Today, cell surface
engineering continues to have a major impact on diabetes treatment [15-17], however,
2
advances in this technology has expanded the application space of cellular coatings. More
sophisticated coating processes now allow for individual mammalian cells to be modified
with polymers [7, 18, 19]. Over the last few decades, cellular coatings have found use in
immunosuppression [20, 21], cell isolation [22-24], drug delivery [8, 9] and the adhesion
of cells [6, 11, 25].
In this work we utilize cell surface engineering as a method to increase the
retention of MSC to the peri-infract region. In our approach, we first modify the cell
surface using a biotin-streptavidin complex. Cell surface engineering commonly relies on
the use of non-covalent binding of biotin and streptavidin due to their high biological
affinity [26] and interchangeability. Commonly, biotin is bound to the cell surface
followed by streptavidin, which can be easily modified with other functional groups. As
biotin-streptavidin binding is the backbone to our overall goal of MSC adhesion to the
peri-infract region, we first study the strength of this complex in the presence of shear
flow.
Here, we model the strong biotin-streptavidin affinity through immobilization of
cardiovascular cells to a glass substrate. Optically monitoring cellular responses to an
environmental stimulus requires the cell to be immobilized within a microscope’s field of
view while the liquid environment is changed. The extreme environmental sensitivity of
primary, adult cardiomyocytes presents a significant challenge to immobilizing functional
cells for observation. Here, we engineer the peripheral membrane of cardiomyocytes
(CMs) with biotin to anchor these challenging cells to borosilicate microscope glass
substrates that have been functionalized with a streptavidin linkage. Use of the
streptavidin-biotin system allows for more than 80% cell retention under conventional
3
rinsing procedures while unmodified cells are largely rinsed away. Additionally, CMs
remain functional despite cell modification. Overall, we show that the high affinity
between biotin and streptavidin allowed for an adhesion force capable of immobilizing
cardiac cells onto a glass substrate after exposure to fluidic flow rates up to 160 mL/min.
Our research lab commonly applies this biotin-streptavidin binding in the coating of
individual MSC.
Gelatin based cellular coatings increase the retention of MSC inside the left
ventricle (LV) of an infarcted mouse heart model [6]. Increased retention of the coated
MSC resulted in decreased scar tissue expansion and improved cardiac function when
compared to uncoated MSC. These gelatin coatings are formed on the MSC surface
through photopolymerization. Eosin, a photoinitiator, is conjugated to streptavidin and
then bound to the cell surface through a biotin-streptavidin complex.
While our lab has preliminary data showing the therapeutic impact of gelatin
methacrylate (GelMA) coated cells in vivo, these experiments did not elucidate the
binding mode of GelMA to the myocardium. Further improvements upon these
promising in vivo results requires us to determine which of the two potential binding
partners for GelMA coated cells and myocardial tissue drive improved retention: the
extracellular matrix (ECM) or resident cells that make up the myocardium. In these
studies, we utilize the bone marrow cells (BMCs) that MSC are originally derived from.
While cardiac resident cells are of interest due to their expression of b1integrins that
mediate cell-ECM adhesion in vivo, resident cells did not promote binding to our GelMA
coated BMC. Alternatively, BMC incubation with decellularized heart tissue resulted in
higher adhesion of coated BMCs versus uncoated BMCs supporting our GelMA-ECM
4
binding mode. Further evaluation of the ECM binding mode including three different
heart ECM components: collagen, laminin and fibronectin. While all three components
promoted higher adhesion than unmodified glass, GelMA coated BMCs were retained at
a significantly higher rate to collagen over laminin and fibronectin. Overall, GelMA
cellular coatings significantly increase the binding of BMCs to collagen within the ECM.
Our previous in vivo model also showed that the significant improvement in MSC
retention was achieved with less than 50% of the MSCs being coated with GelMA [6]. As
increased retention of MSC was matched with improved cardiac function and reduced
scaring, it is reasonable to conclude that efforts in increasing the amount of GelMA
coated MSC will further increase the therapeutic benefit of MSCs for post-MI treatment.
Critically, we grow the adhesive, photoinitiated polymers from eosin placed on the
surface of the cell through by conjugating streptavidin-eosin to a biotinylated cell. In a
model where cells were coated with PEGDA, a higher surface density of photoinitiator
was correlated with a higher yield of polymer coated cells [23]. However, efforts
increasing the number of photoinitiator per streptavidin have been largely unable to
improve the polymerization process [27]. In order to increase the number of coated cells,
we have redirected our focus to increasing the number of biotin-streptavidin complexes
on the cell surface. Overall, biotin acts a limiting factor in the amount of eosin we can
bind to the cell surface.
In the majority of the prior studies from the Berron lab, we utilized the
covalent attachment of sulfo-NHS-Biotin (NHS-Bio) to cell surface amines for various
cell surface engineering applications [7, 9, 28], including coating MSCs [5, 6]. However,
efforts to increase the concentration of NHS-Bio resulted in widespread cell damage [9].
5
Lipid insertion, on the other hand, allows for biotinylating of the cell surface through
inserting lipids into the cells phospholipid bilayer. Specifically, cholesterol-PEG-biotin
(CholBio) directly inserts into lipid rafts in the peripheral membrane. By increasing the
density of biotin on the cell surface, we expect an increase the streptavidin and
photoinitiator density on the cell surface. In this work, we show that CholBio can be used
to achieve a cell surface streptavidin density on the order of 105 streptavidin molecules
per square micron. Due to the increased cell surface streptavidin density, a significant
increase in GelMA positive MSC was achieved when biotinylating the cell surface
through lipid insertion of CholBio (83%) over covalent modification with NHS-Bio
(52%). In addition to the increase in GelMA positive MSC, CholBio resulted in a more
uniform coating around the MSC surface. MSC were utilized in this study to show that
increased GelMA coating efficiency had not impact on MSC function.
While we utilize gelatin coatings for myocardial infarction therapy,
adhesion of cells has wide-spread application making this a desirable therapy. Difficulties
in MSC retention is a common issue that has been well studied [29, 30]. For instance,
MSC have also shown to be promising in children bone growth despite MSC retentions
being reported below 1% [31]. Application of MSC adhesion has also expanded to areas
of inflammation [32]and damaged brain tissue [33]. Optimization of the gelatin-based
coating allows us to move one step closer to human trials and mitigation of patient risk of
continued heart failure.
Chapter 2 BACKGROUND
2.1 Myocardial Infarction and Stem Cell Treatment
Acute myocardial infarction (AMI) remains a prevalent threat to American lives
[34]. MI is caused by a blockage of blood flow, and thus oxygen, to the coronary artery
resulting in localized cell death and scar tissue formation. Overload of pro-inflammatory
response cause expansion of the scar tissue well beyond the initial infarct region.
Expansion of the scar tissue results in loss contractile tissue leaving patients at an
increased risk for decreased left ventricle ejection fraction (LVEF) and potential heart
failure [35, 36].
The immune system immediately responds to MI by increasing the number of
circulating macrophages within in the damaged myocardium, the peri-infarct region in
particular [37]. While the macrophages occupy a complex spectrum of phenotypes,
macrophages are often grouped into groups of categories of M1-like (pro-inflammatory)
or M2-like (anti-inflammatory). In the post-MI environment, pro-inflammatory
macrophages and neutrophils first arrive at the scene and begin to repair damaged tissue
through phagocytosis and the release of enzymes. Afterwards, anti-inflammatory
macrophages arrive and begin to repair the tissue [38]. Unchecked pro-inflammatory
macrophages, arriving in the first wave of immune response, are the leading cause of
infarct expansion and heart failure progression [35, 36]. To reduce the risk of heart failure
and minimize scar tissue, the pro-inflammatory immune response must be controlled.
As the pro-inflammatory signal is the culprit in this damaging physiological
response, complete inflammatory suppression was thought to be a promising approach.
However, complete suppression of the immune response is just as harmful in other ways
7
[39]. For instance, Bulkey, et al. studied inflammatory suppression when a patient was
administrated glucocorticosteroids to hormonally drive an anti-inflammatory response
[40]. They showed the use of steroids lead to harmful effects such as delayed healing and
ventricular aneurysms. Olsen, et al. compared the use of both selective COX-2 inhibitors
and non-selective anti-inflammatory non-steroidal drugs, overall non-steroidal
approaches were found to be associated with higher mortality and reoccurring AMI [41].
Additionally, total removal of the inflammatory cells also showed no therapeutic benefit
[42]. As a result, there is a significant need for a therapeutic approach that does not result
in additional heart complications.
Many studies have shown that mesenchymal stem cells (MSCs) are effective
regulators of pro-inflammatory macrophages. For instance, MSCs have been used to
effectively control inflammation in therapy-resistant graft-versus-host disease [43]. MSC
work by driving macrophages from a pro-inflammatory state towards anti-inflammatory
state while still promoting tissue healing [39, 44]. Both the immunomodulatory and
immunosuppressive properties of MSCs make them ideal candidates for allogeneic
transplantation [45, 46]. Transplanted MSCs have also been shown to improve cardiac
protection, remodeling and perfusion [47, 48], as well as improve cardiomyocyte
circulation [49].
While there are many benefits of MSCs in MI therapy, clinical translation of this
promising therapy has been limited by poor retention of MSC within the patient’s left
ventricle [50, 51]. With MSC retention as low as 0.44% in some studies [2], there is a
critical need for innovative approaches to increase MSC engraftment and protection
inside the post-MI environment. Current methods have failed due to MSC washout into
8
the capillaries or alternate risk associated with large cell number injections [52-54]. The
Berron Lab has developed a gelatin-based cellular coating that increases the number of
implanted MSCs in heart tissue up to 7 days post infarction (dpi) [6]. This coating
technology also provides a significant improvement in cardiac preservation and
functional recovery 30 dpi. An in vivo mouse model also showed that there was a
significant decrease in scar tissue and an increase in LVEF for those administrated coated
MSC over uncoated MSC [55].
2.2 Streptavidin and Biotin Binding
One of the most common binding systems used in biological engineering is that
between streptavidin and biotin [26, 56, 57]. Avidin, streptavidin, and neutravidin all
have a high affinity for biotin (Kd=10-15M), and this creates an adaptable linkage between
a coating technology and cell surface binding antibodies [7, 23, 58]. Specifically, the
avidin-biotin complex has one of the strongest known non-covalent interactions between
a protein and a ligand [59]. Additionally, streptavidin, and similar avidin components, all
have the capability to bind four biotin molecules (Figure 2.1). While researchers use
avidin, streptavidin and neutravidin for various applications, streptavidin provides the
highest trade-off between specific and non-specific binding.
Figure 2.1: Biotin and Streptavidin Binding Diagram
9
Today, the biotin and streptavidin are used in a wide range of applications.
One of the most common applications is the detection of proteins. For instance,
biotinylated antibodies can be used in ELISA and immunohistochemistry staining [60,
61] Of particular interest, it is also commonly used to modify the surface of a cell. The
Mitragotri lab used biotinylated antibodies to anchor drug loaded patches on monocytes
for cell-directed cell therapies [58]. Their “cellular backpack” approach uses layer-by-
layer (LBL) fabrication with a magnetic NP-loaded layer for drug payload and an
attachment layer terminated with biotin groups that was exposed to streptavidin followed
by an antibody of choice allowing for monocyte attachment. Patch-adhered, purified
monocytes were then released from the solid substrate to yield monocytes with a LBL
coating on one side of the cell. The Berron lab has also used biotin-streptavidin in many
cellular coating applications [6, 7, 9, 18, 23]. Biotin, localized on the cell surface, is
bound to streptavidin conjugated eosin that when exposed to green light in a monomer
mixture, creates a polymer hydrogel coating around the cell. In the following work we
utilize this surface polymerization through biotin-streptavidin affinity to apply gelatin-
based coatings to the surface of MSCs.
2.3 Cell Surface Modification
The cell coating community benefits from several well-developed bioconjugation
techniques that have been developed for the broader concept of cell surface engineering.
These are categorized into covalent and noncovalent strategies (Figure 2.2) and are
extensively reviewed elsewhere [56] [57]. In this section, we specifically discuss the
conjugation methods as they relate to the design of a cell coating, and we emphasize the
advantages and disadvantages of each technique (Table 2.1). Many additional resources
10
are currently published for greater detail on chemical modification of cell surface in the
absence of a synthetic coating [56, 62-65].
Figure 2.2: Common methods of anchoring a coating to the mammalian cell surface.
2.3.1 Antibody Conjugation
Cell surface proteins act as convenient functional handles to engineer a cell’s
surface. Antibodies are commercially available for most abundant proteins found on
mammalian cells. Antibodies localize on the cell surface by antigen attachment and can
then be further modified to allow polymers or nanoparticles to attach to these antibodies
functionalizing the cell surface. Biotinylated antibodies are commercially available and
are commonly used for the attachment of cell coatings and commonly rely on following
streptavidin modification. For example, Anselmo et al. used biotinylated antibodies to
11
anchor drug loaded patches on monocytes for cell-directed cell therapies [58]. Their
“cellular backpack” approach uses layer-by-layer fabrication with a magnetic NP-loaded
layer for drug payload and an attachment layer terminated with biotin groups that was
exposed to streptavidin followed by an antibody of choice allowing for monocyte
attachment.
2.3.2 Electrostatic Interactions
The negative charge of a mammalian cell surface helps to drive necessary ions
such as potassium or calcium through the cell. The negative charge on cells results from
the presence of sialic acid residues on glycoproteins, and the negative charge is also
important in preventing cell aggregation [56]. For cell modification, the negative charge
on peripheral membranes serves as a convenient electrostatic binding site for positively
charged polymers, polycations, and nanoparticles. Studies have explored many different
polycations to modify mammalian cells including poly-L-lysine [66], polyallylamine
hydrochloride (PAH) [67, 68], and polyethyleneimine [21]. The most popular use of
electrostatic binding of a polymer to the cell surface is in the field of layer-by-layer cell
coatings [68-70].
2.3.3 Lipid Insertion and Hydrophobic Modification
Hydrophobic interactions are crucial in biology as they play a role in lipid bilayer
formation and protein folding and interactions [57]. The structure of the phospholipid
bilayer in the peripheral membrane creates a stable barrier between the cytoplasm and the
environment. By mimicking the hydrophobic structure of the bilayer, it is possible to
12
stably link unnatural components to the cell surface [57]. Given the abundance of
phospholipids in the peripheral membrane, modified lipids are rational choices for bilayer
integration. Lipid conjugations can be formed through covalent binding of polymers and
proteins to the lipid [56]. These lipid conjugates are readily incorporated into the bilayer
and have minimal impact of cell viability and function. Zhang et al. created drug-loaded,
multilayer coatings anchored to the cell surface with cholesterol [71]. A uniform coating
was formed on individual cells that resulted in controlled release of curcumin that
protected cells from reactive nitrogen and oxygen species while maintaining high cell
viability.
2.3.4 Receptor-Ligand Affinity:
Surface receptors offer a natural target for the modification of a cell’s surface. In
addition, this technique is commonly used to modify specific cell types. Hyaluronic acid
selectively binds to CD44 receptors on lymphocytes and has been utilized as an
anchoring strategy in a variety of layer-by-layer cell coatings [72, 73]. This strategy was
used to attach hyaluronic acid-coatings to B-lymphocytes, and the cell-coating
association is strong enough to resist mechanical separation [73].
2.3.5 Covalent Binding
Covalent binding is one of the most common methods of cell coating attachment.
Proteins, carbohydrates, and lipids on cell surfaces contain functional groups that serve as
covalent binding sites for materials. In contrast to the techniques for generating non-
natural functional groups on the peripheral membrane through biosynthetic or metabolic
13
pathways [57, 74], this section focuses only on the naturally-available binding sites on
the cell surface. The most commonly used, naturally present functional groups are amine
derivatives and thiols found in cell surface proteins.
Amine groups found in lysine groups and at the N-terminus of proteins are easily
coupled to carboxylic acid containing polymers and nanoparticles. The most common
form of amine group modifcation is through the N-hydroxysuccinimide (NHS) activation
of a carboxylic acid. A sulfonated derivative of NHS (sulfo-NHS) is frequently used in
cell surface modification for improved water solubility and minimal cellular
internalization [57]. Beyond NHS, other methods include cyanic chloride, which
covalently links to amines on the cell surface [57, 75], and succinimidyl valerates activate
carboxylic acids for amide bond formation [20, 76].
Thiol groups from cysteine in cell surface proteins are present from the reduction
of disulfide bridges or, less frequently, from naturally occurring free thiols. Stephan et al.
showed high levels of free thiols could be found on T-cells, B-cells, and hematopoetic
stem cells, but only minimal levels of free thiols were found on RBCs [77]. With high
levels of free thiols not being found in all cells lines, this leads to amino groups being
more commonly used due to their high presecence and ease of modification. When thiol
groups are avalailable they still serve as good covlaent modification sites due to their
higher nucelophilicity than amino groups [78].
14
Table 2.1 Advantages and disadvantages of various cell surface modification techniques. Attachment
Method Advantages (+) Disadvantages (-)
Antibody Conjugation
- Target Specific Cells [7, 23] - Great Cell Viability
- Cost of Production [79]
Electrostatic Interactions
- Simple and Inexpensive - Polycation Toxicity [57, 62] - Polymer Specific Drawbacks
[21, 66] Lipid Insertion
and Hydrophobic Modification
- Less Toxic than non-covalent approaches [56]
- Palmitoylation is reversible [57]
- Cholesterol modification hard to make and not always strong [57]
Receptor-Ligand - Non-robust [73] - Cell Specific Modification
- Uncontrolled aggregation of coated cells [73]
Covalent (Amine and Thiol)
- High Binding Strength [57] - High Stability [57]
- Non-reversible [56] - Concentration dependent cell
toxicity [9]
2.4 The Myocardium Extracellular Matrix
During MI, widespread damage to the left ventricle results in fibrotic scar tissue
and irreversible damage. Within the physiological environment, the ECM provides a
support for cellular structure and biochemical and biomechanical response [80].
Particularly, the ECM plays a crucial role in the development of the myocardium,
vascular system, valves and mechanical properties of the myocardium [81, 82]. Overall,
the ECM is a complex 3D structure made up of various substrates such as collagen,
proteins, enzymes and more.
Collagen is by far the most abundant component in the myocardium making up
33% of the cardiac ECM [82]. While there are numerous amounts of collagen found in
the body, Collagen 1 is the most abundant form in the human myocardium. While
collagen is the most abundant, laminin is the first glycoprotein found in the developing
heart and plays a crucial role in rebuild following MI. On the other hand, fibronectin has
15
a high abundance of collagen binding sites and is thus studied in the work. Fibronectin
plays a crucial role in the damaged heart as it directs macrophage movement and
stabilizes the ECM.
Following MI, the body attempts to repair itself by sending an out an immune
response. During this time of unchecked immune system, scar tissue expands, and critical
ECM damage occurs. Myofibroblasts are essential as they slow the spread of scar tissue
and reinforce contractile strength. However, throughout the first few weeks following MI,
collagen expression will increase from roughly 30% to 60% [83]. A study by Nawata, et
al showed a steady increase in collagen up to day 42 following MI, whereas fibronectin
increased initially but began to decrease after day 7 [84]. Laminin was also found to
increase and then return to baseline after 11 days post-MI [82].
2.5 Integrins
The myocardium is made up of a complex system of varying cell types and ECM,
and this complex system is mediated through two adhesion mechanisms: cell-cell
adhesion and cell-ECM adhesion. Integrins are the primary source of cell-ECM adhesion
and play one of the most crucial roles in cell-cell adhesion [85, 86]. These specialized
transmembrane proteins are unique in their ability to signal in both directions across the
plasma membrane [86]
Integrins are heterodimers, meaning they are made up of two main
subunits (Figure 2.3): α (alpha) and β (beta). Of the two subunits, the α chain is most
responsible for the ligand specificity while the β chain is connected to the cytoplasm and
has a direct impact on signaling [87]. Altogether, there exist 18 α-groups and 8 different
β-groups that can be found in 24 different configurations in human tissues [86, 87]. From
16
these 24 configurations, their ligand specificity can be narrowed down to four specific
receptors: collagen, laminin, RGD, and leukocyte-specific receptors (Figure 2.3). Overall,
integrins have been extensively studied with more than 1000 papers a year have been
published on integrins over the last few decades [86].
Figure 2.3: Integrin configuration and human heterodimers.
While there are 24 integrins within the human body, only a few have been
identified within the myocardium. Expression of integrins inside the heart varies based on
several factors including cell type, heart development, and trauma. For example, Chen, et
al. found that cardiac fibroblast expressed β1 subunits in conjugate with α1, α2, α3, α4,
α6, α16, and αv as well as αvβ3 and αvβ5 while endothelial cells expressed only α1β1,
α2β1, α5β1 and αvβ3 [88]. Hauselmann, et al. reported that β1 was expressed more than
17
any other β-integrin post-natal and was responsible for protecting cardiomyocytes during
oxidative stress [89].
In response to MI, integrin expression in the peri-infarct region is subject to
change due to cell death, scar tissue expansion and macrophage circulation. For example,
Nawata, et al. studied the change in expression of α1, α3, and α5, as well as collagen and
fibronectin, for 42 days following myocardial infarction in a rat model [84]. Sham-
operated and pre-operated rats were used as a control while the peri-infarct LV, non-
infarct LV and non-infarct right ventricle were studied. Within the first week through day
42 they observed a significant increase in α1 expression, as well as collagen, only within
the peri-infarct region. However, α5 expression was increased in all regions only at days
4 and 7 and there were no significant changes in α3 across any day or region. Fibronectin
was increased within the peri-infarct LV up to day 7. Altogether, these results show that
MI has a major impact on the spatial and temporal expression of specific integrins and
ECM components within the heart.
18
Chapter 3 INCREASED RETENTION OF CARDIAC CELLS TO GLASS SUBSTRATES THROUGH STREPTAVIDIN-BIOTIN AFFINITY
3.1 Introduction
In a critical effort to displace heart disease from its perennial position atop the
leading causes of death in the developed world [34], cardiac research has turned its focus
to cellular function. The syncytial nature of the heart allows the use of isolated
cardiomyocytes as surrogates for heart chamber function, and for the interrogation of
cellular and molecular mechanisms. For example, cytosolic Ca2+ is a central determinant
of heart function, and Ca2+ dyshomeostasis is associated with reduced contraction and
arrythmias [90]. Therefore, methods to inhibit and regulate the currents through the L-
type calcium channels in cardiomyocytes (CMs) are of great interest [90-93]. Multiple,
interdependent processes regulate sarcoplasmic reticulum Ca2+ load. Hence, precise
measurement of Ca2+ load serves as a key integrative surrogate measure for heart
function, and prediction of disease. A standard bioassay of sarcoplasmic reticulum Ca2+
load requires CM immobilization onto an optically clear glass surface.
CMs offer an added level of complexity compared to other primary cells due to
their unique elongated shape and rigidity [94]. CMs are prone to damage when exposed
to changes in temperature [95], prolonged studies, and routine separation and mixing
techniques. As a result, CMs require long separations by gravity or very gentle
centrifugation [95], while many other primary cells are readily pelleted with minimal
impact on cellular function [96]. In addition to the unique sensitivity of CMs, working
times are extremely limited as the loss of basic CM properties such as morphology [95]
and electrical and contractile forces are lost in culture [97]. These short CM working
times do not facilitate natural cell-substrate adhesion formed over several hours in culture
19
common to other cell types. While many labs have adapted the use of “cellular glues”,
Dvornikov, et al. noted that while performing cellular stretching any stretch beyond 20%
resulted in cell detachment despite the use of MyoTak cell glue [98]. In this work we
develop a method where CMs remain immobilized despite the rapid exchange of
stimulating solutions.
Cell surface engineering has emerged across many biological applications as a
way to adhere cells to targeted areas. Specifically, it has been used to adhere or target
cells to sites of inflammation [32, 99, 100], the myocardium [11, 101, 102], and various
substrates [103-105]. One of the most common interactions used in cell surface
engineering is that between biotin and streptavidin [7, 9, 18, 58, 104] due to their high
affinity and ability to form a single streptavidin bridge between up to four biotinylated
molecules. In particular, Iwasaki et al. showed that microarrays of functional, adhered
cells could be achieved by patterning streptavidin on a silicon based surface followed by
incubation with biotinylated cells [104]. In this work, we applied a simpler concept to
biotinylated cardiac cells onto streptavidin functionalized surfaces and demonstrated a
significant increase in cell immobilization despite high fluidic flow rates (Figure 3.1).
20
Figure 3.1: Functionalization of a glass substrate to increase cell retention. A) Epoxy coating of a glass substrate followed by streptavidin conjugation and binding of biotinylated cells. B) Fluorescent imagining of SA-Cy3 coating on epoxy slides and negative control / unmodified slides. Streptavidin density determined using Cy3 calibration slide (scale bar = 1mm).
Specifically, we studied the adhesive force of biotin-streptavidin when applied to
a cell-to-glass adhesion system. To accelerate the development of the adhesion system,
we first developed the principles of this system using a relative robust and easy to culture
cardiomyoblast cell line (H9C2). We determined the density of streptavidin bound to the
glass substrate and the impact of biotin on cardiomyoblast viability. Then, we determined
the percent retention of cells when subjected to increasing fluidic flow rates. We then
applied the adhesion system to freshly isolated primary CMs, where the adhesion enables
reliable in situ microscopic observation of CM functional assays. The health and function
of the CMs was confirmed through morphological assessments and the shear-induced
contraction of adhered CMs. Overall, the high affinity between biotin and streptavidin
allowed for an adhesion force capable of immobilizing cells onto a glass substrate while
exposed to fluidic flow rates up to 160 mL/min. Using a microfluidic device with the
adhesion system, cells were retained in the field of view while the buffer in contact with
the cells is exchanged in less than a second. Finally, we show that following the
21
introduction of caffeine effecting flow, CMs remain fully immobilized, and a calcium
transient can be obtained at an 80% success rate.
3.2 Materials and Methods
3.2.1 Glass Substrate Preparation.
Both glass microscope slides (VWR) and cover glass (Fisher) were cleansed in
ethanol for 30 minutes followed by plasma etching. Following cleansing, an epoxy
coating was applied to the glass substrates (adapted from Tsukruk, et al. [106]). Briefly,
glass substrates were incubated in an epoxysilane solution of 1% (3-glycidoxypropyl)
trimethoxysilane (Sigma Aldrich) in toluene (Sigma Aldrich) for 24 hours. Following
incubation, glass was rinsed several times with toluene and ethanol then allowed to dry.
After drying, slides were blocked for nonspecific binding with 5 mg/mL bovine serum
albumin (BSA) in PBS for 40 minutes. In order to verify epoxy coating and the ability to
bind SA, varying concentrations of Streptavidin-Cy3 (SA-Cy3, Invitrogen) in PBS (0.1,
02, 1, 2, 10, 20, 100, and 200 μg/mL) was incubated for 40 minutes onto epoxy-coated
slides using altering wells and then washed with PBS. Freshly prepared epoxy coated
glass was compared to uncoated microscope slides. Arrays were analyzed using an
Affymetrix 428 Array Scanner and Image J software. Cy3 calibration slide (Full Moon
Biosystems) was used to determine the average fluorophore density for each sample
(Appendix 2). An optimal concentration of 20 μg/mL SA-Cy3 was determined and used
for all further studies.
To compare retention of cells on unmodified glass to retention on various
functionalized glass substrates, cover glass was also coated in laminin and poly-L-lysine.
22
Cover glass was sterilized by ethanol and allowed to dry. Following, cover glass was
incubated with either 0.1 mg/mL poly-L-lysine (Sigma Aldrich) in PBS for 20 min or 5
µg/mL laminin (Gibco) in PBS for 2 hours both at 37˚C. Cover glass was rinsed with
molecular grade water and allowed to dry.
3.2.2 H9C2 Biotinylating and Viability.
Rat cardiac myoblasts (H9C2, ATCC CLR-1446) were cultured in Dulbecco
modified eagle medium (DMEM, VWR) supplemented with 10% fetal bovine serum
(FBS, VWR) and 1% penicillin/streptomycin (VWR) at 37˚C and 5% CO2. Cells were
seeded in T-182 cm2 tissue culture flasks (VWR) and grown to approximately 80%
confluence prior to use. Cells were aspirated using 0.25% Trypsin-EDTA 1X (VWR) and
then resuspended in 5 mL medium to neutralize the trypsin. Cells were pelleted at 400xg
and 4 ˚C for 3 min, washed three times in 1 mL PBS and then resuspended in PBS for
experiments.
Biotin was nonspecifically, covalently bound to proteins on the surface of the cell
using a biotin-succinimidyl ester conjugate (NHS-Biotin; EZ-link Sulfo-NHS-LC-Biotin,
Thermo Fisher). NHS-biotin was used at a concentration of 1 mM in PBS for 40 min with
a volume of 500 µL per million cells. Biotin conjugation was performed at room
temperature for primary CMs or on ice for H9C2. Cells were then rinsed and resuspended
with PBS or complete tyrodes for H9C2 or primary CMs, respectively.
In order to verify the ability to coat biotinylated cells with SA, biotinylated cells
were incubated with SA-PE at a concentration of 1 µL to 50 µL of PBS for 40 minutes.
Samples of 10,000 events were analyzed by flow cytometry, and the PE mean
23
fluorescence of SA-PE tagged samples was recorded. BD QuantiBRITE PE quantitation
beads were then used to quantify PE fluorescence by vortexing in 500 μL of PBS and
analyzing the PE signal. The linear relationship between PE amount and PE florescence
was determined using QuanitBRITE PE (Appendix 3) counting beads. Per Invitrogen SA-
PE is conjugated at 1:1 ratio of SA to PE. Using the ratio and MSC diameter recorded
after cell culture, SA density on the MSC surface was determined.
Following biotin conjugation, viability of H9C2 was tested using three different
methods: Calcein AM, Ethidium Homodimer-1, and CellEvent Caspase- 3/7. Unmodified
H9C2, used as a control population, and biotinylated H9C2 were separated into samples
of approximately 25,000 cells and incubated at room temperature with each of the
viability assays: Calcein AM at 1 μL in 10 mL PBS for 30 min, Eth-1 at 2 μL in 1 mL
PBS for 10 min, and Caspase at 1 μL in 1 mL PBS for 25 min. Cells were then rinsed,
resuspended in PBS and imaged using a Nikon Ti-U inverted microscope. Unmodified
cardiomyocyte viability was also observed across five hours using Trypan Blue at 100 uL
in 100 uL cell suspension then imaged immediately
3.2.3 Ventricular Myocyte Isolation and Biotinylation.
Adult ventricular cardiomyocyte isolation was performed as in previous studies
[107]. Prior to heart excision, mice were anesthetized with ketamine+xylazine
(90+10 mg/kg intraperitoneally). Hearts were excised and immediately perfused on a
Langendorff apparatus with a high‐potassium Tyrode buffer and then digested with 5 to
7 mg liberase (Roche). After digestion, atria were removed and left ventricular myocytes
24
were mechanically dispersed in stop buffer to quench enzymatic reaction (High K basal
perfusion buffer, FBS, 10 mM CaCl2). Calcium concentrations were gradually restored to
physiological levels in a stepwise fashion (10 mM CaCl2 to 100 mM CaCl2), and
only quiescent ventricular myocytes with visible striations and absence of membrane
blebs were used for adhesion studies. A vial containing a suspension of CMs was placed
at approximately 45° and allowed to separate by gravity for 40 minutes. Supernatant was
removed by pipetting, and cells were resuspended in 1 mM NHS-Bio, gently pipetted and
allowed to incubate for 40 minutes. CMs were again placed at approximately 45° and
allowed to separate by gravity. CMs were resuspended for further analysis.
Unmodified CMs, used as a control population, and biotinylated CMs were
incubated at room temperature with each viability assay: Calcein AM at 1 μL in 10 mL
PBS for 30 min and Trypan Blue at 100 uL in 100 uL cell suspension imaged
immediately. Unmodified CM viability was also observed across five hours using Trypan
Blue as described above.
All experiments were approved by the University of Kentucky IACUC in
accordance with the NIH Guide for the Care and Use of Laboratory Animals (DHHS
publication No. [NIH] 85-23, rev. 1996).
3.2.4 Cell Incubation with Cover Glass.
Biotinylated H9C2 cells were allowed to incubate with eSA cover glass for
varying amounts of time (5, 10, 20, and 40 minutes). For comparison, unmodified H9C2
were incubated with laminin or poly-l-lysine coated cover glass for the same time.
Unmodified H9C2 incubated with BSA blocked cover glass was used as a control.
25
Following incubation time, cells were subject to fluidic shear by dispensing 500 µL of
PBS in ~1 second using an Eppendorf fixed volume pipettor. Images before and after
PBS rinsing were captured with a Nikon Ti-U inverted microscope and used to quantify
retention of cells. Unless otherwise stated, all studies used a 20 minute incubation time
for cells on modified cover glass. The study was repeated using unmodified CM on BSA
blocked cover glass versus biotinylated CMs on eSA cover glass
3.2.5 Rapid Change in Cell Surface Concentration.
Biotinylated H9C2 were incubated on BSA blocked, eSA microscope slides for
20 minutes. Unmodified cells on BSA blocked microscope slides were used as a control.
Following incubation, 500 µL of a blue dye solution, 0.05% Trypan Blue (Sigma
Aldrich) in PBS, was introduced to cells using an Eppendorf fixed volume pipettor.
TinyTake software was used to record video from a Nikon Ti-U inverted microscope as
the change in dye concentration reached the cell surface. A relative concentration of the
dye was then determined by analyzing images at every second and determining the
relative intensity of pixels using Image J. The pixel level was then normalized for each
sample and a percent change from initial to final dye level was determined.
We further analyzed the CM adhesion in a flow chamber model. GlycoTech
Model 31-010 parallel-plate flow chamber, with a 0.005 inch gasket thickness, was used
to increase the speed at which dye is introduced to the cell surface. Due to the
configuration of the microfluidic device, the blue dye concentration was increased to
0.1% Trypan Blue in PBS to allow for optimal visualization of the change in dye
26
concentration. The microfluidic device was operated at a maximum flow rate of 160
mL/min.
3.2.6 Calcium Pathway Change Measurement.
Both biotinylated and unmodified wild-type mouse ventricular CMs were used to
obtain calcium transients. For both cases, cells were loaded with cell permeable fura2-
AM (Invitrogen) at 1.0 Hz to determine transient amplitude, upstroke velocity, and rate of
decay. All measurements were made following more than 2 min of conditioning of 1-Hz
field stimuli to induce steady-state. Transients were recorded at 1 Hz. For caffeine
induced transients, 10 mM caffeine in Tryode’s Solution was introduced to the cell
surface. All Ca2+ transient data were analyzed using IonOptix IonWizard 6.3.
Background fluorescence for 380 nm and 340 nm wavelength were determined from cell-
free regions. Data are expressed as F340/380 and were corrected for background noise.
3.3 Results and Discussion
Our goal was to immobilize cardiac cells on glass microscope slides so that they
would remain in a fixed location during the rapid exchange of liquid solutions. Rapid
solution exchange is necessary in primary cardiomyocyte experiments when studying
contraction, but high fluid shear stresses generated during exchange steps normally cause
cardiomyocytes to separate from the surface. This problem is exacerbated by the short
working time of primary CMs, where cell viability begins to rapidly diminish around 3
hours (Figure 3.2).
27
Figure 3.2: Viability of cardiomyocytes determined by Trypan Blue staining during the first 5 hours following retrieval. Black arrows indicate CM with intact membrane.
Due to their short working times, there is a significant need to perform studies
rapidly with as many of these precious cells as possible. Any loss of cells from
experimental view due to shear requires additional experimental runs and lost functional
time. By applying our adhesion system (Figure 3.1), this allows environmental changes in
microscopy-enabled cell function analyses without losing sight of the target cell. A
biotin/streptavidin system was chosen because the biotin and streptavidin linkage has one
of the strongest known non-covalent biological affinities [57]. By utilizing glycidyl-
silane monolayer chemistry, a glass surface can covalently bind free amine groups on
streptavidin, and the resulting streptavidin coated glass substrate can then strongly bind
biotinylated cells (Figure 3.1).
In order to confirm streptavidin was covalently bound to epoxy-functionalized
glass surfaces, Cy3-conjugated streptavidin (SA-Cy3) was deposited onto freshly
prepared epoxy. Using a Cy3 Full Moon Biosystems scanner calibration slide, the
28
fluorescent signal obtained from an Affymetrix 428 Array Scanner could be directly
related to the fluorophore density on the glass surface. Concentration of SA-Cy3 was
varied and it was determined the optimum concentration of SA-Cy3 was 20 µg/mL
(Appendix 2). While unmodified glass surfaces had a fluorophore density below the level
of detection, freshly prepared epoxy coated slides modified with 20 µg/mL SA-Cy3
showed effective conjugation (Figure 3.1) with a fluorophore density of 1.91E+3
Cy3/µm2. This loading density of SA-Cy3 is well above the observed maximum SA
binding of a biotinylated H9C2 (114 ± 21 SA/µm2, Appendix 3).
3.3.1 Viability of Biotinylated H9C2.
In order to immobilize cardiac cells onto the epoxy-streptavidin (eSA) glass
substrate, cells were first biotinylated using NHS-Biotin. NHS-Biotin binds to the cells
surface through a covalent linkage with free amine groups found on cell surface proteins
(Figure 3.1). In order to evaluate the impact of NHS-Biotin on the viability of H9C2, we
performed a Calcein assay of esterase activity, an ethidium assay of membrane integrity,
and a caspase assay to evaluate apoptosis (Figure 3.3). Statistical analysis showed no
significant difference between uncoated cells and NHS-biotin modified H9C2 in terms of
esterase activity, membrane integrity, and apoptosis activation.
29
Figure 3.3: Viability of unmodified H9C2 versus NHS-Biotin modified H9C2 through Calcein, ethidium and caspase assays.
3.3.2 Retention of Cells on Functionalized Surface.
Both laminin and poly-L-lysine coated glass cover glass are commonly used for
the adherence of unmodified cells in cell culture [95]. While these surfaces allow for
increased cell retention over uncoated cover glass, retention is still limited. After
pipetting 500 µL of PBS, the highest observed retention of unmodified H9C2 on BSA,
laminin or poly-lysine coated cover glass was approximately 0%, 30% and 50%,
respectively (Figure 3.4). By contrast, cell retention above 80% was achieved for
biotinylated H9C2 on eSA cover glass.
30
Figure 3.4: Retention of H9C2 following PBS rinsing when incubated with laminin and poly-L-lysine cover glass compared to biotinylated H9C2 on an eSA functionalized epoxy cover. BSA blocked glass substrate was used as a control. A) Microscopic images of cell retention at incubation times of 5, 10, 20 and 40 minutes (scale bar = 50 µm). B) Summary plot of cells retention versus incubation time. Retention results are an average of 3 replicates with 3 areas counted per coverslip.
Retention of cells versus incubation time was tested next. H9C2 cells were
incubated on each substrate at varying time (5, 10, 20 and 40 minutes. For the laminin
group, statistically relevant changes in retention were observed only when increasing
incubation from 10 to 20 minutes (Appendix 1). Cell retention with the eSA-Biotin
system improved significantly up to a 20 minute incubation, while no statistical
difference in cell retention was observed with longer incubations of cardiomyoblasts on
the glass substrate. Overall, the eSA-Biotin system achieved the highest retention rate at
any given duration of incubation.
31
Figure 3.5: A) Retention of cardiomyocytes following the introduction of PBS for unmodified cells on BSA blocked substrate versus biotinylated cells on eSA functionalized substrate (scale bar = 50 µm). B) Viability of unmodified CM versus NHS-Bio modified CM determined by Trypan blue and calcein assay. C) Contraction of adhered cardiomyocytes following fluidic stimuli. Arrow indicates the position of the deformation wave during cell contraction.
Based on the previous results using H9C2 cardiomyoblasts, we tested our
engineered adhesion system on primary CMs using a 20 minute incubation time. Given
that primary CMs are functional for only a few hours in vitro, the short incubation time
required to achieve adhesion is advantageous. Adult CMs were incubated with both BSA
blocked and eSA functionalized cover glass. While CMs did not adhere to the BSA
blocked cover glass (Figure 3.5A), high retention was observed for the eSA-biotin system
following rapid pipetting 500 µL of PBS, demonstrating that the use of biotin-
32
streptavidin affinity is sufficient to immobilize cardiac cells onto a glass substrate under
the flow conditions common to calcium studies on CMs.
Harsh processing for CM retrieval results in a lower overall viability than
observed with the H9C2 cell line used in the studies from Figure 3.3. However, the
viability of CMs was not statistically impacted by NHS-Bio modification based on
Trypan Blue (membrane integrity), and Calcein (esterase activity) assays (Figure 3.5B).
Another concern is contractile function: healthy adult CMs will commonly spontaneously
contract when exposed to fluidic shear [108]. In our experiments, NHS-Biotin modified
cells remained quiescent until stimulated by the fluidic shear of PBS pipetted onto the
surface (Figure 3.5C). Following the introduction of shear with a stream of PBS, CMs
spontaneously contracted. This observation supports the hypothesis that adult CMs may
be modified with NHS-Biotin and immobilized onto an eSA surface and retain contractile
cell function.
3.3.3 Change in Concentration versus Cell Retention.
A variety of experimental protocols call for rapid superfusion of a bolus of drugs,
inhibitors, or release agents (such as caffeine). However, if the cell is not immobilized
onto the surface or the shear from the superfusion is too strong, the cell can be washed
away from the viewing surface. The required superfusion rates can be higher than what
would be experienced under normal media change conditions; these rapid concentration
changes lead to increased shear stress that attempt to peel cells from their adhered
surfaces. To analyze this phenomenon, we assessed the rate in which concentration
changes as it reaches the cell surface and its impact on cell retention.
33
In order to see a clear change from the suspension buffer to the
introduction of a concentrated solution, trypan blue dye was used as a model agent for its
ease of optical measurement. H9C2 were incubated for 20 min with either unmodified or
eSA modified coerglass then placed on a Nikon Ti-U inverted microscope as 0.05%
Trypan Blue was introduced to the cell surface using a traditional fixed volume pipettor.
Change in color was used to determine the concentration of Trypan Blue as a function of
time (Figure 3.6). While the concentration change trend was the same for unmodified
cells versus the eSA-Biotin system, cell retention can only be observed in the eSA-Biotin
case (Appendix 1).
Additionally, we wanted to show that H9C2 would remain adhered with our eSA-
Biotin system when subject to more rapid changes in concentration. Here, a parallel-plate
flow chamber was used to control the rate at which solution was introduced to the cell
surface. Specifically, 0.1% Trypan Blue was pumped across both an unmodified surface
and the eSA-Biotin system at a rate of 160 mL/min (Figure 3.6). As seen in Figure 3.5,
the rate change of dye concentration was significantly increased (Appendix 1) while no
statistical change in cell retention was observed. Indeed, Figure 3.6 shows that even
under high fluidic shear (260 dynes/cm2), ~82% of the initially deposited cells remain in
the field of view.
34
Figure 3.6: Time for dye to reach full concentration through injection with hand pipetting versus perfusion with 160 mL/min parallel-plate flow chamber. Retention of unmodified H9C2 on plain glass substrate versus NHS-Biotin modified H9C2 on eSA. Microscopic images of cell retention at t=1, 4, 6 and 9 seconds.
3.3.4 Retaining Cells during Calcium Pathway Transient.
We challenged our adhesion system with a representative assay of CM
health which compares the amount of calcium released as a result of an electrical signal
to the total calcium present in the CM. The calcium released following electrical pulses
was quantified, and then each CM was induced to release all of its calcium stored in the
sarcoplasmic reticulum through rapid addition of caffeine [107, 109]. In a typical
experiment on unmodified CMs, only ~20% of the CMs were retained in the field of view
for measurement (Figure 3.7). More commonly, there was partial movement of the CM,
which induced artifacts in the calcium measurement, or the rapid addition of caffeine
moved the CM from the focal plane which precluded quantitation. In contrast, when our
eSA-biotin system was used to anchor the CM to the slides, all of the CMs were held in
the field of view and 80% of CMs yielded artifact-free calcium transients.
35
Figure 3.7: Calcium transients obtained from single ventricular cardiomyocytes. 1 Hz electrical field stimulated to steady state followed by cessation of pacing followed by a 10mM caffeine puff (blue vertical arrow). for unmodified CM/cover glass system (n=5) versus eSA-Biotin system (n=5). Systems are compared based on CM movement and observed Ca2+ transient. (y-axis=F340/F380, x-axis=time(s))
The short functional lifetime of CMs following a lengthy retrieval process
motivates the need for an efficient and rapid measurement of CM function. High failure
rates due to the movement of unmodified CMs across the slide will reduce the number of
CM function measurements per animal and reduce the statistical power of a given study.
Based on the results in Figure 3.7, a CM calcium analysis will yield data four times as
often when using eSA-Biotin than for an unmodified CM.
36
3.4 Conclusion.
Studies of many CM functions are hampered by their limited retention on
experimental surfaces and movement of the cells during observation. We have shown that
by utilizing biotin-streptavidin affinity we can immobilize cells onto a surface without
impacting their viability or functionality. Overall, cells were retained at a rate of 80% in
an analytical microfluidic device under flow rates of up to 160 mL/min and magnitudes
of fluidic shear exceeding physiological conditions. When considering the study of CMs
under high magnification, the field of view will typically only contain a single cell. This
adhesion system allows the rapid (<1s) exchange of an analyte solution while supporting
the likelihood that the cell under observation will be retained. Overall, we demonstrated
that this method can be applied directly to the measurement of calcium signaling in
individual CMs, and results in increased data collection efficiency from 20% to 80% by
binding a biotinylated CM to a streptavidin surface. The overarching significance of these
methods extends beyond studying primary CMs to the increasingly used induced
pluripotent derived CMs in mechanistic and therapeutic studies [110-112].
37
Chapter 4 GELATIN COATING ENHANCES THERAPEUTIC CELL ADHESION TO THE INFARCTED MYOCARDIUM VIA ECM BINDING
4.1 Introduction
Cardiovascular disease, including acute myocardial infarction (AMI), remains one
of the leading causes of death worldwide [1, 113]. During an AMI, the myocardium
undergoes irreversible damage to healthy tissue which is replaced by scar tissue thus
leaving patients at increased risk for developing heart failure [34]. The majority of the
damage caused to the heart tissue is caused by the body’s own immune response. For this
reason, many have explored the use of bone marrow derived cells (BMCs) that can
regulate the body’s immune system to facilitate myocardial repair [51, 114]. However,
many of these efforts have been limited due to historically poor cell retentions, typically
below 1% [2].
Cell surface engineering has emerged as a novel approach for addressing
cell adhesion issues in a wide variety of applications [26]. Specifically, our lab developed
gelatin-based cellular coatings that allow for a 3-fold increase of therapeutic
mesenchymal stem cells (MSCs) retention within the infarct myocardium [6].
Additionally, this increased retention of MSCs resulted in a significant reduction in scar
size compared to mice treated with uncoated MSCs. Overall, the use of gelatin based
cellular coatings promotes long-term patient health and decreased risk of repeated
cardiovascular disease.
The gelatin-based coatings used in previous studies are made up of a PEG
cross-linker and gelatin methacrylate (GelMA). One of the main advantages of using
GelMA is its well-established production and regulatory landscape [115]. While our lab
has already demonstrated an increase in in vivo retention for coated cells over uncoated
38
cells [6], the mechanism for GelMA adhesion in vivo has yet to be determined. The stem
cell injection site consists of two major components: the extracellular matrix (ECM) and
resident cells. Here, we compare two potential biding modes, gelatin-ECM and gelatin-
cell, to determine how GelMA coating increases the retention of BMCs in the infarcted
myocardium (Figure 4.1). Resident heart cells serve as potential binding sites for GelMA
coating due the interaction between b1 integrins and collagen [87]. The gelatin in the
coatings is a degraded collagen and will likely retain many binding motifs for b1
integrins. To evaluate the role of gelatin-cell interactions, we evaluate the adhesion of
GelMA coated BMCs to two different cell lines containing either high b1 expressing or
low b1 expressing.
Alternatively, the possibility exists that gelatin in the coatings interacts
with the ECM components of the myocardium. We first broadly probe gelatin-ECM
adhesion by evaluating coated cell retention on decellularized (decell) heart tissue. We
further investigate the role of specific ECM components in cell adhesion. Collagen is a
good candidate for GelMA binding due to the fact that it makes up 33% of cardiac ECM
proteins and that GelMA is a biological component derived from the breakdown of
collagen [82]. We also selected laminin as a target ECM protein, and fibronectin due to
the abundance of collagen binding sites [82].
39
Figure 4.1: Integrins and ECM components serve as potential binding modes for GelMA coated BMC in vivo. A) Schematic of potential binding modes for GelMA coated BMC within the myocardium. B) Immunofluorescent imagining of infarct heart tissue stained with primary b1 antibody followed by Donkey anti-rabbit Alexa Fluor 568 No primary antibody was used as a negative control.
4.2 Materials and Methods
4.2.1 Study Design.
Male C57BL/6J WT and C57BL/6-Tg(CAG-EGFP)1Osb/J mice (Jackson
Laboratory, BarHarbor, ME, strain code 003291), aged 8-10 weeks, were used in this
study. Mice were fed ad libitum with a normal chow diet (Harlan Teklad, 2018) and
randomly assigned to experimental groups. All procedures were conducted under the
approval of the University of Kentucky IACUC in accordance with the NIH Guide for the
Care and Use of Laboratory Animals (DHHS publication No. [NIH] 85-23, rev. 1996).
H9C2 (rat cardiac myoblast) were obtained from ATCC (CRL-1446). HUVEC (human
umbilical vein endothelial cells) were obtained from Lonza (C2517A).
40
4.2.2 Murine Model of Myocardial Infarction and Injection of BMC.
Myocardial infarction was induced under isoflurane anesthesia by permanent
ligation of the left anterior descending coronary artery (LAD). After making a small
incision over the fourth intercostal space, the heart was exposed, pushed out of the thorax
and the LAD was ligated at a site approximately 3 mm from its origin using a 6-0 silk
suture as previously described [116, 117]. Immediately following LAD ligation, 3 x
106 cells were injected in 2 different injections (25 μL total) in the peri-infarct region
using a 25 gauge needle. After injection, the heart was returned to intrathoracic space and
air was manually evacuated before suturing the incision site. Buprenorphine SR
(1.5mg/kg, ZOOPharm) was administered once immediately after surgery. Surgery and
cell injections were performed by a blinded experimenter.
4.2.3 BMC Encapsulation.
BMCs were nonspecifically covalently modified using a biotin-succinimidyl ester
conjugate (NHS-biotin; EZ-link Sulfo-NHS-LC-Biotin, Thermo Fisher). NHS-biotin was
used at a concentration of 1 mM in PBS for 40 min on ice with a volume of 500 µL per
million cells. Following biotin conjugation, BMC were rinsed with PBS twice and
aliquoted into 1.5 × 106 cell samples. The cell pellet was then resuspended in 1 mL of
PBS containing 35 μg/mL streptavidin-eosin isothiocyanate (SAEITC) and incubated for
30 min on ice. SAEITC conjugation was performed in-house, through formation of a
thiourea bond [118]. Briefly, SA (SigmaAldrich), dissolved in a carbonate buffer, and
eosin isothiocyanate (SigmaAldrich), dissolved in DMSO, were reacted at a 10:1 ratio for
41
8 hours at 4°C. The solution was then concentrated to 1 mg/mL in PBS and purified
through Zeba spin desalting column.
The polymer mixture buffer was prepared using 1% PEGDA 3400, 35 mM
triethanol amine, and 35 mM 1-vinyl-2-pyrrolidinone in PBS. PEGDA 3400 is added as a
low Molecular Weight crosslinking agent to promote cell coating uniformity [6]. Next, 3
wt% gelatin methacrylate (Allevi) was added to the buffer and sonicated for 30 min. The
final macromer solution (GelMA) was filtered through a 0.2 μm syringe filter and stored
at 37°C. Following filtration, 350 μL of the polymer mixture was added to the cell pellet,
gently vortexed and loaded into a chip-clip well (Whatman) with a standard microscopy
slide. The chip-clip was incubated in a dark, N2 purged, sealed, zip lock bag for 3 min
before polymerization. The cells were polymerized for 10 min at 30 mW/cm2 under 530
nm green light while purging with N2 followed by rinsing twice with 1 mL of PBS. To
determine the presence of a GelMA coating, BMC were stained with primary anti-
collagen 1 antibody (Abcam) followed by secondary Alexa Fluor 647 antibody
(Invitrogen). Samples of 10,000 events were analyzed by flow cytometry, and the mean
fluorescence for secondary antibody tagged cells was recorded. GFP-BMC were used for
flow cytometry analysis to reduce bleed over of FITC channel.
4.2.4 Flow Cytometry Analysis of Heart Tissue
Heart tissue was harvested at 7 days and placed in ice cold PBS instantly. Heart
tissue was minced then digested using a Collagenase B (Roche) and Dispase II (Roche)
solution for 30 minutes at 37°C with mixing every 5 minutes. The enzymatic reaction
was stopped by dilution with Flow Buffer (PBS + 5 % normal goat serum + 0.1 %
42
sodium azide) and the heart cell suspensions were passed through 40 μm cell strainers.
Cells were centrifuged at 400×g for 5 min at 4 °C, then suspended in flow buffer. Cells
were incubated directly for 30 minutes with eFluor 660 conjugated GFP Antibody Clone
5F12.4 (eBioscience, Thermo Fisher Scientific) and APC-CY7-conjugated CD45
(Biolegend). After incubation, cells were washed twice using flow buffer and analyzed
using an LSR II (Becton Dickinson) in the University of Kentucky Flow Cytometry Core.
Laser calibration and compensation were carried out utilizing unstained and single
fluorescent control beads (eBioscience). FlowJo v7 software was used to generate dot
plots and analyze the data.
4.2.5 Immunohistochemistry.
Immunohistochemical assessments were carried out on de-paraffinized and
rehydrated sections as previously described [117]. Briefly, sections were exposed to heat-
mediated epitope retrieval in citrate buffer, pH 6.0 (Vector Laboratories, Burlingame
CA)) for 20 mins, then blocked with normal goat serum for 10 minutes at 37 °C. Slides
were incubated with primary antibodies: rabbit anti-GFP (Abcam, Cambridge, United
Kingdom) or rat anti-mouse CD68 (Abcam). After washing, sections were incubated with
secondary antibodies conjugated to APC or FITC, respectively. The sections were finally
incubated with Sudan Black B (Sigma Aldrich, St. Louis, MO) for 30 minutes. Adjacent
areas in the peri-infarct and remote zones were analyzed (1 section/animal, n= 2
animals/group) at 40x magnification using Nikon Confocal Microscope A1 (Nikon,
Tokyo, Japan) in the University of Kentucky Confocal Microscopy facility. Calculations
43
were performed using the Cell Counter plugin for ImageJ, version 1.51d (NIH, Rockville,
MD).
For Beta 1 immunofluorescent analysis of infarcted tissue slices,
immunohistochemical assessments were carried out on de-paraffinized and rehydrated
sections as previously described [117]. Briefly, sections were exposed to heat-mediated
epitope retrieval in tris EDTA buffer, pH 9.0 (Vector Laboratories, Burlingame CA)) for
20 mins, then blocked (3% normal donkey serum + 0.3% Triton X-100 in 1x PBS) for 30
minutes at RT. Slides were incubated with primary antibody (1.5% normal donkey serum
+0.3% Triton X-100 in 1x PBS): rabbit anti-Beta 1 Integrin (Bioss, 1:500, #BS-0486R)
overnight at 4•C. After washing, sections were incubated with donkey anti rabbit Alexa
Fluor 568 (Abcam, 1:500, #ab175692) secondary antibody for 30 minutes at RT. Slides
were washed and coverslipped with Vectastain mounting medium with DAPI (Vector #
H-1200). Slides were imaged at 40X magnification on a Nikon AR1 confocal imaging
system in the University of Kentucky Confocal Microscopy Core.
4.2.6 Statistical Analysis of in vivo Results.
Values are expressed as mean ± standard error of mean (SEM). We used unpaired
Student t-test or analysis of variance (one-way or multiple comparisons) to estimate
differences, as appropriate. We utilized two-sided Dunnett or Dunn tests for post hoc
multiple comparison procedures, with control samples as the control category.
Throughout the analyses, a p-value less than 0.05 was considered statistically significant.
All statistical analyses for in vivo studies were performed using the Prism 7 software
package (GraphPad, La Jolla, CA).
44
4.2.7 Cell b1 Staining and BMC Adhesion.
H9C2 cells were cultured in Dulbecco modified eagle medium (DMEM, VWR)
supplemented with 10% fetal bovine serum (FBS, VWR) and 1% penicillin/streptomycin
(VWR) at 37˚C and 5% CO2. HUVEC cells were cultured in EGM-2 BulletKit (Lonza)
supplemented with 1% penicillin/streptomycin at 37˚C and 5% CO2. Cells were seeded in
T-182 cm2 tissue culture flasks (VWR) and grown to approximately 80% confluence.
Cells were aspirated using 0.25% Trypsin-EDTA 1X (VWR) and then resuspended in 5
mL medium to neutralize the trypsin. Cells were pelleted at 400xg at 4 ˚C for 3 min,
washed three times in 1 mL PBS and then resuspended in PBS for experiments.
H9C2 and HUVEC cells were stained for the presence of b1 integrin using anti-
b1 integrin (Bioss Antibodies) followed by secondary Alexa Fluor 647 antibody
(Invitrogen). Samples of 10,000 events were analyzed by flow cytometry, and the mean
fluorescence for secondary antibody tagged cells was recorded. FlowJo v7 software was
used to generate dot plots and analyze the data.
H9C2 and HUVEC cells were stained using cell tracker red (Invitrogen) and then
cultured in 4-well culture slides (Thermo Fisher) at 37°C and 5% CO2 until they reached
90% confluency. Following incubation, media was removed, and cells were rinsed with
PBS. Coated or uncoated GFP+BMC were added to culture slides at 2.5 × 105 cells/mL
and allowed to incubate for 40 minutes. Following incubation, BMC adherence was
evaluated by placing slides in a 50 mL centrifuge tube filled to the top with PBS and
centrifuging for 5 minutes at 800xg. Images before and after centrifugation were captured
with a Nikon Ti-U inverted microscope and used to quantify retention of cells.
45
4.2.8 Preparation of Decellularized (Decell) Tissue.
A mouse heart was obtained from the University of Kentucky, department of
animal and food sciences. The heart was frozen at -20°C for a minimum of 24 h prior
decellularization. After thawing, the heart was sectioned into pieces approximately 2 cm
in diameter and 1 mm thickness. These pieces were decellularized by room temperature
immersion in 1% sodium dodecyl sulfate in deionized water mixed with 0.5% penicillin-
streptomycin for 96 h, followed by 1% Triton X-100 in deionized water for 48 h, with
solution replaced every 24 h. Tissue was continuously shaking on a shaker table. The
scaffolds were washed with deionized water for two days. Decellularized heart tissue was
confirmed by Hematoxylin and eosin staining and DNA quantification by
PureLinkTM Genomic DNA Mini Kit (Thermo Fisher) according to the manufacturer’s
protocol.
4.2.9 Cell Incubation with Modified Substrates.
GFP+BMCs were used in these studies to detect BMC within the ridges of decell
tissue. GelMA coated BMCs were allowed to incubate with glass substrates coated with
either collagen (VWR), fibronectin (VWR), laminin (Gibco) or Decell Tissue for 40
minutes. Laminin coated slides were prepared in house by covering sterile glass
microscope slides with 5 µg/mL laminin in PBS for 2 hours at 37˚C. Following
incubation, BMC adherence was evaluated by placing slides in a 50 mL centrifuge tube
filled to the top with PBS and centrifuging for 5 minutes at 800xg. Images before and
after centrifugation were captured with a Nikon Ti-U inverted microscope and used to
46
quantify retention of cells. BMC were imaged at 40x and counted using Image J to
determine before and after cell counts. GFP+BMC were also imaged at 4x to verify
uniformity of adhesion across substrate.
4.3 Results and Discussion
While our lab has previously demonstrated the ability to increase the retention of
Mesenchymal Stem Cells (MSCs) in the infarcted myocardium through these gelatin-
based coatings [6], the mechanism of coated cell retention in the tissue has only been
hypothesized. Our goal in this work was to identify the adhesion mechanism behind
increase retention of gelatin coated therapeutic cells in vivo. Here, we first determine if
increased retention for coated cells inside the infarcted myocardium is generalizable to
unfractionated bone marrow derived cells. We then separate the BMC injection site into
its two major components, ECM and cells, and determine the ability of BMC to adhere to
each.
4.3.1 GelMA coating confirmed through collagen antibody and flow cytometry.
The gelatin-based cellular coatings used in this study are deposited through
sequential cell surface modifications. BMCs are modified through biotinylating free cell
surface amines followed by binding of streptavidin that is pre-conjugated to an eosin
photoinitiator (Figure 4.2A). Cells are then suspended in the GelMA monomer and the
coatings are cross-linked near the eosin under 530 nm green light. We further confirm our
ability to coat BMC by staining with a primary anti-collagen antibody followed by
47
secondary AF-647. In all we achieved an effective coating in approximately 29% of
BMCs (Figure 4.2B).
Figure 4.2:Flow cytometry confirms effective GelMA coating of BMC. A) Schematic of GelMA coating process through modification of cell surface amines. B) Flow cytometry analysis of coated versus uncoated BMCs stained with primary collagen antibody followed by secondary AF-647 (EITC, eosin isothiocyanate).
48
4.3.2 GelMA coated BMCs demonstrate enhanced retention in the heart over uncoated BMCs.
We next investigated cell retention in the heart after acute myocardial infarction. Immediately after LAD ligation, GFP expressing unfractionated bone marrow murine cells were injected into the surrounding peri-infarct areas. The use of
GFP donor cells allowed us to track the retention and fate of transplanted cells. Seven days later, we examined the number of BMCs retained in the heart using immunohistochemistry and flow cytometry. Hearts were isolated, digested to
single cell suspension then stained against CD45 and GFP. Overall, we observed a significant increase in the number of coated cells inside the heart
compared to uncoated cells (uncoated = 96.5±45.3, coated = 302±58.9 cells/mg heart tissue, P = 0.0017,
49
Figure 4.3: Increased retention was observed for coated versus uncoated cells. A) Flow
cytometry analysis of digested heart tissue demonstrates higher percentage of GFP+ cells in mice treated with coated cells comparted to mice transplanted with uncoated cells. Panel A shows the expression of CD45 and GFP in digested heart tissue where the majority of GFP cells were CD45+. B) Quantitative analysis of GPF+BMC in digested heart tissue for coated versus uncoated cells by flow cytometry (signal in the PBS arm represent background). C) Immunohistochemical analysis demonstrated higher numbers of retained GFP+ BMC in mice treated with coated cells. D) Quantitative analysis of GPF+BMC in digested heart tissue for coated versus uncoated cells based on immunohistochemical analysis (signal in the PBS arm represent background). (N = 3 mice/group; *P<0.05, **P<0.01).
4.3.3 Gelatin coated BMCs showed no adherence to representative cardiac cells.
The cells in the myocardium adhere to the ECM through integrins expressed on
the surface of cells [85, 86], and β1 integrins are the major physiological receptors for
collagen in cardiac tissue [119, 120]. As GelMA is derived from collagen [121], we
explored the potential for gelatin coated BMC to adhere to β1-positive and β1-negative
cells within the infarcted myocardium.
50
To qualitatively evaluate the presence of b1 integrins within the infarcted
myocardium, heart tissue sections were stained with a primary antibody against b1
followed by Alexa Fluor 568 secondary. Heart tissue stained in the absence of primary
antibody was used as a negative control. Immunofluorescence analysis showed high b1
staining for both macrophages and cardiomyocytes expressed within the heart tissue
(Figure 4.1B). Additionally, it was found that b1 expression was highest in areas of
infarct. Based on these findings, it is reasonable to consider b1 integrins at a potential
binding mode for GelMA inside the infarcted myocardium.
To determine if coated or uncoated cells adhere to cardiac resident cells, we
incubated BMCs with cells of differing b1 expression (Figure 4.4A). In this study,
HUVEC were chosen to represent a cell line with high b1 expression and H9C2 were
chosen as a cell line with low b1 expression (Figure 4.4B). HUVEC and H9C2 cells,
stained with cell tracker red, were cultured onto tissue treated slides until they reached a
confluency of approximately 90%. Coated or uncoated GFP+BMCs were then added in
suspension and allowed to incubate with cultured, adhered cells for 40 minutes.
Microscopic images were obtained and then cells were rinsed through centrifugation at
800xg. Number of cells retained after centrifugation was compared. Figure 4.4C shows
that, following rinsing, BMCs showed no adhesion to HUVEC or H9C2 cells.
51
Figure 4.4:Expression of b1 integrins does not promote GelMA adhesion. A) experimental protocol for experiments examining the role of integrins in cell adhesion of GelMA coated cells. B) HUVEC and H9C2 cells analyzed by flow cytometry for b1 integrins by primary anti-b1 followed by secondary anti-AF647. C) Microscopic imagining of BMC incubated with cultured cells (HUVEC or H9C2) then subjected to rinsing by centrifugation at 800xg for 5 minutes. (n=5 slides per test group).
While β1 integrins are known for collagen adhesion, the lack of GelMA coated
cell retention on the β1-positive HUVECs and β1-negative H9C2 proved this integrin to
not be an effective binding mechanism for GelMA-cell adhesion. While previous studies
52
have shown the adhesion of cultured cells to GelMA scaffolds [122, 123], these models
were not subject to fluid shear. Arterial wall shear stress is one of the highest
physiological shear environments at 10-20 dyne/cm2 [124]. Our lab previously
demonstrated MSC adherence to HUVEC through ICAM antibody labeling could only
withstand this physiological shear when HUVEC were stimulated TNF-a [5]. This
further confirms our hypothesis that any integrin-GelMA adhesion is too weak to
withstand fluid shear.
4.3.4 GelMA coated BMCs show higher adhesion to Decell Heart Tissue than uncoated BMC.
In order to investigate cardiac ECM in the absence of cells, we used surfactants to
strip the cells from mouse heart tissue. Decell tissue allows us to focus specifically on the
ECM – our other potential binding mode for GelMA. Briefly, coated and uncoated
GFP+BMCs were allowed to incubate for 40 minutes with Decell heart tissue. After
rinsing by centrifugation, a statistically higher (p**<0.01) number of coated BMCs were
retained on the decell tissue compared with uncoated BMCs (Figure 4.5A). For the
uncoated group, nearly all adhesion could be attributed to ridges within the decell tissue.
On the other hand, GelMA coated BMCs showed adhesion across the whole decell tissue
(Figure 4.5B). Results suggest that the ECM is critical for the increased retention of
GelMA coated cells.
53
Figure 4.5: A greater number of GelMA coated BMCs are retained on decell heart tissue than uncoated BMC. A) Quantitative analysis of the number of BMC per area determined by Image J before and after washing for coated versus uncoated BMC. B) Microscopic images of coated versus uncoated GFP+BMC on decellularized tissue before and after rinsing. (n=5 decell tissue per test group; p**<0.01)
4.3.5 GelMA coating showed highest adhesion to collagen substrate.
As we demonstrated, our gelatin-based coating increases the retention of
therapeutic cells to heart ECM. In order to better understand this binding mode, we broke
the ECM down into three main components: collagen, laminin and fibronectin. Collagen
type I was used for these studies as it makes up over 80% of the total collagen in the
cardiac ECM [125]. Unmodified glass was used as a control in this study.
54
GelMA coated BMCs were incubated with glass slides coated with either
laminin, collagen, fibronectin or unmodified glass. Then, cells were subjected to a
delamination force via centrifugation at 800xg. All ECM components tested resulted in
significantly higher number of retained cells than that of glass alone (Figure 4.6). While
fibronectin increased the number of GelMA coated cells retained on the surface, the
number of retained cells was significantly less than that of both collagen and laminin.
Overall, GelMA-coated BMC incubated on collagen coated glass substrate showed the
highest level of retention following centrifugal washing. These results indicate that, while
all three ECM components promote binding to GelMA coated BMCs, collagen is the
most effective substrate for retention of GelMA coated cells.
Collagen subunits are well known to self-assemble into triple helices of
procollagen, where they are later enzymatically crosslinked through lysine residues to
form functional collagen fibrils [82]. As GelMA is formed from the hydrolytic
degradation of collagen, it is reasonable to assume the preservation of many structural
elements that would promote the formation of procollagen, and these noncovalent
interactions promote enhanced retention of gelatin coated cells on collagen surfaces.
55
Figure 4.6:Collagen surfaces significantly contribute to the retention of GelMA coated cells when compare to other ECM components. A) Mean number of BMC per area determined by Image J before and after washing for coated versus uncoated BMC. B) Microscopic images of GelMA coated GFP+BMC on various ECM substrates before and after rinsing. (n=5 slides per test group; p*<0.05, p**<0.01)
4.3.6 Uncoated BMCs do not bind to ECM components.
We also repeated this study with uncoated BMCs to verify there was no
interaction between the BMC and ECM. Our results show that collagen, laminin or
fibronectin did not promote binding to BMC without a coating (Figure 4.7). In comparing
Figure 4.6 and Figure 4.7, results clearly demonstrate that GelMA coating significantly
56
increases the retention of BMC to ECM components. This further substantiates our
conclusion that BMCs alone do not promote adhesion within the myocardium, as
evidenced by the litany of poor retention of therapeutic stem cells in clinical studies [2,
50].
Figure 4.7: Coating is essential for BMC adhesion. A) Average amount of BMCs per area determined by ImageJ before and after washing for coated versus uncoated BMC. B) Microscopic images of uncoated GFP+BMC on various ECM substrates before and after centrifugal rinsing. (n=5 slides per test group; p*<0.05, p**<0.01)
4.4 Conclusion
While transplantation of therapeutic stem cells has been greatly limited by poor
uptake, gelatin-based cellular coatings allow for a significant increase in stem cell
57
retention [6]. A recent meta-analysis of human clinical trials concluded that the number
of therapeutic stem cells is positively correlated with therapeutic benefit post-MI [114].
GelMA is an appealing hydrogel for cell encapsulation due to its known biocompatibility,
biodegradation and tunable physical characteristics [121]. Through an in vivo mouse
model, we demonstrated a 3-fold increase in BMC retention for coated versus uncoated
cells (Figure 4.2). While cell imaging and flow cytometry strategies allow us to identify
retained cells [6], they do not indicate what interactions are promoting this increased
retention. We initially divided the potential adhesion sites into cell-based and ECM-based
groups. In this work, we determine the increased retention of GelMA coated cells inside
the myocardium was attributed to GelMA-ECM binding. Based on these results, we seek
to further modify our GelMA coatings to enhance these interactions with the ECM and
promote further increases in the retention of therapeutic cells. Future work for this study
includes the addition of collagen specific antibodies to our GelMA coating.
4.5
58
Chapter 5 INCREASED YIELD OF GELATIN COATED THERAPEUTIC CELLS THROUGH CHOLESTEROL INSERTION
5.1 Introduction
Cellular therapies utilized in cancer treatment and cardiovascular repair struggle
to achieve optimal cell retention for therapeutic effect [26, 32, 126, 127]. For the
placement of therapeutic stem cells in the heart following a heart attack, fewer than 1% of
administrated stem cells are retained in the heart following the direct injection of the cells
into the heart wall [2]. Without the persistence of cells in the area of therapeutic need, the
cell’s effect is diluted across the body, and therapeutic efficacy is reduced [3, 4].
Cell surface engineering is emerging as a tool to intentionally position cells into
tissues for optimal therapeutic effect [26, 128, 129]. One recent strategy uses gelatin-
based cellular coatings to increase retention of therapeutic cells inside the heart. These
gelatin coatings support a significant, three-fold increase in viable bone marrow derived
mesenchymal stem cells (BM-MSCs) retention in the heart wall while also supporting
BM-MSC immunomodulatory function [6], Importantly, these dramatic improvements in
cell retention are achieved with a yield of coated cells of ~50%. As a result, the gains in
cell retention are attributed to the fraction of cells that are actually modified, while the
remaining MSCs are expected to behave like the unmodified control MSCs. Additional
steps to enrich the coated MSCs are discouraged to promote the health and sterility of the
injected MSC population.
Based on this low coating yield, we sought a method capable of achieving
higher yield of gelatin coated cells. The gelatin coating is grown from the MSC surface
by decorating the cell surface with biotin, attaching an eosin photoinitiator using a
streptavidin-eosin conjugate, and polymerizing in the presence of a gelatin methacrylate
59
macromer (Figure 5.1). Prior studies indicate that a higher surface density of
photoinitiator is correlated with a higher yield of coated cells, however, efforts to
increase the streptavidin-photoinitiator binding ratio failed to improve cell
polymerization [130]. As a result, the present study focuses on the role of directly loading
biotin onto the cell surface, thus allowing for an increase in streptavidin-eosin. Most
commonly, sulfo-NHS-LC-biotin (NHS-Bio) is used to covalently bind biotin to free
amines on cell surface proteins [6, 23, 26, 32]. While this method is limited to the amount
of free amine groups on the cell surface, viability also becomes a concern with increasing
concentration [9]. Alternatively, lipid insertion allows for easy modification of the cell
surface through direct insertion into the phospholipid bilayer. Specifically, cholesterol
was used for its ability to intercalate into lipid rafts [57, 131]. Sun, et al. showed
cholesterol-PEG can be used to modify 99% of cells in a population [132]. As a result,
we hypothesized that lipid insertion would allow for modification of a higher fraction of
cells than covalent methods.
Here, we explored the use of functionalized lipids that insert directly into the lipid
bilayer to support higher yields of gelatin coatings on MSCs. We first determined the
impact of MSC incubation in cholesterol-PEG-biotin (CholBio) conjugates on MSC
function and SA surface density. Then, we evaluated the yield and morphology of the
coatings on MSCs. In all, the distinct manner in which the biotinylated lipids load into
the MSC’s peripheral membrane drives higher coating yields with greater uniformity and
viability than that of the covalent NHS-Bio approaches.
60
Figure 5.1: Cell surface engineering methods to introduce biotin onto cell surface. A) Cell surface amines covalently bound to sulfo-NHS-Biotin (NHS-Bio). B) Cholesterol-PEG-Biotin (CholBio) inserted directly into phospholipid bilayer.
5.2 Materials and Methods
5.2.1 Cell Culture.
Bone marrow cells were isolated and cultured at 37 °C in hypoxic conditions (5%
O2). Briefly, both non-GFP and GFP+ 6–8 weeks old C57BL/6-Tg (CAG-EGFP) mice
(Jackson Laboratory, BarHarbor, ME) were sacrificed, femurs, tibias and hip bones were
collected and then flushed with PBS supplemented with 10% FBS. Cells were cultured in
complete MesenCult (StemCell Technology) supplemented with 1% penicillin-
streptomycin (VWR), and 1% L-Glutamine (StemCell Technology). Cells were seeded
into 182 cm2 culture flasks (VWR) and cultured until approximately 90% confluency.
Cells were then passaged at using 8 mL TrypLE (Thermo Fisher) for 2-3 min. The cell
61
suspension was centrifuged at 400 x g at 4 ˚C for 10 min, supernatant aspirated and
resuspended in MesenCult. Cells were then counted and frozen in liquid nitrogen until
needed or re-seeded at 1.0 × 105 cells per flask. For all studies, MSCs were used at
passages 6-8 and aspirated at 90% confluency. Prior to biotin conjugation, cells were
rinsed with 1 mL PBS twice and cell diameter and density were measured using a
Cellometer (Bioscience).
5.2.2 Biotin Conjugation.
MSC were split into aliquots of 1 million cells for further processing. Biotin was
nonspecifically covalently bound to proteins on the surface of the cell using a biotin-
succinimidyl ester conjugate (NHS-biotin; EZ-link Sulfo-NHS-LC-Biotin, Thermo
Fisher). NHS-biotin was used at a concentration of 0.55 mg/mL in PBS for 40 min on ice
with a volume of 500 µL per million cells. Biotin was also bound to MSC through lipid
modification by CholBio(x) (NanoSoft Polymer, Item 3084), where x stands for the
number average molecular weight of the PEG chain (x=1000, 2000, 3400 and 5000). This
lipid insertion approach utilizes a biological cholesterol PEG, where the PEG group is
further functionalized with biotin. CholBio(x) was used at concentrations of 1 and 10 µM
in PBS for 30 min at 37˚C with a volume of 1 mL per million cells. Following biotin
conjugation, MSC were rinsed with PBS twice.
5.2.3 Streptavidin-Phycoerythrin.
Using biotin-streptavidin affinity, streptavidin-phycoerythrin (SA-PE; Invitrogen)
was conjugated to CholBio, NHS-Biotin and unmodified MSC. MSCs that remained
62
suspended in PBS and on ice for the duration of the study (unmodified MSCs), were used
as control in all studies. Both 25 µL of 0.2 mg/mL SA-PE and 25 µL of 1 mg/mL
streptavidin (SA) were combined and diluted to 1 mL in PBS. Competitive binding with
the 1 mg/mL SA was necessary as SA-PE alone resulted in fluorescence above the
detection level of BD Accuri C6 flow cytometry. Samples of 10,000 events were
analyzed by flow cytometry, and the FL2 mean fluorescence of SA-PE tagged samples
was recorded. BD QuantiBRITE PE quantitation beads were then used to quantify PE
fluorescence by vortexing in 500 μL of PBS and analyzing the FL2 signal. The linear
relationship between FL2 signal and PE florescence was determined using QuanitBRITE
PE (Appendix 3) counting beads. Per Invitrogen SA-PE is conjugated at 1:1 ratio of SA
to PE. Using the ratio and MSC diameter recorded after cell culture, SA density on the
MSC surface was determined.
5.2.4 Cell Viability.
Following biotin conjugation, cell viability was tested using three different
methods: Calcein AM, Ethidium Homodimer-1 (Eth-1), and CellEvent Caspase- 3/7
(Invitrogen). CholBio(x), NHS-biotin and unmodified MSC were separated into samples
of approximately 25,000 cells and incubated at room temperature with each of the
viability assays and then measured using flow cytometry: Calcein at 1 μL in 10 mL PBS
for 30 min, Eth-1 at 2 μL in 1 mL PBS for 10 min, and Caspase at 1 μL in 1 mL PBS for
25 min. Unmodified cells were used as a control population for all viability assays.
63
5.2.5 Cell Encapsulation and Antibody Staining.
GFP+ MSC were washed twice with PBS by centrifugation at 500 ×g, 3 min at 4
°C and 1.5 × 106 cells were collected for each polymerization trial. PBS buffer was
removed, and the cell pellet was resuspended in NHS-Biotin or CholBio1k as previously
specified in Biotin Conjugation. Following incubation, cells were rinsed three times with
PBS by centrifuging at 500×g for 3 min at 4 °C. The cell pellet was then resuspended in 1
mL of PBS containing 35 μg/mL streptavidin-eosin isothiocyanate (SAEITC) and
incubated for 30 min on ice. SAEITC conjugation was performed in-house, through
formation of a thiourea bond [118]. Briefly, SA (SigmaAldrich), dissolved in a carbonate
buffer, and eosin isothiocyanate (SigmaAldrich), dissolved in DMSO, are reacted at a
10:1 ratio for 8 hours at 4°C. The solution is then concentrated to 1 mg/mL in PBS and
purified through Zeba spin desalting column.
The polymer mixture buffer was prepared using 1% PEGDA 3400, 35 mM
triethanol amine, and 35 mM 1-vinyl-2-pyrrolidinone in PBS. PEGDA 3400 is added as a
low MW crosslinking agent to promote cell coating uniformity [6]. This mixture was
purged with ultra-pure N2 for 10 min. Next, 3 wt% gelatin methacrylate (Allevi) was
added to the buffer and vortexed and sonicated for 10 min. The final macromer solution
(GelMA) was filtered through a 0.2 μm syringe filter and stored at 37°C. Following
filtration, 350 μL of the polymer mixture was added to the cell pellet, gently vortexed and
loaded into a chip-clip well (Whatman) with a standard microscopy slide. The chip-clip
was incubated in a dark, N2 purged, sealed, zip lock bag for 3 min before polymerization.
The cells were polymerized for 10 min at 30 mW/cm2 under 530 nm green light while
purging with N2 followed by rinsing twice with 1 mL of PBS. To determine the presence
64
of a GelMA coating, GFP+MSC were stained with primary anti-collagen 1 antibody
(Abcam) followed by secondary Alexa 647 (AF-647) antibody (Invitrogen). Samples of
10,000 events were analyzed by flow cytometry, and the FL4 mean fluorescence for
secondary antibody tagged cells was recorded. These cell suspensions were also imaged
using a Nikon Ti-U inverted microscope.
5.2.6 MSC Migration Assay.
GFP+MSC were modified with CholBio1k-10 followed by SAEITC and gelatin
coating as specified above. Uncoated GFP+MSC were used as a control. The migration
potential of modified and unmodified MSCs was evaluated in an in vitro scratch assay.
Cells were seeded on 24 well-plates (VWR) at a concentration of 5x104 cells per well in
MesenCult. Cells were allowed to attach and form uniformed monolayer. A 200 μL
pipette tip was used to introduce a scratch at the center of each well. The plates were
washed with PBS solution after scratch, followed by serial measurement of the distance
between the leading cell edge at the scratch site. Image analysis software (ImageJ,
National Institutes of Health, Bethesda, MD, USA) was used to quantify the area of the
wound closure.
5.2.7 MSC Proliferation Assay.
Colony-forming unit (CFU) assay was used to determine the proliferation
efficiency of CholBio1k-10 coated and uncoated MSCs. Modified and un-modified
MSCs were seeded separately in cell culture dish (Fisher Scientific Cat#07-000-331) at
the concentration of 30 cells/cm2 and cultured in MesenCult. Eight days after initial
65
seeding, cells were fixed and stained with 1% crystal-violet in 100% methanol. Cell
cluster consist of 10 or more cells were counted as a colony.
5.2.8 Cytokine Secretion Assay.
Conditioned media (CM) from uncoated MSC and CholBio1k-10 coated MSCs
was collected 24 hours after culturing in Opti-MEM I Reduced Serum Media
(ThermoFisher, cat#11058021). The immunomodulatory effect of CM on primary
macrophage were texted in LPS challenging assay. Briefly, primary macrophages were
isolated from the bone marrow of 12 weeks old mice. One week after initial isolation,
macrophages were seeded in 24 well plates at a density of 0.3 million cells/well.
Macrophage were challenged with LPS (10 ng/mL) in the presence of CM collected from
coated or uncoated MSCs. The culture media from macrophage was collected 6 hours
after LPS stimulation. Cytokine levels were analyzed by ELISA (BD Biosciences,
cat#BD555252 and cat#555268) according to manufacturer’s protocol.
5.3 Results
The overall goal of this study was to determine if lipid insertion of biotin groups
increased the yield of gelatin coated cells over that of a covalent approach. Biotin is
commonly used in cell surface engineering due to its high affinity for SA allowing for
continued functionalization [26]. For this reason, cell surface SA density was used as a
direct measure of a cell’s ability to be functionalized with streptavidin and other
compounds. In this study, MSCs were directly modified with either NHS-Bio (covalent
modification) or CholBio (lipid insertion method).
66
5.3.1 Impact of lipid concentration on streptavidin surface density and cell health.
Cell biotinylating with CholBio occurs through the insertion of individual
CholBio molecules into the peripheral membrane of the cell. To quantify this transfer of
CholBio to the MSC peripheral membrane, MSCs were incubated with 0, 1, 5, 10, 50,
and 100 µM CholBio1k (CholBio with a 1 kDa PEG chain) where the 0 µM was used as
a negative control. The available biotin on the surface was tagged with a fluorescent SA-
PE and the fluorescence was quantified with flow cytometry. The available SA density on
the cells was calculated with a calibration curve generated using QUANITBRITE PE
calibration beads (Appendix 3). Then surface density was estimated using a mean MSC
diameter of 19 µm and compiled for the different concentrations of CholBio1k (Figure
5.2A). Unmodified control cells were incubated in buffer without CholBio1k followed by
SA-PE. In general, the cell surface SA density increases with an increasing solution
concentration of CholBio1k between 1 and 50 µM. Above 50 µM CholBio1k, there was
no statistically significant difference in cell surface SA density. When compared to NHS-
Bio, a significantly higher SA density was achieved at CholBio concentration at or above
10 µM.
The viability of MSC following modification with CholBio1k was evaluated
using a calcein assay of esterase activity, an ethidium assay of membrane integrity, and a
caspase assay to evaluate apoptosis (Figure 5.2B). When compared to unmodified control
cells (0 µM), immersion of <10 µM CholBio1k on the cells did not have a significant
impact on cell necrosis, as determined by calcein and ethidium assay, but there was a
modest increase in activation of the caspase 3 apoptotic pathways. At and above 50 µM
67
CholBio1k, there was widespread damage to the cells. The viability of MSC modified
with CholBio at all concentrations was also compared to the viability of MSc modified
with NHS-Bio. Activation of apoptosis pathway was significantly higher starting at 5
µM. However, membrane integrity was not significantly impacted until 50 µM and
esterase activity only at 100 µM when compared to NHS-Bio.
Figure 5.2: High concentrations of CholBio allow for a significant increase in cell surface SA over NHS-Bio. A) Dependency of biotin surface density on solution concentration of CholBio1k. B) Impact of increased concentration of CholBio1k on MSC viability.
5.3.2 Impact of PEG molecular weight on streptavidin surface density and cell function.
The CholBio molecule is composed of a cholesterol head group, a PEG linker,
and a biotin tail group. We evaluated the role of the PEG linker molecular weight on SA
density and cell function at the maximum practical concentration for CholBio1k of 10
μM (Figure 5.3). While maintaining an overall CholBio concentration of 10 μM, the
molecular weight of the PEG linker varied (1, 2, 3, and 5 kDa). No significant difference
in cell surface SA density was observed between all molecular weights (Figure 5.3A).
68
When compared to an unmodified MSC incubated with SA-PE alone, a 103-fold increase
in cell surface SA density was observed for CholBio at 10 μM concentration (Figure
5.3A). When contrasted with the more commonly used method to covalently conjugate
biotin to the surface of the cell, sulfo-NHS-LC-biotin (NHS-Bio) [6, 7, 9, 26, 32], a
solution of 10 μM CholBio achieved a significantly higher SA surface density on MSCs
across all molecular weights studied (1 kDa and 2 kDa P<0.05, 3 kDa and 5 kDa P<0.01).
Impact of PEG chain linker molecular weight on cell viability was also evaluated
based on calcein, ethidium and caspase assays. No significant changes in viability were
observed at lower PEG molecular weights (1 and 2 kDa, Figure 5.3). However, as the
molecular weight of the PEG linker was increased to 3 and 5 kDa, an increase in necrosis
or late apoptosis as measured by a decrease in calcein or ethidium was observed. Results
showed molecular weight of the PEG chain had no significant impact on activation of the
apoptosis pathway. When the CholBio groups were compared to covalent method, NHS-
Bio, no significant difference was observed for calcein viability. However, ethidium did
show damage to the peripheral membrane using the 5 kDa linker. In contrast, the 5 kDa
linker was the only molecular weight to not result in significantly higher activation of
caspase 3 pathways than NHS-Bio (Figure 5.3B).
69
Figure 5.3: Molecular weight of CholBio PEG chain did not significantly impact SA loading. A) CholBio at 10µM resulted in a significant increase in cell surface biotin density when compared to NHS-Bio (P*<0.05, P**<0.01, n=5). Data obtained through PE QUANTIBRITE calibration beads. B) Viability of CholBio(x) at 10 µM versus NHS-Bio by Calcein, Ethidium and Caspase assays. Higher molecular weight of CholBio resulted in significantly lower viability than NHS-Bio (P*<0.05, n=5). C) CholBio at 1 µM resulted in significantly less biotin on the cell surface when compared to NHS-Bio (P*<0.05, P**<0.01, n=5). Data obtained through PE QUANTIBRITE calibration beads. D) Viability of CholBio(x) at 1 µM versus NHS-Bio by Calcein, Ethidium and Caspase assays.
When using a 1 μM concentration of CholBio, the SA density was still insensitive
to PEG molecular weight, but this SA loading was significantly lower than that of NHS-
Bio (Figure 5.3C). Viability trends at 1 μM (Figure 5.3D) largely parallel the trends at 10
μM. No significant effect was seen through calcein assays, and damage to the peripheral
70
membrane becomes apparent with a 5kDa linker. The fraction of cells with increased
caspase activity was significantly higher for 3kDa and 5kDa linkers when compared to
NHS-Bio and control MSC.
5.3.3 Yield of gelatin coated MSCs from CholBio anchoring.
The ultimate goal of this work was to increase the yield of gelatin coated MSCs
for cellular therapy following a heart attack. In this coating approach, biotinylated cells
are labeled with streptavidin-eosin [7, 9, 23, 133], and the eosin coated cells are
photopolymerized in a GelMA macromer solution [6]. To directly compare the yield of
coated MSCs for each biotinylation strategy, biotin coated GFP+ MSCs were prepared
using 1 mM NHS-Bio [6, 23, 133], 1 μM CholBio1k, or 10 μM CholBio1k. Following
biotinylation, GelMA coatings were formed through photopolymerization. Unmodified
GFP+ MSCs, which received no cell surface modification, were used as control. The
number of gelatin coated cells were determined through staining the cells for collagen-I
and flow cytometric analysis (Figure 5.4). Collagen-I antibodies specifically stain for the
gelatin coatings [6], owing to the presence of residual collagen epitopes in the gelatin.
71
Figure 5.4: Cell surface modification through lipid insertion increased the amount of MSC effectively coated with GelMA. A) Flow Cytometry images of GelMA processed GFP+MSC stained with antibody against collagen followed by secondary AF-647 antibody. B) CholBio1k-y GelMA processed GFP+MSC resulted in significantly higher gelatin coated cells than NHS-Bio processing (n=5, P*<0.01).
Through quantitative analysis of GFP+ MSCs with increased gelatin labeling
(Figure 5.4B), the percent of GelMA coated cells is significantly increased for 1 μM
CholBio1k (P<.05) and 10 μM CholBio1k (P<.01) over the conventional NHS-Bio
approach. While the NHS-Bio strategy positively stains only 40-50% of MSCs for
gelatin, the 10 μM CholBio1k approach yields over 80% coated MSCs. These results are
qualitatively confirmed through epifluorescent microscopic imaging of coated cells
(Figure 5.5). In addition, microscopic images also show NHS-Bio resulted in patchy,
uneven coatings while CholBio1k groups have a more uniform circumferential coating.
72
Figure 5.5: Cell surface modification through lipid insertion resulted in more uniform coatings on the MSC surface. A) Representative microscopic images of CholBio(1k) anchored GelMA coating resulting in more gelatin positive MSC, by secondary AF-647 antibody, than the NHS-Bio anchored GFP+MSC. Zoomed in image of individual cell shows morphology of cell coating. (scale bar = 50 µm)
5.3.4 Impact of gelatin coating on MSC function.
While our lab has already shown that gelatin coatings may be used to increase
retention of MSCs [6], we also continue to show that these coatings do not hinder the
natural function of MSCs. Here, we compare the function of coated and uncoated MSCs.
As stated above, biotinylation of MSC with CHolBio1k-10 resulted in the highest level of
73
gelatin positive MSCs (83%). Therefore, we used CholBio1k-10 as the coating method to
determine the highest impact gelatin coatings may have on MSC function.
Commonly, the ability of MSCs to proliferate and migrate are used as a surrogate
of their overall functionality. The migratory function of coated and uncoated MSCs was
examined using in vitro scratch assay. As seen in Figure 5.6A, there was no statistical
difference between coated and uncoated MSC in the time necessary to migrate and close
the gap. In addition, MSCs ability to proliferate was defined as their ability to form
colonies (CFUs). Figure 5.6B shows that coating of MSC did not statistically change
their ability to proliferate and form colonies.
Figure 5.6: Increased coating efficiency did not impact MSC ability to migrate and proliferate. A) Microscopic and quantitative representation of CholBio1k-10 coated MSC versus uncoated MSC migration across an introduced gap. B) Microscopic and quantitative comparison of colonies formed by crystal-violet stained coated MSCs versus uncoated MSC after 8 days in culture.
74
MSCs modulate the immune system through secretion of cytokines. Here, we
demonstrate that despite the high level of uniform gelatin coating observed using the
CholBio1k-10 strategy, MSCs are still able to release the cytokines necessary for
immunomodulation. Figure 5.7A shows that there was no statistical difference (n=3,
P=0.962) in the pro-inflammatory, TNFa, production for macrophages co-cultured with
CM from CholBio1k-10 coated or uncoated MSCs. Additionally, gelatin coating had no
statistical impact (n=3, P=0.234) on anti-inflammatory signal, IL-10, production when
cultured with coated or uncoated MSC CM (Figure 5.7B).
Figure 5.7: 6 hour LPS stimulation. A) Macrophage TNFa signal following incubation with cultured media (CM) from CholBioak-10 coated MSC and uncoated MSC. B) Macrophage IL-10 signal following incubation with cultured media (CM) from CholBioak-10 coated MSC and uncoated MSC.
5.4 Discussion
Biotin has been used to functionalize the mammalian cell surface for cellular
adhesion [6, 32, 104], cell mediated drug delivery [9, 58] and isolation [7, 22, 23]. In
these cases, biotin serves as a handle for further functionalization with streptavidin. In
75
this work, we show that lipid insertion of cholesterol can be used to achieve higher cell
surface modification than traditional covalent modification of cell surface amines. The
amount of SA loaded onto the cell surface was directly related to the concentration in
solution (Figure 5.2A). While lipid bilayers could expand in surface area through the
integration of additional lipids into the peripheral membrane, the theoretical limit of
accessible biotin groups is approximated by the planar density of binding sites in a 2D
crystal of streptavidin (3.1x104 molecules/μm2) [134]. This theoretical maximum agrees
reasonably well with our saturation limit when considering the characteristic membrane
folds of mesenchymal cells [135]. The increased surface area introduced by the ruffled
morphology is not accounted for in our assumption of a cell being a perfect sphere with a
19 μm diameter, and the quantity and accessibility of this added surface area is
challenging to estimate. In all, the observed saturation phenomenon near 105 SA
molecules/μm2 is plausible relative to the limiting theoretical surface density, and it is
unlikely that the surface density of SA can be significantly increased beyond this level
without significantly more topographical complexity.
In this work, we present an increase in SA density using the CholBio approach
over NHS-Bio. In our final application, we use SA to bind a photoinitiator to the surface
of the cell. Biotin-SA affinity is commonly used for many cell surface engineering
techniques such as the addition of adhesion factors [32], cellular drug delivery [58],
cellular coatings [6, 7, 9], among others [26]. This increased SA density is achieved by
loading more biotin binding molecules onto the surface of the cell through direct lipid
insertion. However, as some researchers may be interested in biotin modification without
the addition of a SA step, it is important to consider the amount of biotin loaded onto the
76
cell surface. A single SA molecule can bind up to four biotin molecules, thus making a
direct correlation between SA density and biotin density challenging. We estimate that
the biotin density is up to four times higher the reported SA density. However, a true SA
to biotin ratio was not determined for this application.
Previous data has shown viability to be a limiting factor in increasing NHS-Bio
solution concentration,[9] and data from this study confirms cell viability as a limiting
factor for CholBio as well (Figure 5.2B, Figure 5.3B, Figure 5.4B). While the
concentration of CholBio used is limited by cell viability, a concentration of 10 µM
allowed viability to be maintained while still achieving higher biotin levels than covalent
biotinylation (Figure 5.3A).
Based on the mass of the molecule’s constituent groups (cholesterol: 387
Da [136], biotin: 244 Da [137], and PEG: MN 1,000-5,000 Da [NanoSoft Polymer]),
many properties of the CholBio structure are likely dominated by the contribution of the
PEG group. In particular, the PEG linker is expected to govern the overall hydrophobicity
of the molecule, the toxicity, and the accessibility of the biotin group. The viability trend
seen in Figure 5.2B appears to contradict previous data from tissue engineering scaffolds
showing toxicity to cells is inversely related to the molecular weight of the PEG chain
[138]. While our viability holds the molar concentration of the PEG chains constant, the
tissue engineering data holds the mass fraction of the macromer constant. This departure
suggests that the greater toxicity of shorter PEG chains is overcome in our study by the
larger amounts of PEG mass in the higher molecular weight groups.
For all viability data, the type of damage is distinct between the CholBio and the
NHS-Bio treated cells. CholBio data (Figure 5.3B and Figure 5.4B) consistently shows
77
increased permeabilization of the peripheral membrane, while the NHS-Bio groups show
a decrease in esterase activity. The CholBio increase in permeability is more likely than
for the NHS-Bio due to massive insertion of CholBio molecules into the peripheral
membrane driving an increase in membrane surface area.
As expected, the increase in SA density on the surface of MSCs treated with 10
μM CholBio1k over that of NHS-Bio treated cells has a corresponding increase in the
fraction of coated cells following interfacial polymerization. Previous studies using PEG
diacrylate polymerization on NHS-Bio modified cells concluded that the number of
polymer coated cells was positively correlated with to the eosin photoinitiator density
[23]. Since the photoinitiator is coupled to the cell surface using streptavidin, the
available SA density is also positively correlated with the photoinitiator density [23]. The
presence of a polymer coating is related to the gelation of the polymer coating through
surface mediated polymerization, and this biotin and streptavidin dose response has been
successfully leveraged in multiple biodetection settings [27, 139-141]. Similarly, this
response on the surface of a cell has been leveraged to protect specific cells in suspension
[7, 23, 142].
The higher yield of coated cells for the 1 μM CholBio1k group than the NHS-Bio
group (Figure 5.4B) was not anticipated on the basis of the lower SA density for the 1
μM CholBio1k group (Figure 5.3C). The increased coating yield with a lower SA density
suggests the CholBio system offers an advantage during polymerization over the covalent
NHS-Bio system. The greater uniformity of coatings formed from CholBio tethers
(Figure 5.5) also suggests a fundamental difference between the two anchors for
78
photoinitiators that likely results from uniformity of the biotin on the peripheral
membrane.
MSC are commonly used for their immunosuppressive and
immunomodulating properties [45, 46]. We previously demonstrated in an in vivo MI
model that enhanced MSC cardiac retention with gelatin coating significantly reduced
cardiac inflammation leading to better cardiac functional recovery and smaller scar size
[6]. This increased therapeutic benefit was attributed to the adhesive properties of the
gelatin coating allowing for a higher retention of MSC inside the cardiac muscle.
However, in these studies, MSCs were coated using the NHS-Bio approach which results
in less uniform coatings and fewer cells coated. The work presented herein suggest that
the efficacy of coating can be significantly increased using the CholBio approach.
Critically, the enhanced cell coating using this novel approach did not affect the secretory
or migratory functions of MSCs which are critical for their beneficial effects in
myocardial preservation.
5.5 Conclusion.
These GelMA coatings have been shown to increase cell retention in the heart
post-MI [6], and CholBio resulted in a more conformal coatings than NHS-Bio with an
increase in coated cell yield from 52% to 83%. It is reasonable to expect that injection of
MSC coated with lipid anchored GelMA will result in even higher cell retention in the
post-MI environment (Figure 5.1). However, further in vivo studies are needed to support
this additional hypothesis.
79
Chapter 6 CONCLUSIONS AND FUTURE DIRECTION
6.1 Conclusions
Modifications to the cell surface have widespread application and are being
heavily studied [32, 55, 58, 76]. By modifying the cell surface with gelatin-based
coatings, we are able to increase the retention of MSC inside the myocardium and reduce
patient risk of further heart failure [6, 55]. However, preliminary studies did not allow
full investigation of GelMA binding mechanism and showed poor coating efficiency. .
This dissertation focused on optimizing gelatin-based coatings on the surface of MSC for
MI therapy.
Initially, in Chapter 3 we begin by looking closely at the biotin-SA affinity that is
central to our overall coating process. We utilize a cardiac model where we aim to
immobilize cardiac cells onto a glass substrate so they can withstand flows of stimulant
solutions including caffeine. Use of the biotin-SA system allows for more than 80%
retention of cardiac myoblasts under conventional rinsing procedures while unmodified
cells are largely rinsed away. We also demonstrate that the biotin-SA model allows
retention despite shears of up to 160 mL/min. When applying this adhesion system to
cardiomyocyte functional analysis, we measure calcium release transients by caffeine
induction at an 80% success rate compared to 20% without surface engineering. In all, we
prove the strength of anchoring our coatings to the cell surface through biotin-SA is
stronger than common cell adhesion promotors. We also demonstrate further application
of cell surface engineering beyond in vivo MI therapies.
After demonstrating the strength of our cell anchoring technique, Chapter 4 and
Chapter 5 focus on optimizing the GelMA coating for therapeutic benefit. In Chapter 4,
80
we investigate the potential binding mode of GelMA inside the myocardium. While we
looked at both resident cells and the ECM, our results demonstrated that ECM drives
GelMA binding while resident cells did not significantly contribute to binding. We
conclude that GelMA cellular coatings significantly increase the binding of cells to
collagen within the ECM. As collagen subunits in the GelMA coating are likely binding
to collagen in the ECM, Chapter 5 seeks to increase the yield of coated cells during the
processing of therapeutic cells. In Chapter 5, we step away from traditional covalent
methods of cell biotinylating and utilize lipid insertion with CholBio. The use of
cholesterol anchors increased streptavidin density on the surface of MSCs and increased
the fraction of MSCs coated with gelatin (83%) when compared to NHS-biotin (52%).
Additionally, the cholesterol anchors increased the uniformity of the coating on the MSC
surface. Critically, this improvement in gelatin coating efficiency did not impact cytokine
secretion and other critical MSC functions. In all, these cholesterol anchors offer an
effective path towards the formation of conformal coatings on the majority of MSCs to
improve the retention of MSCs in the heart following AMI.
According to a recent meta-analysis, in vivo human trials showed that the more
MSC administrated to patients, the higher the therapeutic impact [114]. Administration of
high MSC numbers could potentially result in clogged arteries. Therefore, there is a
critical need to retain as many MSCs without increasing administration dosage. In this
work, we take an already proven method of increasing MSC retention in vivo [6, 55] and
further optimize the process. By optimizing our coating process, we take one step closer
to making MSC therapy effective and safe for human trials. By applying our GelMA
coated MSC to human trials we aim to reduce patient risk for continued heart disease.
81
6.2 Future Direction
Throughout this work we focused on optimizing our coating process to achieve
the highest therapeutic impact. However, there is a critical need to take basic science and
translate it to clinically applicable research. The future direction proposed below takes
this research one step closer to human clinical trials. I first propose a method to track the
coating in vivo and ensure safe clearance of GelMA from the body. Secondly, I discuss
how to scale out current process to meet human trial needs.
6.2.1 Tracking of GelMA Coating in vivo
As we optimize our GelMA coating process, we move closer to introducing this
therapy to human clinical trials. While GelMA is degraded from naturally derived
collagen and FDA approved for other various applications, it is still critical to understand
the degradation of our coatings in vivo. GelMA is most likely degraded in vivo through
matrix metalloproteinases (MMPs). One approach to studying this possibility would be
introducing MMPs at varying concentrations to GelMA coated MSC in vitro and
monitoring degradation.
However, MMP concentration in the body is unknown and this still does not
indicate where the coating goes once it is degraded. Here, I propose the use of FDA
approved iron nanoparticle, Ferumoxytol, for the tracking of our GelMA coating in vivo.
Ferumoxytol may be embedded into GelMA coating and tracked in vivo through
magnetic resonance imagining (MRI). In order to do so, Ferumoxytol is introduced to our
monomer solution at varying concentrations (400 or 800 ug/mL). Afterward, cells are
suspended in the monomer solution per normal protocol and crosslinked under 530 nm
82
green light. Ferumoxytol embedding could then be confirmed through histology staining
with Ferrocyanide. Coated versus uncoated cells would then be administrated and an
initial MRI obtained. MRI would then be continued periodically for the first 30 days
following injection. In this model, we assume the body flushes out the iron nanoparticles
the same as it does the GelMA. By understanding how GelMA is flushed out of the body
we move closer to proving the safety of this therapy for human trials.
6.2.2 Freezing GelMA Coated MSC for Future Use
Our coating process allows cells to be coated in batches of 1.5 million MSCs.
While the amount of MSC needed for our mouse models is minimal, this number does
not compare to the minimum of 50 million cells needed for human injection [114]. Our
lab has made some attempts to scale up the batch coating process, however, scale-up is
largely limited due to the formation of bulk gel as opposed to individual cellular coatings.
The formation of bulk gel is undesirable as MSC must now secrete the necessary
stimulants through a thicker gel wall and, most importantly, bulk gel puts the patient at
risk for clogging of the arteries. There is a clinical need to modify the MSC coating
process to fit human trial numbers.
One alternate approach to scaling up the coating process is to continue with
multiple smaller cell batches made in advance. Cells are routinely preserved through
cryopreservation, either short-term at -80 C or long term in the vapor phase of liquid
nitrogen. In this proposed approach, allogenic MSC are obtained from donor patients and
then coated in the gelatin process. Following, cells are suspended in a 10% DMSO to
90% cell media mixture then preserved in liquid nitrogen until further use. Once a patient
83
is admitted to the hospital for MI, the necessary number of allogenic cells are thawed and
ready for use. In this approach we minimize the risk of bulk gel formation and utilize a
well-established freeze-thaw technique for cells.
Prior to applying this methodology to human trials, MSC function following the
freeze-thaw method would need to be evaluated in vitro. In this approach, we evaluate the
ability for MSC to proliferate and migrate according to their normal function. Most
importantly, we would study the ability of MSC to secrete immunomodulating factors.
This is commonly done by taking cultured media from MSC, adding it a macrophage
culture than measuring inflammatory signal through flow cytometry. Lastly, we would
want to determine if the MSC coating is retained after freeze-thaw cycles. This may be
done through collagen antibodies or embedding Ferumoxytol as previously mentioned.
An added benefit to this approach is enabling broader temporal control over
therapeutic cell interventions. Human clinical trials involving bone marrow cells injected
into heart disease patients showed the optimal time for MSC administration is between 3-
10 days [114]. This 3–10 day window is not enough time to culture the autologous
MSCs, thus an existing culture of heterologous MSC must be ready. While previous
hospital data can be used to predict how frequently a patient is admitted for MI, the
clinical research team can never be absolutely certain when a patient will appear and
when allogenic cells need to be ready. The ability to prepare GelMA coated MSC in
advance meets a critical need for human trials.
84
APPENDIX
Appendix 1. Chapter 3 Statistical Values
In chapter 3 we take an in-depth analysis at a new adhesion system. In this system
we take normally unmodified cells and biotinylate them using NHS-Bio. Cells are then
incubated with an epoxy coated glass substrate. Below summarizes statistical
comparisons for the epoxy-SA system and biotinylating of cells to traditional systems.
Table A 1. Statistical comparison using Student t-test for Figures 1-3. Figure Sample 1 Sample 2 Argument P -value
1 Unmodified Glass Epoxy Coated Glass Cy3/µm2
2 Untouched H9C2 Biotinylated H9C2 Viability: Calcein 0.6965
Viability: Ethidium 0.1414 Viability: Caspase 0.6395
3
PBSA Untouched H9C2 eSA-Biotin System
Retention: 5 min 0.0045 Retention: 10 min 0.0062 Retention: 20 min 0.0028 Retention: 40 min 0.0004
Laminin Untouched H9C2 eSA-Biotin System
Retention: 5 min 0.0011 Retention: 10 min 0.0019 Retention: 20 min 0.0008 Retention: 40 min 0.0031
Poly-L-Lysine Untouched H9C2 eSA-Biotin System
Retention: 5 min 0.0010 Retention: 10 min 0.0337 Retention: 20 min 0.0405 Retention: 40 min 0.0045
PBSA @ t
PBSA @ t+∆t
Retention: 5 to 10 min N/A* Retention: 10 to 20 min N/A* Retention: 5 to 10 min N/A*
Laminin @ t Laminin @ t+∆t
Retention: 5 to 10 min 0.2977 Retention: 10 to 20 min 0.0394 Retention: 5 to 10 min 0.7049
Poly-L-Lysine @ t
Poly-L-Lysine @ t+∆t
Retention: 5 to 10 min 0.0380 Retention: 10 to 20 min 0.8637 Retention: 5 to 10 min 0.4604
eSA-Biotin @ t
eSA-Biotin @ t+∆t
Retention: 5 to 10 min 0.0026 Retention: 10 to 20 min 0.0480 Retention: 5 to 10 min 0.9075
Red Text: P*<0.05 Red Text and Background: P*<0.01 N/A*: No change. P-value cannot be calculated
85
Appendix 2. Calibration of Cy3 on Epoxy Slide
In Chapter 3, utilizing surface chemistry, we modify the surface of glass with an
epoxy coating that can be used to bind SA. The ability to modify the glass with SA is
particular useful as biotinylated cells can be immobilized onto the glass through high SA-
biotin affinity.
In order to test the ability to bind SA to an epoxy surface, varying concentrations
of Streptavidin-Cy3 (SA-Cy3, Invitrogen) in PBS (0.1, 02, 1, 2, 10, 20, 100, and 200
μg/mL) was incubated for 40 minutes onto epoxy-coated slides using altering wells and
then washed with PBS. Freshly prepared epoxy coated glass was compared to uncoated
microscope slides. Arrays were analyzed using an Affymetrix 428 Array Scanner and
Image J software. Cy3 calibration slide (Full Moon Biosystems, Figure A. 1A and Figure
A. 1B)) was used to determine the average fluorophore density for each sample. An
optimal concentration of 20 μg/mL SA-Cy3 was determined and used for all further
studies (Figure A. 1C).
86
Figure A.1: Cy3 can be effectively loaded onto epoxy coated glass substrate. A) Higher Limit Array Scanner calibration curve for density of Cy3 per image pixels. B) Lower Limit Array Scanner calibration curve for density of Cy3 per image pixels. C) Density of Cy3 conjugated to a freshly prepared epoxy slide at varying concentrations of SA-Cy3.
87
Appendix 3. Streptavidin-Phycoerythrin Calibration
BD QuantiBRITE PE quantitation beads were used to quantify PE fluorescence
on the cell surface by vortexing in 500 μL of PBS and analyzing the PE signal using
Flow Cytometry. QuantiBRITE beads generate PE signal at four different intensities that
are referenced to a known concertation of PE.
The linear relationship between PE signal and PE florescence was determined
using QuantiBRITE PE (Figure A. 2) counting beads. Per Invitrogen, SA-PE is
conjugated at 1:1 ratio of SA to PE. Using the ratio and diameter of MSC or H9C2
determined using a cellometer, SA density on the cell surface was determined.
Figure A. 2: Calibration curve of PE signal measured by Flow Cytometry versus total PE molecules. Calibration determined using BD QuantiBRITE PE quantitation beads.
88
REFRENCES
1.BENJAMIN, E.J., ET AL., HEART DISEASE AND STROKE STATISTICS—2019 UPDATE: A REPORT FROM THE AMERICAN HEART ASSOCIATION. CIRCULATION, 2019. 139(10): P. E56-E528. 2.TOMA, C., ET AL., HUMAN MESENCHYMAL STEM CELLS DIFFERENTIATE TO A CARDIOMYOCYTE PHENOTYPE IN THE ADULT MURINE HEART. CIRCULATION, 2002. 105(1): P. 93-98. 3. MÜLLER-EHMSEN, J., ET AL., EFFECTIVE ENGRAFTMENT BUT POOR MID-TERM PERSISTENCE OF MONONUCLEAR AND MESENCHYMAL BONE MARROW CELLS IN ACUTE AND CHRONIC RAT MYOCARDIAL INFARCTION. JOURNAL OF MOLECULAR AND CELLULAR CARDIOLOGY, 2006. 41(5): P. 876-884. 4. QUEVEDO, H.C., ET AL., ALLOGENEIC MESENCHYMAL STEM CELLS RESTORE CARDIAC FUNCTION IN CHRONIC ISCHEMIC CARDIOMYOPATHY VIA TRILINEAGE DIFFERENTIATING CAPACITY. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA, 2009. 106(33): P. 14022-14027. 5. WU, P.-J., ET AL., ADHESIVE STEM CELL COATINGS FOR ENHANCED RETENTION IN THE HEART TISSUE. ACS APPLIED BIO MATERIALS, 2020. 6. GOTTIPATI, A., ET AL., GELATIN BASED POLYMER CELL COATING IMPROVES BONE MARROW-DERIVED CELL RETENTION IN THE HEART AFTER MYOCARDIAL INFARCTION. STEM CELL REVIEWS AND REPORTS, 2019. 15(3): P. 404-414. 7. ROMERO, G., ET AL., PROTECTIVE POLYMER COATINGS FOR HIGH-THROUGHPUT, HIGH-PURITY CELLULAR ISOLATION. ACS APPLIED MATERIALS & INTERFACES, 2015. 7(32): P. 17598-17602. 8. ROSENBERG, S.A., ET AL., USE OF TUMOR-INFILTRATING LYMPHOCYTES AND INTERLEUKIN-2 IN THE IMMUNOTHERAPY OF PATIENTS WITH METASTATIC MELANOMA. NEW ENGLAND JOURNAL OF MEDICINE, 1988. 319(25): P. 1676-1680. 9. WU, P.-J., ET AL., HYDROGEL PATCHES ON LIVE CELLS THROUGH SURFACE-MEDIATED POLYMERIZATION. LANGMUIR, 2017. 33(27): P. 6778-6784. 10. DONG, X., D. CHU, AND Z. WANG, LEUKOCYTE-MEDIATED DELIVERY OF NANOTHERAPEUTICS IN INFLAMMATORY AND TUMOR SITES. THERANOSTICS, 2017. 7(3): P. 751-763. 11. WON, Y.-W., A.N. PATEL, AND D.A. BULL, CELL SURFACE ENGINEERING TO ENHANCE MESENCHYMAL STEM CELL MIGRATION TOWARD AN SDF-1 GRADIENT. BIOMATERIALS, 2014. 35(21): P. 5627-5635. 12. PATHAK, C.P., A.S. SAWHNEY, AND J.A. HUBBELL, RAPID PHOTOPOLYMERIZATION OF IMMUNOPROTECTIVE GELS IN CONTACT WITH CELLS AND TISSUE. JOURNAL OF THE AMERICAN CHEMICAL SOCIETY, 1992. 114(21): P. 8311-8312. 13. SAWHNEY, A.S., C.P. PATHAK, AND J.A. HUBBELL, MODIFICATION OF ISLET OF LANGERHANS SURFACES WITH IMMUNOPROTECTIVE POLY(ETHYLENE GLYCOL) COATINGS VIA INTERFACIAL PHOTOPOLYMERIZATION. BIOTECHNOLOGY AND BIOENGINEERING, 1994. 44(3): P. 383-386. 14. CRUISE, G.M., D.S. SCHARP, AND J.A. HUBBELL, CHARACTERIZATION OF PERMEABILITY AND NETWORK STRUCTURE OF INTERFACIALLY PHOTOPOLYMERIZED
89
POLY(ETHYLENE GLYCOL) DIACRYLATE HYDROGELS. BIOMATERIALS, 1998. 19(14): P. 1287-1294. 15. WILSON, J.T., W. CUI, AND E.L. CHAIKOF, LAYER-BY-LAYER ASSEMBLY OF A CONFORMAL NANOTHIN PEG COATING FOR INTRAPORTAL ISLET TRANSPLANTATION. NANO LETTERS, 2008. 8(7): P. 1940-1948. 16. VAITHILINGAM, V. AND B.E. TUCH, ISLET TRANSPLANTATION AND ENCAPSULATION: AN UPDATE ON RECENT DEVELOPMENTS. THE REVIEW OF DIABETIC STUDIES : RDS, 2011. 8(1): P. 51-67. 17. SHIH, H., H.-Y. LIU, AND C.-C. LIN, IMPROVING GELATION EFFICIENCY AND CYTOCOMPATIBILITY OF VISIBLE LIGHT POLYMERIZED THIOL-NORBORNENE HYDROGELS VIA ADDITION OF SOLUBLE TYROSINE. BIOMATERIALS SCIENCE, 2017. 5(3): P. 589-599. 18. LILLY, J.L., ET AL., CHARACTERIZATION OF MOLECULAR TRANSPORT IN ULTRATHIN HYDROGEL COATINGS FOR CELLULAR IMMUNOPROTECTION. BIOMACROMOLECULES, 2015. 16(2): P. 541-549. 19. ROSSI, N.A.A., ET AL., RED BLOOD CELL MEMBRANE GRAFTING OF MULTI-FUNCTIONAL HYPERBRANCHED POLYGLYCEROLS. BIOMATERIALS, 2010. 31(14): P. 4167-4178. 20. WANG, D., W.M. TOYOFUKU, AND M.D. SCOTT, THE POTENTIAL UTILITY OF METHOXYPOLY(ETHYLENE GLYCOL)-MEDIATED PREVENTION OF RHESUS BLOOD GROUP ANTIGEN RHD RECOGNITION IN TRANSFUSION MEDICINE. BIOMATERIALS, 2012. 33(10): P. 3002-3012. 21. LEE, J., ET AL., CYTOPROTECTIVE SILICA COATING OF INDIVIDUAL MAMMALIAN CELLS THROUGH BIOINSPIRED SILICIFICATION. ANGEWANDTE CHEMIE INTERNATIONAL EDITION, 2014. 53(31): P. 8056-8059. 22. CAHALL, C.F., ET AL., A QUANTITATIVE PERSPECTIVE ON SURFACE MARKER SELECTION FOR THE ISOLATION OF FUNCTIONAL TUMOR CELLS. BREAST CANCER: BASIC AND CLINICAL RESEARCH, 2015. 9: P. BCBCR. S25461. 23. LILLY, J.L. AND B.J. BERRON, THE ROLE OF SURFACE RECEPTOR DENSITY IN SURFACE-INITIATED POLYMERIZATIONS FOR CANCER CELL ISOLATION. LANGMUIR, 2016. 32(22): P. 5681-5689. 24. ŠAFAŘÍK, I. AND M. ŠAFAŘÍKOVÁ, USE OF MAGNETIC TECHNIQUES FOR THE ISOLATION OF CELLS. JOURNAL OF CHROMATOGRAPHY B: BIOMEDICAL SCIENCES AND APPLICATIONS, 1999. 722(1): P. 33-53. 25. PENN, M.S., ET AL., SDF-1 IN MYOCARDIAL REPAIR. GENE THERAPY, 2012. 19: P. 583. 26. DAVIS, K.A., ET AL., COATINGS ON MAMMALIAN CELLS: INTERFACING CELLS WITH THEIR ENVIRONMENT. JOURNAL OF BIOLOGICAL ENGINEERING, 2019. 13(1): P. 5. 27. KAASTRUP, K. AND H. SIKES, USING PHOTO-INITIATED POLYMERIZATION REACTIONS TO DETECT MOLECULAR RECOGNITION. CHEMICAL SOCIETY REVIEWS, 2016. 45(3): P. 532-545. 28. CAHALL, C.F., ET AL., CELL DEATH PERSISTS IN RAPID EXTRUSION OF LYSIS-RESISTANT COATED CARDIAC MYOBLASTS. BIOPRINTING, 2020. 18: P. E00072. 29. KARP, J.M. AND G.S. LENG TEO, MESENCHYMAL STEM CELL HOMING: THE DEVIL IS IN THE DETAILS. CELL STEM CELL, 2009. 4(3): P. 206-216.
90
30. PHINNEY, D.G. AND D.J. PROCKOP, CONCISE REVIEW: MESENCHYMAL STEM/MULTIPOTENT STROMAL CELLS: THE STATE OF TRANSDIFFERENTIATION AND MODES OF TISSUE REPAIR—CURRENT VIEWS. STEM CELLS, 2007. 25(11): P. 2896-2902. 31. HORWITZ, E.M., ET AL., ISOLATED ALLOGENEIC BONE MARROW-DERIVED MESENCHYMAL CELLS ENGRAFT AND STIMULATE GROWTH IN CHILDREN WITH OSTEOGENESIS IMPERFECTA: IMPLICATIONS FOR CELL THERAPY OF BONE. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, 2002. 99(13): P. 8932-8937. 32. SARKAR, D., ET AL., ENGINEERED CELL HOMING. BLOOD, 2011. 118(25): P. E184-E191. 33. WU, J., ET AL., INTRAVENOUSLY ADMINISTERED BONE MARROW CELLS MIGRATE TO DAMAGED BRAIN TISSUE AND IMPROVE NEURAL FUNCTION IN ISCHEMIC RATS. CELL TRANSPLANTATION, 2007. 16(10): P. 993-1005. 34. BENJAMIN, E.J., ET AL., HEART DISEASE AND STROKE STATISTICS-2017 UPDATE: A REPORT FROM THE AMERICAN HEART ASSOCIATION. CIRCULATION, 2017. 135(10): P. E146-E603. 35. ZOUGGARI, Y., ET AL., B LYMPHOCYTES TRIGGER MONOCYTE MOBILIZATION AND IMPAIR HEART FUNCTION AFTER ACUTE MYOCARDIAL INFARCTION. NATURE MEDICINE, 2013. 19(10): P. 1273. 36. PANIZZI, P., ET AL., IMPAIRED INFARCT HEALING IN ATHEROSCLEROTIC MICE WITH LY-6CHIMONOCYTOSIS. JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY, 2010. 55(15): P. 1629-1638. 37. EPELMAN, S., P.P. LIU, AND D.L. MANN, ROLE OF INNATE AND ADAPTIVE IMMUNE MECHANISMS IN CARDIAC INJURY AND REPAIR. NATURE REVIEWS IMMUNOLOGY, 2015. 15(2): P. 117. 38. NAHRENDORF, M., ET AL., THE HEALING MYOCARDIUM SEQUENTIALLY MOBILIZES TWO MONOCYTE SUBSETS WITH DIVERGENT AND COMPLEMENTARY FUNCTIONS. JOURNAL OF EXPERIMENTAL MEDICINE, 2007. 204(12): P. 3037-3047. 39. FRANGOGIANNIS, N.G., THE INFLAMMATORY RESPONSE IN MYOCARDIAL INJURY, REPAIR, AND REMODELLING. NATURE REVIEWS CARDIOLOGY, 2014. 11(5): P. 255. 40. BULKLEY, B.H. AND W.C. ROBERTS, STEROID THERAPY DURING ACUTE MYOCARDIAL INFARCTION: A CAUSE OF DELAYED HEALING AND OF VENTRICULAR ANEURYSM. THE AMERICAN JOURNAL OF MEDICINE, 1974. 56(2): P. 244-250. 41. OLSEN, A.-M.S., ET AL., DURATION OF TREATMENT WITH NONSTEROIDAL ANTI-INFLAMMATORY DRUGS AND IMPACT ON RISK OF DEATH AND RECURRENT MYOCARDIAL INFARCTION IN PATIENTS WITH PRIOR MYOCARDIAL INFARCTION. CIRCULATION, 2011. 123(20): P. 2226-2235. 42. HORCKMANS, M., ET AL., NEUTROPHILS ORCHESTRATE POST-MYOCARDIAL INFARCTION HEALING BY POLARIZING MACROPHAGES TOWARDS A REPARATIVE PHENOTYPE. EUROPEAN HEART JOURNAL, 2016. 38(3): P. 187-197. 43. DI NICOLA, M., ET AL., HUMAN BONE MARROW STROMAL CELLS SUPPRESS T-LYMPHOCYTE PROLIFERATION INDUCED BY CELLULAR OR NONSPECIFIC MITOGENIC STIMULI. BLOOD, THE JOURNAL OF THE AMERICAN SOCIETY OF HEMATOLOGY, 2002. 99(10): P. 3838-3843.
91
44. MAGGINI, J., ET AL., MOUSE BONE MARROW-DERIVED MESENCHYMAL STROMAL CELLS TURN ACTIVATED MACROPHAGES INTO A REGULATORY-LIKE PROFILE. PLOS ONE, 2010. 5(2): P. E9252-E9252. 45. ASAHARA, T., A. KAWAMOTO, AND H. MASUDA, CONCISE REVIEW: CIRCULATING ENDOTHELIAL PROGENITOR CELLS FOR VASCULAR MEDICINE. STEM CELLS, 2011. 29(11): P. 1650-1655. 46. HARE, J.M., ET AL., COMPARISON OF ALLOGENEIC VS AUTOLOGOUS BONE MARROW–DERIVED MESENCHYMAL STEM CELLS DELIVERED BY TRANSENDOCARDIAL INJECTION IN PATIENTS WITH ISCHEMIC CARDIOMYOPATHY: THE POSEIDON RANDOMIZED TRIAL. JAMA, 2012. 308(22): P. 2369-2379. 47. PREMER, C., ET AL., ALLOGENEIC MESENCHYMAL STEM CELLS RESTORE ENDOTHELIAL FUNCTION IN HEART FAILURE BY STIMULATING ENDOTHELIAL PROGENITOR CELLS. EBIOMEDICINE, 2015. 2(5): P. 467-475. 48. WILLIAMS, A.R. AND J.M. HARE, MESENCHYMAL STEM CELLS: BIOLOGY, PATHOPHYSIOLOGY, TRANSLATIONAL FINDINGS, AND THERAPEUTIC IMPLICATIONS FOR CARDIAC DISEASE. CIRCULATION RESEARCH, 2011. 109(8): P. 923-940. 49. HATZISTERGOS, K.E., ET AL., BONE MARROW MESENCHYMAL STEM CELLS STIMULATE CARDIAC STEM CELL PROLIFERATION AND DIFFERENTIATION. CIRCULATION RESEARCH, 2010. 107(7): P. 913-922. 50. MÜLLER-EHMSEN, J., ET AL., EFFECTIVE ENGRAFTMENT BUT POOR MID-TERM PERSISTENCE OF MONONUCLEAR AND MESENCHYMAL BONE MARROW CELLS IN ACUTE AND CHRONIC RAT MYOCARDIAL INFARCTION. JOURNAL OF MOLECULAR AND CELLULAR CARDIOLOGY, 2006. 41(5): P. 876-884. 51. QUEVEDO, H.C., ET AL., ALLOGENEIC MESENCHYMAL STEM CELLS RESTORE CARDIAC FUNCTION IN CHRONIC ISCHEMIC CARDIOMYOPATHY VIA TRILINEAGE DIFFERENTIATING CAPACITY. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, 2009. 106(33): P. 14022-14027. 52. GRIEVE, S.M., ET AL., MICROVASCULAR OBSTRUCTION BY INTRACORONARY DELIVERY OF MESENCHYMAL STEM CELLS AND QUANTIFICATION OF RESULTING MYOCARDIAL INFARCTION BY CARDIAC MAGNETIC RESONANCE. CIRCULATION: HEART FAILURE, 2010. 3(3): P. E5-E6. 53. LIPOWSKY, H.H., ET AL., MESENCHYMAL STEM CELL DEFORMABILITY AND IMPLICATIONS FOR MICROVASCULAR SEQUESTRATION. ANNALS OF BIOMEDICAL ENGINEERING, 2018. 46(4): P. 640-654. 54. SILVA, G.V., ET AL., MESENCHYMAL STEM CELLS DIFFERENTIATE INTO AN ENDOTHELIAL PHENOTYPE, ENHANCE VASCULAR DENSITY, AND IMPROVE HEART FUNCTION IN A CANINE CHRONIC ISCHEMIA MODEL. CIRCULATION, 2005. 111(2): P. 150-156. 55. PENG, H., ET AL., POLYMER CELL SURFACE COATING ENHANCES MESENCHYMAL STEM CELL RETENTION AND CARDIAC PROTECTION. ACS APPLIED BIO MATERIALS, 2021. 4(2): P. 1655-1667. 56. ABBINA, S., ET AL., SURFACE ENGINEERING FOR CELL-BASED THERAPIES: TECHNIQUES FOR MANIPULATING MAMMALIAN CELL SURFACES. ACS BIOMATERIALS SCIENCE & ENGINEERING, 2017. 57. MICRO- AND NANOENGINEERING OF THE CELL SURFACE. 1 ED. MICRO & NANO TECHNOLOGIES. 2014: ELSEVIER. 400.
92
58. ANSELMO, A.C., ET AL., MONOCYTE-MEDIATED DELIVERY OF POLYMERIC BACKPACKS TO INFLAMED TISSUES: A GENERALIZED STRATEGY TO DELIVER DRUGS TO TREAT INFLAMMATION. JOURNAL OF CONTROLLED RELEASE, 2015. 199: P. 29-36. 59. LIVNAH, O., ET AL., THREE-DIMENSIONAL STRUCTURES OF AVIDIN AND THE AVIDIN-BIOTIN COMPLEX. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, 1993. 90(11): P. 5076-5080. 60. DE LA RICA, R. AND M.M. STEVENS, PLASMONIC ELISA FOR THE ULTRASENSITIVE DETECTION OF DISEASE BIOMARKERS WITH THE NAKED EYE. NATURE NANOTECHNOLOGY, 2012. 7(12): P. 821-824. 61. NEMZEK, J.A., J. SIDDIQUI, AND D.G. REMICK, DEVELOPMENT AND OPTIMIZATION OF CYTOKINE ELISAS USING COMMERCIAL ANTIBODY PAIRS. JOURNAL OF IMMUNOLOGICAL METHODS, 2001. 255(1): P. 149-157. 62. WILSON, J.T. AND E.L. CHAIKOF, CHALLENGES AND EMERGING TECHNOLOGIES IN THE IMMUNOISOLATION OF CELLS AND TISSUES. ADVANCED DRUG DELIVERY REVIEWS, 2008. 60(2): P. 124-145. 63. SCOTT, M.D. AND A.M. CHEN, BEYOND THE RED CELL: PEGYLATION OF OTHER BLOOD CELLS AND TISSUES. TRANSFUSION CLINIQUE ET BIOLOGIQUE, 2004. 11(1): P. 40-46. 64. VILLA, C.H., ET AL., RED BLOOD CELLS: SUPERCARRIERS FOR DRUGS, BIOLOGICALS, AND NANOPARTICLES AND INSPIRATION FOR ADVANCED DELIVERY SYSTEMS. ADVANCED DRUG DELIVERY REVIEWS, 2016. 106: P. 88-103. 65. ROBERTUS, J., W.R. BROWNE, AND B.L. FERINGA, DYNAMIC CONTROL OVER CELL ADHESIVE PROPERTIES USING MOLECULAR-BASED SURFACE ENGINEERING STRATEGIES. CHEMICAL SOCIETY REVIEWS, 2010. 39(1): P. 354-378. 66. RIBEIRO, R.D.C., ET AL., TEMPORARY SINGLE-CELL COATING FOR BIOPROCESSING WITH A CATIONIC POLYMER. ACS APPLIED MATERIALS & INTERFACES, 2017. 9(15): P. 12967-12974. 67. DZAMUKOVA, M.R., ET AL., A DIRECT TECHNIQUE FOR MAGNETIC FUNCTIONALIZATION OF LIVING HUMAN CELLS. LANGMUIR, 2011. 27(23): P. 14386-14393. 68. ZHANG, P., ET AL., TOP-DOWN FABRICATION OF POLYELECTROLYTE-THERMOPLASTIC HYBRID MICROPARTICLES FOR UNIDIRECTIONAL DRUG DELIVERY TO SINGLE CELLS. ADVANCED HEALTHCARE MATERIALS, 2013. 2(4): P. 540-545. 69. DOSHI, N., ET AL., CELL-BASED DRUG DELIVERY DEVICES USING PHAGOCYTOSIS-RESISTANT BACKPACKS. ADVANCED MATERIALS, 2011. 23(12): P. H105-H109. 70. BORKOWSKA, M., ET AL., PERFORMANCE AND DETECTION OF NANO-THIN POLYELECTROLYTE SHELL FOR CELL COATING. JOURNAL OF NANOPARTICLE RESEARCH, 2014. 16(7): P. 2488. 71. ZHANG, Y., ET AL., DRUG-ELUTING CONFORMAL COATINGS ON INDIVIDUAL CELLS. CELLULAR AND MOLECULAR BIOENGINEERING, 2016. 9(3): P. 382-397. 72. SWISTON, A.J., ET AL., SURFACE FUNCTIONALIZATION OF LIVING CELLS WITH MULTILAYER PATCHES. NANO LETTERS, 2008. 8(12): P. 4446-4453. 73. SWISTON, A.J., ET AL., FREELY SUSPENDED CELLULAR “BACKPACKS” LEAD TO CELL AGGREGATE SELF-ASSEMBLY. BIOMACROMOLECULES, 2010. 11(7): P. 1826-1832.
93
74. MAHAL, L.K., K.J. YAREMA, AND C.R. BERTOZZI, ENGINEERING CHEMICAL REACTIVITY ON CELL SURFACES THROUGH OLIGOSACCHARIDE BIOSYNTHESIS. SCIENCE, 1997. 276(5315): P. 1125. 75. HASHEMI-NAJAFABADI, S., ET AL., A METHOD TO OPTIMIZE PEG-COATING OF RED BLOOD CELLS. BIOCONJUGATE CHEMISTRY, 2006. 17(5): P. 1288-1293. 76. WANG, D., ET AL., INDUCTION OF IMMUNOTOLERANCE VIA MPEG GRAFTING TO ALLOGENEIC LEUKOCYTES. BIOMATERIALS, 2011. 32(35): P. 9494-9503. 77. STEPHAN, M.T., ET AL., THERAPEUTIC CELL ENGINEERING USING SURFACE-CONJUGATED SYNTHETIC NANOPARTICLES. NATURE MEDICINE, 2010. 16(9): P. 1035-1041. 78. BASLE, E., N. JOUBERT, AND M. PUCHEAULT, PROTEIN CHEMICAL MODIFICATION ON ENDOGENOUS AMINO ACIDS. CHEMISTRY & BIOLOGY, 2010. 17(3): P. 213-227. 79. COUNCIL, N.R., MONOCLONAL ANTIBODY PRODUCTION. 1999, WASHINGTON, DC: THE NATIONAL ACADEMIES PRESS. 74. 80. FRANTZ, C., K.M. STEWART, AND V.M. WEAVER, THE EXTRACELLULAR MATRIX AT A GLANCE. JOURNAL OF CELL SCIENCE, 2010. 123(24): P. 4195-4200. 81. LOCKHART, M., ET AL., EXTRACELLULAR MATRIX AND HEART DEVELOPMENT. BIRTH DEFECTS RESEARCH. PART A, CLINICAL AND MOLECULAR TERATOLOGY, 2011. 91(6): P. 535-550. 82. JOURDAN-LESAUX, C., J. ZHANG, AND M.L. LINDSEY, EXTRACELLULAR MATRIX ROLES DURING CARDIAC REPAIR. LIFE SCIENCES, 2010. 87(13): P. 391-400. 83. LINDSEY, M.L., ET AL., EFFECT OF A CLEAVAGE-RESISTANT COLLAGEN MUTATION ON LEFT VENTRICULAR REMODELING. CIRCULATION RESEARCH, 2003. 93(3): P. 238-245. 84. NAWATA, J., ET AL., DIFFERENTIAL EXPRESSION OF Α1, Α3 AND Α5 INTEGRIN SUBUNITS IN ACUTE AND CHRONIC STAGES OF MYOCARDIAL INFARCTION IN RATS. CARDIOVASCULAR RESEARCH, 1999. 43(2): P. 371-381. 85. RUOSLAHTI, E., INTEGRINS. THE JOURNAL OF CLINICAL INVESTIGATION, 1991. 87(1): P. 1-5. 86. HYNES, R.O., INTEGRINS: BIDIRECTIONAL, ALLOSTERIC SIGNALING MACHINES. CELL, 2002. 110(6): P. 673-687. 87. BARCZYK, M., S. CARRACEDO, AND D. GULLBERG, INTEGRINS. CELL AND TISSUE RESEARCH, 2010. 339(1): P. 269. 88. CHEN, C., ET AL., INTEGRINS AND INTEGRIN-RELATED PROTEINS IN CARDIAC FIBROSIS. JOURNAL OF MOLECULAR AND CELLULAR CARDIOLOGY, 2016. 93: P. 162-174. 89. HÄUSELMANN, S.P., ET AL., Β1-INTEGRIN IS UP-REGULATED VIA RAC1-DEPENDENT REACTIVE OXYGEN SPECIES AS PART OF THE HYPERTROPHIC CARDIOMYOCYTE RESPONSE. FREE RADICAL BIOLOGY AND MEDICINE, 2011. 51(3): P. 609-618. 90. BERS, D.M., CARDIAC EXCITATION–CONTRACTION COUPLING. NATURE, 2002. 415(6868): P. 198. 91. CRUMP, S.M., ET AL., THE CARDIAC L-TYPE CALCIUM CHANNEL DISTAL CARBOXY TERMINUS AUTOINHIBITION IS REGULATED BY CALCIUM. AMERICAN
94
JOURNAL OF PHYSIOLOGY-HEART AND CIRCULATORY PHYSIOLOGY, 2012. 304(3): P. H455-H464. 92. MATTHES, J., ET AL., CA2+-DEPENDENT MODULATION OF SINGLE HUMAN CARDIAC L-TYPE CALCIUM CHANNELS BY THE CALCINEURIN INHIBITOR CYCLOSPORINE. JOURNAL OF MOLECULAR AND CELLULAR CARDIOLOGY, 2004. 36(2): P. 241-255. 93. AHERN, B.M. AND J. SATIN, THE L-TYPE CALCIUM CHANNEL CURRENT MODULATION MECHANISM: THE PLOT THICKENS AND FOGS. THE JOURNAL OF CLINICAL INVESTIGATION, 2019. 129(2). 94. CHEN, C.Y., ET AL., SUPPRESSION OF DETYROSINATED MICROTUBULES IMPROVES CARDIOMYOCYTE FUNCTION IN HUMAN HEART FAILURE. NATURE MEDICINE, 2018. 24(8): P. 1225-1233. 95. LOUCH, W.E., K.A. SHEEHAN, AND B.M. WOLSKA, METHODS IN CARDIOMYOCYTE ISOLATION, CULTURE, AND GENE TRANSFER. JOURNAL OF MOLECULAR AND CELLULAR CARDIOLOGY, 2011. 51(3): P. 288-298. 96. PHELAN, K. AND K.M. MAY, BASIC TECHNIQUES IN MAMMALIAN CELL TISSUE CULTURE. CURRENT PROTOCOLS IN CELL BIOLOGY, 2015. 66(1): P. 1.1.1-1.1.22. 97. MITCHESON, J.S., J.C. HANCOX, AND A.J. LEVI, ACTION POTENTIALS, ION CHANNEL CURRENTS AND TRANSVERSE TUBULE DENSITY IN ADULT RABBIT VENTRICULAR MYOCYTES MAINTAINED FOR 6 DAYS IN CELL CULTURE. PFLÜGERS ARCHIV, 1996. 431(6): P. 814-827. 98. DVORNIKOV, A.V., ET AL., NOVEL APPROACHES TO DETERMINE CONTRACTILE FUNCTION OF THE ISOLATED ADULT ZEBRAFISH VENTRICULAR CARDIAC MYOCYTE. THE JOURNAL OF PHYSIOLOGY, 2014. 592(9): P. 1949-1956. 99. KO, I.K., T.J. KEAN, AND J.E. DENNIS, TARGETING MESENCHYMAL STEM CELLS TO ACTIVATED ENDOTHELIAL CELLS. BIOMATERIALS, 2009. 30(22): P. 3702-3710. 100. KO, I.K., ET AL., TARGETING IMPROVES MSC TREATMENT OF INFLAMMATORY BOWEL DISEASE. MOLECULAR THERAPY. 18(7): P. 1365-1372. 101. LEE, R.H., ET AL., INTRAVENOUS HMSCS IMPROVE MYOCARDIAL INFARCTION IN MICE BECAUSE CELLS EMBOLIZED IN LUNG ARE ACTIVATED TO SECRETE THE ANTI-INFLAMMATORY PROTEIN TSG-6. CELL STEM CELL, 2009. 5(1): P. 54-63. 102. SANTOSO, M.R. AND P.C. YANG, MAGNETIC NANOPARTICLES FOR TARGETING AND IMAGING OF STEM CELLS IN MYOCARDIAL INFARCTION. STEM CELLS INTERNATIONAL, 2016. 2016: P. 4198790. 103. KIM, H., ET AL., LARGE AREA TWO-DIMENSIONAL B CELL ARRAYS FOR SENSING AND CELL-SORTING APPLICATIONS. BIOMACROMOLECULES, 2004. 5(3): P. 822-827. 104. IWASAKI, Y. AND T. OTA, EFFICIENT BIOTINYLATION OF METHACRYLOYL-FUNCTIONALIZED NONADHERENT CELLS FOR FORMATION OF CELL MICROARRAYS. CHEMICAL COMMUNICATIONS, 2011. 47(37): P. 10329-10331. 105. HSIAO, S.C., ET AL., DIRECT CELL SURFACE MODIFICATION WITH DNA FOR THE CAPTURE OF PRIMARY CELLS AND THE INVESTIGATION OF MYOTUBE FORMATION ON DEFINED PATTERNS. LANGMUIR, 2009. 25(12): P. 6985-6991. 106. TSUKRUK, V.V., I. LUZINOV, AND D. JULTHONGPIPUT, STICKY MOLECULAR SURFACES: EPOXYSILANE SELF-ASSEMBLED MONOLAYERS. LANGMUIR, 1999. 15(9): P. 3029-3032.
95
107. MAGYAR, J., ET AL., REM-GTPASE REGULATES CARDIAC MYOCYTE L-TYPE CALCIUM CURRENT. CHANNELS, 2012. 6(3): P. 166-173. 108. KIM, J.C. AND S.H. WOO, SHEAR STRESS INDUCES A LONGITUDINAL CA2+ WAVE VIA AUTOCRINE ACTIVATION OF P2Y1 PURINERGIC SIGNALLING IN RAT ATRIAL MYOCYTES. THE JOURNAL OF PHYSIOLOGY, 2015. 593(23): P. 5091-5109. 109. HENDERSON, S.A., ET AL., FUNCTIONAL ADULT MYOCARDIUM IN THE ABSENCE OF NA+-CA2+ EXCHANGE: CARDIAC-SPECIFIC KNOCKOUT OF NCX1. CIRCULATION RESEARCH, 2004. 95(6): P. 604-611. 110. OBAL, D. AND J.C. WU, INDUCED PLURIPOTENT STEM CELLS AS A PLATFORM TO UNDERSTAND PATIENT‐SPECIFIC RESPONSES TO OPIOIDS AND ANAESTHETICS. BRITISH JOURNAL OF PHARMACOLOGY, 2020. 111. PAIK, D.T., M. CHANDY, AND J.C. WU, PATIENT AND DISEASE–SPECIFIC INDUCED PLURIPOTENT STEM CELLS FOR DISCOVERY OF PERSONALIZED CARDIOVASCULAR DRUGS AND THERAPEUTICS. PHARMACOLOGICAL REVIEWS, 2020. 72(1): P. 320-342. 112. RAMACHANDRA, C.J.A., ET AL., HUMAN INDUCED PLURIPOTENT STEM CELLS FOR MODELLING METABOLIC PERTURBATIONS AND IMPAIRED BIOENERGETICS UNDERLYING CARDIOMYOPATHIES. CARDIOVASCULAR RESEARCH, 2020. 113. RITCHIE, H. AND M. ROSER, CAUSES OF DEATH. OUR WORLD IN DATA, 2018. 114. AFZAL, M.R., ET AL., ADULT BONE MARROW CELL THERAPY FOR ISCHEMIC HEART DISEASE: EVIDENCE AND INSIGHTS FROM RANDOMIZED CONTROLLED TRIALS. CIRCULATION RESEARCH, 2015. 117(6): P. 558-575. 115. YUE, K., ET AL., SYNTHESIS, PROPERTIES, AND BIOMEDICAL APPLICATIONS OF GELATIN METHACRYLOYL (GELMA) HYDROGELS. BIOMATERIALS, 2015. 73: P. 254-271. 116. GAO, E., ET AL., A NOVEL AND EFFICIENT MODEL OF CORONARY ARTERY LIGATION AND MYOCARDIAL INFARCTION IN THE MOUSE. CIRC RES, 2010. 107(12): P. 1445-53. 117. KLYACHKIN, Y.M., ET AL., PHARMACOLOGICAL ELEVATION OF CIRCULATING BIOACTIVE PHOSPHOSPHINGOLIPIDS ENHANCES MYOCARDIAL RECOVERY AFTER ACUTE INFARCTION. STEM CELLS TRANSL MED, 2015. 118. HANSEN, R.R., H.D. SIKES, AND C.N. BOWMAN, VISUAL DETECTION OF LABELED OLIGONUCLEOTIDES USING VISIBLE-LIGHT-POLYMERIZATION-BASED AMPLIFICATION. BIOMACROMOLECULES, 2007. 9(1): P. 355-362. 119. DOETSCHMAN, T. AND M. AZHAR, CARDIAC-SPECIFIC INDUCIBLE AND CONDITIONAL GENE TARGETING IN MICE. CIRCULATION RESEARCH, 2012. 110(11): P. 1498-1512. 120. XU, H., ET AL., DISCOIDIN DOMAIN RECEPTORS PROMOTE Α1Β1-AND Α2Β1-INTEGRIN MEDIATED CELL ADHESION TO COLLAGEN BY ENHANCING INTEGRIN ACTIVATION. PLOS ONE, 2012. 7(12): P. E52209. 121. KLOTZ, B.J., ET AL., GELATIN-METHACRYLOYL HYDROGELS: TOWARDS BIOFABRICATION-BASED TISSUE REPAIR. TRENDS IN BIOTECHNOLOGY, 2016. 34(5): P. 394-407. 122. CHEN, Y.C., ET AL., FUNCTIONAL HUMAN VASCULAR NETWORK GENERATED IN PHOTOCROSSLINKABLE GELATIN METHACRYLATE HYDROGELS. ADVANCED FUNCTIONAL MATERIALS, 2012. 22(10): P. 2027-2039.
96
123. LIU, B., ET AL., HYDROGEN BONDS AUTONOMOUSLY POWERED GELATIN METHACRYLATE HYDROGELS WITH SUPER-ELASTICITY, SELF-HEAL AND UNDERWATER SELF-ADHESION FOR SUTURELESS SKIN AND STOMACH SURGERY AND E-SKIN. BIOMATERIALS, 2018. 171: P. 83-96. 124. MONGRAIN, R. AND J. RODÉS-CABAU, ROLE OF SHEAR STRESS IN ATHEROSCLEROSIS AND RESTENOSIS AFTER CORONARY STENT IMPLANTATION. REVISTA ESPANOLA DE CARDIOLOGIA, 2006. 59(1): P. 1-4. 125. MESCHIARI, C.A., ET AL., THE IMPACT OF AGING ON CARDIAC EXTRACELLULAR MATRIX. GEROSCIENCE, 2017. 39(1): P. 7-18. 126. BRABLETZ, T., EMT AND MET IN METASTASIS: WHERE ARE THE CANCER STEM CELLS? CANCER CELL, 2012. 22(6): P. 699-701. 127. BEAUCHAMP, J.R., ET AL., DYNAMICS OF MYOBLAST TRANSPLANTATION REVEAL A DISCRETE MINORITY OF PRECURSORS WITH STEM CELL–LIKE PROPERTIES AS THE MYOGENIC SOURCE. THE JOURNAL OF CELL BIOLOGY, 1999. 144(6): P. 1113-1122. 128. D'SOUZA, S., ET AL., ENGINEERING OF CELL MEMBRANES WITH A BISPHOSPHONATE-CONTAINING POLYMER USING ATRP SYNTHESIS FOR BONE TARGETING. BIOMATERIALS, 2014. 35(35): P. 9447-9458. 129. MAO, A.S., ET AL., PROGRAMMABLE MICROENCAPSULATION FOR ENHANCED MESENCHYMAL STEM CELL PERSISTENCE AND IMMUNOMODULATION. PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES, 2019. 116(31): P. 15392-15397. 130. KAASTRUP, K. AND H. SIKES, INVESTIGATION OF DENDRIMERS FUNCTIONALIZED WITH EOSIN AS MACROPHOTOINITIATORS FOR POLYMERIZATION-BASED SIGNAL AMPLIFICATION REACTIONS. RSC ADVANCES, 2015. 5(20): P. 15652-15659. 131. SIMONS, K. AND E. IKONEN, HOW CELLS HANDLE CHOLESTEROL. SCIENCE, 2000. 290(5497): P. 1721-1726. 132. VABBILISETTY, P., ET AL., CHEMICAL REACTIVE ANCHORING LIPIDS WITH DIFFERENT PERFORMANCE FOR CELL SURFACE RE-ENGINEERING APPLICATION. ACS OMEGA, 2018. 3(2): P. 1589-1599. 133. LILLY, J.L., ET AL., COMPARISON OF EOSIN AND FLUORESCEIN CONJUGATES FOR THE PHOTOINITIATION OF CELL-COMPATIBLE POLYMER COATINGS. PLOS ONE, 2018. 13(1): P. E0190880. 134. FUKUTO, M., ET AL., EFFECTS OF SURFACE LIGAND DENSITY ON LIPID-MONOLAYER-MEDIATED 2D ASSEMBLY OF PROTEINS. SOFT MATTER, 2010. 6(7): P. 1513-1519. 135. HOON, J.-L., W.-K. WONG, AND C.-G. KOH, FUNCTIONS AND REGULATION OF CIRCULAR DORSAL RUFFLES. MOLECULAR AND CELLULAR BIOLOGY, 2012. 32(21): P. 4246-4257. 136. CHOLESTEROL, CID=5997, IN NATIONAL CENTER FOR BIOTECHNOLOGY INFORMATION. SEPT. 12, 2019: PUBCHEM DATABASE. 137. BIOTIN, CID=171548, N.C.F.B. INFORMATION, EDITOR.: PUBCHEM DATABASE. 138. LIU, G., ET AL., CYTOTOXICITY STUDY OF POLYETHYLENE GLYCOL DERIVATIVES. RSC ADVANCES, 2017. 7(30): P. 18252-18259. 139. KIM, S. AND H.D. SIKES, PHENOLPHTHALEIN-CONJUGATED HYDROGEL FORMATION UNDER VISIBLE-LIGHT IRRADIATION FOR REDUCING VARIABILITY OF COLORIMETRIC BIODETECTION. ACS APPLIED BIO MATERIALS, 2018. 1(2): P. 216-220.
97
140. SIKES, H.D., ET AL., USING POLYMERIC MATERIALS TO GENERATE AN AMPLIFIED RESPONSE TO MOLECULAR RECOGNITION EVENTS. NATURE MATERIALS, 2008. 7(1): P. 52. 141. BERRON, B.J., ET AL., ANTIGEN-RESPONSIVE, MICROFLUIDIC VALVES FOR SINGLE USE DIAGNOSTICS. LAB ON A CHIP, 2012. 12(4): P. 708-710. 142. SAKAI, S., ET AL., CELL-SELECTIVE ENCAPSULATION IN HYDROGEL SHEATHS VIA BIOSPECIFIC IDENTIFICATION AND BIOCHEMICAL CROSS-LINKING. BIOMATERIALS, 2015. 53(SUPPLEMENT C): P. 494-501.
98
KARA DAVIS VITA
EDUCATION - PhD in Chemical Engineering, Defense May 2021, Graduation August 2021 –
UNIVERSITY OF Kentucky, Lexington, KY - BS in Chemical Engineering, May 2016 – Rose-Hulman Institute of Technology,
Terre Haute, IN
HONORS / AWARDS - NIH TL1 Institutional Training Fellow, 2019-2021, Center for Clinical and
Translation Science, University of Kentucky, Lexington, KY
EXPERINCE - PhD Thesis, Gelatin-Based Coatings for Increased Retention of Therapeutic Cells
Following Myocardial Infarction, 2017-2021, University of Kentucky, Lexington, KY
- Manufacturing Technology Specialist, Cook Pharmica, 2016-2017 , Bloomington, IN
PUBLICATIONS 1. K.A. Davis^, P. Wu^, C.F. Cahall, C. Li, A. Gottipati and B.J. Berron*; “Coatings on
Mammalian Cells: Interfacing Cells with Their Environment.” Journal of Biological Engineering, 2019.
2. K.A. Davis, H. Peng, C. Lakshman, A. Abdel-Latif*, B.J. Berron*; “Increased Yield of Gelatin Coated Therapeutic Cells Through Cholesterol Insertion.” Journal of Biomedical Materials Research: Part A, 2020.
3. C.F. Cahall, A. Preet Kaur, K.A. Davis, J.T. Pham, H.Y. Shin, B.J. Berron*; “Cell Death Persists in Rapid Extrusion of Lysis-Resistant Coated Cardiac Myoblasts.” Bioprinting, 18: e00072, 2020.
4. H. Peng, L. Chelvarajan, A. Gottipato, C. F Cahall, K. A. Davis, H. Tripathi, A. Al-Darraji; E. Elsawalhy, N. Dobrozsi, A. Srinivasan, B. Levitan, R. Kong, E. Gao, A. Abdel-Latif and B. J Berron, “Polymer Cell Surface Coating Enhances Mesenchymal Stem Cell Retention and Cardiac Protection” ACS Applied Biomaterials, 2021.
5. K.A. Davis, A.H. Sebastian, B.M. Ahern, C.A. Trinkle; J. Satin, A. Abdel-Latif, B.J. Berron*; “Increased Retention of Cardiac Cells to Glass Substrate through Streptavidin-Biotin Affinity” ACS Omega, Submitted April 2021
6. K.A. Davis, A. Gottipati, R. Donahue, H. Peng, L. Chelvarajan, C.F. Cahall, H. Tripathi, A. Al-Darraji, A. Abdel-Latif* and B.J. Berron* “Increased Retention of Gelatin Coated Therapeutic Cells through ECM binding” Anticipated Submission April 202.
7. Lauren Mehanna, Kara Davis, Shankar Miller-Murthy, Tracy Gastineau-Stevens, Bert Lynn, Brad J. Berron*,” Investigation of Appropriate Cleaning Solutions for
99
Removal of Denatonium Benzoate from Distillery Equipment”, Journal of Distilling Science, Submitted April 2021.
PRESENTATIONS 1. K.A. Davis, H. Peng, C. Lakshman, A. Abdel-Latif*, B.J. Berron*; “Increased Yield
of Gelatin Coated Therapeutic Cells Through Cholesterol Insertion.” Thin Films Symposium, University of Tennessee. 2019.
2. K.A. Davis, “Life after Rose: Industry or Grad School,” Rose-Hulman Chemical Engineering Seminar Series, 2020, Terre Haute, IN.
3. K.A. Davis, * “Engineering Stem Cells to Heart Attack Patients.” Materials and Chemical Engineering Graduate Conference, Three Minute Thesis, First Place. 2020
4. K.A. Davis, A. Gottipati, R. Donahue, L. Chelvarajan, C.F. Cahall, H. Tripathi, A. Al-Darraji, A. Abdel-Latif* and B.J. Berron* “Increased Retention of Gelatin Coated Therapeutic Cells through ECM binding” Clinical and Translational Science Virtual Conference, March 2021.