characterization of human bone marrow mesenchymal stem … · mesenchymal stem cells into...
TRANSCRIPT
CHARACTERIZATION OF ADULT HUMAN
BONE MARROW MESENCHYMAL STEM CELLS
FOR EFFECTIVE MYOCARDIAL REPAIR
GENEVIEVE TAN MEI YUN
(B.Sc. (Hons.), NUS)
A THESIS SUBMITTED
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
DEPARTMENT OF SURGERY
NATIONAL UNIVERSITY OF SINGAPORE
2009
i
Summary
Cardiovascular disease is a prevalent cause of death in the world. Cell transplantation therapy
has recently been developed as an alternative therapy for cardiovascular disease. However,
current studies employing the use of undifferentiated bone marrow stem cells have resulted in
variable clinical outcomes with modest efficacies. Differentiating primitive adult bone
marrow stem cells into a stable and committed cardiac (-like) phenotype ex vivo prior to
transplantation into an injured myocardium may be more effective for the treatment of
cardiovascular disease.
Fibrosis and ventricular remodeling following a myocardial infarction begins with elevated
extracellular matrix (ECM) deposition, which stiffens the myocardium and inadvertently
contributes to ventricular dysfunction. Notwithstanding, ECMs reportedly influence critical
cellular processes such as survival, proliferation and differentiation in many cell types via the
engagement of specific integrins. However, it is not well understood if ECMs exert a
significant influence on the proliferation and cardiac transdifferentiation of primitive
mesenchymal stem cells. Additionally, the effect/s of myocardial fibrosis and post-infarct
remodeling on stem cell differentiation in vivo is not well studied.
This project has 2 specific objectives:
1. To explore the role/s of extracellular matrices and their integrin partners on the cell
fate and development of CLCs vs, MSCs in vitro and in vivo and
2. To compare the relative therapeutic efficacies of the 2 cell types
Adult human bone marrow mesenchymal stem cells (MSCs) can differentiate into
cardiomyocyte-like cells (CLCs) with the concomitant use of collagen V extracellular
matrices and a simple non-toxic culture medium in vitro. Importantly, this distinct cardiac-
like phenotype is stable in prolonged cultures. In contrast, MSCs exhibited a spontaneous but
transient expression of cardiac-specific proteins.
Objective 1: Cell-ECM interactions are mediated via the engagement of integrins, which in turn activates a
cascade of downstream intracellular signaling events that lead to the expression of multiple
proteins among other biological processes. CLCs demonstrated preferential interaction with
specific collagen subtypes. Specifically, Collagen -V, but not collagen -I, promoted the
cellular adhesion and cardiac differentiation of MSC-derived CLCs. More remarkably,
collagen V matrices promoted the large-scale production of CLCs that was valuable for
subsequent transplantation therapies. Initial cellular adhesion to collagen V but not collagen I,
was dependent on the 21 integrin but independent of the v3 and v integrins. However,
inhibition of v3 integrin, but not 21 integrin reduced gene expression levels of troponin T,
sarcomeric -actin and RyR2 in CLCs cultured on collagen V ECM. Importantly, the
engraftment of CLCs within close proximity of collagen V-expressing myofibers promoted
their integration into the cardiac syncytium. More remarkably, CLCs demonstrated distinct
striations that were indistinguishable from host cardiomyocytes in collagen V-enriched areas
in the infarcted myocardium, while CLCs that engrafted in collagen I-enriched areas in the
infarct borders did not. Thus, collagens -I and -V may play pivotal roles in the cell fate
development of CLCs in vivo, although it remains to be elucidated if the colocalization of
CLCs with the collagen V-enriched, endomysial-lined myofibers correlates with a specific
interaction between the endogenous ECM and the transplanted cells, that is reminiscent of
their affinity in vitro. It is also unclear if the localization of CLCs in the interstitial tissues
enriched in collagens -I and -III prevented their integration into the host myofibrillar
architecture. Notwithstanding, significant improvements in cardiac function were observed in
rats administered with low-dose CLC therapy despite low incidences of cell integration in the
ii
host myocardium. However, it is highly unlikely that such remarkable benefits were attributed
to myocyte replacement. Instead, the introduction of an exogenous supply of viable cells may
possibly improve cardiac function by modulating the ECM architecture in vivo to retain a
certain degree of pliancy in the post-infarcted myocardium, thus reducing overall tissue
stiffness in the compromised myocardium. Hence, CLCs may facilitate functional recovery
by preserving tissue compliance in the peri-infarct borders, which in turn sustains the
contractile efficiency for long-term functional recovery in the infarcted myocardium.
Objective 2: CLC-therapy was more effective when administered in higher doses as demonstrated by
increasingly evident improvements in cardiac function. Additionally, high- dose CLC therapy
resulted in enhanced cell integration with the host myocardium. Notwithstanding similar
trends in functional improvements observed in both low- and high-dose cell therapy groups,
the high-dose therapy groups were relatively better reflections of the direct cellular effects of
cell- transplantation on the infarcted myocardium. Correspondingly, cell-treated rats exhibited
smaller cardiac volumes and LV internal diameters. Cell therapy generally improves the
injured myocardium by restraining progressive wall thinning and ventricular dilatation. Thus,
cell-therapy alleviated adverse remodeling effects, possibly by sustaining myocardial tissue
compliance. However, CLCs were more effective than MSCs in improving cardiac function
as CLC-treated rats demonstrated persistently superior systolic activities with respect to
control and in particular, MSC-treated rats. Echocardiography assessments showed that high-
dose CLC-therapy mediated a significant 9.9 ± 12.1% improvement in LV fractional
shortening (FS) as compared to a decrease of 14.4 ± 13.6% in control rats (p<0.001 vs.
control). Similarly, CLC-treated rats showed a 7.14 ± 8.39 % improvement in ejection
fraction as compared a deterioration of -11.3 ± 11.4 % in control rats (p< 0.001). In contrast,
MSC -therapy appeared only to prevent further deterioration in LVFS (LVFS = -0.1 ±
10.1%) and EF (LVEF = -0.76 ± 7.1 %); p< 0.005 vs. control, p< 0.05 vs. CLC-treated rats.
Additionally, only CLC-treated rats (-13.8 ± 23.9%) demonstrated a significant improvement
in the heart rate- independent myocardial performance index with respect to control rats (10.2
± 18.3%). More remarkably, PV catheterization shows that CLC-therapy restores myocardial
systolic performance as end-systolic pressure-volume relationships (2.10 ± 0.889mmHg/l) in
CLC-treated rats reverted to baseline levels (all pairwise comparisons, p < 0.05). MSC-treated
rats consistently exhibited significant cardiac relief with respect to control rats. These
functional improvements may be attributed to possible cardioprotective paracrine-mediated
effects, as the MSC-treated myocardium persistently demonstrated enhanced angiogenesis in
the infarcted myocardium with respect to sham-operated control myocardium. In contrast,
engrafted CLCs exhibited mature cross-striated fibers in vivo that aligned with and were
indistinguishable from native cardiac myofibers in the myocardium. This further translated
into a superior regional and global contractility to that in MSC-treated rats. A significant
increase in neoangiogenesis in regions proximal to the sites of cell engraftment relative to
control myocardium indicates that CLCs may also confer a paracrine-mediated
cardioprotective influence on the compromised myocardium in vivo. Thus, CLCs may confer
cardiac relief via distinct myogenic and non-myogenic repair mechanisms. These coexistent
mechanisms may bring about a synergistic improvement in cardiac function, which may
further explain CLCs‟ exceptional recovery of contractile performance in the infarcted
myocardium. To date, most clinical studies have employed the use of undifferentiated bone
marrow stem cell therapy with variable outcomes and modest efficacies. This study shows
that the pre-differentiation of MSCs into a stable and committed cardiac lineage prior to
transplantation is a more effective and efficacious treatment for cardiovascular disease.
iii
Publications
1. Genevieve Tan, Yingying Chung, Sze Yun Lim, Pearly Yong, Ling Qian,
Eugene Sim, Philip Wong and Winston Shim. Myocardial Matrix-driven Cardiac
Differentiation and Integration of Human Mesenchymal Stem Cells [Manuscript
submitted to Stem Cells and Development].
2. Genevieve MY Tan, Jack WC Tan, Yee Jim Loh, Terrance Chua, Tian Hai Koh,
Yong Seng Tan, Yoong Kong Sin, Chong-Hee Lim, Eugene KW Sim, Philip EH
Wong and Winston SN Shim. Collagen V Extracellular Matrices Promotes the
Large Scale Expansion of Human Bone Marrow Mesenchymal Stem Cell-
Derived Cardiomyocyte-Like Cells [Manuscript submitted to Asian
Cardiovascular and Thoracic Annals].
3. Genevieve MY Tan, Yingying Chung, Yacui Gu, Shiqi Li, Ling Qian, Yee Jim
Loh, Jack Tan, Terrance Chua, Tian Hai Koh, Yeow Leng Chua, Yong Send Tan,
Chong Hee Lim, Yoong Kong Sin, Eugene Sim, Philip Wong and Winston Shim.
Predifferentiating Bone Marrow-Derived Mesenchymal Stem Cells into
Cariomyocyte-like Cells Significantly Improves the Efficiency of Myocardial
Repair [Manuscript in preparation]. (Presented at the 17th
ASEAN Congress of
Cardiology, Young Investigator‟s Award)
Book Chapters
4. Genevieve M.Y. Tan, Lei Ye, Winston S.N. Shim, Husnain Kh, Haider, Alexis
B.C. Heng, Terrance Chua, Tian Hai Koh and Eugene K.W. Sim. (2007). Tissue
Engineering for the Infarcted Heart: Cell Transplantation Therapy. In: Dhanjoo N
Ghista & Eddie Yin-Kwee Ng (Eds.) Cardiac Perfusion and Pumping
Engineering. Singapore. World Scientific Publishers. 477-540.
5. Genevieve M. Y. Tan, Lay Poh Tan, N.N. Quang, Winston S.N. Shim, Alfred
Chia, Subbu V. Venkatramen and Philip E. H. Wong. (2007). Tissue Engineering
of Artificial Heart Tissue. In: Dhanjoo N Ghista & Eddie Yin-Kwee Ng (Eds.)
Cardiac Perfusion and Pumping Engineering. Singapore. World Scientific
Publishers. 541-578.
iv
Awards/ Honors/ Recognition October 2008
17th
ASEAN Congress of Cardiology, Young Investigator’s Award, Awarded 1st
place for the abstract entitled, “Predifferentiating Bone Marrow-Derived
Mesenchymal Stem Cells into Cariomyocyte-like Cells Significantly Improves the
Efficiency of Myocardial Repair”
April 2007
SGH 16th
Annual Scientific Meeting, Awarded Best Oral Paper (Scientist) for the
abstract entitled, “Human Bone Marrow- Derived Cardiomyocyte-Like Cells Improve
Cardiac Performance In The Infarcted Myocardium"
March 2007
Singapore Cardiac Society 19th
Annual Scientific Meeting, Young Investigator’s
Award, Awarded 1st place for the abstract entitled, "Human Bone Marrow-Derived
Cardiomyocyte-like cells improve left ventricular remodelling and contractile
function in the infarcted myocardium"
March 2006
Singapore Cardiac Society 18th
Annual Scientific Meeting, Young Investigator’s
Award, Awarded 2nd
place for the abstract entitled, “Large Scale Expansion of
Cardiomyocytye-Like Cells for Cell Transplantation Therapy”
November 2005
American Heart Association Scientific Session 2005, Presenting author for the
abstract, entitled, “Large Scale Expansion of Human Cardiomyocyte-Like Cells for
Cell Therapy” and 1 of 5 poster finalists in the Basic Science category of “Stem/
Progenitor Cells in Cardiac Repair”
Conference papers 1. W. Shim, G. Tan, Y.L Chua, Y.S Tan, Y.K. Sin, C.H. Lim, J. Tan, P. Wong
(2006). Scale-up production of human cardiomyocyte-like cells for cell therapy.
European Heart Journal 28 Suppl 1:548
2. Winston Shim, Genevieve Tan, Eugene Sim and Philip Wong (2005). Large
Scale Expansion of Cardiomyocyte-like Cells for Cell Therapy. Circulation
112(17): (Suppl II) II-14.
3. Winston Shim, Genevieve Tan and Philip Wong (2005). Cardiac Differentiated
Adult Human Bone Marrow Stem Cells Express Sarcomeric and Structural
Proteins of Cardiomyocytes. Microscopy and Microanalysis 11 (Suppl 1):140.
4. Winston Shim, Genevieve Tan, Eugene Sim and Philip Wong (2005). Collagen V
Matrix Supports Proliferation and Differentiation of Cardiomyocyte-like Cells
Derived from Adult Human Bone Marrow. Cytotherapy 7 (Suppl 1):194.
5. Genevieve Tan, Philip Wong, Terrance Chua, Te Chih Liu, Ming Teh, Eugene
Sim and Winston Shim (2004) Directed Differentiation of Adult Human Bone
Marrow Mesenchymal Stem Cells towards Cardiomyocytes. Annals Academy of
Medicine Singapore 33(5): S182.
6. Genevieve Tan, Philip Wong, Terrance Chua, Jack Tan, Yeow Leng Chua, Yong
Seng Tan, Yoong Kong Sin, Chong Hee Lim, Te Chih Liu, Ming Teh, Eugene
v
Sim and Winston Shim (2004) ECM-dependent proliferation of Adult Bone
Marrow Mesenchymal Stem Cells. Proceedings of the 1st International
BioEngineering Conference (ISBN: 981-05-1946-X), BioEngineering: Challenges
and Innovations, pp49-50, 8 – 10 September 2004, Singapore
7. Genevieve Tan, Philip Wong, Terrance Chua, Jack Tan, Yeow Leng Chua, Yong
Seng Tan, Yoong Kong Sin, Chong Hee Lim, Te Chih Liu, Ming Teh, Eugene
Sim and Winston Shim (2004) Cardiac Differentiation of Adult Bone Marrow
Mesenchymal Stem Cells. Proceedings of the 1st International BioEngineering
Conference (ISBN: 981-05-1946-X), BioEngineering: Challenges and
Innovations, pp30-32, 8 – 10 September 2004, Singapore.
8. Tan GMY, Wong P, Law ACS and Shim WSN (2005) Adult Bone Marrow
Mesenchymal Stem Cells for Cardiac Tissue Engineering. 7th
Annual NTU-SGH
Symposium (ISBN: 981-05-3996-7), Moving Technology Towards Better Patient
Care, pp 9-12, 11-12 August 2005, Singapore.
Abstract Presentations
Oral Presentations
1. Genevieve Tan, Yingying Chung, Yacui Gu, Shi Qi Li, Ling Qian, Yee Jim Loh,
Jack Tan, Terrance Chua, Tian Hai Koh, Yeow Leng Chua, Yong Seng Tan,
Chong Hee Lim, Yoong Kong Sin, Eugene Sim, Philip Wong and Winston Shim.
Predifferentiating Bone Marrow-Derived Mesenchymal Stem Cells into
Cariomyocyte-like Cells Significantly Improves the Efficiency of Myocardial
Repair. (Young Investigator’s Award, 1st prize). 17
th ASEAN Congress of
Cardiology, 18-21 October, Hanoi, Vietnam.
2. Genevieve Tan, Yingying Chung, Jack Tan, Yee Jim Loh, Terrance Chua, Yeow
Leng Chua, Chong Hee Lim, Yoong Kong Sin, Seng Chye Chuah, Tian Hai Koh,
Eugene Sim, Philip Wong and Winston Shim. Collagen V Matrix Supports
Differentiation of Human Bone Marrow Stem Cells Towards Cardiomyocytes. 4th
International Cardiac Bio-Assist Association Congress, 12-13 March 2008,
Singapore.
3. Genevieve Tan, Ling Qian, Sze Yun Lim, Yacui Gu, Shi Qi Li, Jack Tan, Yeow
Leng Chua, Chong Hee Lim, Yoong Kong Sin, Terrance Chua, Tian Hai Koh,
Eugene Sim, Philip Wong and Winston Shim. Cardiac Differentiated
Mesenchymal Stem Cells Protect Against Diastolic Dysfunction and Negative
Post-Infarct Remodelling. 4th
International Cardiac Bio-Assist Association
Congress, 12-13 March 2008, Singapore.
4. Genevieve MY Tan, Ling Qian, Ying Ying Chung, Yacui Gu, Shi Qi Li, Yee
Jim Loh, Jack Tan, Terrance SJ Chua, Yeow Leng Chua, Yong Seng Tan, Chong
Hee Lim, Kenny YK Sin, Eugene KW Sim, Philip EH Wong and Winston SN
Shim. Human Bone Marrow-Derived Cardiomyocyte-Like Cells Improve Cardiac
Performance in the Infarcted Myocardium. (Best Oral Paper (Scientist), 1st
prize). 16th
SGH Annual Scientific Meeting incorporating 14th SGH-Stanford
Annual Joint Update and Annual Evidence-Based Medicine Seminar, 27-28 April
2007, Singapore
vi
5. Genevieve MY Tan, Ling Qian, Ying Ying Chung, Yacui Gu, Shi Qi Li, Yee
Jim Loh, Jack Tan, Terrance SJ Chua, Yeow Leng Chua, Yong Seng Tan, Chong
Hee Lim, Kenny YK Sin, Eugene KW Sim, Philip EH Wong and Winston SN
Shim (2007). Human Bone-Marrow-Derived Cardiomyocyte-Like Cells Improve
Left Ventricular Remodelling and Contractile Function in the Infarcted
Myocardium. (Young Investigator’s Award, 1st prize). The 19
th Singapore
Cardiac Society Annual Scientific Meeting: Cardiovascular Disease: The
Metabolic Age. 17-18 March 1007, Singapore.
6. Genevieve Tan, Philip Wong, Eugene Sim, Jack Tan, Terrance Chua, Yeow
Leng Chua, Yong Seng Tan, Chong Hee Lim, Yoong Kong Sin and Winston
Shim (2006). Large-Scale Expansion of Cardiomyocyte-like Cells for Cell
Transplantation Therapy (Young Investigator’s Award, 2nd
Prize). The 18th
Singapore Cardiac Society Annual Scientific Meeting: The Growing Burden of
Cardiovascular Disease in the Aging Population. 25 – 26 March 2006, Singapore.
7. Genevieve Tan, Philip Wong, Anthony Law and Winston Shim (2005). Adult
Bone Marrow Mesenchymal Stem Cells For Cardiac Tissue Engineering. The 7th
Annual NTU-SGH Symposium: Moving Technology Towards Better Patient
Care. 11 – 12 August 2005, Singapore.
8. Genevieve Tan, Eugene Sim, Terrance Chua, Jack Tan, Yeow Leng Chua, Yong
Seng Tan, Yoong Kong Sin, Chong Hwee Lim, Te Chih Liu, Ming Teh, Eugene
Sim and Winston Shim (2005). Extracellular Matrices For Proliferating
Cardiomyogenic Adult Bone Marrow Mesenchymal Stem Cells. The 17th
Singapore Cardiac Society Annual Scientific Meeting; Cardiology: From
Beginning to the End. 26 – 27 March 2005, Singapore.
9. Tan GMY, Shim WSN, Wong P, Tan Jack, Chua T, Liu TC, Aye WMM and Sim
EKW, Adult Bone Marrow Mesenchymal Stem cells for Cardiomyogenesis, 16th
Annual Scientific Meeting, Singapore Cardiac Society, 2004
10. Genevieve Tan, Philip Wong, Terrance Chua, Jack Tan, Yeow Leng Chua, Yong
Seng Tan, Yoong Kong Sin, Chong Hee Lim, Te Chih Liu, Ming Teh, Eugene
Sim and Winston Shim, ECM-dependent proliferation of Adult Bone Marrow
Mesenchymal Stem Cells, 1st International BioEngineering Conference in
Conjunction with the 6th
Annual NTU-SGH Biomedical Engineering Symposium,
September 2004
11. Genevieve Tan, Philip Wong, Terrance Chua, Jack Tan, Yeow Leng Chua, Yong
Seng Tan, Yoong Kong Sin, Chong Hee Lim, Te Chih Liu, Ming Teh, Eugene
Sim and Winston Shim, Cardiac Differentiation of Adult Bone Marrow
Mesenchymal Stem Cells, 1st International BioEngineering Conference in
Conjunction with the 6th
Annual NTU-SGH Biomedical Engineering Symposium,
September 2004
12. Tan GMY, Sim EKW, Chua TSJ, Wong P, Tan J, Chua YL, Tan YS, Sin YK,
Lim CH and Shim WSN, Extracellular Matrices for proliferating cardiomyogenic
adult bone marrow mesenchymal stem cells, 17th
Annual Scientific Meeting,
Singapore Cardiac Society, 2005
vii
Poster Presentations
1. Genevieve Tan, Yingying Chung, Sze Yun Lim, Ling Qian, Yacui Gu, Shiqi Li,
Yee Jim Loh, Terrance Chua, Yeow Leng Chua, Chong Hee Lim, Yoong Kong
Sin, Tian Hai Koh, Eugene Sim4, Philip Wong and Winston Shim. Integration of
Transplanted Human Cardiomyocyte-like Cells in Infarcted Myocardium. (Best
Poster (Scientist), 1st prize), 17th SGH Annual Scientific Meeting, incorporating
Annual Evidence-Based Medicine Seminar, 25-26 April 2008, Singapore.
2. Genevieve Tan, Ling Qian, Shi Qi Li, Yacui Gu, Yee Jim Loh, Yeow Leng Chua,
Chong Hee Lim, Yoong Kong Sin, Terrance Chua, Tian Hai Koh, Eugene Sim,
Philip Wong, Winston Shim. Cardiac Differentiated Mesenchymal Stem Cells
Protect Against Diastolic Dysfunction by Preventing Post-Infarct Remodeling.
American College of Cardiology, 57th Annual Scientific Session, 29 March – 1
April 2008, Chicago, Illinois, USA.
3. Genevieve Tan, Ling Qian, Yacui Gu, Shiqi Li, Yee Jim Loh, Terrance Chua,
Yeow Leng Chua, Chong Hee Lim, Yoong Kong Sin, Seng Chye Chuah, Tian Hai
Koh, Eugene Sim, Philip Wong and Winston Shim. Post-Infarct Myocardial
Function Recovery Is Preserved by Stabilizing Left Ventricular Negative
Remodeling by Cardiac Differentiated Stem Cells but not Undifferentiated Stem
Cells, Cardiovascular Research Therapies, 11-13 February 2008, Washington
D.C.
4. Gu Yacui, Winston Shim, Li Shiqi, Genevieve Tan, Qian Ling, Tan Ru San,
Philip Wong. Usefulness of tissue Doppler imaging for quantifying regional
myocardial function in a rat heart infarct model. 12th Asian Pacific Congress
Doppler & Echocardiography. 28-30 October 2007
5. Gu Yacui, Winston Shim, Li Shiqi, Genevieve Tan, Tan Ru San, Philip Wong.
Usefulness of tissue Doppler imaging for quantifying regional myocardial
function in a rat heart infarct model. SGH 16th
Annual Scientific Meeting. 27-28
April 2007
6. Winston Shim, Genevieve Tan, Shiqi Li, Hwee Choo Ong, In Chin Song, Eugene
Sim and Philip Wong. Scale-up production of human cardiomyocyte-like cells for
cell therapy, World Congress of Cardiology, 2006. 2-6 September 2006,
Barcelona, Spain.
7. Winston Shim, Genevieve Tan, Shiqi Li, Hwee Choo Ong, In Chin Song, Eugene
Sim and Philip Wong. Collagen Matrix Supports Differentiation of Human Bone
Marrow Stem Cells Towards Cardiomyocytes, 8th International Congress of the
Cell Transplant Society 2006
8. Winston Shim, Genevieve Tan, Shi Qi Li, Hwee Choo Ong, In Chin Song,
Eugene Sim and Philip Wong (2006). Collagen Matrix Supports Differentiation of
Human Bone Marrow Stem Cells Towards Cardiomyocytes. The 15th
Singapore
General Hospital Annual Scientific Meeting: Blending Borders, Merging Science,
Healthcare and Education. 21 – 22 April 2006, Singapore.
9. Winston Shim, Genevieve Tan, Eugene Sim and Philip Wong (2005). Large
Scale Expansion of Cardiomyocyte-like Cells for Cell Therapy. Presenting author
at the American Heart Association Scientific Sessions 2005. Poster finalist in the
Basic Science category, "Stem/Progenitor Cells in Cardiac Repair".
viii
10. Genevieve MY Tan, Eugene KW Sim, Terrance Chua, Philip EH Wong, and
Winston Shim, Extracellular Matrices For Proliferating Cardiomyogenic Adult
Bone Marrow Mesenchymal Stem Cells, 14th
Singapore LIVE 2005.
11. Genevieve Tan, Philip Wong, Terrance Chua, Te Chih Liu, Ming Teh, Eugene
Sim and Winston Shim, Directed Differentiation of Adult Human Bone Marrow
Mesenchymal Stem Cells towards Cardiomyocytes, NHG Annual Scientific
Congress 2004
12. Tan GMY, Wong P, Chua T, Tan J, Chua YL, Tan YS, Sin YK, Lim CH, Liu
TC, Teh M, Sim EKW and Shim WSN, Cardiac Differentiation of Adult Bone
Marrow Mesenchymal Stem cells, SingHealth Scientific Meeting 2004.
ix
Acknowledgements
My senior in the university once said to me that it takes perseverance, and not so
much ingenuity, to survive a PhD experience. As a fresh graduate brimming with
idealism, I did not believe him entirely until 6 years later, when I finally realized the
wisdom behind his words. My journey towards a PhD was fraught, intense but
fulfilling. I was gratified to be entrusted with full independence to address the
scientific challenges presented with each seeming deadlock. Just as no man is an
island, so no project, especially one of this magnitude, could have been brought to
fruition without the integral efforts of key role players. I would like to thank my
supervisors A/P Eugene Sim, Dr. Winston Shim and Dr. Philip Wong for their
valuable insights and the cherished opportunities to be a part of the pioneering efforts
at the National Heart Centre, Singapore to bring adult stem cell therapy from the
bench to the bedside. I am also grateful to Dr. Ratha Mahendran for her continued
guidance throughout my candidature and Dr. Ye Lei for taking time to evaluate this
thesis. Special thanks also to Dr. Li Shiqi, the resident animal surgeon; Dr. Gu Yacui,
our ultrasound sonographer; Dr. Jason Villiano and Ms. Cindy Phua of the
Department of Experimental Surgery, Singapore General Hospital, for their
dedication to the care and well being of the rats used in this study; Dr. Jack Tan, Dr.
Loh Yee Jim and the team of cardiothoracic surgeons at the National Heart Centre,
without whose efforts this project would never have taken flight. I would also like to
express my appreciation to the research team at the Stem Cell Laboratory, especially
Ms. Chung Yingying and Ms. Lim Sze Yun for their kind assistance in the massive
number of histological examinations of frozen tissue cryosections and microvessel
counting that was necessary in this study; and Ms. Pearly Yong of the Flow
Cytometry facility at the Research Department, National Heart Centre for her patient
assistance in flow cytometry experimentation.
x
Abbreviations
AMI Acute Myocardial Infarction
ANF Atrial Natriuretic Factor
ANOVA Analysis of Variance
AWT Anterior Wall Thickening
BNP Brain Natriuretic Peptide
C1/Col. I Collagen I
C43/cxn43 Connexin 43
C5/Col. V Collagen V
CABG Coronary artery bypass grafting
CFDA-SE Carboxy-fluorescein diacetate- succinimidyl ester
CLCs Cardiomyocyte-like cells
CM-DiI Chloromethylbenzamido 1,1'-dioctadecyl-3,3,3'3'
Tetramethylindocarbocyanine
Ctrl Control
DAPI 4',6-diamidino-2-phenylindole
CVD Cardiovascular Disease
dP/dt differential changes in pressure with time
dP/dtmax Maximum dP/dt
ECM Extracellular Matrix
ED End-diastole
EDPVR End diastolic Pressure-Volume relationships
EDV End Diastolic Volume
EF Ejection Fraction
Emax Time Varying Maximal Elastance
ES End-systole
ESPVR End systolic Pressure-Volume Relationships
ESV End Systolic Volume
FN Fibronectin
FS Fractional Shortening
IP3R Inositol-triphosphate receptor
IVC Inferior Vena Cava
IVS Interventricular Septum
LAD Left Anterior Descending
LV Left Ventricle
LVID Left Ventricular Internal Diameter
MEF Myocyte Enhancer Factor
MHC Myosin heavy chain
MI Myocardial Infarctions
MLC Myosin light chain
xi
MPI Myocardial Performance Index (aka Tei Index)
MSC Mesenchymal stem cells
PCNA Proliferating Cell Nuclear Antigen
PCR Polymerase Chain Reaction
PRSW Preload Recruitable Stroke Work
PV Pressure-Volume
RT-PCR Reverse Transcription-PCR
RyR Ryanodine Receptor
Sk. M Skeletal muscle
SMA Smooth Muscle Actin
SWT Systolic Wall Thickening
TGN trans Golgi Network
TnC Troponin C
TnT Troponin T
Tukey‟s HSD Tukey‟s Honest Significant Difference
UN Uncoated
Vcfc Circumferential Fibre shortening velocity
VEGF Vascular endothelial growth factor
VEGFR Vascular endothelial growth factor receptor
vWF von Willebrand Factor
xii
Table of Contents Publications ............................................................................................................ iii
Book Chapters ........................................................................................................ iii
Awards/ Honors/ Recognition ................................................................................ iv
Conference papers .................................................................................................. iv
Abstract Presentations............................................................................................. v Oral Presentations .................................................................................................. v
Poster Presentations ............................................................................................ vii
Chapter One: Introduction .................................................................. 1 1.1 Atherosclerosis ................................................................................................... 1
1.2 Ischemic Cardiomyopathy ................................................................................ 2
1.3 Cardiovascular Heart Failure ........................................................................... 2
1.3.1 Myocardial systolic dysfunction ................................................................... 2
1.3.2 Myocardial diastolic dysfunction .................................................................. 3
1.3.3 Ventricular Remodeling ................................................................................ 4
1.4 Cardiac myofibrillogenesis ................................................................................ 4
1.5 Novel therapy ..................................................................................................... 6
1.5.1 Stem cells ...................................................................................................... 7
1.5.1.1 Embryonic stem cells ............................................................................. 7
1.5.1.2 ES cell stem therapy is not ideal for clinical use ................................... 9
1.5.2 Adult stem cells........................................................................................... 10
1.5.2.1 Bone marrow stem cells ....................................................................... 10
1.5.2.1.1 Hematopoietic stem cells .............................................................. 11
1.5.2.1.2 Mesenchymal stem cells ............................................................... 11
1.5.3 Bone marrow stem cell-transplantation therapy ........................................ 13
1.6 Skeletal myoblasts ............................................................................................ 15
1.7 Clinical trials in BM stem cell transplantation for treatment of MI ........... 17
1.7.1 Clinical outcomes........................................................................................ 17
1.7.1.1 Bone marrow transfer to enhance ST-elevation infarct regeneration
(BOOST) .......................................................................................................... 17
1.7.1.2 Autologous Stem Cell Transplantation in Acute Myocardial Infarction
(ASTAMI)........................................................................................................ 18
1.7.1.3 LEUVEN-AMI .................................................................................... 18
1.7.1.4 Reinfusion of Enriched Progenitor cells and Infarct Remodeling in
Acute Myocardial Infarction (REPAIR-AMI) ................................................. 18
1.7.2 Problems encountered ................................................................................. 19
1.8 Extracellular matrices ..................................................................................... 22
1.8.1 Collagens..................................................................................................... 23
1.8.2 Integrins and Integrin- mediated Cell Signaling ......................................... 24
1.8.2.1 Integrins ............................................................................................... 24
1.8.2.2 Focal Adhesions ................................................................................... 26
1.9 Cardiac Tissue Engineering ............................................................................ 28
xiii
1.10 Significance of study ...................................................................................... 30
1.11 Specific Aims .................................................................................................. 31
Chapter Two: Materials and Methods .............................................. 32
2.1 Recruitment of patients ................................................................................... 32 2.1.1 Isolation and Cell culture ............................................................................ 32
2.2 Isolation and culture of neonatal rat cardiomyocytes .................................. 33
2.3 Flow Cytometry ................................................................................................ 33
2.4 Gene expression profiling via RT-PCR analyses .......................................... 33
2.5 Cell proliferation assays .................................................................................. 35
2.6 Integrin inhibition assays ................................................................................ 35
2.7 Indirect Immunofluorescence Microscopy .................................................... 36 2.7.1 Detection of collagen-coated tissue culture coverslips ............................... 36
2.7.2 Detection of cardiac myofibrillar proteins .................................................. 37
2.8 Golgi Disruption ............................................................................................... 38
2.9 Cell-labeling and detection .............................................................................. 38
2.10 Rat myocardial infarction models ................................................................ 39 2.10.1 Ultrasound echocardiography assessments ............................................... 40
2.10. 2 In vivo hemodynamic measurements ....................................................... 40
2.10.3 Detection of human nuclei ........................................................................ 41
2.10.4 Fluorescence microscopy of frozen tissue sections .................................. 42
2.10.5 Microvessel density count......................................................................... 42
2.10.6 Statistical analysis ..................................................................................... 43
Chapter Three: Results ....................................................................... 44 3.1 Isolating Sternum-Derived Mesenchymal Stem Cells .................................. 46
3.1.1 Isolation of Bone Marrow Mesenchymal Stem Cells (MSCs) ................... 46
3.1.2 Patient-derived cells can differentiate into osteoblasts and adipocytes ...... 46
3.2 Developing a myogenic development medium .............................................. 49
3.3 In vitro characterization of MSCs and MSC-derived CLCs ........................ 51 3.3.1 Golgi-Localization of GATA4 and b Myosin Heavy Chain ................... 58
3.4 The roles of extracellular matrices on the cell fate and development of
CLCs........................................................................................................................ 65 3.4.1 Collagen V enhances cellular adhesion and proliferation of CLCs ............ 66
3.4.2 CLCs demonstrated enhanced cardiac differentiation on Collagen V
matrices ................................................................................................................ 71
3.4.3 Study of relevant integrin signaling ............................................................ 74
3.5 In vivo functional studies ................................................................................. 82 3.5.1 BrdU labeling of cells in the low-dose therapy group ................................ 96
3.5.2 Cell fate and development of BrdU labeled cells in vivo............................ 96
3.5.3 High-Dose Cell Therapy ............................................................................. 98
3.5.3.1 Fluorescent Labeling of cells in the high-dose therapy groups ........... 98
3.5.3.2 2D M-mode ultrasound echocardiography assessments and Tissue
Doppler Imaging of Cardiac Function ........................................................... 100
3.5.3.3 In vivo Hemodynamics via Pressure-Volume Catheterization .......... 107
3.5.4 Dose-dependent effects of cell therapy on cardiac function ..................... 112
3.5.5 Cell-mediated Cardiac Repair Mechanisms In vivo. ............................... 119 3.5.5.1 Donor cell engraftment and survival ...................................................... 119
xiv
3.5.5.2 CLCs but not MSCs enhanced cardiac contractility via myocyte-
replacement ........................................................................................................ 119
3.5.5.3 Cell therapy mediates cardiac repair via alternative non-myogenic
mechansims in the infarcted myocardium ......................................................... 127
3.5.6 The role of collagen in the cell fate and development of CLCs in vivo .. 134
3.5.7 Key findings ................................................................................................. 138
Chapter 4: Discussion ....................................................................... 140 4.1 In vitro characterization studies ....................................................................... 140
4.1.1 MSC commitment to a distinct cardiac lineage before cell transplantation
............................................................................................................................ 140
4.1.2 Cardiac differentiation of MSCs into CLCs ............................................. 140
4.1.3 Expression profiling of CLCs . ................................................................. 141
4.1.4 Golgi-localization of GATA4 and -MHC in patient-derived MSCs and
CLCs. ................................................................................................................. 143
4.1.5 CLCs‟ contractile apparatus resemble primitive premyofibrils in early ... 145
myofibrillogenesis.............................................................................................. 145
4.1.6 Effect of ECM in vitro on CLCs ............................................................... 146
4.2 In vivo functional studies ............................................................................... 150 4.2.1 Optimal time for cell transplantation ........................................................ 150
4.2.2 Dose dependent contribution of cell therapy ............................................ 151
4.2.2.1 Donor cell survival ............................................................................. 151
4.2.2.2 Postulated role of myocardial ECM on donor cell survival ............... 153
4.2.2.3 Cell mediated attenuation of adverse LV remodeling ....................... 155
4.2.2.4 Dose-dependent contribution of cell therapy towards myocardial .... 159
contractility .................................................................................................... 159
4.3 Relative therapeutic efficacies of CLCs vs. MSCs ...................................... 164 4.3.1 Assessment of myocardial systolic activities ............................................ 164
4.3.1.1 Echocardiography assessments .......................................................... 164
4.3.1.2 Real-time pressure-volume catheterization ........................................ 166
4.3.1.2.1 Load-sensitive hemodynamic measurements ............................. 166
4.3.1.2.2 Load-insensitive contractility measures ...................................... 166
4.3.1.3 CLCs contribute actively towards myocardial contractile performance
via myocyte replacement ............................................................................... 169
4.3.1.4 MSCs contribute passively to myocardial systolic activities ............. 170
4.3.2 Assessment of myocardial diastolic activities .......................................... 173
4.3.2.1 Echocardiography assessments .......................................................... 173
4.3.2.2 Conductance catheterization .............................................................. 173
4.4 Cell-mediated paracrine effects .................................................................... 176 4.4.1 MSC-mediated neovascularization in the infarcted myocardium ............. 176
4.4.2 CLCs also elicit alternative non-myogenic mechanisms of cardiac repair
............................................................................................................................ 176
4.5 Postulated role of myocardial ECM on in vivo cell fate and development179
Chapter 5: Conclusion ...................................................................... 182
References .......................................................................................... 185
xv
List of Tables
TABLE 1. CLINICAL OUTCOMES OF CURRENT MAJOR TRIALS ON CARDIAC
REGENERATION FOR TREATMENT OF ACUTE MYOCARDIAL INFARCTION ....... 21
TABLE 2. HUMAN CLCS EXPRESS 2, 1, V, 3, 21, AND V3 INTEGRINS. THE
GENE EXPRESSION LEVELS OF THE V AND 1 INTEGRINS WERE MARKEDLY
HIGHER IN CLCS CULTURED ON COLLAGEN V ECM. ..................................... 78
TABLE 3. MEAN LVEF (%) OF RATS IN THE RESPECTIVE TREATMENT GROUPS AT
BASELINE, POST-LIGATION AND 8 WEEKS POST-THERAPY. .............................. 86
TABLE 4. ULTRASOUND ECHOCARDIOGRAPHY ASSESSMENT OF CELL-TREATED
RATS. .................................................................................................................. 92
TABLE 5.CONTRACTILITY MEASUREMENTS OBTAINED VIA REAL-TIME OCCLUSION
OF THE INFERIOR VENA CAVA AT 6 WEEKS POST-THERAPY IN THE HIGH-DOSE
THERAPY GROUPS. ........................................................................................... 109
TABLE 6. ULTRASOUND ECHOCARDIOGRAPHY ASSESSMENT OF RATS TREATED WITH
HIGH- AND LOW-DOSE CELL THERAPY. ........................................................... 114
TABLE 7. STEADY STATE HEMODYNAMIC MEASUREMENTS VIA REAL-TIME IN VIVO
PRESSURE-VOLUME CATHETHERIZATION SHOWING TRENDS IN DIASTOLIC
ACTIVITIES. ...................................................................................................... 118
xvi
List of Figures
FIGURE 1. EMBRYONIC STEM CELLS ARE TOTIPOTENT AND CAN DIFFERENTIATE
INTO VARIOUS CELL TYPES. ................................................................................. 8
FIGURE 2. BONE MARROW STEM CELLS ARE MULTIPOTENT AND CAN DIFFERENTIATE
INTO CELLS BELONGING TO VARIOUS MESODERMAL LINEAGES. ..................... 12
FIGURE 3. KNOWN -INTEGRIN HETERODIMERS. ................................................. 25
FIGURE 4. SCHEMATIC ILLUSTRATION OF A FOCAL ADHESION COMPLEX FOLLOWING
INTEGRIN RECEPTOR CLUSTERING AT A FOCAL POINT ON THE CELL SURFACE.27
FIGURE 5. SCHEMATIC MAP SHOWING AN OVERVIEW OF HOW THE VARIOUS
PROJECT OBJECTIVES AND SPECIFIC AIMS INTEGRATE WITH ONE ANOTHER AND
COLLECTIVELY FORM THE FRAMEWORK OF THIS STUDY. ............................... 45
FIGURE 6. PATIENT-DERIVED CELLS ISOLATED VIA STANDARD PURIFICATION
TECHNIQUES DEMONSTRATED A LOW EXPRESSION OF CD34, AND WERE
POSITIVE FOR CD44, CD90, CD105 AND CD106, 40X MAGNIFICATION. ........ 47
FIGURE 7. PATIENT-DERIVED BONE MARROW CELLS CAN BE INDUCED TO UNDERGO
OSTEOGENIC AND ADIPOGENIC DIFFERENTIATION. .......................................... 48
FIGURE 8. GENE EXPRESSION PROFILE OF DIFFERENTIATING MESENCHYMAL STEM
CELLS (MSCS) FOLLOWING EXPOSURE TO 5-AZACYTIDINE (AZA), BUTYRIC
ACID (BA) OR PREVIOUSLY DEVELOPED MYOGENIC DEVELOPMENT MEDIUM,
MDM. ................................................................................................................ 50
FIGURE 9. INDUCTION OF MSCS IN MDM COINCIDED WITH A CHANGE IN
MORPHOLOGY FROM SPINDLE TO STAR-LIKE. .................................................. 52
FIGURE 10. DIFFERENTIAL CELL PROLIFERATION RATES IN MSCS AND CLCS.
MDM PROMOTED THE SIGNIFICANT EXPANSION OF DIFFERENTIATING MSCS.
............................................................................................................................ 52
FIGURE 11. GENE EXPRESSION PROFILES OF MSCS AND CLCS AFTER 7 AND 14 DAYS
IN THE RESPECTIVE CULTURE MEDIA. ............................................................... 53
FIGURE 12. SPATIAL EXPRESSION AND FUNCTIONAL ACTIVITY OF CARDIAC
TRANSCRIPTION FACTORS IN CLC CULTURES.................................................. 56
FIGURE 13. CHARACTERIZATION OF PROTEIN EXPRESSION IN CLCS. .................... 61
FIGURE 14. GOLGI-LOCALIZATION OF GATA4 IN MSCS AND CLCS. ................... 62
FIGURE 15 SPATIAL EXPRESSION OF MHC IN MSCS VS. CLCS. ......................... 63
xvii
FIGURE 16. CARDIAC MYOSIN HEAVY CHAIN (MHC) TAKES ON A PARALLEL
SWOLLEN APPEARANCE AS THE GOLGI APPARATUS IS DISPERSED IN MSCS AND
CLCS WITH 500NM MONENSIN. ........................................................................ 64
FIGURE 17. CLCS WERE HARVESTED AND SEEDED ON TISSUE CULTURE PLASTIC
SURFACES PRE-COATED WITH VARIOUS ECM SUBSTRATES. ........................... 68
FIGURE 18. CLCS DEMONSTRATE PREFERENTIAL AFFINITY FOR COLLAGEN V ECM.
............................................................................................................................ 68
FIGURE 19. COLLAGEN V ECM PROMOTES ENHANCED ADHERENT-DEPENDENT
SURVIVAL OF CLCS. .......................................................................................... 69
FIGURE 20. LARGE SCALE EXPANSION OF CLCS ON COLLAGEN V ECM. .............. 69
FIGURE 21. RED FLUORESCENCE VERIFIED THAT THE TISSUE CULTURE PLASTICS
WERE INDEED PRE-COATED WITH COLLAGENS -I AND -V MOLECULES
RESPECTIVELY. .................................................................................................. 70
FIGURE 22. QUANTITATIVE ASSESSMENT OF CARDIAC GENE EXPRESSION LEVELS IN
CLCS EXPANDED ON COLLAGENS -I AND -V SUBSTRATA VIA DENSITOMETRY.70
FIGURE 23. RELATIVE CARDIAC GENE EXPRESSION IN CLCS CULTURED ON
COMPETING COLLAGEN I: COLLAGEN V SUBSTRATA. ..................................... 73
FIGURE 24. STABLE (GENE) EXPRESSION OF CARDIAC SPECIFIC MYOFIBRILLAR
PROTEINS IN PROLONGED CLC CULTURES. ...................................................... 73
FIGURE 25. FLOW CYTOMETRY ANALYSIS OF INTEGRIN EXPRESSION IN MSCS AND
CLCS. ................................................................................................................ 78
FIGURE 26. INITIAL CELLULAR-ATTACHMENT OF CLCS ON COLLAGEN MATRICES IS
MEDIATED VIA ENGAGEMENT OF SPECIFIC INTEGRINS. ................................... 79
FIGURE 27. DOSE-DEPENDENT EFFECT/S OF 21 AND V3 INTEGRIN INHIBITION ON
EXPRESSION LEVELS OF CARDIAC-SPECIFIC PROTEINS IN CLCS THAT WERE
EXPANDED ON COLLAGEN V EXTRACELLULAR MATRICES............................... 80
FIGURE 28. CLCS EXPANDED ON COLLAGEN V EXTRACELLULAR MATRICES
RETAINED EXPRESSION OF SEVERAL MATURE CARDIAC MYOFIBRILLAR
PROTEINS. .......................................................................................................... 81
FIGURE 29. FLOWCHART SUMMARIZING EXPERIMENTAL METHODOLOGY FOR IN
VIVO STUDIES...................................................................................................... 83
FIGURE 30. SURVIVING NUMBER OF RATS IN THE RESPECTIVE CELL THERAPY
GROUPS PRE- AND POST TRANSPLANTATION. .................................................... 84
FIGURE 31. 2-DIMENSIONAL (2D) M-MODE ECHOCARDIOGRAPHICAL ANALYSES
SHOW THE DISEASE PROGRESSION OF A SHAM-OPERATED CONTROL RAT. ..... 85
FIGURE 32. FUNCTIONAL IMPROVEMENT IN CELL TRANSPLANTED MYOCARDIUM.93
xviii
FIGURE 33. EFFICIENCY OF BRDU-LABELING IN MSCS AND CLCS. ...................... 94
FIGURE 34. IDENTIFICATION OF TRANSPLANTED HUMAN STEM CELLS IN RAT
MYOCARDIUM 6 WEEKS POST TRANSPLANTATION. .......................................... 94
FIGURE 35. ENGRAFTMENT OF BRDU-LABELED CLCS IN THE CARDIAC SYNCYTIUM
IN HOST MYOCARDIUM. ..................................................................................... 95
FIGURE 36. ULTRASOUND ECHOCARDIOGRAPHY ASSESSMENTS SHOW THAT CELL-
BASED THERAPY LED TO SIGNIFICANT IMPROVEMENTS IN CARDIAC FUNCTION
AS COMPARED TO CONTROL RATS. .................................................................. 106
FIGURE 37. PRESSURE-VOLUME LOOPS OF A REPRESENTATIVE NORMAL AND
(SERUM-FREE) MEDIUM INJECTED CONTROL RAT POST MYOCARDIAL
INFARCTION (POST-MI) RESPECTIVELY. ........................................................ 109
FIGURE 38. PRESSURE VOLUME RELATIONSHIPS OF A REPRESENTATIVE HEALTHY
RAT VS. A SHAM- OPERATED RAT. ................................................................... 110
FIGURE 39. CLCS PERSISTENTLY MEDIATED SUPERIOR ENHANCEMENT OF PRELOAD
INDEPENDENT CONTRACTILITY INDICES WITH RESPECT TO CONTROL RATS.111
FIGURE 40. 2D M-MODE ECHOCARDIOGRAPHY ASSESSMENTS SHOW THAT HIGH
DOSE CELL THERAPY LED TO SIGNIFICANTLY LV SMALL INTERNAL DIAMETERS.
.......................................................................................................................... 115
FIGURE 41. CELL THERAPY IS GENERALLY MORE EFFECTIVE WHEN ADMINISTERED
IN A HIGHER DOSAGE AS HIGH-DOSE THERAPY EFFECTED PERSISTENTLY
BETTER IMPROVEMENTS IN CARDIAC FUNCTION. ........................................... 117
FIGURE 42. ENGRAFTMENT OF THE RESPECTIVE CELLS TYPES IN A REPRESENTATIVE
RAT IN EACH RESPECTIVE HIGH-DOSE THERAPY GROUP. ............................... 122
FIGURE 43. CELL FATE AND DEVELOPMENT OF DONOR MSCS VS. CLCS IN THE
INJURED MYOCARDIUM. .................................................................................. 123
FIGURE 44. A 3-DIMENSIONAL MODEL OF AN ENGRAFTED CM-DII LABELED CLC
CELL IN THE RAT MYOCARDIUM...................................................................... 125
FIGURE 45. CLCS INTEGRATED IN SITES PROXIMAL TO CONNEXIN 43 WHILE MSCS
DID NOT. ........................................................................................................... 126
FIGURE 46. ENGRAFTED MSCS PERSISTENTLY COLOCALIZED WITH THE VON
WILLEBRAND FACTOR AND SMA IN VIVO. ..................................................... 131
FIGURE 47. ENGRAFTED CLCS MEDIATED CARDIOPROTECTIVE PARACRINE
EFFECTS, WHICH STIMULATED EXTENDED ENDOTHELIAL DIFFERENTIATION
AND ENDOGENOUS CELLULAR PROLIFERATION. ............................................ 132
FIGURE 48. MSC-TREATED RATS SHOW SIGNIFICANTLY HIGHER NUMBERS OF
SMALL ARTERIOLES (<20UM) IN THE MYOCARDIUM WITH RESPECT TO CLC-
TREATED AND CONTROL RATS. ........................................................................ 133
xix
FIGURE 49. SPATIAL EXPRESSION OF ENDOGENOUS COLLAGEN V IN THE PERI-
INFARCTED REGIONS OF THE MYOCARDIUM................................................... 136
FIGURE 50. CLCS (YELLOW ARROWHEADS) PRIMARILY ENGRAFTED IN THE
COLLAGEN V-ENRICHED ENDOMYSIUM THAT ENWRAPS VIABLE MYOFIBERS IN
THE TREATED MYOCARDIUM. .......................................................................... 137
FIGURE 51. CLCS ENGRAFTED IN COLLAGENOUS DOMAINS IN THE MYOCARDIUM.
.......................................................................................................................... 137
FIGURE 52. POSSIBLE MSC- AND CLC- MEDIATED REPAIR MECHANISMS IN THE
COMPROMISED MYOCARDIUM. ........................................................................ 184
1
Chapter One: Introduction
Cardiovascular Disease (CVD) is a leading cause of death in the world. Current trends
in epidemiology and rising incidences of diabetes and obesity among others indicate
that CVD will continue to remain widespread. Transcending geographical and
socioeconomic boundaries and gender differences, CVD was estimated to afflict 80.7
million people in the United States in 20051. In that year, CVD also accounted for an
estimated 17.5 million deaths, and is representative of 30% of all global deaths. Of
these mortalities, 7.6 million were due to heart attack and 5.7 million were attributed
to stroke. Locally, CVD is the cause for 1 in 3 deaths in Singapore or 32.3% of all
deaths in 20072.
1.1 Atherosclerosis
Atherosclerosis is a form of arteriosclerosis and is a condition in which patchy fatty
deposits or atherosclerotic plaques develop within the walls of arteries, leading to
reduced or blocked blood flow. Atherosclerosis is caused by repeated injury to the
arterial wall and many factors such as high blood pressure, smoke, diabetes and high
cholesterol levels in the blood contribute to this injury. Atherosclerosis is one of the
leading causes of illness and death in America as well as most other developed
countries. Approximately 16 million people have atherosclerotic heart disease in
20053. Clinical manifestations of artherosclerosis vary and depend on the location of
the affected artery and whether it is gradually constricted or occluded. Complications
of artherosclerosis include angina, heart attack, abnormal heart rhythms and heart
failure.
2
1.2 Ischemic Cardiomyopathy
Coronary artery disease or ischemic heart disease is the most common underlying
cause of heart failure. An acute myocardial infarction (AMI) is the consequence of a
sudden and complete occlusion of a coronary artery that supplies blood to the
myocardium. This leads to a significant amount of cardiomyocyte death within the
myocardium. Resuscitation of cardiomyocytes is not possible if blood supply is not
restored within minutes. This decreases the bulk of functional cardiomyocytes and
leads to decreased cardiac contractility and an irreversible dilatation of the ventricular
chambers, as the left ventricle undergoes pathological remodeling or enlargement.
1.3 Cardiovascular Heart Failure
Heart failure is a multisystem disorder that is characterized by aberrant cardiac,
skeletal muscle, renal dysfunction and complex neurohormonal changes4. Heart
failure can result from systolic dysfunction or diastolic dysfunction. Heart failure due
to systolic dysfunction can be seen in two thirds of patient population and is caused
primarily by ischemic heart disease. Patients with this type of dysfunction
demonstrate a left ventricular ejection fraction (EF) < 30%. Heart failure resulting
from diastolic dysfunction on the other hand is observed in the remaining one third of
the patient population and is generally brought about by hypertension, ventricular
hypertrophy and possibly diabetes.
1.3.1 Myocardial systolic dysfunction
Systolic dysfunction can be simplistically described as failure of the heart to pump
blood into the circulation. This may be attributed to impairment or loss of cardiac
3
myocytes and/or their molecular components. In congenital diseases such as
Duchenne muscular dystrophy, the molecular structure of individual myocytes is
affected. Myocytes and their components can be damaged by inflammation, as in the
case of myocarditis5. However, the most common mechanism of damage is ischemia
causing infarction and scar formation. After a myocardial infarction, dead myocytes
are replaced by scar tissues that affect the function of the myocardium.
Myocardial systolic dysfunction is characterized by a decrease in ejection fraction as
the strength of ventricular contractions is attenuated and generates an inadequate
stroke volume. This results in an inadequate cardiac output. Ventricular end-diastolic
pressures and volumes are correspondingly increased as the ventricle is inadequately
emptied. This pressure is in turn transmitted to the atrium, whereupon increased
pressure on the left side of the heart is transmitted to the pulmonary vasculature in left
heart failure. The resultant hydrostatic pressure favors extravasation of fluid into the
lung parenchyma, leading to pulmonary edema6. On the other hand, increased
pressure on the right side of the heart is transmitted to the systemic venous circulation
and systemic capillary beds in right heart failure. This in turn leads to fluid
extravasation into the tissues of target organs and results in peripheral edema6.
1.3.2 Myocardial diastolic dysfunction
Heart failure caused by diastolic dysfunction is generally described as the failure of
the ventricle to relax adequately. This generally denotes a stiffer myocardium that
leads to inadequate filling of the ventricle, and further results in an inadequate stroke
volume. Impaired chamber relaxation also results in elevated end-diastolic pressures
and the end result is identical to the case of systolic dysfunction (i.e. pulmonary
edema in left heart failure and/or peripheral edema in right heart failure).
4
1.3.3 Ventricular Remodeling
Cardiac contractility is often impaired following a myocardial infarction.
Neurohormonal activation leads to regional hypertrophy of the infarct zone in a
process known as remodeling. In ventricular remodeling and dysfunction, the
geometric shape of the ventricle is altered and becomes more spherical and dilated7.
This geometric alteration adversely affects hemodynamic function and increases the
risk of ventricular arrhythmias so that sudden cardiac death is seen in 40% to 60% of
persons with systolic dysfunction8. These aberrant physiological effects are associated
with ventricular remodeling and involves neurohormonal release which affect the
structure and metabolism of the heart, increases hypertrophy, and ultimately leads to
chamber dilatation via adverse alterations in preload, afterload, stretch, wall stress,
interstitial collagen deposits and other direct inflammatory and apoptotic effects9.
Adverse consequences of remodeling include increased wall tension, increased
oxygen consumption, decreased subendocardial perfusion and decreased myocyte
shortening9.
1.4 Cardiac myofibrillogenesis
In order to develop efficient therapies for cardiovascular disease, it is important to
understand the basic physiology of muscle development. Myofibrillogenesis
essentially involves the ordered integration of actin, myosin and other accessory
proteins into sarcomeres, the basic functional units responsible for contraction in
mature striated muscle. The incorporation of actin and myosin proteins at episodic
intervals characteristic of sarcomeres requires
specific and dynamic interactions with
other cytoskeletal components10-12. The initial assembly
of myofibrils, Z bodies
composed of -actinin, the N-terminus of titin, nebulin, and telethonin contribute to
5
the polarized organization of the thin actin filaments to form I-Z-I
bands in
developing sarcomeres11, 13-18
. Likewise, M-band proteins such as myomesin, M
protein and the C-terminus of titin play a significant role in incorporating the myosin
thick filaments into regular A bands
19-25. Importantly, the coincident association of the
giant protein titin (3–4 MDa) with thick and thin filaments as well as its persistent
localization to nascent Z and M bands strongly indicates that titin is of central
importance for the coordinated assembly of the key elements of sarcomeres during
early myofibrillogenesis13, 21, 22, 26, 27
.
There are currently two models of sarcomerogenesis. Both models are similar in that
structural proteins are essential for the assembly and incorporation of actin and
myosin into mature myofibrils. In the first, thin and thick filaments
assemble
independently on stress fiber-like structures during early myofibrillogenesis. These
stress fibre-like structures further develop into nonstriated myofibrils, which progress
to nascent striated myofibrils that in turn develop into fully
mature, striated
myofibrils12
. Adjacent strands of thin and thick filaments are initially aligned at the
cell borders and subsequently throughout the cytoplasm during the transition from
nonstriated to striated myofibrils. Thus, the earliest precursors of mature thin and
thick filaments are formed independently within the myoplasm of developing muscle
but proceed to integrate along common filaments during the later stages of myocyte
development.
The second model describes three distinct structures during myocyte development: (i)
premyofibrils28
, (ii) nascent myofibrils and (iii) mature myofibrils. Premyofibrils
contain transitory arrays of I-Z-I complexes consisting of sarcomeric actin occupying
6
primitive I bands that are attached to precursors of Z disks, termed Z bodies, that are
enriched in -actinin and interact with miniature A bands composed of nonmuscle
myosin II. These primitive premyofibrils develop into nascent myofibrils, which
further differentiates into mature myofibrils with the simultaneous replacement of
nonmuscle myosin II with muscle myosin II
29. In this model therefore, the precursors
of thick and thin myofilaments are formed along the same structures and develops
concurrently into mature sarcomeres.
1.5 Novel therapy
Organ transplantation or orthotopic heart transplant is the gold standard of treatment
for heart failure, but remains impractical in reality due to a severe shortage of donor
organs. Successful treatment of cardiovascular disease is therefore restricted in many
situations by the lack of suitable autologous tissue to restore injured cardiac muscle or
tissue.
There is a pressing need to develop innovative and alternative therapies that may
replace irreversibly damaged heart tissue. The emergence of new drugs and
innovative devices have improved the quality of life for many patients diagnosed with
heart failure, however these do not necessarily prolong life expectancy or decrease
mortality and morbidity. End-stage heart failure eventually becomes refractory to
current treatments.
Current treatments for loss or failure of cardiovascular function include organ
transplantation, surgical reconstruction, mechanical assistance via the use of synthetic
devices, or the administration of drugs and other metabolic products. Although
routinely used, these treatments are not without constraints or complications. Most of
7
these modalities do not replace existing damaged areas of the heart but rather work
towards enhancing the viable areas in the heart that are still functional. To date, the
only other therapy that addresses the underlying problem of extensive cardiomyocyte
loss in AMI is cardiac cell transplantation. Recent discoveries in the past decade have
led to the revelation that various stem/progenitor cells may regenerate the
myocardium. These have ignited an explosive implementation of numerous clinical
trials. Most of these studies however have reported variable outcomes and modest
efficacies.
1.5.1 Stem cells
1.5.1.1 Embryonic stem cells
Human embryonic stem (ES) cells are totipotent and are distinguished by their ability
to proliferate in their primitive state while retaining their ability to differentiate into a
myriad of cell types arising from all 3 germ layers (See Figure 1). Derived from the
inner cell mass (ICM) of pre-implantation or peri-implantation embryos in the
blastocyst stage, these cells remain in their native state only when cultured on a feeder
layer of mitotically inactivated mouse embryonic fibroblasts30, 31
. The use of
xenosupport systems has raised concerns of possible inter-species pathogen infection
that may prove problematic in downstream clinical applications. However, original
concerns surrounding the use of such a system have been quelled with the
development of serum-free support systems and human feeder layers.
Differentiation may be induced upon removal from feeder layers30, 31
. Most ES cell
lines will spontaneously differentiate into a vast array of cell lineages upon the
formation of an embryoid body. This includes spontaneously beating cells, which
have since been determined to be cells belonging to a cardiomyogenic lineage.
8
Figure 1. Embryonic stem cells are totipotent and can differentiate into various cell
types.
ES cell-derived embryoid bodies have been reported to develop into early-stage
cardiomyocytes that recapitulate the ontogeny of cardiomyogenesis by sequential
expression of cardiac-specific transcription factors and sarcomeric proteins30, 31
.
Messenger RNAs encoding cardiac transcription factors, GATA-4 and Nkx2.5 are
expressed in EBs before messenger RNAs encoding the atrial natriuretic factor
(ANF), myosin light chain (MLC)-2v, - and ß-myosin heavy chain (-MHC and ß-
MHC respectively), Na+-Ca
2+ exchanger, and phospholamban. Sarcomeric proteins
are also expressed in the following order: titin (Z disk), -actinin, myomesin, titin (M
band), MHC, -actin, cardiac troponin T and M protein
30, 31. Myofibrillogenesis and
organization into sarcomeric structures are subsequently observed that accompany
cardiac-like excitability and contractility with nodal, atrial, and ventricular
characteristics. The nascent myofibrils in these terminally differentiated cells are
sparse and irregularly organized, although parallel bundles of myofibrils demonstrate
evidence of
A and I bands that are characteristic of sarcomeric structures32
.
Additionally, these mature cells reportedly express intercalated disks, fascia adherens,
desmosomes and functional gap junctions. This ES-derived cardiomyocytes are
capable of being electrically coupled with the host myocardium upon their
9
engraftment in the compromised myocardium32, 33
. Furthermore, they form stable
intracardiac grafts for 7 to 32 weeks following their engraftment in the ventricular
wall. This in turn led to significantly reduced left ventricular remodeling and
enhanced cardiac contractile performance in the impaired myocardium.
1.5.1.2 ES cell stem therapy is not ideal for clinical use
The derivation of human ES cell lines and the establishment of a cardiomyocyte
differentiation system are promising for the development of heart regenerative
therapy. ES cells‟ pluripotent nature marks their main attraction as therapeutic agents.
However, this same property confers ES cells with a similar propensity for
tumorigenesis. It was recently reported that as little as 500 mouse or human ES cells
were sufficient to elicit reproducible tumor formations in vivo 34, 35
.
Consistent with a tumor-like phenotype, ES cells express a low level of the human
leukocyte antigen (HLA) class I molecules and do not express class II molecules. Just
like cancer cells, cells with a negligible expression of HLA molecules would be able
to evade the immune system. Thus, ES cells may exhibit immunotolerance following
engraftment in the myocardium. However, while ES cells may be immunoprivileged
in their undifferentiated state, it is likely that these cells may express HLA molecules
upon differentiation. To circumvent this problem, it has been suggested that ES cells
from a bank of only 150 donors with distinct HLA types be used for therapy36
.
However, these 150 cell lines may be inadequate to match the highly variable
genotypes in a multiracial patient population. Moreover, this „solution‟ was conceived
on the assumption that each of the 150 cell lines posses the same inherent
differentiation potential, a concept far from clinical reality.
10
The implantation of undifferentiated ES cells or other inappropriate ES cell-
derivatives may also lead to an inadvertent perturbation of tissue function. Ironically,
it is ES cells‟ immense plasticity coupled with the fact that there are presently no
differentiation techniques that would yield a 100% pure population of mature progeny
that further underscores their limitations as a choice cardiovascular therapeutic agent.
It is not surprising therefore that ES cells are currently restricted from clinical use for
their questionable immunogenicity, innate propensity for teratoma formation and
controversial ethical issues arising from their use.
1.5.2 Adult stem cells
Adult stem cells are found in many organs and tissues and may replenish cells that are
lost during physiological homeostasis. Specific gene expressions are imprinted at the
fetal stage so that their ability to differentiate is restricted to the tissue in which they
reside. However, recent discoveries report that adult stem cells retain a high degree of
developmental plasticity, and can transdifferentiate into brain, gut, lung, liver,
pancreas, kidney and cardiac cells, when placed under specific conditions in vitro and
in vivo 37-44
. This new concept challenges the traditional dogma of adult stem cell
commitment and presents the impetus for a burgeoning interest in adult stem cell
therapeutics. The use of adult bone marrow stem cells in cellular transplantation is
perhaps the most progressive in current cardiovascular research.
1.5.2.1 Bone marrow stem cells
There are 2 most common types of stem cells in bone marrow: (i) hematopoietic stem
cells and (ii) mesenchymal stem cells. These cell types are believed to be ideal for
myocardial transplantation for their ability to develop into cell types representing
11
various mesodermal lineages. These include smooth muscles, angioblasts and cardiac
muscle, which are the three major cell constituents in the heart.
1.5.2.1.1 Hematopoietic stem cells
Hematopoietic stem cells (HSCs) are multipotent and responsible for reconstituting
and maintaining all blood lineages. This population of adult stem cells is also
conferred with natural self-renewing properties. In vitro studies have reported that
human hematopoietic stem cells possess an inherent regenerative capacity and are
particularly predisposed to adipocyte and osteoblast differentiation (See Figure 2).
However, it has since been believed that hematopoietic stem cells may not
differentiate into cardiomyocytes45
.
1.5.2.1.2 Mesenchymal stem cells
Mesenchymal stem cells (MSCs) encompass a heterogeneous population of
undifferentiated, lineage-primed cells with differential phenotypic characteristics and
differentiation propensities. Found in bone marrow, muscle, skin and adipose tissue,
MSCs may be isolated from bone marrow aspirates in culture as they may adhere
differentially to the plastic culture dish. These cells can differentiate into muscle,
fibroblasts, bone, tendon, ligament and adipose tissue46
(See Figure 2). Cells
belonging to mature blood lineages may be removed using specific antibodies to
hematopoietic markers by fluorescence- or magnetic- activated cell sorting and
further purified in a Histopaq-Ficoll density gradient, prior to adherence on tissue
culture plastic surfaces. This unique property allows for an additional purification
measure, which enhances the specificity of this isolation method47-49
. Single cells of
spindle morphology may be observed 3-4 days after initial plating in a normal growth
medium. These cells exhibit a high proliferation rate and rapidly expand into colonies
12
of confluent spindle cells by 10-14 days. The cultures consist mostly of
homogeneously needle-like cells typical of mesenchymal stem cells, although a
minority population of cells may reflect a polygonal morphology.
Figure 2. Bone marrow stem cells are multipotent and can
differentiate into cells belonging to various mesodermal lineages.
The MSCs used in cardiac research are mainly from the heterogeneous cell
population. Several studies have demonstrated that these adult stem cells can
transdifferentiate into cardiomyocytes and vascular-like structures37, 50, 51
.
5-azacytidine is a demethylating agent52
reported to induce the transdifferentiation of
murine bone marrow stem cells into spontaneously beating myocytes53
. Human MSCs
have also been reported to exhibit a cardiomyocyte-like phenotype when treated with
5-azacytidine. These cells reportedly express multiple cardiac specific proteins and
display distinct ultrastructural and electrophysiological characteristics that resemble
13
native cardiomyocytes54, 55
. However, this phenomenon is inefficient and not readily
reproducible. Moreover, 5-azacytidine may induce a random demethylation of non-
specific genes that may in turn elicit toxic and adverse consequences on a systemic
basis. Thus, its clinical application has been cast with doubts.
On the other hand, butyric acid inhibits histone deacetylases, thus favoring histones in
an acetylated state in the cell56. Acetylated histones can neutralize electrostatic
interactions and thus have a lower affinity for DNA than non-acetylated histones.
Transcription factors are thus unable to access regions where non-acetylated histones
are tightly associated with DNA (i.e. heterochromatin). Therefore, butyric acid may
enhance the transcriptional activity at promoters that are otherwise usually silenced or
downregulated due to histone deacetylase activity. However, like 5-azacytidine,
butyric acid has a random mode of action and may lead to undesirable complications
on a systemic basis in downstream clinical applications.
Milieu-dependent differentiation forms the basis for undifferentiated bone marrow
stem cell therapy. It is believed that these multipotent stem cells undergo site-directed
differentiation to become resident cells of the recipient organ, specifically
cardiomyocytes and endothelial cells in regenerating myocardium. However, there is
growing evidence that cardiac transdifferentiation may be a rare and inefficient
process in MSCs.
1.5.3 Bone marrow stem cell-transplantation therapy
Bone marrow stem cell transplantation was originally conceived as a means to
repopulate and regenerate the non-viable areas of the infarcted myocardium to
compensate for the extensive loss in cardiomyocytes following a myocardial
14
infarction. Moreover, bone marrow stem cells present as an attractive modality from
the perspective of autologous cardiovascular therapeutics.
The rationale for engrafting cardiomyocytes onto a compromised myocardium is
intuitive and has prompted the initiative for subsequent “proof-of-concept”
experimentation. Early success in these preclinical investigations formed the impetus
for the explosive, albeit premature initiation of numerous clinical trials in the last
decade. To date, there are more than 1000 patients receiving undifferentiated bone
marrow stem cell therapy in clinical trials. However, most of these studies have
reported mixed outcomes and modest efficacies57, 58
.
Bone marrow-derived cardiac transdifferentiation is inefficient and rare, but
existent59
. This inefficient process may be attributed to the heterogeneous
subpopulations in the bone marrow and ironically further confounded by the
multipotent plasticity of stem cell precursors in the bone marrow. Breitback et al
recently reported the presence of encapsulated deposits containing calcifications
and/or ossifications following cell transplantation in a murine myocardial infarction
model60
. These findings highlight the disturbing possibility of extended bone
formation in vivo following undifferentiated bone marrow stem cell therapy. To
significantly enhance clinical outcomes, it may be safer and more efficient to engineer
primitive bone marrow stem cells in such a way that they may play a more active role
in recovering the overall contractile performance in the infarcted myocardium. This
may be achieved if undifferentiated bone marrow stem cells were first committed to a
stable and distinct cardiac lineage ex vivo prior to actual transplantation therapy.
15
1.6 Skeletal myoblasts
Skeletal myoblasts are resident mononucleated progenitor cells within the skeletal
muscle that unlike cardiac muscle possess the ability to regenerate upon injury.
Animal studies have shown that grafted myoblasts form myotubes in the myocardium
and eventually develop into distinct myofibers with contractile properties61
. However,
cardiac arrhythmia is an inadvertent early postoperative complication following
skeletal myoblast transplantation62
.
One of the earliest reports described the implantation of autologous skeletal myoblasts
into the postinfarction scar during a coronary bypass. Menasche et al measured
contraction and viability in the grafted scar. A mean of 871 x 106 autologous
myoblasts were injected into the scar during bypass grafting. There was improvement
in contraction and viability in the cell-implanted scars at an average follow-up of 10.9
months. However, 4 of the 10 patients suffered from sustained ventricular tachycardia
(VT) that was resistant to treatment with amiodarone and beta-blockers and was
managed with automated defibrillator implantation, while 4 of 5 patients suffered
from atrial fibrillation or VT63-65
. However, most episodes of arrhythmias were
clinically well tolerated and did not result in mortality. Nonetheless, the reasons for
arrhythmia following myoblast transplantation are multifactorial and may be
attributed to:
(i) A lack of gap junctions between transplanted myoblasts and resident
cardiomyocytes,
(ii) A difference in the action potential of the 2 cell types,
(iii) An inhomogeneous distribution of gap junctions in skeletal muscle as
compared to cardiac muscle,
16
(iv) A differential expression of ion channel isotypes between skeletal muscle cells
and cardiomyocytes and
(v) The release of inflammatory mediators after needle puncture66
.
Myoblast transplantation is also plagued with problems of survival of donor cells
post-transplantation67
. It has been shown in animals that up to 90% of grafted cells die
within the first 1-2 days after transplantation68, 69
so that total myoblast survival is
approximately <1% in humans. This poor cell survival has been attributed to
inflammatory changes at the site of implantation, which has been attributed to the
result of trauma due to needle puncture, immune-mediated rejection of myoblasts, or
release of immune modulators as a result of myoblast cell death70
. Furthermore, a
recent randomized trial conducted by Menasche et al did not show any significant
benefit in patients with ischemic cardiomyopathy62
. Efforts are underway to
genetically modify skeletal myoblasts to enhance their therapeutic efficacy. The
future of skeletal myoblast as a cardiovascular treatment modality remains to be
elucidated.
17
1.7 Clinical trials in BM stem cell transplantation for treatment of
MI
1.7.1 Clinical outcomes
Interest in bone-marrow stem cell- mediated cardiac regeneration began in 2001 with
Orlic et al‟s observation of induced cardiac repair in murine myocardial infarcts
following the introduction of Lin- cKit+ bone marrow cells in the compromised
myocardium71
. These mice demonstrated significant functional recovery within 9
days of transplantation via the de novo regeneration of cardiomyocytes and blood
vessels. Orlic‟s newly acquired knowledge then embraced a new concept that directly
challenges the traditionally accepted developmental paradigm of adult stem cell
commitment, as bone marrow cells were able to generate new cardiomyocytes,
smooth muscle cells and endothelial cells72-77
. This has inspired numerous
nonrandomized phase I clinical trials that test the feasibility and safety of autologous
cell transplantation. The results of the first clinical trial78
were published 1 year after
Orlic et al‟s discovery71
and almost 1000 patients have received similar treatments to
date. Table 1 summarizes the outcomes of the 4 major clinical trials79-86
.
1.7.1.1 Bone marrow transfer to enhance ST-elevation infarct regeneration (BOOST)
The BOOST trial was a randomized controlled trial, which recruited 60 patients
diagnosed with an ST-elevation myocardial infarction with subsequent reperfusion87
.
Patients were randomized into a conventional therapy group vs. a conventional
therapy combined with intracoronary transfer of autologous bone marrow cells
approximately 5 days following percutaneous coronary intervention. Cardiac function
was assessed via magnetic resonance imaging (MRI) and post transplantation
assessments show smaller infarct sizes, and no significant difference between scar
18
tissue remodeling following cell transplantation therapy. More importantly, the
initially promising results at 6 months post therapy was not sustained in a longer
follow-up study at 18 months post-therapy, as there are no significant differences in
LV contractile performance between the 2 treatment groups. This in turn raises
concerns that transplantation of bone marrow cells may only bring about a transient
recovery of myocardial function but not permanent cardiac repair.
1.7.1.2 Autologous Stem Cell Transplantation in Acute Myocardial Infarction
(ASTAMI)
This trial recruited 100 patients into the study, of which 47 patients underwent
intracoronary injection of BM cells 6 days after an ST elevated myocardial infarction.
In a similar experimental design to the BOOST trial, cardiac function was assessed
with single photon emission computed tomography (SPECT) echocardiography and
MRI. As in the BOOST trials, transplantation of bone marrow cells did not
significantly affect LV myocardial function, infarct size or scar tissue remodeling.
1.7.1.3 LEUVEN-AMI
The LEUVEN-AMI trial conducted by Janssens et al was a double-blind randomized,
placebo-controlled study involving 67 patients84
. Bone marrow cell transplantation
was conducted the day after successful percutaneous coronary intervention for
STEMI. Functional recovery of cardiac function following transplantation of BM
included a significant reduction in infarct size and an enhanced recovery of regional
systolic activities. However, there were no significant differences in cardiac function
4 months post-therapy.
1.7.1.4 Reinfusion of Enriched Progenitor cells and Infarct Remodeling in Acute
Myocardial Infarction (REPAIR-AMI)
To date, this randomized, double-blind and placebo-controlled trial is currently the
19
largest trial involving bone marrow stem cell transplantation in the world. BM cells
were transplanted 3-7 days following successful percutaneous coronary interventions
in patients with STEMI. Stem cell transplantation led to an increase of 2.5% in EF 4
months post therapy. This trial highlighted the fact that timing of cell transplantation
may be important in mediating myocardial improvement, as BM cells administered
from days 5-7 led to an increase of approximately 5.1% EF, whereas there no
observable recovery of cardiac performance with BM cell treatment up to 4 days after
reperfusion.
1.7.2 Problems encountered
Current clinical trials have shown that bone marrow stem cell transplantation for the
treatment of heart failure is feasible and relatively safe. However, early benefits in
cardiac function have been overshadowed by consistent trends showing only a modest
recovery of cardiac function in patients at longer follow-ups. These cast doubts on
bone marrow stem cells as ideal candidates for the treatment of heart failure as these
undifferentiated bone marrow stem cells appear only to confer temporary recovery of
myocardial function but not permanent cardiac repair. However, the source, number
and type of cells, mode and time of cell application and other details of design differ
between trials and this makes comparisons difficult. The critical problem is that there
is currently no mechanistic explanation for the clinical data. The initial results were
interpreted in terms of transdifferentiation of BMC to myocytes and vessels, as in the
animal models, but alternative angiogenic and paracrine hypotheses have also been
proposed. For example, in angiogenesis, endothelial cell precursors would
differentiate into blood vessels in vivo. This enhances myocardial perfusion, which in
turn keeps the remaining myocytes alive. In the paracrine hypothesis, it is believed
that the exogenous cells produce certain anti-apoptotic factors that can prevent cell
20
death and/or stimulate stemness of other cells72, 73, 76, 88
.
Several other key concerns of BM stem cell transplantation in the clinical setting that
remain to be addressed in the clinical setting include the longevity of intracardiac
grafts, (engrafted) cell propensity for cardiac transdifferentiation, integration with
host myocardium, (engrafted) cell response to physiological and pathological stimuli,
possible tumorigenesis at the site of transplant and asynchronus contractions leading
to arrhythmias. Current phase I trials provide an insight into how BM stem cells may
contribute towards improving cardiac function. However, the long-term benefits of
BM stem cell transplantation remain less conclusive than those established in
preclinical studies.
21
Table 1. Clinical outcomes of current major trials on cardiac regeneration for
treatment of acute myocardial infarction
Trial Design No. of cells
injected
No. of
patients/control
Change in EF
(%)
BOOST82, 83
Randomized 2.5 x 109
Unfractionated
BM stem cells
30/30 +6.0 (6 months
follow up)
+2.8% (18
months follow-
up);
p = NS
LEUVEN-
AMI84
Randomized,
double-blind,
placebo
controlled
3 x 108
Fractionated
BM-MNCs
33/34 +1.2 (3 months
follow up)
p = NS
REPAIR-
AMI80, 86
Randomized,
double-blind,
placebo
controlled
2.4 x 108
Fractionated
BM-MNCs
101/103 + 5.5 vs. +3.0
in placebo (4
months follow
up)
ASTAMI85, 89
Randomized 7 x 107
Fractionated
BM-MNCs
47/50 -3.0 (6 months)
p = NS
0.0 (12 months
follow-up)
p = NS
MAGIC62
Randomized,
double-blind,
placebo
controlled
4x108 (low-
dose)
8x108 (high-
dose)
autologous
skeletal
myoblast
33 (low-
dose)/34 (high
dose)/30
+3.4 (low-dose)
+5.2 (high-
dose)
+4.4 (placebo)
p = NS
22
1.8 Extracellular matrices
The intimate and dynamic relationship between stem cells and their niches is
invaluable to the design of cardiovascular therapeutics but is not fully understood.
However, it is known that extracellular matrices (ECM) constitute a major part of
stem cell niches. These niches sustain the stem cell pool within each tissue or organ
by maintaining a tight balance of self-renewal, proliferation and differentiation.
Collagen, proteoglycan, glycoaminoglycans and glycoproteins are among the primary
constituents of the ECM. ECM collectively performs diverse physiological functions
and is elementary to maintaining the structural integrity of tissues and organs.
Cardiomyocytes in the healthy myocardium usually adhere to a collagen-based ECM
that is adequately compliant to external and internal physical forces to facilitate
cardiac contractility. This presents the „cue‟ for many cells, and influences many
physiological functions such as morphology, protein expression and organization,
proliferation, repair, senescence as well as differentiation, via the engagement of
relevant cell surface receptors and subsequent downstream signaling events. These
surface receptors include laminin receptors and syndecans, matrix metalloproteinases
and integrins.
Extracellular matrices may also provide a means to manipulate differentiation of stem
cells into well-defined lineages in vitro, by introducing ECM molecules into the
culture milieu. In addition, cell-ECM interactions may promote the proliferation of
these adult stem cells while simultaneously retaining their differentiation potential.
Although adult mesenchymal stem cells (MSCs) are amenable to in vitro
manipulation, they progressively lose their proliferative potential and multipotent
nature with continuous passages in culture. One way to harness stem cells into well-
23
defined lineages in vitro would be to introduce ECM substrates within the culture
milieu.
1.8.1 Collagens
Collagen fibrils and proteoglycans are the most ubiquitous component of the ECM
and are present in all connective tissues. Collagens are homo- or hetero- trimeric
proteins assembled from type-specific- chains. The basic structure of collagen sees
α - chains are wound around one another in a coil-coiled conformation to form a triple
helix of collagenous domains.
The collagen family is a large heterogeneous class of proteins, which contain a
myriad of other protein-binding domains, in addition to the collagenous domain.
Collagens are also known to form supramolecular structures with other collagens
and/or other components of the ECM. Depending on their ability to form organized
fibers, collagens can be classified into fibrillar and non-fibrillar collagens90-92
. The
fibrillar collagens are a rather homogeneous group comprising of collagens type –I, -
II, -III, -V and –XI. These collagens are characterized by a single collagenous domain
and form fibrils, fibers via intermolecular interactions and are ultimately organized
into fiber bundles that constitute the primary structure rendering mechanical support
in the ECM93
.
Collagen type I is the most dominant form of collagen in the vertebrate body
including the heart. Collagen type I typically acts in concert with other collagens and
proteoglycans to confer tissues with optimal tensile strength and structural integrity.
Upon cellular interactions, collagen type I elicits vital biological functions such as
survival, motility and proliferation among other cellular events. The type I collagen
fibrils comprises of a core of collagen type V fibrils encapsulated by polymerized
24
collagen types I and III fibrils94, 95
. The ubiquitous collagens types I-III confers
mechanical strength onto the collagen fibril, while the core collagens contribute
towards overall fibril morphology and thus tissue- specificity. Importantly, these
collagens serve as both a structural scaffold and attachment sites for other ECM
constituents and cell surface receptors such as integrins.
The overexpression of collagen type I is common observation among a variety of
fibrotic disorders. These disorders arise from a defective regulation of collagen
synthesis rather than mutation in the collagen structure. Following a myocardial
infarction, beating cardiomyocytes are replaced by a scar tissue that is infinitely stiffer
and less elastic than the normal myocardium. Elevated collagen deposition replaces
functional contractile cardiomyocytes in the scar areas of the infarcted myocardium
and correlates with a corresponding decrease in myocardial tissue compliance. Just as
a scar-like rigidity inhibits spontaneous beating, so the collagen-based domains in the
scar may be responsible for the low frequencies of marrow-derived cardiac myogenic
transdifferentiation in vivo.
1.8.2 Integrins and Integrin- mediated Cell Signaling
1.8.2.1 Integrins
Integrins are a major class of ECM receptor molecules with critical roles in cellular
differentiation and tissue development. The binding of distinct - and - subunits
induces integrin clustering at focal point on the cell membrane. This leads to the
formation of macromolecular signaling complexes defined as focal adhesions, which
trigger a cascade of downstream intracellular signaling events. Thus, integrins
essentially translate exogenous signals released from a cell‟s microenvironment into
25
intracellular signaling events to regulate and influence subsequent cell fate and
function.
Figure 3. Known -integrin heterodimers.
To date, there are 24 identified human - and 9 -subunits96
. Established - and -
integrin dimers are presented in Figure 3. Integrin activation occurs via positive
allosteric modulation of the receptor. Upon ligand binding, integrins undergo
conformational changes, which increase ligand affinity. However, integrin activation
can also be accomplished via an upregulation of relevant integrin expression levels
and/ or induced clustering at focal adhesion points on the cell surface, which leads to
enhanced integrin avidity97-99
.
26
1.8.2.2 Focal Adhesions
Focal adhesions act as intracellular sites of signal transduction and form a dynamic
„bridge‟ between the ECM and the cytoskeleton and influence adhesion-dependent
cellular events such as survival, proliferation (via MAPK activation) and
differentiation. Ligand-occupied integrins connect to the actin cytoskeleton and
participate in its reorganization, thereby coupling integrin receptors to the actomyosin
apparatus that is responsible for cellular contractility, while sequestering a myriad of
signaling molecules at a focal point (Figure 4).
The engagement of integrin receptors activates critical a cascade of intracellular
signaling events which influence cell survival and proliferation100-102
. Additionally,
integrin-stimulated signaling pathways have also been implicated in the
differentiation of stem cells. Notably, integrin-mediated pathways reportedly play
significant roles in the transdifferentiation of ES cells into the neuroectodermal and
mesodermal lineages103
. They may also be responsible for the differentiation of bone
marrow stromal precursors into functional osteoblasts104
. For instance, laminin is
known to direct the osteogenic differentiation in vitro105
.
Integrins that bind to the interstitial collagens include the 11, 21 101 111 and
v3 integrins106, 107
. Of these integrins heterodimers, the 21 and the v3 integrins
have been implicated in diverse physiological and pathological conditions. Several
reports have described 11 to inhibit collagen synthesis, and 21 to be a positive
regulator of collagen gene expression although they may function more as a
competitive inhibitor of the 11 integrin than a direct activator of collagen gene
expression108, 109
.
27
Figure 4. Schematic illustration of a focal adhesion complex following integrin receptor clustering at a focal point on the cell surface.
Following ligand binding, distinct a/b subunits undergo receptor heterodimerization, which in turn leads to clustering at point of ligand
engagement. This further recruits cytosolic proteins to the site of clustering and subsequent formation of an intracellular macromolecular
signaling complex that further engages the cytoskeletal structure in the cell and activates critical cellular processes such as survival,
proliferation (activation of Ras/MAP Kinase pathways) and differentiation. Thus, integrins act as the physical and dynamic link between the
ECM and the cytoskeleton.
28
The v3 integrin is an indiscriminate receptor with multiple specificities to a diverse
array of ligands including fibrinogen, fibronectin, thrombospondin and bone
sialoproteins among others. The principal binding motif for v3 interaction is the
Arg-Gly-Asp (RGD) sequence110
. Some ECM molecules such as collagens and
laminins possess intrinsic RGD sequences that are exposed upon allosteric
conformational changes following receptor binding. Notwithstanding, it is important
to highlight that the v3 integrin binds denatured but not native collagen. The v3
integrins have been implicated in participating in collagen- mediated processes
including remodeling of the collagen matrix111-114
.
1.9 Cardiac Tissue Engineering
The main disadvantage of cell-based transplantation therapies for the treatment of
cardiovascular disease is that only a limited number of cells being delivered to the
myocardium at any one time. However, this may be circumvented by efforts in tissue
engineering, in which cells were concentrated in a hydrogel-based and patch-like
heart tissue construct to replace irreversibly damaged portions of the heart.
Collagen type I is the most ubiquitous collagen in the human body. It is readily
available commercially and has been the most studied matrix for tissue engineering.
Zimmermann et al demonstrated the feasibility of collagen-based scaffold for tissue
engineering of an artificial cardiac tissue construct. They further demonstrated that
collagen matrix supported mechanical load that enabled development of
spontaneously contracting heart tissue constructs. These collagen-based cardiac
constructs were mechanically robust yet flexible and was quickly vascularized after
implantation115
. Radisic M and colleagues have also successfully seeded preformed
29
collagen sponges/foam with neonatal rat heart cells suspended in Matrigel while
stimulating them electrically for extended time116
m while Simpson and colleagues
observed that neonatal rat cardiac myocytes aligned with the collagen to adopt a
tissue-like organization117
when a scraper was used to remove a part of the coated
surface of culture dishes pre-coated with freshly neutralized collagen type I. This led
to the generation of cardiac muscle constructs with improved cardiac tissue
morphology, contractile function, and molecular marker expression when compared
with non-stimulated cultures. Therefore, the underlying substrate of extracellular
matrix (ECM) plays an important role in cell growth and differentiation.
Commercially available ECMs substrates can be acquired as a purified powder to
form tissue construct with enhanced interaction and adhesion of stem cells.
Tissue engineered cardiac construct is an emerging field with real potential to benefit
a large number of patients with cardiovascular disease. Understanding the complex
interactions between the stem cells and their underlying collagen substrate will be
crucial in developing this into a clinical relevant application.
30
1.10 Significance of study
Plasticity of mesenchymal stem cells may present an added advantage for cell
therapy. However, like their embryonic counterparts, this may inadvertently lead to
tumorigenesis or transdifferentiation into undesirable cell types following
transplantation therapy in the injured myocardium. Recent findings have indicated
current undifferentiated bone marrow stem cell therapies mediate their improvements
via non-myogenic mechanisms, while actual marrow-derived cardiac
transdifferentiation in vivo is inefficient and rare59
. Thus, it may be a safer and more
efficacious to commit these undifferentiated stem cells to a stable cardiac-like lineage
in vitro before transplantation into infarcted myocardium. This study reports that adult
human bone marrow mesenchymal stem cells (MSCs) can be differentiated into
cardiomyocyte-like cells (CLCs) via the simultaneous use of a non-toxic myogenic
induction medium (MDM) and collagen V matrix. This preclinical investigation aims
to develop a more effective cell-based therapy for the treatment of heart failure.
31
1.11 Specific Aims
This study aims to establish a large-scale production of cardiomyocyte-like cells by
exploring stem cell commitment via milieu-dependent cardiac transdifferentiation of
human mesenchymal stem cells. Specifically, this study explores
(1) The effect of extracellular matrices like fibronectin, collagen, laminin, gelatin
and vitronectin, on the proliferation and cardiomyogenic differentiation
(2) Functional contribution of the newly established human cardiomyocyte-like
cells
(3) Investigate the dose-dependent effect of cardiomyocyte-like cells in the
systolic and diastolic performance of post-infarcted myocardium.
32
Chapter Two: Materials and Methods
2.1 Recruitment of patients
Forty-five patients who were scheduled open-heart surgery were recruited into the
study. The study was approved by the Institutional Review Board of the Singapore
General Hospital. Informed consent was obtained from recruited patients.
2.1.1 Isolation and Cell culture
Cell suspensions were isolated from the sternum of patients undergoing open-heart
surgery. Approximately 0.5 -2 ml of bone marrow aspirates were collected in a sterile
tube containing 17IU/ml lithium heparin using a 23-gauge needle. The aspirates were
subsequently depleted of mature blood lineages by using specific antibodies against
CD3, CD14, CD19, CD38, CD66b and Glycophorin-A (Stem Cell Technologies,
Vancouver, Canada) and further purified by centrifugation in a 1.077 g/ml
Histopaque-Ficoll (Sigma, St. Louis, MO) density gradient. The enriched cell fraction
was collected from the interphase and subsequently transferred onto tissue culture
flasks in basal medium for 9-11 days to yield plastic adherent MSCs. The basal
medium, designated NGM, comprises of Dulbecco‟s modified Eagle‟s medium-low
glucose ([DMEM-LG] GIBCO; Grand Island, NY) supplemented with 10% fetal
bovine serum (Hyclone, Logan, UT) and 1% penicillin-streptomycin. Media was
changed every 3-4 days thereafter. Subconfluent cells were harvested by
trypsinization in 0.25% Trypsin-EDTA (Sigma) 14-21 days after initial plating and
further expanded as BM-MSCs in basal medium or differentiated towards
cardiomyocyte-like cells in a myogenic differentiation medium (MDM). All primary
cell cultures were maintained at 37 in 5% CO2 in air.
33
2.2 Isolation and culture of neonatal rat cardiomyocytes
Cardiomyocytes were isolated from neonatal rats using a cardiomyocyte-isolation kit
(Worthington Biochemical Corporation). The cardiomyocytes were subsequently
cultured in the presence of arabinosylcytosine (Ara-C) to inhibit the growth of
fibroblast in the primary cultures. Cardiomyocyte cultures were maintained in NGM
at 37 in 5% CO2 in air, and medium was changed every 3-4 days.
2.3 Flow Cytometry
Cells were stained with a selection of specific anti-human antibodies. A comparative
histogram analyses was performed using the mean fluorescence intensities of MSCs
and CLCs. Early passage cells (passage 4-6) were trypsinized, washed with 5% BSA
and 1% fetal bovine serum in PBS and subsequently stained with a panel of the
following fluorescence- conjugated anti-human antibodies and the corresponding IgG
isotype controls for 2 hours at 4°C: FITC-conjugated CD34, CD90, CD106 and
integrin v3 (Chemicon), PE-conjugated CD31, CD44, CD45 and APC-conjugated
KDR (R&D Systems), CD105 and VEGF Receptor 2 (VEGFR2). Unconjugated
mouse anti-human antibodies to TIE2 (USBiological), 21 (Chemicon), 2 (Santa
Cruz), v (Research Diagnostics Inc), 1 (Chemicon) and 3 integrins (Cell Signaling
Technology) were detected using an allophycocyanin-conjugated rat anti-mouse
secondary antibodies (all BD Biosciences, unless otherwise stated). FACS was
performed using FACSDiva (BD Biosciences) and analyzed with FlowJo software.
2.4 Gene expression profiling via RT-PCR analyses
BM-MSCs were cultured separately in the basal or myogenic development medium
for 7 and 14 days. The expression of specific cardiomyogenic markers was assessed
by RT-PCR on mRNAs extracted from the respective cell types on Days 7 and 14.
34
BM-MSCs were also proliferated on uncoated, collagen -I and -V extracellular
matrices for 14 days, whereupon mRNAs were subsequently extracted and probed for
gene expression of cardiac-specific markers.
Total RNA was extracted using TRIzol reagent (Invitrogen Life Technologies, Inc),
and RNA was precipitated with isopropanol. One microgram of RNA, extracted from
the respective cells, was reversed transcribed using 500ng oligo-dT and the
SuperScript® III First-Strand Synthesis System for RT-PCR (oligo-dT primers and
reverse transcription kits both from Invitrogen Life Technologies, Inc). Control
reactions were carried out in the absence of:
i. Superscript III reverse transcriptase (Invitrogen Life Technologies, Inc.) to
control for genomic DNA contamination and
ii. cDNA (template) to control for primer specificity.
iii. PCR amplification was carried out on 200ng of cDNA in a Biometra thermal
cycler, using Taq polymerase (Promega Corp.) Cycling parameters are
described as follows: 95C for 2 minutes; 40 cycles of Denaturation at 95C
for 1 minute, 60C for 30 seconds and 72C for 1 minute; 72C for 7 minutes.
Nested PCR was employed to detect the presence of transcripts expressed in low
abundance. These transcripts include GATA4, csx/nkx2.5, atrial natriuretic factor,
brain natriuretic peptide, cardiac troponin t, - myosin light chain 2a, and the - and
- myosin heavy chains. The thermal cycling profile is described as follows:
i. Primary PCR: 95C for 2 minutes; 35 cycles of 95C for 1 minute, 58C for
30 seconds and 72C for 1 minute; 72C for 7 minutes.
35
ii. Secondary PCR: 95C for 2 minutes; 40 cycles of 95C for 1 minute, 60C for
30 seconds and 72C for 1 minute; 72C for 7 minutes.
2.5 Cell proliferation assays
Individual wells were coated overnight with various ECM substrates, according to the
manufacturer‟s recommendations. Early passage (Passage numbers 1-4) BM-MSCs
were cultured in DMEM or a myogenic development medium (MDM) and
proliferated on Collagen I, III, IV, V, gelatin, fibronectin, laminin, poly-L-lysine and
vitronectin matrices, at a seeding density of 1000 cells per well. BM-MSC and MSC-
derived CLC proliferation were monitored over 11 days, and a total cell count was
obtained with the use of a hemocytometer at days 1, 3, 5, 7, 9 and 11. The assay was
repeated 3 times with 5 replicates per respective treatment.
Gene expression profiling of cells expanded on competing collagen I:V substratum.
CLCs were cultured on tissue culture plastic surfaces pre-coated with competing
collagens I:V substrata in the respective ratios of 5:0, 1:4, 2:3, 3:2, 4:1 and 0:5,
overnight at 4°C. All samples were made up to a final concentration of 10mg/cm2.
Cells were trypsinized upon attaining 90% confluency, and re-seeded on freshly
coated surfaces with consistent collagens I:V ratios. Cells were harvested on the 14th
day of culture and total RNA was extracted from the cell samples as described for
subsequent gene expression profiling.
2.6 Integrin inhibition assays
Neutralizing antibodies were used to inhibit the v (Millipore) and 1 integrin
subunits as well as 21 and v3 heterodimers on CLCs (All antibodies were
purchased from Chemicon unless otherwise stated). To determine the effect of
functional integrin inhibition on initial cellular adhesion on collagen- ECM, CLCs
36
were harvested and allowed to recover overnight on low-adhesion TC plates.
Subsequently, CLCs were cultured with neutralizing antibodies towards 21, v3,
v and 1 integrins for 1 hr on ice. These treated cells were then seeded on the
respective uncoated surfaces and collagens -I and -V substrata. A total cell count was
obtained the following day.
0.4 x 106 CLCs were typsinized and incubated with 1mg/ml, 5mg/ ml, 10 mg/ml,
25mg/ml or 50 mg/ml of the respective neutralizing antibodies for 2 hours on ice in a
final volume of 400mL to investigate the effects of dosage-dependence on gene
expression. Control samples included treatment with 10ug/ml IgG isotype antibodies
and untreated CLCs. Treated CLCs were subsequently plated on collagen V surfaces.
One thousand treated CLCs/200uL were plated 96 wells plates pre-coated with
collagen V with 5 replicates per treatment. A total cell count was obtained the
following day to compare the effects of integrin blocking on initial cell adhesion. At
the same time, the remaining (treated) cells were allowed to grow till approximately
90% confluency on tissue culture flasks pre-coated with collagen V ECM, thereafter
CLCs were trypsinized and the inhibition assay repeated 7 days after the first
inhibition, before cells were re-plated on fresh collagen V-coated surfaces. CLCs
were harvested on the 14 days following initial neutralization, and pellets were
collected for gene expression profiling.
2.7 Indirect Immunofluorescence Microscopy
2.7.1 Detection of collagen-coated tissue culture coverslips
Tissue culture coverslips (Electron Microscopy Sciences) were coated with 10ug/cm2
Collagen I and Collagen V (Sigma Aldrich Pte. Ltd), and their presence detected with
the use of specific antibodies. The coverslips were incubated with anti-Collagen I,
37
anti-Collagen V and the control mouse IgG antibodies respectively, overnight at 4C.
After washing three times with PBST for 5 minutes each time, the cells were
incubated with Alexa Fluor 555-conjugated secondary antibodies in 0.1% bovine
serum albumin in PBS buffer for 1 h at room temperature, and subsequently mounted
with 70% Glycerol in PBS. Immunofluorescence microscopy was performed with a
Zeiss Axiovert 200M microscope.
2.7.2 Detection of cardiac myofibrillar proteins
CLCs were cultured on round glass coverslips (Electron Microscopy Science) at a
seeding density of 5000 cells per (24-) well for 14 days in the myogenic development
medium, MDM. The cells were first washed in a phosphate buffered saline containing
0.025% Tween 20 (PBST) and fixed in 4% paraformaldehyde and permeabilized with
0.1% Triton X-100 for 10 min each time, at room temperature. The cells were then
washed in 5% bovine serum albumin for 1 hour to block nonspecific binding. The
permeabilized cells were incubated overnight at 4C with primary or control IgG
antibodies in 0.1% bovine serum albumin. After washing three times with PBST for 5
minutes each time, the cells were incubated with Alexa Fluor 488-conjugated
secondary antibodies in 0.1% bovine serum albumin in PBS buffer for 1 h at room
temperature. Nuclei were stained with DAPI; cells were washed and subsequently
mounted with prolong Gold. Immunofluorescence microscopy was performed with a
Zeiss Axiovert 200M microscope equipped with 40x and 63x oil immersion lens.
2.7.3 Cardiac-specific promoter-driven EGFP reporter assay
The -myosin heavy chain promoter driven construct was a generous gift from
Professor Loren J. Field (The Indiana University School of Medicine), while the
myosin light chain 2v promoter driven EGFP plasmid was kind donation by Dr.
38
Jennifer Moore (Department of Physiology, The McKnight Brain Institute). The
265bp fragment of the myosin heavy chain promoter (kindly provided by Professor
Giuseppe Inesi at the School of Medicine, University of Maryland), was subcloned
into the pEGFP-2 (Clontech) using BamHI/EcoRI restriction sites. All constructs
were verified with restriction digestion and sequencing analysis. The promoter-less
plasmid was transfected as a control. All plasmids were transiently transfected into
MSC-derived cardiomyocyte-like cells with Fugene6 (Roche). Subsequently, the
transfected CLCs were selected with 25ug/ml G418 for 2 weeks to a month.
Neomycine-resistant clones were subsequently analyzed for the presence of green
fluorescence under a Zeiss Axiovert 200M confocal microscope at 40x magnification.
2.8 Golgi Disruption
Cells were trypsinized and replated on coverslips in 24-well plates at a seeding
density of 10k cells per well. Cells were incubated with 500nM sodium monensin for
an hour at room temperature before washing with ice-cold PBS and fixing with ice-
cold 4% paraformaldehyde. Treated cells were subsequently stained sequentially for
the cardiac MHC and human Golgin97 and visualized under a Zeiss Axiovert 200M
microscope at 20-40x magnification.
2.9 Cell-labeling and detection
For the low-dose cell therapies, MSCs and CLCs were incubated with 10uM BrdU
(Roche Applied Sciences) at 37C for 48 hours. Subsequently, cells were washed
thoroughly with PBS for 15 minutes, and resuspended in a final concentration of 1
million cells/ 250 mL. This volume compensates for dead space. For the high-dose
cell therapies, CLCs were incubated with 1uM Vybrant CellTracker CM-DiI at 37C
for 48 hours, while MSCs were trypsinzed and incuated with 10uM Vybrant CFDA-
SE at 37C for 15 minutes. CM-DiI labeled CLCs were excited at 546nm using a
39
Helium-Xenon lam and emission was acquired at 615-665nm. CFDA-labeled MSCs
on the other hand were verified with an anti-fluorescein AlexaFluor 488-conjugate
antibody which was excited at 492nm using a Helium-Xenon lamp. Emission for
CFDA fluorescence was acquired at 500-530nm. All fluorescence were detected with
a Zeiss Axiovert 200M fluorescence microscope and visualized with Metamorph
version 6.2 software.
2.10 Rat myocardial infarction models
A rat model of myocardial infarction was developed in female Wistar rats weighing
200-250g in body weight. Animals were anesthetized with a ketamine (100mg/ml)/
valium (5mg/ml) cocktail at a dose of 0.1ml per 100g body weight via intramuscular
administration. A left thoracotomy was performed while the rats were intubated
through intratracheal ventilation at a tidal volume of 2.5 mL (85 cycles/min). The left
anterior descending (LAD) coronary artery was identified and permanently suture-
ligated with a 7-0 polypropylene snare. One week post infarction, the rats were
subjected to an ultrasound echocardiography assessment prior to another thoracotomy
before 1 million BrdU-labled cells, 5 million CFDA-labeled MSCs (n=20 rats) or
CM-DiI labeled CLCs (n=20 rats) or serium-free medium (n=23 rats) were injected
intramyocardially around the peri-infarct regions of the compromised myocardium.
BrdU- and fluorescence labeled cells were administered in a final dose of 1x106 and
5x106 cells per 0.2ml serum free medium (NGM) respectively. Cyclosporine was
administered daily at a dose of 5mg/kg following cell transplantation. Six weeks post
LAD-ligation, the rats were subjected to another ultrasound echocardiography
assessment as well as a real-time in vivo pressure-volume catheterization before hearts
were explanted. Post-mortem was conducted immediately following death of animals.
40
2.10.1 Ultrasound echocardiography assessments
All rats were gas anesthetized with 1.5% isoflurane and positioned left lateral
decubitus on a wooden table. The chests were shaved and cleaned with methylated
spirit. Transthoracic echocardiographical assessments were performed with a
commercially available ultrasound system equipped with a 13-MHz linear array
transducer (i13L) (Vivid 7 GE-Vingmed). Cardiac function was assessed as baseline
determinations, 1 week post-LAD ligation and 6 weeks post therapy. Recordings were
made under continuous ECG monitoring by securing the electrodes on the back, Grey
scale images were recorded in a parasternal short axis view at a depth of 2 cm. M-
mode tracings were recorded from the parasternal short axis view at the papillary
muscle (mid-LV) level at a speed of 200mm/s. Data was analyzed in the offline
system (Echopac). LV dimensions were measured from 3 consecutive cardiac cycles
on the M-mode tracings. LV ejection fraction was calculated using the modified
Quinones method, while LV fractional shortening was calculated as (LVIDed-
LVIDes)x100%/LVIDed.
Multiple pairwise comparisons were conducted between rats in the respective
treatment groups for in vivo assessment of cardiac function. For ultrasound
echocardiography assessment, normalized changes in respective cardiac parameters
measured were used as statistical variables and analyzed with Tukey‟s HSD or the
Games-Howell post-hoc ANOVA analysis for samples with equal and unequal
variances respectively.
2.10. 2 In vivo hemodynamic measurements
Surviving rats were allowed to recover for 6 weeks before being subjected to real-
time pressure volume catheterization. Animals were anesthetized with a ketamine
41
(100mg/ml)/ valium (5mg/ml) cocktail at a dose of 0.1ml per 100g body weight via
intramuscular administration, and placed supine on a table. A 2-Fr microtip pressure-
volume conductance catheter specially designed for small animals (Millar
Instruments, Houston, TX) was inserted into the right carotid artery and advanced into
the left ventricle. The position of the catheter was visualized and guided by ultrasound
echocardiography. The volume calibration of the conductance system was performed
according to Pacher et al‟s methodology. At experimental endpoint, 50 uL of 15%
hypertonic saline was injected intravenously, and the shift of PV relationships parallel
conductance volume (Vp) was derived using PVAN3.2 (Millar Instruments) to correct
for cardiac mass volume and interference from baseline interference from the
immediate peripheral tissues. Preload independent indices were obtained via the
depression of the inferior vena cave with a sterile cotton swab for 7 seconds. Upon
completion of all steady state and preload-independent hemodynamic measurements,
rats were injected intraperitoneally with an overdose application of 70mg/kg of
sodium pentobarbitals while still subjected to general anesthesia. Hearts were
perfused with 0.9% saline, explanted and embedded in OCT, frozen in liquid nitrogen
and stored at -80C. Tissue heart sections were prepared with a cryostat and cut to
5um.
2.10.3 Detection of human nuclei
Explanted hearts were frozen fresh and sectioned to 5m. Cryosections were
rehydrated in PBS for 10 minutes and post fixed in 4% paraformaldehyde for 15
minutes before permeabilization with 0.5% Triton X-100 for 30 minute and blocking
in 5% bovine serum albumin containing 0.25% Triton X-100 (PBSTx). Cells were
incubated with a specific anti-human nuclei antibody (Millipore) in PBSTx for 4
hours, washed thrice with PBSTx for 15 minutes at room temperature and incubated
42
with Alexa Fluor 488-conjugated secondary antibodies in PBSTx for 1 hour at room
temperature. Visualization of nuclei and general tissue sections was performed as
described in the following section.
2.10.4 Fluorescence microscopy of frozen tissue sections
For detection of other cardiac myofibrillar proteins, frozen tissue sections were fixed
in 4% paraformaldehyde, permeabilized with 0.1% Triton X-100 for 10 minutes, and
blocked in 5% bovine serum albumin (BSA) before overnight incubation with
primary antibodies against fluorescein, cardiac troponins –I and –T, -actinin,
MHC, connexin 43, von Willebrand factor, smooth muscle actin, the proliferating
cell nuclear antigen (PCNA) and collagens –I, -III and –V diluted in 1% BSA at 4C.
Washing steps were carried out with phosphate buffered saline supplemented with
0.25% Triton X100. Cells were incubated with Alexa Fluor 488/555/660-conjugated
secondary antibodies in 0.1% bovine serum albumin in PBS buffer for 2 hours at
room temperature. Nuclei were stained with DAPI. Immunofluorescence microscopy
was performed with a Zeiss Axiovert 200M fluorescence microscope and visualized
with Metamorph version 6.2 software.
2.10.5 Microvessel density count
Following von Willeband factor staining, tissue sections were visualized with
fluorescence microscopy. Random fields in the locale of cell engraftment in the mid-
anterior section of the infarcted myocardium were selected for vessel counting at
200x magnification (20x objective lens and 10x ocular lens). A microvessel was
defined as any immunofluorescent stained endothelial cell, while presence of a lumen
was not necessarily defined as a microvessel. Microvessel counts were derived from
an average of 3 areas per stained tissue section. Counts were further taken from 3
43
fields in each selected area. Microvessel density count was expressed as the average
number of microvessels identified within the 200x fields.
2.10.6 Statistical analysis
All statistics were analyzed with SPSS version 13. A one-way ANOVA was used to
determine statistical significance between groups. Tukey‟s Highly Significant
Difference (HSD) post-hoc analyses were used to determine statistical significance
between treatment groups for samples with equal variances, while significance for
samples with unequal variances was calculated with the Games-Howell post-hoc
analyses. Homogeneity of variances was tested using the Levenne‟s statistic.
44
Chapter Three: Results
This project investigated the relative efficacies of undifferentiated human
mesenchymal stem cells (MSCs) vs. MSC-derived cardiomyocyte-like cells (CLCs)
as a cell-based therapy for the treatment of chronic heart failure in a rat myocardial
infarction model. Furthermore, ECM that plays central roles in many vital aspects of
the electrically active myocardium including cell survival, differentiation,
proliferation, metabolism and regulation of resident cardiomyocytes, cardiac
progenitor cells and stem cells were examined. This chapter is sectioned
systematically into:
i. Comparative in vitro characterization of sternum-derived mesenchymal stem
cells (MSCs) vs. MSC-derived cardiomyocyte-like cells (CLCs);
ii. Examination of physiological roles of extracellular matrices on the cell fate
and development of CLCs;
iii. Comparative analyses of the respective therapeutic efficacies of high- and
low- dose MSC- vs. CLC- therapy via ultrasound echocardiographical
assessments, real-time in-vivo pressure volume hemodynamic measurements
and comparative histological assessments.
iv. Investigation into possible cell-mediated cardiac repair mechanisms in vivo
45
Figure 5. Schematic map showing an overview of how the various project objectives and specific aims integrate with one
another and collectively form the framework of this study.
46
3.1 Isolating Sternum-Derived Mesenchymal Stem Cells
3.1.1 Isolation of Bone Marrow Mesenchymal Stem Cells (MSCs)
After informed consent, bone marrow aspirates were obtained from 30 male and 15
female patients (mean age = 58.9 ± 11.4 years) undergoing open heart surgery. MSCs
were isolated from bone marrow mononuclear cells via Histopaque-Ficoll density
gradient centrifugation. Sternum-derived cells isolated via this standard purification
method were negative for CD31 expression but exhibited low CD34 expression
levels. Consistent with adult bone marrow mesenchymal stem cell, these cells were
also positive for CD44, CD90, CD105 and CD106 (Figure 6).
MSCs were further identified by colony forming units comprising of cells with a
fibroblastic-like morphology after adhering to tissue culture plastic following
purification in a standard density gradient. This passage was designated P0.
Subsequent MSC passages were characterized by a rapid proliferation rate, which
declined progressively in the mid passages (6-9). The patient-derived bone marrow
stem cells demonstrated plasticity consistent with mesenchymal stem cells since they
differentiated into bone and fat (Figure 7) upon induction in specific induction media.
3.1.2 Patient-derived cells can differentiate into osteoblasts and adipocytes
The hallmark of bone marrow stem cells is their multipotent plasticity. CD34+
CD44+ CD90+ CD105+ and CD106+ bone marrow cells were induced to become
osteoblasts and adipocytes (Figure 7), and subsequently cardiomyocyte-like cells
(discussed extensively in later sections) and were established as bone marrow MSCs.
47
Figure 6. Patient-derived cells isolated via standard purification techniques demonstrated a low
expression of CD34, and were positive for CD44, CD90, CD105 and CD106, 40x magnification.
48
Figure 7. Patient-derived bone marrow cells can be induced to undergo osteogenic and adipogenic differentiation.
Cells isolated from the sternum of patients are bone marrow stem cells. Von Kossa stainings in Figure 7A(i) shows mineral deposition (black
deposits) in these cells upon induction with an osteogenic induction medium, while green fluorescence in Figure 7A(ii) shows positive alkaline
phosphatase activities in the sternum-derived cells. Mineral deposition and high alkaline phosphatase activity are characteristic of osteocytes. B. The
presence of red oil deposits indicates that these cells have been successfully induced to become adipocytes. Cells were counterstained with
hematoxylin to visualize the nuclei.
49
3.2 Developing a myogenic development medium
This project extends from an earlier study, which had reported that MSCs can be
differentiated into cardiomyocyte-like cells118
that expressed multiple cardiac
myofibrillar proteins and transcription factors.
RT-PCR showed that the expression profile of such differentiated CLC expressed
cardiac transcripts consistent with human adult cardiomyocytes. Moreover, cells
cultures induced with MDM promoted the enhanced expression of cardiac transcripts
such as Troponin -C and -T throughout 28 days in vitro as compared to other
differentiation methods using demethylation or acetylation techniques (Figure 8).
Further characterization studies show that different isolates of MSCs treated with this
simple and non-toxic myogenic development medium, MDM, showed a stable and
reproducible expression of many cardiac-specific myofibrillar proteins and
transcription factors that were consistent with the previously characterized CLCs.
Furthermore, the cardiomyogenic profile of these CLCs was stable after prolonged
passages in culture.
50
Figure 8. Gene expression profile of differentiating mesenchymal stem cells (MSCs)
following exposure to 5-azacytidine (AZA), butyric acid (BA) or previously developed
myogenic development medium, MDM.
51
3.3 In vitro characterization of MSCs and MSC-derived CLCs
Bone marrow aspirates were isolated from the sternum of 45 patients undergoing
open- heart surgery. These were subjected to negative selection with specific
antibodies to cells belonging to the mature blood lineages, and further purified in a
Histopaque-Ficoll density gradient. Plastic- adherent MSCs were used in subsequent
studies. In all experiments, MSCs were cultured in MDM to derive cardiomyocyte-
like cells for at least 7 days. Differentiation of early-passage MSCs (passage 0-3) into
cardiomyocyte-like cells after 2 weeks of culture in MDM was accompanied by a
change from spindle to stellate morphology (Figure 9). Importantly, MDM had a
significant effect on cell proliferation from the 5th day in culture, and eventually led
to a significant 4-fold increase in cell numbers at the end of 11 days (Figure 10).
CD34 and CD45 are cell surface markers typically associated with hematopoietic cell
types, while CD31 and CD106 are typical endothelial cell markers. Consistent with
histological stainings in Figure 6, MSCs did not express any of these cell surface
proteins but expressed high levels of CD44, CD90 and CD105 as consistent with a
cells belonging to mesenchymal lineage. MSCs of similar passage number were
cultured in MDM for 7 and 14 days.
52
Figure 9. Induction of MSCs in MDM coincided with a change in morphology from
spindle to star-like.
Figure 10. Differential cell proliferation rates in MSCs and CLCs. MDM promoted the
significant expansion of differentiating MSCs.
53
Figure 11. Gene expression profiles of MSCs and CLCs after 7 and 14 days in the
respective culture media.
Figure (i) shows the spontaneous expression of several cardiac-specific proteins in MSCs,
which disappeared after 14 days in culture. In contrast, the expression of the same markers
was sustained in MDM-treated MSCs after 2 weeks in culture. In addition, Figure (ii) shows
the expression of additional cardiac markers in vitro after 14 days in MDM. However, the
expression of the late-stage cardiac myofibrillar protein, Myomesin and the calcium channel,
Ryanodine Receptor 2 was notably attenuated by the 14th day in culture.
54
Figure 11 shows that freshly isolated human MSCs spontaneously express some
cardiac-specific proteins such as the cardiac-specific transcription factors GATA4 and
Nkx2.5, cardiac troponin C and MLC2a as well as the cardiac-specific calcium
channels IP3R2 and the L-type voltage gated 1c calcium channel. However, the
expression of these proteins quickly diminished with time and disappeared by the 2nd
week in culture. In contrast, MSCs differentiated in MDM showed sustained
expression of these proteins. In addition, MDM also promoted the expression of other
cardiac-specific proteins such as the cardiac transcription factors, Nkx2.3, Nkx2.8 and
GATA6 as well as the cardiac myofilament, skeletal muscle actin. MDM also
induced expression of the mature cardiac myofibrillar protein cardiac, myomesin and
the specific calcium channel, ryanodine receptor 2 (See Figure 11).
Further characterization studies showed that MSC-derived cardiomyocyte-like cells or
CLCs, expressed functional Nkx2.5 and GATA4. Both transcription factors displayed
a strong nuclear localization (Figure 12A), which strongly suggests an “activated”
state that is required to drive the expression of relevant downstream cardiac-specific
proteins. GATA4 and Nkx2.5 are reportedly required for the lineage development
preceding the expression of genes such as atrial natriuretic protein (ANP), myosin
light chain (MLC)-2v, -myosin heavy chain (-MHC), -myosin heavy chain
(MHC), and connexin 43 that are indicative of distinct maturation stages within the
developing organ in vivo 30, 31, 119, 120
.
These transcription factors interact with other proteins to form a macromolecular
protein complex that binds to specific sequences in the promoter regions upstream of
the cardiac myosin heavy chains and the MLC2v mRNA121, 122
. However, an
aberration in any of the proteins forming the transcriptional machinery complex will
55
not drive the expression of any cardiac genes downstream of the promoter, regardless
of their respective expression levels. Cardiac-specific promoter driven reporter assays
were used to test for the functionality of these transcription factors responsible for the
expression of the myosin heavy and light chains. In this assay, the MHC-, MHC-
and MLC2v- promoters were cloned upstream of a reporter enhanced GFP gene
(expression vectors were kindly provided by Dr LJ Field, Dr. G. Inesi and Dr.
Jennifer Moore). Thus, green fluorescence in Figure 12 indicates that the cardiac
transcription factors driving the expression of myosin heavy and light chains are both
expressed and functional.
56
Figure 12. Spatial expression and functional activity of cardiac transcription factors in CLC cultures.
A. Prevalent nuclear localization of Nkx2.5 and GATA4 in CLC cultures. B. Green fluorescence strongly indicates that
the transcription machinery, which includes Nkx2.5 and GATA4, driving the expression of the cardiac myosin heavy
and light chains are expressed and fully functional.
B
A
57
CLCs exhibited intense stainings of endothelial cell markers such as the vascular
endothelial growth factor receptors, VEGFR2 and the smooth muscle -actin, that are
typically associated with blood vessel formation (Figure 13A). Additionally, CLCs
also expressed several calcium channels such as SERCA and especially the cardiac-
exclusive voltage gated 1c calcium channel that are critical in muscle contractility
(Figure 13B).
CLCs exhibit a distinct cardiac myogenic-like phenotype as they express a multitude
of cardiac myofibrillar proteins including the cardiac troponins -C and –T,
tropomodulin, phalloidin actin filaments as well as mature myofibrillar proteins like
titin and the myosin heavy and light chains. More importantly, the punctate assembly
of myofibrillar protein aggregates or Z-bodies, especially those of -actinin, into
well-defined cross striations consistent with sarcomeres strongly suggests a
developing contractile phenotype (Figure 13C). Nonetheless, CLCs did not exhibit
spontaneous beating. This suggested that CLCs while committed to a stable and
distinct cardiac-like phenotype, they did not achieve terminal differentiation. This in
turn implies that CLCs may still retain a certain degree of multipotency reminiscent of
their stem cell origins and these cells may undergo dedifferentiation to develop into
other mesenchymal-derivatives such as chrondrocytes and osteocytes among others.
However, dedifferentiation was neither observed in prolonged CLC cultures (up to 11
passages in vitro) nor in vivo, as fluorescent-labeled CLCs in the treated myocardium
demonstrated distinct cross striations that were not distinguishable from host
cardiomyocytes 8 weeks post-transplantation. Nonetheless, future preclinical
investigations should including tracing the cell fate and development of CLCs in the
58
treated myocardium at 4, 6, 12 and 24 months post-transplantation to establish a
stable CLC-derived cardiac development in vivo.
3.3.1 Golgi-Localization of GATA4 and Myosin Heavy Chain
GATA4 is an early transcription factor that critically regulates embryonic
development, cardiomyocyte differentiation and stress responsiveness of the adult
heart. GATA4 also promotes MSCs to transdifferentiate into a partial cardiac
myogenic phenotype via an upregulation of the insulin growth factor binding protein
4123
. These cells have been observed to exhibit nuclear translocation of GATA4124
,
which correlates with -sarcomeric actinin expression during cardiac
transdifferentiation. Consistent with a developing cardiac phenotype, CLCs but not
MSCs, demonstrated an increased nuclear localization of GATA4 and
correspondingly an enhanced cytoplasmic accumulation of MHC. More specifically,
MDM induced protein transport into the relevant spatial compartments within the cell.
Moreover, MHC assumed a myofibrillar appearance in the cytoplasm that was
consistent with a cardiac differentiated state. This further supported MDM-treated
MSCs differentiation into cardiomyocyte-like cells. Interestingly however, patient-
derived MSCs and CLCs also exhibited a persistent Golgi-localization of GATA4 and
myosin heavy chain (Figure 14-15).
It has been reported that certain MHC antibodies may recognize the -COP protein
that is ubiquitous in the Golgi Apparatus125
. However, the Golgi-retention of MHC
in particular was consistent with 3 different antibodies to MHC. Moreover, when
applied to neonatal rat cardiomyocytes, these same antibodies did not illustrate the
Golgi-retention of MHC. Moreover, MHC took on a parallel swollen appearance
when the Golgi Apparatus was disrupted with sodium monensin (Figure 16), which
59
further verified a specific interaction between MHC and the Golgi-apparatus.
However, Figure 15 also shows that the cytoplasmic localization of MHC in CLCs
was not affected by Golgi disruption with 500nM of monensin. This in turn suggests
that cytoplasmic trafficking and fibrillogenesis of MHC may be independent of the
Golgi apparatus.
The golgi-localization of cytoplasmic proteins, myosin in particular, is not new.
However, the myosins that associate with the Golgi apparatus are the non-
conventional myosins such as myosin II, which are implicated in vesicular transport
of secretory proteins or in the maintenance of structural integrity of the Golgi
Apparatus. It remains to be determined if MHC also played an alternative
developmental role in maintaining the structural integrity of the Golgi Apparatus or
the cytoskeletal architecture of the increasingly cardiac-like CLC.
The golgi-localization of these proteins is recurring phenomenon in patient-derived
MSCs and MSC-derived CLCs. GATA4 has been implicated in driving the
expression of MHC. The golgi-localization of both GATA4 and the myosin heavy
chain implicates the Golgi apparatus in playing an important role in the regulation of
cardiac differentiation and myofibrillogenesis of MHC in differentiating CLCs. This
is discussed further in Chapter 4 of this thesis.
60
A. Endothelial markers
B. (Cardiac) Calcium channels
Figure 13 A-B. CLCs express endothelial marker proteins and several calcium channels that are critical for cell contractility
including the cardiac-exclusive ryanodine receptor 1/ 2and the voltage gated L-type 1c calcium channels. … continued next page
61
C. Cardiac Myofibrillar Proteins
Figure 13. Characterization of protein expression in CLCs.
A. CLCs express endothelial protein markers and may thus possess an inherent propensity for
endothelial differentiation. B-C: CLCs also express a myriad of cardiac specific myofibrillar
proteins such as the cardiac troponins, well developed actin filaments, cardiac myosin heavy and
light chains and titin. C. More importantly, the appearance of well-defined cross-striations,
especially distinct with -actinin stainings, strongly suggests a developing contractile phenotype.
All fluorescence were detected with a Zeiss Axiovert 200M fluorescence microscope and
visualized at 63x magnification (oil immersion lens) with Metamorph version 6.2 software.
62
Figure 14. Golgi-localization of GATA4 in MSCs and CLCs. (A) shows that GATA4 expression in both undifferentiated MSCs and cardiomyogenic CLCs was predominantly localized to the Golgi
Apparatus, although CLCs also exhibited an enhanced nuclear localization of the cardiac transcription factor that is strongly indicative of
an activated cardiac differentiated state. (B) shows the colocalization of the GATA4 transcription factor with a resident trans Golgi network
(TGN) protein, Golgin 97. This further verifies the spatial expression of GATA4 in the trans-Golgi network.
A
B
63
Figure 15 Spatial expression of MHC in MSCs vs. CLCs.
Immunofluorescence analysis shows the colocalization of the cardiac -myosin heavy chain with human golgin 97 in both
MSCs and CLCs. This golgi-localization was consistent with 3 different specific antibodies to MHC, and is a persistent
occurrence in more than 95% of either cell types. There is however increased cytoplasmic localization of MHC in
MDM-treated MSCs. This spatial expression of MHC in the cytoplasm resembles the cytoplasmic accumulation of Z-
bodies prior to their coalescence into Z-bands in early sarcomeric development. Hence, the Golgi-retention of MHC may
be a property of the undifferentiated state. Nonetheless, the differentiated cell types (CLCs) still retained a relatively
intense MHC expression in the golgi apparatus. This may be explained by the fact that MSC-derived CLCs have not yet
achieved terminal differentiation, despite being committed to a distinct cardiac lineage.
64
Figure 16. Cardiac myosin heavy chain (MHC) takes on a parallel swollen appearance as the Golgi
Apparatus is dispersed in MSCs and CLCs with 500nM monensin.
The cytoplasmic localization of myosin heavy in CLCs was not affected by golgi disruption with 500nM monensin. This in
turn suggested that cardiac MHC fibrillogenesis may be independent of the Golgi apparatus.
65
3.4 The roles of extracellular matrices on the cell fate and
development of CLCs
Heart failure is a debilitating clinical manifestation of an acute myocardial infarction
(MI) and is typically characterized by active remodeling of the compromised
myocardium. Fibrosis and ventricular remodeling are inherent pathophysiological
responses to an acute myocardial infarction and begins with elevated extracellular
matrix (ECM) deposition126
. Subsequent deregulation of matrix synthesis occurs as
an inadvertent process of (negative) ventricular modeling post-infarction as changes
to the composition, organization and fibril orientation of the myocardial extracellular
matrices (ECMs) constitute principal processes in restructuring cardiac geometry
during active ventricular remodeling. This ultimately „stiffens‟ the myocardium and
leads to deleterious consequences as an increasingly noncompliant myocardium
inadvertently exacerbates ventricular dysfunction.
There is increasing evidence that the myocardial ECMs may be a dynamic link
between electromechanical function and the macroscopic architecture of the
myocardium. It is possible that remodeled myocardium may secrete interstitial ECM
substrates (or other secretory proteins such as inflammatory cytokines, etc) into the 3-
dimensional microenvironment of resident/stem cells in the cardiac syncytium. In
turn, these myocardial ECMs may influence critical cellular processes such as cell-
survival, proliferation and differentiation via the engagement of specific integrins on
resident/stem cells in vivo. The interstitial collagens are integral constituents of the
myocardial ECMs. Specifically, Collagen I and III represent the major components of
the myocardial ECMs while collagen V is a minor but important collagen subtype that
regulates matrix assembly and tissue stiffness127, 128
. It is not well studied if ECMs
play significant roles in the proliferation and cardiac transdifferentiation of primitive
66
human mesenchymal stem cells (MSCs). Furthermore, the pathophysiological
effect/s of myocardial fibrosis and post-infarct ventricular remodeling on stem cell
survival, engraftment and differentiation in the infarcted myocardium also remains
poorly understood.
This section seeks to understand the roles ECM substrate molecules and their
extracellular matrix ligands in the cellular physiology of MSC-derived
cardiomyocyte-like cells (CLCs). More specifically, this study aims to identify an
appropriate extracellular matrix that would
i. Promote the large-scale expansion of MSC-derived cardiomyocyte-like cells
in vitro and
ii. Sustain/ promote a distinct cardiac-like phenotype in vitro and in vivo.
3.4.1 Collagen V enhances cellular adhesion and proliferation of CLCs
To test for the effects of extracellular matrices on human mesenchymal stem cells,
patient-derived MSCs were cultured on various ECM substrata including fibronectin,
gelatin, laminin, poly-L-lysine, vitronectin and the collagen types I, III, IV and V. As
is evident in Figures 17-20 and further verified with ANOVA analyses, CLCs
expanded significantly on collagen V matrix through enhanced initial cell attachment
and cell survival. In contrast, laminin was among the least supportive substratum for
cell attachment, p > 0.05 (See Figure 17). This result is consistent with Conget et al‟s
and Phinney et al‟s findings that laminin matrices do not encourage cell adherence129,
130.
Collagens –I, -III and –V constitute the predominant components of the interstitial
ECM in the myocardium. CLC-attachment on the collagen extracellular matrices in
67
vitro showed that the number of total cells adhered to pre-coated collagen surfaces
increased with respect to those cultured on uncoated tissue culture (TC) plastic
surfaces. CLCs (1.39 ± 0.23 fold) cultured on Collagen I extracellular matrices
showed a trend towards enhanced initial cellular attachment. However, this did not
achieve statistical significance. Similarly, collagen III matrix, another major collagen
subtype found in the heart, did not support a significant adhesion of CLCs and MSCs
(Figures 17-18). In contrast, CLCs showed preferential interaction with collagen V
substrate molecules, resulting in a significantly enhanced initial cellular attachment
(1.97 ± 0.54 fold, p<0.01) (Figures 17-18).
Collagen V supported the highest attachment of CLCs as compared to those cultured
on collagens -I (p=0.06) matrices. The simultaneous effect of MDM and collagen V
extracellular matrices promoted a synergistic increase in CLC proliferation that
persisted throughout 11 culture days (Figure 19-20), as CLCs expanded a further 8-
fold from a simple culture in MDM on uncoated TC plastic surfaces (Figure 19-20).
In addition, CLCs cultured on collagen V ECM persistently demonstrated enhanced
expansive rates (in relation to CLCs cultured on collagen I matrices). As seen in
Figure 19, this benefit persisted throughout the entire study period, and resulted in a
29.3 ± 13.5 fold increase (p<0.05) in CLCs expanded on collagen V, as compared to a
7.4 ± 2.9 fold expansion in CLCs cultured on collagen I matrices on the 11th culture
day. Collectively, these observations suggest that a significant initial attachment-
dependent survival, which is achieved via CLCs‟ preferential interaction with
collagen V substrates, is important for subsequent expansion of CLCs on a large
scale.
68
Figure 17. CLCs were harvested and seeded on tissue culture plastic surfaces pre-coated
with various ECM substrates.
These include fibronection, collagens I, III, IV and V, gelatin, Laminin, poly-L-lysine and
vitronectin. Of the 9 ECM substrates tested, only Collagen V promoted a significantly
enhanced adherence-dependent cell survival in CLCs with respect to CLCs that were
expanded on uncoated tissue-plastic surfaces.
Figure 18. CLCs demonstrate preferential affinity for collagen V ECM. Collagen V, but not collagens -I and -III, promotes a significant attachment off the
differentiating cells 1 day after cell seeding on surfaces pre-coated with the respective
collagen ECM. In contrast, MSCs did not demonstrate preferential interactions with Collagen
V ECM. *p<0.01 vs. MSCs on collagen V.
69
Figure 19. Collagen V ECM promotes enhanced adherent-dependent survival of CLCs.
CLCs‟ significant initial attachment on collagen V extracellular matrices resulted in
significant expansions that persisted throughout 11 days. In contrast, CLCs did not proliferate
significantly on collagen I. Data were normalized to day 3 control that was seeded on
uncoated tissue culture surface. *p<0.005 vs. CLCs on collagen I, #p<0.05 vs. CLCs on
collagen I.
Figure 20. Large scale expansion of CLCs on collagen V ECM.
Collagen V was the only 1 of the 9 ECM substrates tested to promote a significant increase in
expansive rate as compared to CLCs that were expanded on uncoated tissue culture plastic
surfaces. This benefit was initiated as early as the first day and persisted for 2 weeks in
cultures.
70
Figure 21. Red fluorescence verified that the tissue culture plastics were indeed
pre-coated with Collagens -I and -V molecules respectively.
Figure 22. Quantitative assessment of cardiac gene expression levels in CLCs expanded
on collagens -I and -V substrata via densitometry.
All gene expression levels were normalized to the expression level of the housekeeping gene,
-actin. CLCs cultured on collagen I matrices exhibited persistently attenuated expression
levels of several cardiac-specific proteins, while collagen V matrices enhance selective
cardiac gene expression in cardiomyocyte-like cells. Gene expression of cardiac transcription
factors and sarcomeric myofilaments, but not ion channels, was upregulated in collagen V
cultured cells. Densitometric assessments of cardiac gene expression were measured as Mean
SD arbituary units normalized to the actin expression levels in CLCs cultured on
Collagen I and V ECM respectively.
71
3.4.2 CLCs demonstrated enhanced cardiac differentiation on Collagen V
matrices
As shown in Figure 21, red fluorescence verified that the pre-coated surfaces were
indeed coated with the respective collagen substrate molecules. CLCs cultured on
collagen V matrices demonstrated enhanced expression levels of the cardiac
transcription factors, GATA4 and Nkx2.5, as compared to collagen I-cultured CLCs
(Figure 22). Consistently, downstream gene expression of ryanodine receptor-2
(RyR2), skeletal muscle α actin, troponin C and troponin T was similarly enhanced in
CLCs that were expanded on collagen V matrices.
Images obtained via DNA gel electrophoresis were subjected further to densitometry
to obtain a quantitative assessment of the relative gene expression levels of the
respective cardiac-specific markers in CLCs that were expanded on collagen I vs.
collagen V ECMs (Figure 22). CLC-interaction with collagen I ECMs resulted in the
attenuated expression of several cardiac specific transcription factors such as GATA4
and Nkx2.5, the ryanodine receptor 2 (RyR2), which is in turn a cardiac-exclusive
intracellular calcium channel that releases calcium from the sarcoplasmic reticulum;
and several myofibrillar proteins including troponins -C and –T, the cardiac -actin
and the cardiac skeletal -actin. However, the gene expression of MEF2C,
connexin43, IP3R2, L-type calcium channel α1c subunit and cardiac α actin remained
relatively unchanged in relation to the 2 collagen subtypes (Figure 22). The same
observations were further vindicated in a comparative analysis of gene expression
profiles of CLCs cultured on competing ratios of collagen I: collagen V substratum
(Figure 23).
72
Figure 23 shows enhanced expression levels of cardiac-specific proteins with
increasing CLC-collagen V interaction. Conversely, CLCs demonstrated decreasing
cardiac expression with increasing CLC- collagen I interaction (Figure 23). Thus,
collagen V matrices promote and support the distinct cardiac-like gene expression in
CLCs. More importantly, the expression of these cardiac-specific proteins remained
stable in prolonged cultures, thus supporting the persistence of a distinct cardiac-like
phenotype in the differentiating CLCs on Collagen V ECM (Figure 24).
73
Figure 23. Relative cardiac gene expression in CLCs cultured on competing
Collagen I: collagen V substrata.
CLCs were expanded on competing collagen I: collagen V substrata with a
final total collagen concentration of 10ug/cm2 in any 1 sample. Data is
presented to show increasing collagen V concentrations toward the right, and
increasing collagen I concentrations towards the left.
Figure 24. Stable (gene) expression of cardiac specific myofibrillar
proteins in prolonged CLC cultures.
74
3.4.3 Study of relevant integrin signaling
Cell-ECM interactions activate downstream intracellular signaling processes via the
heterodimerization of relevant integrin subunits. Cellular interaction with collagens
reportedly involves the engagement of the 11, 21 and V3 heterodimeric
integrins 104
. The human-specific antibody against the 11 integrin heterodimer was
commercially unavailable at the point of writing and studies investigating the role of
11 in cellular adhesion on collagen types -I and -V extracellular matrix surfaces
were thus excluded in this thesis.
Flow cytometry analysis showed that more CLCs expressed higher levels of the 21
(50.2 17.2% vs 34.2 23.2% in MSCs) and v3 (56.5 21.4 % vs. 33.1 25.5 %
in MSCs) integrin at basal levels relative to MSCs (Figure 25), while, the v and 3
integrin was upregulated in CLCs on collagen –I and –V extracellular matrices Table
2. In particular, CLC-attachment on collagen -I and especially collagen -V surfaces
was significantly attenuated upon the functional blocking of the 21, v3 and 1
integrins (Figure 26B). These observations suggest that cellular attachment of CLCs
on collagen –I and especially collagen -V surfaces involve the engagement of these
integrins. As less MSCs express these integrins at basal levels (Figure 18), they do not
exhibit preferential interaction with collagens I and V extracellular matrices.
Notably, the v and 1 integrin subunits constituted the majority of integrin family in
CLCs regardless of matrix surface. CLCs expressed the 1D integrin, which is
selectively expressed by skeletal and cardiac muscle as well as endogenous levels of
collagens –I and –V, and it is possible that they may secrete varying levels of the
respective collagens when cultured on different ECM surfaces (Figure 26A). This
75
may lead to inadvertent variations in endogenous integrin expression as CLCs interact
with the respective collagen matrix substrates. To circumvent possible confounding
results, each treatment group was normalized to untreated CLCs cultured on the
respective surfaces. A paired t-test was then conducted against the IgG isotype
control in each ECM group to analyze for significant attenuation in CLC attachment
on each ECM surface.
As shown in Figure 26B, CLCs treated with the IgG isotype control did not affect
cellular attachment on any surface as indicated by an approximately equivalent ratio
of IgG-treated CLCs to untreated CLCs on the respective surfaces (i.e. IgG-treated
CLCs on collagen I ECM/ untreated CLCs on collagen I ECM ≈ 1). Initial CLC
attachment on uncoated tissue culture plastic surfaces involves the engagement of the
v3 integrin heterodimer and the 1 integrin subunits, since the inhibition of either
integrin significantly attenuated cellular adhesion.
Functional inhibition of 21 and 1 integrins on CLCs significantly attenuated initial
cellular adhesion on collagen I (p<0.005 and p<0.0005 respectively) and collagen V
(p<0.05 and p=0.06 respectively). Conversely, neutralization of v3 and v integrin
inhibited adhesion of CLCs on collagen I (p<0.05 and p=0.06 respectively) but not on
collagen V (p=0.09 and p=0.6 respectively), although functional blocking of the v3
integrin in CLCs cultured on collagen V ECM trended towards low attachment on
collagen V substratum with respect to IgG-treated CLCs on the same ECM surface.
Figure 23B. Similarly, functional neutralization of the v integrin did not affect
cellular attachment on collagen V surfaces as the number of v integrin- inhibited
CLCs was approximately equivalent to the number of untreated CLCs). CLC
adhesion also appeared to be primarily dependent on the 1 integrin subunit, as
76
inhibition of the 1 integrin mediated lower CLC-attachment on all tested surfaces
(p=0.002 on uncoated surface, p = 0.001 on collagen I, p = 0.06 on collagen V coated
surfaces). Likewise, inhibition of the 21 integrin heterodimer mediated a
correspondingly marked decrease in CLC- attachment on collagens -I and -V matrices
(Figure 26B). This is consistent with current literature that reported that the 21
integrin heterodimer as an important receptor of triple helical interstitial collagens,
such as collagens -I and -V.
Cellular interaction with collagens predominantly involves the engagement of the
11, 21 and V3 heterodimeric integrins104
. This study shows that collagen V
significantly enhanced CLC adhesion as compared to MSCs. This enhanced adhesion
may possibly be attributed to a larger number of CLCs expressing the 21 and V3
integrin heterodimers as compared to MSCs.
With the exception of troponin T expression, functional blocking of the 21 integrin
failed to have any significant impact on cardiac gene expression in CLCs that were
cultured on collagen V matrix (Figure 27). In contrast, neutralizing the v3 integrin
with higher concentrations of the specific function-blocking antibody to the (integrin)
heterodimer significantly reduced expression of troponin T, skeletal muscle actin
and RyR2 expression in CLCs cultured on collagen V matrices (Figure 27). Thus, the
V3 integrin may be a primary effector of a collagen V-centric response to CLCs in
cardiac differentiation. However, GATA4 expression in collagen V-expanded CLCs
remained inert to the functional neutralization of the v3 integrins, even after
exposure to the highest concentration of v3 neutralizing antibody. This suggests that
GATA4 expression is independent of 21 and v3 integrin-mediated signaling.
77
It remains to be elucidated if differential integrin signaling was responsible for CLCs‟
significantly enhanced adherence and expansion rates when exposed to collagen V
matrix. More importantly, CLCs cultured on collagen V matrices retained the
expression of cardiac-specific myofilamental proteins, as these partially cardiac
differentiated stem cells retained the appearance of nascent Z-body aggregation
reminiscent of developing cross striations that are in turn characteristic of sarcomeres
(Figure 28). Therefore, collagen V extracellular matrices support an emergent
contractile phenotype in the differentiating CLCs.
78
Figure 25. Flow cytometry analysis of integrin expression in MSCs
and CLCs.
MSCs and CLCs also express the 21 and v3 integrin heterodimers,
which reportedly interact with the interstitial collagens, including the
collagen subtypes -I and -V respectively. CLC expressed higher levels of
v3 and v3 integrin heterodimers than MSCs. Flow analyses derived
from 3 independent experiments.
Table 2. Human CLCs express 2, 1, v, 3, 21, and v3 integrins. The gene
expression levels of the v and 1 integrins were markedly higher in CLCs cultured on
collagen V ECM.
Integrin subunits CLCs
Collagen I Collagen V
2 33.3% 37.1%
v 81.7% 79.2%
1 81.4% 75.1%
3 43.6% 38.7%
79
Figure 26. Initial cellular-attachment of CLCs on collagen matrices is mediated via engagement of specific integrins.
A. CLCs expressed the cardiac-specific integrin 1d as well as endogenous levels of collagens I- and V.
All cell counts were normalized to the total number of untreated CLCs on the respective ECMs to circumvent possible confounding experimental
variations due to endogenous collagen or integrin expression. B. CLC-attachment on uncoated tissue culture plastic surfaces is mediated via the
engagement of the v3 and 1 integrins (p<0.05 vs. IgG control in either inhibition), while CLCs interact with collagen substrates via ligation of the
21 integrin heterodimer (p= 0.026 vs. IgG control on Collagen V ECM, p = 0.001 vs. IgG control on Collagen I ECM). Additionally, CLC-
adhesion on collagen V matrix primarily engages 1 integrins (p=0.06 vs. IgG control), but not the v and v3 integrins (p>0.05 vs. IgG control on
collagen V ECM). In contrast, CLCs‟ interaction with collagen I substrates involves the v3 (p=0.016 vs. IgG control) and 1 integrins (p<0.001 vs.
IgG control). CLC attachment on any of the tested surfaces did not demonstrate selective integrin engagement in cell attachment. Data presented here
is derived from 5 replicates.
A
B
80
Figure 27. Dose-dependent effect/s of 21 and v3 integrin inhibition on expression
levels of cardiac-specific proteins in CLCs that were expanded on collagen V
extracellular matrices.
Functional integrin blocking neutralized the effect of collagen V on cardiac gene expression
in cardiomyocyte-like cells when administered in higher doses. In particular, inhibition of the
v3 but not the 21 integrin heterodimer attenuated expression of troponin T, skeletal
actin and RyR2 receptor.
81
Figure 28. CLCs expanded on
collagen V extracellular matrices
retained expression of several mature
cardiac myofibrillar proteins.
More importantly, these cells retained
the distinct cross-striations that are
consistent with a developing myogenic
phenotype.
82
3.5 In vivo functional studies
This section reports the findings of a comparative study investigating the respective
therapeutic efficacies of (i) low- and (ii) high- dose cell therapies employing the use
of (i) undifferentiated MSCs and (ii) CLCs in a rat myocardial infarction model. The
full methodology for in vivo functional studies is summarized in Figure 29. As shown
in Figure 30, the different sample sizes used in echocardiographical assessments and
examination of PV relationships in the respective cell therapy groups was attributed to
the high experimental attrition encountered in this survival model. One week after the
creation of the rat MI model via the ligation of the LAD coronary artery, cell-treated
rats, MSCs or CLCs were injected around the peri-infarct region of the myocardium 7
days post-ligation. MSCs and CLCs were administered in doses of 1x106 or 5x10
6
MSCs or CLCs in a final volume of 0.2ml serum free medium. These rats were
subsequently subjected to another ultrasound echocardiographical assessment as well
as a real-time in vivo pressure-volume catheterization 6 weeks post-transplantation.
83
Figure 29. Flowchart summarizing experimental methodology for in vivo studies
84
Figure 30. Surviving number of rats in the respective cell therapy groups pre- and post transplantation.
The relatively high attrition observed in post-therapy assessments of pressure-volume relationships was attributed to a
learning curve associated with unfamiliarity with a relatively novel experimental technique at the point of
investigation.
85
Figure 31. 2-Dimensional (2D) M-mode
echocardiographical analyses show the disease
progression of a sham-operated control rat.
(A) A 2-Dimensional M-Mode echocardiogram of a rat
myocardium at baseline demonstrates a robust
contractility characteristic of a healthy rat. (B) 2D M-
mode echocardiographs of a representative media-
injected control rat illustrate a severe thinning of the
interventricular septum, which worsens progressively
with time.
A
B
CFDA labeled
CLCs? Or
MSCs???
86
Figure 31C. Successfully myocardial infarction was created in rats in the respective
treatment group as shown by significant functional decreases in LV ejection fraction from
baseline levels post- LAD ligation.
Table 3. Mean LVEF (%) of rats in the respective treatment groups at baseline, post-
ligation and 8 weeks post-therapy.
*p<0.05 among all groups,
#p<0.05
vs. low-dose treatment groups
Cardiac
Function
Treatment
Groups
Baseline Post-ligation Post-therapy
Mean Std.
Deviation Mean
Std.
Deviation Mean
Std.
Deviation
Ejection
Fraction
(%)
Media-injected
control (1M) 71.91 2.74
*55.86 8.23 52.53 8.36
MSC-therapy
(1M) 70.12 2.08
*55.25 6.79 53.99 10.87
CLC-therapy
(1M) 70.65 1.90
*50.51 11.56 51.66 9.52
Media-injected
control (5M) #78.34 2.18
*70.21 6.38 69.52 5.99
MSC-therapy
(5M) #77.85 2.05
*65.51 7.29 70.09 8.82
CLC-therapy
(5M) #75.90 3.31
*68.24 4.02 61.69 7.85
87
Figure 31 shows a 2-dimensional M-mode echocardiographical analysis of a
representative rat in the medium-injected control group. Figure 31A depicts the
healthy myocardium at baseline levels, while Figure 31B shows the myocardium of a
representative serum-free media injected rat 6 weeks after injecting serum-free
medium into the peri-infarct zones of the myocardium. Following ligation of the LAD
coronary artery, control rats demonstrated severely thinned interventricular septa
(IVS) with respect to baseline. Systolic wall thickening in these rats worsened
progressively with time, following injection of serum-free media into myocardium, as
can be seen in Figure 31B. At the same time, Figure 31B also shows that the LV
internal diameter increased with time, suggesting progressive chamber dilatation in
the control rat.
Figure 31C shows that the successful creation of a myocardial infarction in all rats
before they were randomized into the respective treatment groups as LVEF decreased
significantly from baseline levels post LAD ligation. Baseline values were not
significantly different among the different treatment groups. However, post-ligation
values significantly differed between several groups despite randomization into the
respective groups and the sonographer and surgeon being blinded to the study (Table
3). Furthermore, post-ligation and post-therapy values of rats within each treatment
group were plagued with large variations as reflective of inadvertently variable infarct
sizes, which are in turn exemplified by the large standard deviations between groups.
These variations may give rise to large standard errors, which may obscure actual
functional improvements between treatment groups.
88
Normalized treatment-induced changes in multiple cardiac functions were calculated
as follows to circumvent possible confounding factors due to the presence of large
variations between treatment groups.
This method of calculation essentially used each rat as its own control and allowed for
a more meaningful comparison of functional improvements effected by the respective
therapies. This also eliminated the confounding factor of significantly different
statistics that was present in the low dose cell therapy group post ligation and which
may be attributed to the relatively smaller sample sizes in the respective low-dose cell
therapy groups, as compared to the high dose cell therapy groups. In contrast,
conventional comparisons between treatment and placebo groups in current clinical
trials do not normalize measurements in cardiac function to pre-transplantation
values. However, this approach is only applicable if the post-ligation reference values
between treatment groups are not statistically different from one another. This was not
the case in this study as post-ligation EF values were significantly different between
rats in the respective treatments.
Rats treated with a low-dose of MSCs and CLCs had a mean LVEF of 56.7 21.8%
and 54.6 18.8% while control rats had a mean LVEF of 24.0 5.9% (Figure 32iv).
Following normalization to post-ligation values, these translated into (0.76 28.99)%
and (5.87 21.71)% as control rats deteriorated by (3.34 20.23)%. Rats
administered with a high dose of CLCs and MSCs showed a mean improvement of
89
(7.14 8.39)% and (-0.76 7.1)% in LVEF respectively while control rats
deteriorated by (11.3 11.4)%. This is consistent with current literature, which
reports an improvement of 3 - 7% in LVEF in clinical trials (Table 1).
Notwithstanding reportedly modest improvements observed in most clinical studies
(Table 1), marked improvements in LVEF have been described in rare cases. For
example, LVEF improved to 50.1% from 20.9% in a 32 year old man who had
suffered end-stage cardiac failure from haemachromatosis secondary to red blood cell
transfusions and was administered bone marrow stem cell transplantation131
.
Consistent with Figure 31A-C, 2D M-mode ultrasound echocardiography showed that
serum-free media injected control rats injected with an equivalent volume of serum-
free media trended towards a progressive deterioration in cardiac function (Table 4).
Multiple pairwise ANOVA analyses using Tukey‟s HSD post-hoc statistical analysis
showed no significant differences between therapy groups. However, this particular
study showed large experimental variability as shown by the large standard deviations
reported in Table 4. Nonetheless, CLC-treated rats demonstrated consistent trends in
smaller internal diameters and improved global cardiac functions such as ejection
fraction and fractional shortening.
In contrast, hemodynamic measurements derived from real-time intra-ventricular
pressure-volume catheterization showed that low-dose cell- therapy mediated
statistically significant improvements in cardiac functions. Consistently, control rats
demonstrated severely dilated ventricular chambers as represented by enlarged end-
systolic (188.2 69.8 uL) and end-diastolic volumes (224.1 74.4 uL). In contrast,
end-systolic volumes were significantly smaller in both MSC-treated (84.1 50.3 uL,
p<0.01) and CLC-treated rats (71.7 50.8 uL, p<0.005). Similarly, smaller end-
90
diastolic volumes were observed in rats treated with MSCs (148.9 ± 49.1 uL, p<0.05)
and CLCs (139.1 45.2 uL, p<0.05). These smaller cardiac volumes suggest that cell
therapy attenuated negative LV remodeling in the cell-treated myocardium. It remains
to be elucidated if cell-therapy actively or simply acted to prevent the progressive
thinning of the interventricular septum walls (See Figure 32i-ii).
Cell-therapy also mediated improvements in global systolic performance.
Specifically, CLC-treated (56.7 21.8%, p<0.01 vs. control) and MSC-treated (54.6
18.8%, p<0.005 vs. control) rats demonstrated enhanced left ventricular ejection
fraction (LVEF) with respect to control animals (25.0 5.9%). This translated further
into significantly higher cardiac output in rats administered with CLCs (33073.6
15018.3 uL/min, p<0.05) and MSCs (31498.8 11162.1 uL/min, p<0.05)
transplantation, in relation to control animals (18111.2 3353.8 uL/min) that received
serum-free medium injection (See Figure 32 iv-vi).
Additionally, cell-therapy significantly enhanced myocardial systolic function as
shown by the elevated maximum dP/dt levels in CLC- treated (13685.5 3870.7
mmHg/s, p=0.06) and MSC –treated (12929.8 5547.8 mmHg/s, p=0.1) rats with
respect to control animals (8316.6 4788.7 mmHg/s). Similarly, diastolic myocardial
relaxation in rats treated with CLCs (10159.6 2838.5 mmHg, p<0.05) and MSCs
(11118.1 4744.1 mmHg/s, p<0.01) exhibited significantly enhanced minimum dP/dt
levels as compared to control animals (5391.2 2913.7 mmHg/s) (See Figure 32 iii).
However, CLC- (66.5 27.8 mWatts, p<0.05), but not MSC- transplantation (60.5
44.9 mWatts, p=NS), significantly improved maximal cardiac power relative to
91
serum-free media injected control rats (30.8 8.6 mWatts). Although not statistically
significant, there was a trend towards considerably lower end-diastolic pressures in
CLC-treated rats (8.5 ± 2.0 mmHg) as compared to MSC-treated (11.2 ± 9.4 mmHg)
and control rats (17.1 ± 27.7 mmHg). On the average, the Tau (Glantz) relaxation
index in CLC –treated rats (14.9 3.5 msec) was also lower than both MSC- treated
(17.5 7.0 msec) and control (18.5 3.2 msec) animals. These trends collectively
suggest that CLC-therapy mediated better relaxation properties within the ventricular
chambers, in turn mediating better tissue compliance, which ultimately leads to
enhanced biomechanical efficiencies in the CLC-treated myocardium.
92
Table 4. Ultrasound echocardiography assessment of cell-treated rats.
Measurements obtained via 2D ultrasound echocardiographical assessments show that CLC-therapy trended towards improvements in cardiac
function, although these did not achieve statistical significance. In particular, the CLC-treated myocardium demonstrated trends in enhanced
mid-anterior wall velocities and enhanced global contractility such as ejection fraction and fractional shortening with respect to serum-free
media injected control animals. In contrast, MSC-treated rats trended towards little or no improvement in cardiac function with respect to
control rats.
Multiple
pairwise
comparisons
Analysis of cell-mediated changes in cardiac functions by Ultrasound echocardiography
assessments (Mean ± SD)
Medium-injected
control rats
n = 10
MSC-Therapy
(1 million cells)
n = 8
CLC-therapy
(1 million cells)
n = 9
p-value
% changes in:
Heart Rate -10.86 ± 8.98 -1.41 ± 13.69 -6.68 ± 11.51 NS
IVSes 12.86 ± 33.66 14.66 ± 33.38 12.7 ± 18.71 NS
IVSed 2.60 ± 13.90 6.61 ±17.81 2.68 ± 9.59 NS
LVIDes 15.36 ± 20.51 16.09 ± 35.44 9.49 ± 20.60 NS
LVIDed 11.44 ± 12.12 12.23 ± 15.79 10.00 ± 13.18 NS
Systolic Wall
Thickening 98.59 ± 162.3 43.30 ± 86.19 48.88 ± 71.77 NS
Ejection Fraction -4.43 ± 19.03 0.76 ± 28.99 5.87 ± 21.71 NS
Fractional Shortening -3.34 ± 20.23 -333.9 ± 502.2 8.51 ± 24.33 NS
Mid-Anterior Wall
Velocity 38.03 ± 176.82 28.18 ± 87.60 53.82 ± 99.87 NS
93
Figure 32. Functional improvement in cell transplanted myocardium.
In vivo pressure-volume catheterization showed that both MSC and CLC transplantation stabilized cardiac geometry by preventing chamber
dilatation as demonstrated by significantly smaller end-systolic volumes and end-diastolic volumes in relation to control animals that received
serum-free medium injection. Systolic activity and global contractile function was correspondingly significantly improved in MSC and CLC
transplanted myocardium as compared to control animals.
94
Figure 33. Efficiency of BrdU-labeling in MSCs and CLCs.
MSCs and CLCs were labeled with bromodeoxyuridine (BrdU) before cell transplantation.
Labeling efficiency of either cell-type was greater than 90%.
Figure 34. Identification of transplanted human stem cells in rat
myocardium 6 weeks post transplantation.
Co-localization of anti-BrdU (arrows) and anti-human nuclear (arrows)
antibodies confirms the specificity of staining in identifying the transplanted
cardiomyocyte-like cells.
95
Figure 35. Engraftment of BrdU-labeled CLCs in the cardiac syncytium in host
myocardium.
Immunofluorescent cell tracking showed that the integration of some human cardiomyocyte-
like cells (arrows) within the host myofibrillar architecture, while exhibiting distinct
sarcomeric cross-striation indistinguishable from native cardiomyocytes. The engrafted
human cardiomyocyte-like cells were found to connect electrically to the host myofibers
through connexin43 gap junction proteins (arrowheads).
96
3.5.1 BrdU labeling of cells in the low-dose therapy group
Cells in the low -dose cell therapy groups were uniformly labeled with
bromodeoxyuridine (BrdU) nuclear labeling dye before cell transplantation. More
than 99% of the cells were labeled and there were no significant differences in the
labeling efficiencies of either cell types. In addition, viability of the BrdU labeled
MSCs or CLCs was comparable with greater than 95% cell survival (Figure 33).
3.5.2 Cell fate and development of BrdU labeled cells in vivo
The identity of transplanted human cells in the infarcted myocardium was vindicated
by the co-localization of BrdU label and human nuclei staining (Figure 34). These
cells appeared to demonstrate distinct cross-striations characteristic of sarcomeric
structures in vivo that were indistinguishable from host cardiomyocytes (Figure 35).
Furthermore, positive connexin 43 stainings in juxtaposition to the engrafted cells
indicate the formation of gap junctions between the exogenously administered cells
and the host myocardium (Figure 35). Taken together, these findings imply that the
engrafted cells underwent extended cardiac differentiation in vivo to become
cardiomyocytes that were able to couple electrically with host cardiomyocytes.
However, immunofluorescence tracking of the transplanted cells showed only a low
frequency of cellular engraftment in the host myocardium. This may explain the lack
of significant improvements in regional myocardial contractility in 2D ultrasound
echocardiography assessments (Table 4). More importantly, the low engraftment rate
in the cell-treated myocardium is still disproportionate to the significant
improvements in global cardiac function that was observed in cell-treated rats, as it is
highly unlikely that such a small population of exogenously supplied cells could
97
regenerate the infarcted myocardium or contribute meaningfully to overall
contractility.
Persistent improvements in cardiac function in spite of low incidences of cellular
integration in the cell-treated myocardium underscore the likelihood of alternative
non-myogenic and possibly paracrine-induced repair mechanisms in vivo. In
particular, smaller cardiac volumes in cell-treated rats strongly suggest a significant
attenuation of (negative) LV remodeling (Figure 32 i-ii) Thus, the cell-treated
myocardium may prevent the progressive thinning of the ventricular walls by
modulating matrix architecture, thereby reducing tissue „stiffness‟ in an otherwise
increasingly non-pliable tissue mass in the peri-infarct borders of a developing „scar‟.
In so doing, cell therapy may act to sustain tissue compliance in the remaining viable
tissues within the peri-infarct precincts of the infarcted myocardium, while facilitating
a synergistic recovery of biomechanical functions via myocyte replacement, and
sustaining or enhancing overall contractile efficiencies with respect to the untreated
myocardium.
98
3.5.3 High-Dose Cell Therapy
It is important to emphasize that notwithstanding the discrepancies in cell integration
and cell-mediated myogenic repair, CLC- transplantation can contribute to cardiac
contractile activities via myocyte replacement (See Figure 35). However, direct
cellular effects of cell-mediated myogenic cardiac repair in vivo could only be
vindicated with relatively higher frequencies of cellular integration in the
myocardium. It was believed that high-dose cell-therapy would increase the
probability of cell engraftment in the myocardium when administered in a higher
dosage. Specifically, cells were administered in high dosage to determine if CLC-
therapy mediated myogenic replacement/ regeneration. This also allowed for a
subsequent comparative study investigating the dose-dependent effects of cell-therapy
on cardiac function.
3.5.3.1 Fluorescent Labeling of cells in the high-dose therapy groups
Cytoplasmic fluorescent dyes were used to label MSCs and CLCs for rats in the high-
dose cell therapy groups. MSCs were labeled with CFDA-SE, while CLCs were
labeled with CM-DiI. More than 99% of either cell types were labeled and no
significant differences were observed in the respective labeling efficiencies of
cytoplasmic stain. More importantly, viability of either fluorescence-labeled cell types
remained high at 92.8 ± 1.5% in CFDA-labeled MSCs and 98.9 ± 0.5% in CM-DiI
labeled CLCs. The high viability rates indicate that the fluorescence dyes used to
label the respective cell types did not have adverse cytotoxic effects on either cell
types and thus would not be responsible for any cell death in vivo following
transplantation (Figure 30).
99
Low cellular integration in the low-dose cell therapy group may obscure any
discernible improvements in global contractility via ultrasound echocardiography.
The respective cell therapy groups were administered in higher doses to gain an
insight into the mechanistic benefits of cell transplantation in vivo. To determine the
effects of high-dose cell therapy on cardiac function, 5 million CM-DiI-labeled CLCs
or CFDA-labeled MSCs (n= 20 rats per therapy group) or 200 uL of serum-free
medium (n=15 rats) were injected into the infarct border regions of the myocardium 1
week after ligating the left anterior descending coronary artery of
immunocompromised rats. Left ventricular (LV) function was analyzed via
ultrasound echocardiography and pressure-volume (PV) catheterization.
100
3.5.3.2 2D M-mode ultrasound echocardiography assessments and Tissue Doppler
Imaging of Cardiac Function
Consistent with previous findings discussed in the low dose groups, statistical
analyses show that the serum-free media injected control rats demonstrated
significantly larger LV dimensions during end-systole and end-diastole and
correspondingly thinner interventricular septum (IVS) at the anterior walls (Figure
36i-iii). In contrast, cell-treated rats generally exhibited significantly thicker
interventricular septum and anterior walls (Figure 36 i-vi) and smaller LV diameters
during contraction. Correspondingly, cell-treated myocardium demonstrated smaller
LV volumes during end-systole and end-diastole with respect to control rats (Figure
36 vii-x). There was a trend towards systolic and anterior wall thickening in rats
treated with high-dose MSC-therapy. However, only CLC-therapy mediated
statistically significant improvements in systolic (Figure 36iii) and anterior wall
thickening (Figure 36 vi), which were previously obscured by large standard errors
due to large experimental variation in the low dose group. This underscores a strong
possibility that CLC-therapy may confer a more superior restraint on progressive
ventricular wall thinning and chamber dilatation as compared to MSC-therapy.
Control rats exhibited significantly larger internal diameters as was characteristic of
ventricular dilatation (Figure 36 vii-viii). In contrast, cell-treated rats demonstrated
smaller changes in LVIDes measurements with respect to control rats. Thus, cell-
therapy actively restrained the severe progression of chamber dilatation in cell-treated
rats. However, a positive change in LVIDes measurements indicates that post-therapy
LVIDes measurements were larger than post-ligation measurements in MSC-treated
101
rats. This in turn indicates that the MSC-treated myocardium still underwent a fair
degree of ventricular dilatation (Figure 36 vii-viii) following ligation of the anterior
descending artery, although not to such severe extents as characteristic of control rats.
In contrast, CLC-treated myocardium exhibited significantly reduced LV internal
diameters during end-systole and end-diastole following cell transplantation. This
shows that chamber dilation was attenuated in CLC-treated rats and treated
myocardium had undergone positive LV remodeling. Furthermore, Doppler analyses
showed that LVes volumes remained relatively constant following high-dose CLC
transplantation (Figure 36 ix-x). Taken together, CLC-mediated changes towards
thicker walls, smaller diameters and constant volumetric changes strongly suggest that
the CLC-treated myocardium did not undergo ventricular dilation but instead
exhibited a more robust contractility during end-systole. Thus, CLC-therapy may play
active roles in enhancing systolic activities in vivo. More specifically, CLCs may be
more effective than MSCs in promoting systolic recovery in the compromised
myocardium. More significantly, the combined effects of smaller LV volumes and
enhanced contractility in relation to a general preservation of the characteristic
ellipsoidal cardiac geometry of the heart further translate into enhancement in overall
contractile efficiencies such as ejection fraction and fractional shortening, in relation
to the untreated hearts.
Consistent with M-mode findings, tissue Doppler velocities also showed a significant
improvement in regional myocardial systolic wall motion in cell-treated rats (Figure
36 xi). These cell-mediated improvements in regional contractility translated into
significantly enhanced global contractility, with respect to control rats. However,
MSC-therapy appeared only to prevent further deterioration in LV ejection fraction
102
and fractional shortening as no discernible changes in global contractility indices were
observed following high-dose MSC-treatment of infarcted rats (Figure 36 xii-xiii). In
contrast, CLCs worked actively to enhance global systolic functions, thus leading to
superior improvements in EF and FS with respect to MSC-treated rats (Figures 36 xii-
xiii). These results are consistent with previous postulations that CLCs may be more
effective than MSCs in the recovery of systolic activities.
It is important to note that hemodynamic data obtained via pressure-volume
cathetheraization are measured at end point thereafter comparisons were made
between the respective treatment groups. Unlike echocardiography assessments, no
data points were obtained a week following LAD ligation, and thus improvements
made with respect to each rat could not be determined. Correspondingly, each rat
could not be normalized to itself. This also explains the differences between MSC-
mediated improvements via echocardiography assessment and conductance
catheterization approaches. Notwithstanding, similar trends were observed with the
data obtained via echocardiography and PV catheterization.
Ejection fraction is amenable to changes in heart rate that may be affected by external
factors such as stress and anesthesia etc. In particular, EF may be confounded in
small, fast-beating rat hearts. Although recordings of echocardiographical data
acquisition was recorded to circumvent potential confounding effects of anesthesia,
the conventional clinical indicators of cardiac improvement may not present as good
indices of actual improvements in contractility in rat hearts which beat at heart rates
of 400 bpm. ECG tracings in ultrasound assessments showed that heart rate was
consistent across all treatment groups although hemodynamic measurements obtained
103
with an ultra-sensitive 2F catheter designed specifically for small animals detected
significant decreased in heart rate in the control rats.
Heart rate-independent contractility indices were obtained as an alternative to
conventional clinical indicators of heart function. Consistently, these indices showed
similar trends. As shown in Figure 36 xiv, either cell-based therapy effected
significantly improved velocities in circumferential fiber shortening (Vcfc).
Compared to fractional shortening, the rate-corrected Vcfc offers the advantage of
being relatively heart rate and preload-independent over a physiological range.
However, only CLC-treated rats demonstrated significant improvements in the
myocardial performance index (Tei index), which is in turn a reflection of both global
systolic and diastolic functions. The advantage of this index includes less dependence
on heart rate and blood pressure132
(Figure 36 xv). It is also independent of influences
due to geometrical ventricular changes. These HR-independent echocardiography-
derived contractility measures are discussed in greater depth in Chapter 4. Figure 36
showed that the MSC-treated myocardium persistently exhibited no discernible
changes in global contractility from the point of LAD ligation. Thus, while high-dose
MSC therapy prevented further deterioration in overall cardiac performance via active
restraint of negative LV remodeling effects, MSCs did not appear to contribute
actively towards contractile activities.
104
Figure 36 i-vi. Ultrasound M-mode assessments show that high-dose cell therapy alleviates (negative) LV remodeling by limiting progressive
thinning of the interventricular septa and anterior walls. Correspondingly, this resulted in enhanced systolic and anterior wall thickening. However,
only CLC-therapy achieved statistical significance with respect to control rats. Sytolic wall thickening is a regional contractility measure of the IVS
wall motion, while anterior wall thickening is a regional contractility measure of the mid-anterior walls, which were affected upon LAD-ligation.
…continued next page.
105
Figure 36 vii-x. Consistent with thicker wall, rats treated with high-dose cell therapy
demonstrated smaller internal diameters and smaller cardiac volumes as compared to control
rats.
…continued next page.
106
Figure 36. Ultrasound echocardiography assessments show that cell-based therapy led to significant
improvements in cardiac function as compared to control rats.
However, despite significantly enhanced myocardial systolic velocities, MSC-treatment appears only to
prevent further deterioration in function, while CLC-therapy actively improved global contractility such as
ejection fraction and fractional shortening, as well as the heart rate- independent contractility indices
including (xiv) Vcfc (velocities of circumferential fiber shortening) and (xv) the myocardial performance
index, which is a measure of both systolic and diastolic activities.
107
3.5.3.3 In vivo Hemodynamics via Pressure-Volume Catheterization
In vivo hemodynamic measurements obtained with pressure-volume catheterization
coincided with ultrasound echocardiographical assessments. A 2F catheter specially
designed for rats was inserted into the right carotid artery and advanced into the left
ventricle to obtain steady-state hemodynamic measurements in real time. Load-
independent contractility indices were obtained via the occlusion of the inferior vana
cavae (Table 5). The rightward shift of the pressure-volume loops in Figure 32 shows
that cardiac volumes were significantly increased from baseline levels following a
myocardial infarction. This trend was consistent in all infarcted control rats. PV
relationships show that cell-based therapies led to significantly smaller volumes.
These observations are consistent with regional 2D M-mode measurements depicting
smaller LVID (LV dimensions) measurements and thicker interventricular walls
during contraction and are supportive of previous findings, which reported cell-
mediated restraint of LV dilatation. Consistent with low-dose cell therapy, PV
relationships also reported similar trends in cell-mediated improvements in other
parameters of cardiac function (Table 6).
However, ultrasound echocardiography assessments show that high-dose CLC-
therapy may confer superior benefits of global systolic activities with respect to high-
dose MSC-therapy. Consistent trends were observed in contractility measurements
obtained via PV catheterization. To obtain contractility indices, the inferior vena cava
was depressed lightly with a cotton swab for a few seconds in each rat. The pressure
volume loops depicted in Figure 37 were obtained from representative rats in each
treatment group. The blue line shows robust end-systolic PV relationships in a healthy
rat, as represented by its steep gradient. These were severely compromised following
108
a myocardial infarction, as observed by the flatter gradient in the red line (Figure 38
i). Figure 38 shows that cell therapy significantly enhanced end-systolic PV
relationships. However, CLC-treated rats mediated a more enhanced recovery of
ESPVR (Figure 38 ii). More remarkably, CLC-therapy restored myocardial systolic
performance, as ejection fraction and end-systolic pressure-volume relationships in
CLC-treated rats reverted to baseline levels in normal rats (Figure 38 iii).
Consistent trends were observed with preload independent contractility indices such
as maximum dP/dt-EDV, ESPVR and PRSW as CLC-treated rats demonstrated
superior systolic activities with respect to both control and in particular, MSC-treated
rats (Table 5). These persistent trends in ultrasound echocardiography assessments
and PV relationships show that CLC-therapy was more effective in improving
myocardial systolic activities with respect to MSC-therapy. More importantly,
contractility studies further show that CLC-therapy restored myocardial systolic
performance to healthy levels.
In contrast, Figure 39 shows that preload-independent contractility measurements in
MSC-treated rats were not significantly different from control animals, although
MSC-treated rats demonstrated a trend towards enhanced contractility. Consistent
with previous findings, this may be reflective of the possibility that MSC-mediated
cardiac repair is of a passive nature. These indices are discussed further in Chapter 4
of this thesis. Nevertheless, MSC-treated rats showed a trend towards enhanced
contractility with respect to control rats (Figure 38-39).
109
Figure 37. Pressure-Volume loops of a representative normal and
(serum-free) medium injected control rat post myocardial
infarction (post-MI) respectively.
The rightward shift of the pressure-volume loops shows that cardiac
volumes were significantly increased from baseline levels following a
myocardial infarction. This trend was consistent in all infarcted
control rats.
Table 5.Contractility measurements obtained via real-time occlusion of the inferior vena cava
at 6 weeks post-therapy in the high-dose therapy groups.
NS = no statistical significance.
Contractility
Occlusion
Parameters
(Mean ± SD)
Baseline
(n=8 rats)
Medium-
injected
control
(n=11 rats)
MSC-
therapy
(n=13 rats)
CLC-therapy
(n=13 rats) p-value
ESPVR 2.69 ± 1.13 0.77 ± 0.23* 1.36 ± 0.40# 2.10 ± 0.89#,†
*p < 0.02 vs.
baseline
#p ≤ 0.001
vs. control
†p < 0.05 vs.
MSC-therapy
EDPVR 0.06 ± 0.04 0.06 ± 0.06 0.04 ± 0.08 0.05 ± 0.07 p = NS
Preload
Recruitable
Stroke Work
133.3 ± 44.5 53.4 ± 16.6* 83.4 ± 31.8* 121.0 ± 31.3#,†
*p < 0.01 vs.
baseline
#p < 0.05 vs.
control
†p < 0.05 vs.
MSC-therapy
Maximum
dP/dt - EDV 147.7 ± 53.5 44.0 ± 33.8* 77.2 ± 23.9* 120.0 ± 65.4#
*p < 0.02 vs.
baseline
#p ≤ 0.001
vs. control
Emax 3.71 ± 1.49 1.47 ± 0.48* 2.04 ± 0.57 2.59 ± 1.08#
*p < 0.05 vs.
baseline
#p < 0.05 vs.
control
110
Figure 38. Pressure volume relationships of a representative healthy rat vs. a sham- operated rat.
The top-leftmost points of a single pressure-volume loop represent end-systole in a single cardiac cycle, and the gradient of the line connecting all the
top-leftmost points of each PV loop represents the end systolic pressure-volume relationships (ESPVR) and is a measure of myocardial systolic
contractility. The steep gradient of the blue line is consistent with a robust contractility that is typical of a healthy myocardium, while the flatter
gradient in the red line indicates a decrease in contractility in control rats as the PV loops shift towards the right. Figure 38ii shows increasingly
steeper gradients with cell-treated rats with respect to control rats. (iii.) However, CLC-treated rats demonstrated superior myocardial systolic
performance to MSC-treated rats as CLC-treated rats exhibited significantly enhanced ESPVR than MSC-treated rats.
111
Figure 39. CLCs persistently mediated superior enhancement of preload independent contractility indices with respect to control rats.
Similar trends were observed in preload-independent contractility indices such as (i) the preload recruitable stroke work and the (ii.) maximum dP/dt –
EDV and the (iii.) maximum time-varying elastance. CLC-treated rats showed persistently better systolic activities than MSC-treated rats in other
measures of systolic performance, including the preload-independent contractility indices such as the preload recruitable stroke work and the
maximum dP/dt – EDV.
112
3.5.4 Dose-dependent effects of cell therapy on cardiac function
Low-dose cell-therapy was unable to detect significant improvements in regional
myocardial contractility via 2D ultrasound echocardiography assessment. Moreover,
low engraftment frequencies obscured insights into in vivo cell-mediated repair
mechanisms.
A comparative study was thus conducted to compare the dose-dependent effects of
cell therapies on cardiac function. To conduct multiple pairwise comparisons between
the high- and low- dose therapy groups, control animals for the respective treatments
were combined to form a single control group. Figure 30 documents the number of
rats in the respective low- and high- dose cell therapy groups that survived each
analytical/ experimental time point. The relatively high experimental attrition
observed in post-therapy cardiac assessments was not surprising in a survival model
and was further intensified by a learning curve associated with unfamiliarity with
relatively novel experimental techniques in pressure-volume relationships at the point
of investigation.
Consistent trends in improving cardiac function were observed in high- and low- dose
cell therapy. Functional assessments via ultrasound echocardiography and PV
catheterization showed that cell therapy was generally more effective when
administered in higher doses as evidenced by significantly smaller LV internal
diameters and persistently better improvements in ejection fraction, cardiac output,
stroke volume and stroke work with respect to control animals (Table 6 and Figure
40).
113
However, trends in echocardiography and pressure-volume relationships persistently
showed that CLC-therapy mediated superior improvements in cardiac repair with
respect to MSC-therapy. Regional wall assessments via 2D M-mode
echocardiography showed that high-dose CLC-therapy mediated a negative change in
LVIDes following cell transplantation (Figure 36 vii). This suggests smaller internal
diameters post therapy from the previous echocardiographical assessment post-
ligation. Consistently, post-therapy LVIDed measurements were also smaller with
respect to the post- LAD ligation time point. Overall, these trends suggest smaller
cardiac volumes with respect to control rats, which combined with a general
preservation of cardiac geometry would translate into enhanced contractile
efficiencies. Indeed, PV relationships show that high-dose CLC-therapy trended
towards the largest improvements with respect to control animals. Additionally, rats
treated with a high dose of CLCs were the only ones to exhibit statistically significant
improvements in maximum dP/dt levels (Figure 36 iii).
Comparatively, MSC-therapy improved cardiac function with respect to control.
However, unlike CLC-therapy, MSC-therapy appeared only to play passive roles in
vivo. Despite the increase in dosage, MSC-treated rats demonstrated no discernible
changes in ejection fraction, following transplantation therapy (Table 6 and Figure 41
iv). Thus, MSC-therapy appears only to prevent further deterioration in contractile
function (Figure 36 xii-xv). Notwithstanding, MSC-treated rats consistently exhibited
thicker interventricular septum walls and correspondingly smaller LV internal
diameters, which are in turn indicative of possible attenuation in remodeling effects.
Thus, MSC-therapy does not contribute actively to contractile performance but
appears only to partake in preventing progressive wall thinning.
114
Table 6. Ultrasound echocardiography assessment of rats treated with high- and low-dose cell therapy.
Measurements obtained via 2D ultrasound echocardiographical assessments show that high-dose cell therapies were more generally more effective than
low-dose cell therapy with respect to control rats. The low- and high-dose therapy groups were conducted at different time points. The respective control
groups for the respective low- and high- therapy groups were combined to form a single control group. Experimental attrition was responsible for the
different number of rats used in each treatment group. Breakdown of the sample size with time is explained in Chart 3
Multiple
pairwise
comparisons
Analysis of cell-mediated changes in cardiac function by Ultrasound
echocardiography assessments (Mean ± SD) Medium-
injected control
n = 25
MSC-Therapy
(1 million cells)
n = 8
MSC-Therapy
(5 million cells)
n = 20
CLC-therapy
(1 million cells)
n = 9
CLC-therapy
(5 million cells)
n = 19
p-value
% changes in:
Heart Rate -7.64 ± 6.78 -1.41 ± 13.69 -1.82 ±13.03 -6.68 ± 11.51 -5.02 ± 11.35 NS
IVSes -0.73 ± 27.50 14.66 ± 33.38 7.12 ±14.31 12.7 ± 18.71 13.17 ± 12.53 NS
IVSed -1.97 ± 13.5 6.61 ±17.81 3.3 ± 7.13 2.68 ± 9.59 9.49 ± 7.45* *p = 0.008 vs. control
(Tukey‟s HSD)
LVIDes 21.98 ± 22.29 16.09 ± 35.44 4.56 ± 13.46# 9.49 ± 20.60 -3.39 ± 11.41*
#p = 0.034 vs. control
*p = 0.001 vs. control
(Tukey‟s HSD)
LVIDed 12.60 ± 10.57 12.23 ± 15.79 3.49 ± 7.74# 10.00 ± 13.18 2.69 ± 7.81*
#p = 0.033 vs., control
*p = 0.018 vs. control
(Tukey‟s HSD)
Systolic Wall
Thickening 29.79 ± 116.27 43.30 ± 86.19 15.19 ± 39.11 48.88 ± 71.77 17.85 ± 42.97* NS
Ejection
Fraction -8.72 ± 14.74 0.76 ± 28.99 -0.76 ± 7.09 5.87 ± 21.71 7.14 ± 8.39*
*p = 0.001 vs. control
(Games-Howell)
Fractional
Shortening -10.25 ± 16.85 -333.9 ± 502.2 -0.11 ± 10.12 8.51 ± 24.33 9.88 ± 12.08 NS
Mid-Anterior
Wall Velocity -1.85 ± 111.68 28.18 ± 87.60 29.53 ± 24.72 53.82 ± 99.87 25.95 ± 40.67 NS
115
Assessment of dose-dependent effects of cell-therapy on cardiac function by ultrasound echocardiography
Figure 40. 2D M-mode echocardiography assessments show that high dose cell therapy led to significantly LV small internal
diameters.
However, only CLC-treated rats mediated the highest improvement in ejection fraction relative to control animals. Control animals in the
respective treatment groups were combined to form a single control group, to conduct multiple pairwise comparisons between the high- and
low- dose therapy groups.
116
Assessment of dose-dependent effects of cell-therapy on cardiac function by real-
time pressure-volume catheterization
…continued next page
117
Figure 41. Cell therapy is generally more effective when administered in a higher dosage as high-dose therapy effected
persistently better improvements in cardiac function.
118
Table 7. Steady state hemodynamic measurements via real-time in vivo pressure-volume cathetherization showing trends in diastolic activities.
Control rats showed diastolic dysfunction with respect to normal rats. In general, cell-therapy trended towards improved diastolic activities. However, of
the 4 different therapies employed, rat treated with high dose MSC-therapy trended towards the highest EDP, and lowest minimum dP/dt and maximum
dV/dt levels, and thus mediated the worst improvements in diastolic activities as compared to both control as well as the other cell treated rats. A similar
trend was observed with the preload-independent Tau (weiss) relaxation index as high dose MSC-therapy mediated the worst improvement in relaxation
properties with respect to the other cell-therapies. However, unlike the load-sensitive indices, the Tau (weiss) relaxation index in cell-treated rats
showed a trend towards enhanced relaxation with respect to control animals.
Hemodynamic
Measurements
via PV
catheterization
(Steady-state
diastolic
indices)
Normal
(n=8 rats)
Media-
injected
control
(n= 23 rats)
Low dose
MSC-therapy
(n = 9 rats)
High dose
MSC-therapy
(n=18 rats)
Low dose CLC-
therapy
(n=9 rats)
High dose CLC-
therapy
(n= 15 rats)
p-value
Load-dependent indices (Mean SD)
Heart Rate
(bpm) 422.24 29.90 345.04 51.83 385.78 23.68 365.77 57.15 401.90 37.75 368.09 48.41 NS
End-Diastolic
Pressure
(mmHg) 11.03 2.92 12.33 17.34 11.18 9.39 13.96 8.34 8.50 1.94 11.72 2.98 NS
End-Diastolic
Volume (uL) 115.58 23.92 192.55 58.90 148.93 49.02 146.79 28.93 139.11 45.17 145.79 28.34 NS
Minimum dP/dt
(mmHg/s) 12151 3679 7859 3694 11118 4744 9340 3361 10160 2839 10611 2544 NS
Maximum
dV/dt (uL/s) 3670 1202 3634 1714 5101 2473 4516 2251 5694 3675 5842 2547 NS
Load-independent indices (Mean SD)
Tau (weiss)
(Load
independent) 9.26 1.68 13.62 6.40 12.11 3.57 12.49 3.89 10.13 1.43 11.47 2.55 NS
119
3.5.5 Cell-mediated Cardiac Repair Mechanisms In vivo.
3.5.5.1 Donor cell engraftment and survival
It is imperative that the transplanted cell types integrate with the host myocardium in
order to mediate synchronized contractility and avoid potentially lethal variations in
the electrical conductance of the heart133
. Immunofluorescence stainings show that
both cell types engrafted in the host myocardium. Green fluorescence in Figure 42A
denotes the presence of MSCs, while red fluorescence in Figure 42B depicts the
presence of engrafted CLCs in vivo. MSCs were further verified with a specific
antibody (green fluorescence) towards fluorescein (CFDA), while the presence of
CLCs were further affirmed with the use of a specific anti-human nuclei antibody.
MSCs did not stain positive for human nuclei. The absence of human genetic material
strongly suggests that MSCs did not survive well after their initial engraftment into
the host myocardium, although a residual presence of the fluorescein label used to
track the donor MSCs in vivo could still be detected in the host myocardium and
vasculature. Collectively, these observations suggest the immunocompromised host
myocardium may have retained the green fluorescent labels reminiscent of these cells
following their mortality in vivo. Nonetheless, the remaining green fluorescence was
useful in identifying the last sites of MSC engraftment before their subsequent
demise. In contrast, CLC-treated myocardium appeared to show higher incidences of
red fluorescence in vivo that were coincident with human nuclei. This in turn strongly
suggests that CLCs may have a more robust survival rate as compared to MSCs.
3.5.5.2 CLCs but not MSCs enhanced cardiac contractility via myocyte-
replacement
CLCs exhibited mature cross-striated fibers in vivo that aligned with and were
indistinguishable from native cardiac myofibers in the myocardium, as depicted by
120
the respective cross-striating I-band and A-band of cardiac -actinin and -myosin
heavy chain, troponin-I and troponin-T (Figure 43). A 3-dimensional representation
of an engrafted CM-DiI labeled CLC in the host myocardium further shows that the
cross striations, exemplified by cardiac -actinin, were indistinguishable from the
host myocardium (Figure 44). More importantly, positive connexin 43 stainings in the
viable myocardium adjacent to engrafted CLCs indicate the formation of gap
junctions between the transplanted cells and the host myocardium (Figure 45). This
strongly suggests that CLCs may become cardiomyocytes that can couple electrically
with host cardiomyocytes, and are thus likely to contribute actively towards
contractile performance in the infarcted myocardium. This phenomenon was visibly
absent in the MSC-treated myocardium (Figure 43A-B, 44-45), and is consistent with
earlier functional analyzes via ultrasound echocardiography and PV catheterization,
which show that CLCs are more effective than undifferentiated MSCs in enhancing
myocardial systolic performance.
Immunofluorescence microscopy of the MSC-treated myocardium showed an
apparent alignment of green fluorescence with native cardiac myofibers, some of
which were encased by red fluorescence denoting positive myocyte stains (Figure
42B). These observations suggest that some CFDA-labeled MSCs may have engrafted
in alignment with cardiac myocytes in the MSC-treated myocardium. However, only
a minority population of CFDA-labeled MSCs was found engrafted in the
myocardium (Figure 43B). Of the cells that successfully engrafted in the myocardium,
MSC populations exhibiting a singular green fluorescent staining were visibly
prevalent in the collagen I-enriched domains of the compromised myocardium.
Notwithstanding, these staining patterns appeared to assume a vascular-like
appearance (Figures 42-43 and 45). These observations accentuate the lack of
121
coincident myocyte staining patterns that were contiguous with the engrafted cells.
This in turn suggests that most of the donor MSCs did not integrate with the resident
myocardium upon their initial engraftment and may have remained quiescent in vivo
following cell injection into the peri-infarct borders of the myocardium.
Consequently, the MSC-treated rats demonstrated unremarkable frequencies of actual
integration as functional cardiomyocytes (Figure 43), as compared to CLC-treated
animals. Furthermore, MSCs did not engraft at sites that were proximal to connexin
43 (Figure 45). This suggests that MSCs did not form gap junctions with resident
cardiomyocytes and were most likely electrically isolated from the host myocardium.
These findings are consistent with MSCs‟ significantly diminished capacity to
enhance myocardial contractile performance in relation to CLCs. Nonetheless,
functional assessments via echocardiography and PV catheteriazation have shown that
MSC-therapy conferred some benefits with respect to control rats, as exemplified by
significantly enhanced global contractility with respect to control myocardium that
received serum-free media. These results are consistent with current literature which
reports that MSCs do not survive well in the myocardium49, 50
and actual bone marrow
stem cell-derived cardiomyocyte engraftment is rare and an inefficient process59
. Also
consistent with current postulations, MSCs may contribute towards overall
contractility via passive paracrine-mediated non-myogenic mechanisms.
122
Figure 42. Engraftment of the respective cells types in a representative rat in each respective high-dose therapy
group.
Green fluorescence in (A) denotes the presence of MSCs, while red fluorescence depicts the presence of engrafted
CLCs in vivo. CFDA-labeled MSCs were verified with the use of an anti-fluorescein antibody that recognized the
fluorescein conjugate that was used to label MSCs, while donor derived CLCs in (B) were verified with the use of a
specific human-nuclei antibody (green).
123
Figure 43. Cell fate and development of donor MSCs vs. CLCs in the injured myocardium.
Representative micrographs of CLC-treated and MSC-treated rats illustrate the respective developmental fates of
engrafted cells in the infarcted myocardium.
…continued next page
124
Figure 43. Cell fate and development of MSCs vs. CLCs in the injured myocardium (continued) A. The prevailing MSC populations are physically isolated from the myocardium. B. A minority of engrafted MSCs exhibited
an intense staining of the myocardium immediately surrounding the engrafted cells (middle panel). However, these regions did
not exhibit any distinctive cross-striating patterns that were typical cardiac myocytes. C. In contrast, red-fluorescence labeled
CLCs demonstrated well-defined cross striations consistent with a developed myocardial phenotype in vivo.
125
Figure 44. A 3-dimensional model of an engrafted CM-DiI labeled CLC cell in
the rat myocardium. Constructed from Z-sections and observed at different perspectives, three-dimensional
rendering of a z-stack of an engrafted CLC showed that cross striations denoting the I-band of
cardiac a-actinin (green) were indistinguishable from host -actinin fibers.
126
Figure 45. CLCs integrated in sites proximal to connexin 43 while MSCs did not.
This suggests that the myogenic CLCs can couple electrically with the host myocardium via the formation of gap junctions
between the engrafted cells and the resident cardiomyocytes. In contrast, MSCs were not electrically coupled with the
myocardium.
127
3.5.5.3 Cell therapy mediates cardiac repair via alternative non-myogenic
mechansims in the infarcted myocardium
Immunofluorescence examinations showed regular incidences of green fluorescence
tracking label reminiscent of MSCs in vascular structures in vivo. Figure 46 illustrates
the colocalization of fluorescein (CFDA) with the von-Willebrand factor and smooth
muscle actin. In contrast, CLCs did not integrate with any mature vascular structures
despite their occasional engraftment in sites proximal to the vascular system (Figure
46). Despite a seemingly rare incidence in the myocardium, MSC-therapy
significantly enhanced myocardial angiogenesis in the infarcted myocardium with
respect to control myocardium (Figure 48). However, there was no distinct evidence
of extended MSC-differentiation into endothelial cells in the myocardium, although
MSC-treated rats demonstrated a significantly larger number of small arterioles (6.69
± 0.53 vessels/mm2 myocardial tissue, p<0.01) in the myocardium as compared to
CLC- treated (4.48 ± 1.28 vessels/mm2, p<0.05) and control (2.44 ± 1.74
vessels/mm2) rats (Figure 48). It is important to note that the cell-treated myocardium
also trended toward greater vascular expansion into mid-and large-sized arterioles
although these trends did not achieve statistical significance with respect to control
rats. In turn, these increases in the number of arterioles in the MSC-treated
myocardium may lead to significantly enhanced myocardial perfusion of the MSC-
treated heart. Taken together, these findings strongly indicate that MSC-therapy may
act to benefit cardiac function via alternative non-myogenic mechanisms.
Although CLCs were not integrated within vascular structures in the myocardium, the
CLC-treated myocardium stained positive for the von Willebrand factor (vWF) in
focal regions that were proximal to the site of cell engraftment. However, these cells
did not stain positive for the mature blood vessel marker, smooth muscle actin
128
(SMA). Figure 47A shows the colocalization of green fluorescence depicting vWF
expression with CM-Dil labeled CLCs in vivo (yellow arrowheads). This suggests an
alternative non-myogenic differentiation pathway in CLCs, as the engrafted cells may
have differentiated to become endothelial cells in vivo. This is consistent with in vitro
characterization studies showing an inherent propensity for endothelial differentiation
in CLCs. Importantly, multiple singular green fluorescent stains (white arrowheads)
in the immediate vicinity of the engrafted CLCs strongly indicate the expression of
vWF expression in resident host cells (Figure 47A). Thus, the engrafted CLCs may
have stimulated the nascent development of resident endothelial cells, which may in
turn lead to subsequent vasculogenesis. Consistently, the CLC-treated myocardium
showed a trend towards higher counts of arterioles per mm2 of myocardial tissue
section (Figure 48) with respect to control heart sections, although this did not
achieve statistical significance.
Subsequent staining on sequential tissue cryosections also showed positive expression
of the cell proliferation marker, PCNA, in the myocardial interstices proximal to
engraftment sites (Figure 47B). PCNA expression was similarly augmented in the
same vWF-enriched locale of CLC-engraftment in the cell-treated myocardium. The
colocalization of green and red fluorescence representative of CM-DiI labeled CLCs
(yellow arrowheads) in Figure 47B suggests that these CLCs survived well and
proliferated in the treated myocardium. Subsequent immunofluorescent stainings on
sequential tissue sections showed the absence of distinct cardiac myofibrillar protein
expression in the vicinity of these proliferating cell types. Given the focused vWF-
enriched locale of CLC-engraftment (discussed in the following paragraph) as well as
an enhanced vWF expression in the engrafted CLCs, it is likely that the PCNA-
positive CLCs differentiated to become proliferating endothelial cells in vivo.
129
Singular green fluorescent PCNA stains encasing the nuclei (white arrow heads) of
host cells in the immediate vicinity of CLC- engraftment indicate the presence of
proliferating endogenous resident cells in the treated myocardium. Thus, the engrafted
CLCs may have stimulated an uncharacteristically active proliferation in terminally
differentiated host cells in vivo (Figure 47B). Given the spatial localization of these
proliferative cells in the interstitial spaces between host muscle cells and the relatively
absent myocyte expression in these PCNA-positive cells, it is highly unlikely that the
proliferating endogenous cell types were resident cardiac myocytes. The lack of
(obvious) mature vascular structures and an absent smooth muscle actin expression in
the focal regions surrounding the site of CLC-engraftment also precluded the
possibility that these cells were endogenous pericytes or mature blood vessels. On the
contrary, these observations collectively suggest the nascent development of SMA-
vWF+ endothelial cells in the synthetic phase of mitosis in the CLC-treated
myocardium. Thus, CLCs can promote the nascent development of endogenous host
cells in the locale of CLC engraftment into actively proliferating endothelial cells.
Consistently, the CLC-treated myocardium also demonstrated consistent trends in
higher counts of arterioles per mm2 of myocardial tissue section (Figure 42) with
respect to control heart sections, although this did not achieve statistical significance.
This highlights the possibility that CLC-therapy may mediate alternative non-
myogenic cardiac repair mechanisms such as neoangiogenesis in the infarcted
myocardium. Importantly, the active proliferation of donor CLCs and endogenous
resident cells underscore the likely possibility that engrafted CLCs may secrete
cardioprotective paracrine factors like cytokines into the milieu of CLC-engraftment.
These elements may in turn exert a protective effect on the compromised
myocardium. For example, paracrine factors may improve myocardial perfusion in
130
the infarcted myocardium by increasing the number of functionally viable arterioles,
albeit to a lesser extent with respect to MSC-therapy. This observation is not
surprising given that a large number of CLCs have been committed to a cardiac
myogenic lineage in vitro before cell transplantation so that CLCs‟ primary
contribution to cardiac repair is likely through myocyte replacement. Cardioprotective
cytokines may also alleviate adverse remodeling effects, thus enhancing regional
myocardial activities and contributing indirectly towards the recovery of global
cardiac functions via this non-myogenic repair mechanism.
131
Figure 46. Engrafted MSCs persistently colocalized with the von Willebrand factor and SMA in vivo.
This suggests that MSCs may have integrated as immature blood vassals, and thus may promote angiogenesis in the host
myocardium. In contrast, although CLCs may have engrafted in sites proximal to blood vessels, they did not integrate as
mature blood vessels in the treated myocardium.
132
Figure 47. Engrafted CLCs mediated cardioprotective paracrine effects, which stimulated extended endothelial
differentiation and endogenous cellular proliferation.
A. The CLC-treated myocardium promoted expression of the von Willebrand factor in the focal regions proximal to the site
of cell engraftment. B. Subsequent staining for vascular structures on sequential section positive expression of the
proliferating cell nuclear antigen, PCNA in regions proximal to the site of engraftment. The juxtaposition of red (CLCs) and
green (PCNA) fluorescence encasing some nuclei (yellow arrow heads) identifies proliferating CLCs in vivo. Thus, engrafted
CLCs survived well and demonstrated sustained proliferation in vivo. Similarly, singular green fluorescent stains enclosing
other nuclei (white arrow heads) in the interstitial spaces between muscle cells suggest that the endogenous resident cells in
treated myocardium underwent proliferation that was uncharacteristic of terminally differentiated cells.
133
Figure 48. MSC-treated rats show significantly higher numbers of small arterioles (<20um) in the myocardium with
respect to CLC-treated and control rats.
134
3.5.6 The role of collagen in the cell fate and development of CLCs in
vivo
CLCs demonstrated preferential engraftment in collagen V-expressing myofibers in
the peri-infarct borders. Figure 49 shows extensive fibrosis and active remodeling of
the myocardial extracellular matrix in the anterior wall of the left ventricle of a
control rat, following suture-ligation of left anterior descending coronary artery.
Collagen type I ECM was the predominant subtype in the subendocardium of
transmural infarcts, while endogenous collagen type I colocalized with collagen type
III expression in the pericardium and the subepicardium that extended into the
infarcted myocardium. More specifically, collagen types –I were mainly localized to
the perimysium that enwrapped large cardiac muscle bundles in the viable regions of
the infarcted myocardium (Figure 49). In contrast, predominant endogenous collagen
V expression was persistently observed in the endomysium surrounding viable
myofibers in the peri-infarct zones of the infarcted myocardium (Figure 49).
Immunofluorescent cell tracking showed the persistent engraftment of CLCs (arrows)
in the collagen V enriched endomysial space proximal to viable myofibers in the peri-
infarcted region of myocardium (Figure 50). These cells were often intimately
integrated with -actinin-, MHC- and troponin T- positive myofibers of the host
myocardium (Figure 43), and were also electrically connected to the resident
myofibers via the formation of gap junctions in the myocardium (Figures 45). As
previously established, these intimately integrated CLCs demonstrated distinct cross-
striations that were indistinguishable from the native myofibers. Consistent with in
vitro findings, endogenous myocardial collagen V matrices supported the distinct
cardiomyogenic phenotype of CLCs in vivo. At the same time, a minority population
of CLCs also engrafted within the characteristically collagen I-enriched interstitial
135
spaces bordering the infarcted areas in the myocardium. These cells did not show
distinct myogenic actinin striations and were physically isolated from the main
muscle bundles in the myocardium (Figure 51.1-51.2). This observation is consistent
with in vitro studies showing that collagen I ECM does not support a cardiac-like
phenotype in CLCs. It remains unclear if the observed localization of CLCs in the
interstitial tissues enriched in collagen -I extracellular matrices prevented their
engraftment into the host myofibrillar architecture. Nonetheless, these observations
highlight the possibility that a collagen V -directed response may be essential to
lineage differentiation and functional integration of CLCs within the cardiac
syncytium
In contrast, transplanted MSCs mostly engrafted in collagen I-enriched perimysial
tissues or infarcts that were electrically isolated from the native muscle. These stem
cells were absent of cardiac myofilaments, even as neighboring native cardiac
muscles demonstrated an ubiquitous expression of myofibrillar proteins (Figure 43).
Collectively, these findings suggest that myocardial collagen ECMs may play pivotal
roles in cell fate and development of in the infarcted myocardium134, 135
136
Figure 49. Spatial expression of endogenous collagen V in the peri-infarcted regions of the myocardium.
Severe thinning of the infarcted left ventricular anterior wall (yellow box) 6 weeks post ligation of left anterior descending coronary artery.
Magnification of the boxed area showing sub-endocardial expression of collagen I and sub-epicardial expression of both collagen I and collagen III in
the transmural infarcts. Collagen V was detected in the peri-infarcted myocardium with occasional co-localization with collagen III in the infarcted
myocardium. Focal expression of collagen V was detected in the endomysial tissues that surround viable myofibers (boxed ) on the order of the
infarct. Expression of collagen I and III was co-localized in the infarcted areas.
137
Figure 50. CLCs (yellow arrowheads)
primarily engrafted in the collagen V-
enriched endomysium that enwraps
viable myofibers in the treated
myocardium.
Figure 51. CLCs engrafted in collagenous domains in the myocardium.
CLCs that engrafted in the collagen I-enriched (1) interstices or (2) myofibers in the
myocardium did not exhibit positive -actinin stainings. 3. In contrast, CLCs that
engrafted in the collagen V-enriched endomysium sheath that encapsulates viable
myofibers showed cross striations indistinguishable from host cardiomyocytes. As
shown in Figure 43, these cells were often intimately integrated with host -actinin,
MHC- and troponin I- positive myofibers, and were also electrically connected to
the rat myofibers via gap junction formation (connexin 43) in the myocardium.
138
3.5.7 Key findings
1. Bone marrow mesenchymal stem cells (MSCs) can be differentiated into
cardiomyocyte-like cells (CLCs) via the concomitant use of myogenic-
differentiating medium and collagen V matrices in vitro.
2. CLCs expanded more than 30-fold within 2 weeks while still retaining their
distinct cardiac-like phenotype.
3. CLCs interacted preferentially on collagen V extracellular matrices.
4. The v3 and 21 integrins are involved in initial CLC attachment and cardiac
transdifferentiation.
5. Pre-differentiating MSCs into CLCs ex vivo prior to cell transplantation therapy,
commits these multipotent cells to a stable cardiac-like lineage.
6. Cell-based therapies enhanced global contractile function in the infarcted
myocardium by restraining progressive LV dilation and attenuating LV
remodeling. It is likely that the presence of cells in an otherwise non-compliant
tissue „softens‟ the infarct area, which in turn retains myocardial tissue
compliance and indirectly improves overall tissue contractility.
7. Transplanted CLCs appeared to demonstrate enhanced survival rates in host
myocardium as compared to MSCs.
8. Rats treated with CLCs demonstrated superior systolic function to rats
administered with undifferentiated MSCs.
9. CLCs actively confer added mechanical benefit by actively restoring myocardial
systolic performance.
139
10. CLCs often engrafted in the collagen V-enriched peri-infarct domains of the
myocardium that were relatively more viable.
11. These CLCs demonstrated cross-striations that were indistinguishable from host
cardiomyocytes in vivo.
12. CLCs that were localized to collagen I-enriched interstitial domains in the scar
areas of the myocardium did not achieve terminal myocyte differentiation.
13. However, CLCs that were localized in the collagen I-enriched interstices
expressed the proliferating cell nuclear antigen, and were thus actively
proliferative. These cells also expressed the von-Willebrand factor and were thus
angiogenic.
14. These cells also promoted neoangiogenesis within the locale of CLC-engraftment.
15. Transplanted MSCs did not survive well in the host myocardium.
16. MSCs contributed passively towards improved global contractility via alternative
non-myogenic repair mechanisms.
17. Robust angiogenesis in the infarcted and peri-infarcted areas of the MSC-treated
myocardium strongly suggests cardiac improvements via paracrine effects.
18. Predifferentiating MSCs into CLCs ex vivo before transplantation is more
efficacious than current undifferentiated bone marrow stem cell therapy for the
treatment of chronic heart failure. However, CLCs‟ beneficial effects may be
conditional on engraftment in relatively more viable areas in the infarcted
myocardium.
140
Chapter 4: Discussion
4.1 In vitro characterization studies
4.1.1 MSC commitment to a distinct cardiac lineage before cell transplantation
ES-derived cardiomyocytes have huge potential in clinical cell therapy for heart
failure. These differentiated cells are also more efficacious than undifferentiated
BM stem cells in restoring cardiac contractile function136
. Similar to benefits of
committing pluripotent ES cells to a stable and distinct cardiac phenotype for
cell therapy, predifferentiating autologous multipotent BM stem cells into
cardiomyocytes before cell transplantation may significantly increase the
frequency of myogenic replacement in vivo to confer enhanced regional and
global contractile performance in the failing heart. In this study, patient-derived
MSCs were found to transiently express some cardiac gene transcripts such as
GATA4, Nkx2.5, troponin C and MLC2a. Committing these MSCs into CLCs
stabilizes the cardiac-like phenotype in vitro and maximizes their survival and
extended differentiation in vivo.
4.1.2 Cardiac differentiation of MSCs into CLCs
Cardiac differentiated MSCs have also been reported to demonstrate active
cardiac-like action potentials137
, inward rectifying potassium currents135
and
rhythmic calcium transient currents55
when placed in appropriate and conducive
microenvironments. These methods are unlikely to be clinically applicable
given their low reproducibilities and 5-azacytidine‟s potentially toxic
consequences on a systemic level. This study extends from an earlier work
which reported the development of a simple and non-toxic myogenic
development medium, MDM118
. This medium is essentially free of cytotoxic
141
agents and is composed of several growth factors such as L-ascorbic acid138
and
dexamethasone139
, which are known to promote cardiac myogenesis in MSCs
and embryonic stem cells. A preliminary study comparing the respective
differentiation efficacies of MDM vs. azacytidine and butyric acid showed an
enhanced and sustained cardiac troponin -C and -T expression in MDM-treated
MSCs (Figure 8).
4.1.3 Expression profiling of CLCs .
In this study, primitive and unsorted sternum-derived mesenchymal stem cells
(MSCs) were successfully differentiated with MDM into cardiomyocyte-like
cells (CLCs) with a distinct myogenic phenotype118
. These CLCs retained some
characteristics of primitive and untreated MSCs, such as adherent-dependent
expansion and a fibroblastic morphology.
In contrast to the reportedly limited cardiomyogenic potential of MSCs140
,
CLCs exhibited distinctive cardiac-like characteristics that remained stable in
prolonged cultures, as was evident by the persistent gene and protein expression
of multiple cardiac-specific proteins including the Nkx2, GATA and MEF2
family of cardiac transcription factors, cardiac calcium channels including
IP3R2 and the voltage gated L-type calcium channel and sarcomeric
myofilaments such as troponin T and cardiac actin throughout 11 culture
passages. More than 95% of the cardiac differentiated CLC cultures stained
positive for sarcomeric -actin, sarcomeric-actinin, tropomyosin, the cardiac
troponins -I, -C and -T, the myosin light chains 1/2 and titin. Furthermore, the
appearance of nascent Z-lines rich in sarcomeric -actinin in 15-20% of MDM-
treated MSC cultures substantiated a developing contractile phenotype.
142
However, the diffused spatial distribution of the other cardiac sarcomeric
proteins in the cytoplasm and the apparent lack of spontaneous contractions in
CLC cultures suggest that these cells only attained partial differentiated
phenotype in vitro.
In this study, bone marrow stem cells were derived from cardiac patients with a
mean age of 58.9 ± 11.4 years. Although successfully committed to a stable
cardiac phenotype with Z-bands, CLCs did not exhibit more mature A-, and I-
bands in vitro. Furthermore, the absence of spontaneous beatings in patient-
derived CLC- cultures underscore the strong possibility that the bone marrow
MSCs derived from relatively old patients may not possess optimal
differentiation or cardiomyogenic developmental potential. Lin- cKit+ MSCs
have been reported to undergo cardiac transdifferentiation with a higher
efficiency in vitro as compared to primitive and unsorted bone marrow MSC
progenitor cells, although no reports of spontaneously beating Lin- cKit+ MSC
cultures in vitro have been documented. Nonetheless, isolating Lin- cKit+
MSCs from patient-derived bone marrow stem cells before their induction with
MDM may further enhance the cardiomyogenic potential126, 141, 142
of these
MSCs.
143
4.1.4 Golgi-localization of GATA4 and -MHC in patient-derived MSCs and
CLCs.
Interestingly, patient-derived MSCs and CLCs exhibited a persistent Golgi-
localization of GATA4 and MHC. However, the golgi-localization of GATA4
remains to be verified with further studies since the colocalization of GATA4
with Golgin97, a resident golgi protein in the trans Golgi network (TGN), were
only observed with immunohistochemical but not immunofluorescent
examinations. In contrast, golgi-localization of cardiac MHC has been
consistently shown in patient-derived MSCs and CLCs by 3 different cardiac
MHC- specific antibodies. More importantly, this unexpected accumulation of
MHC within the juxta-nuclear TGN complexes assumed a coincident swollen
appearance upon monensin disruption of the golgi apparatus143
. This strongly
suggests that MHC‟s association with the golgi apparatus was not fortuitious.
However, this remains to be verified with stringent coimmunoprecipitation
experimentation with resident golgi proteins. Alternatively, golgi fractions
obtained via ultracentrifugation in a sucrose density gradity may be probed with
a GATA4 and MHC-specific antibodies. Notwithstanding, golgi retention of
GATA4 and MHC did not appear to affect normal cellular function as
exemplified by CLCs‟ robust proliferation (Figure 10) and stable cardiac gene
expression (Figure 13).
Golgi complex encompasses
an actin-based filament system in which
nonmuscle actin and many of its binding partners including myosin are
implicated in Golgi complex-mediated trafficking. In particular, the myosins
that are associated with Golgi apparatus are the non-conventional myosins such
144
as the nonmuscle myosin II and myosin VI, which are not involved in
contractility but are implicated in vesicular transport of secretory/membrane
proteins144, 145
or in the maintenance of structural morphology of the Golgi
Apparatus146
. Therefore, it remains to be determined if MHC has an alternative
nonmuscle isoform which play similar developmental roles in Golgi dynamics.
The in vitro characterization studies in this thesis strongly suggest that the
persistent golgi-localization of the cardiac MHC may be a property of
undifferentiated state, rather than an aberration of MHC development since
MSCs can be induced to increase protein transport into the proper cellular
compartments upon cellular differentiation into CLCs.
Interestingly, GATA4 expression is critical for downstream MHC
expression147
. The golgi-localization of both GATA4 and the myosin heavy
chain implicates the Golgi apparatus in playing an important role in the
developmental regulation of myofibrillogenesis of MHC in quiescent MSCs.
Therefore, it is more likely that the complete maturation of the cardiac MHC
may be dependent on a protein/factors that is expressed later in the cells‟
differentiation/ development. Consistently, MSC-derived conditioned medium
has been shown to upregulate the expression levels of beta-myosin heavy chain
(MHC) in cardiac progenitor cells148
. More investigations are warranted in
elucidating the association between Golgi dynamics and MHC participation in
myofibrillogenesis.
145
4.1.5 CLCs’ contractile apparatus resemble primitive premyofibrils in early
myofibrillogenesis
The golgi-localization of cardiac HMC underscores the strong possibility that
MHC did not attain terminal maturation as a functional contractile protein in
CLCs. In addition, its insertion into nascent myofibrils was independent of -
actinin. Similarly, although other sarcomeric proteins such as myomesin and
titin were expressed, they were not also inserted into the distinct cross striations
of sarcomeres. This diffused (spatial) expression of other sarcomeric proteins
such as titin and myosin is coincident with the distinctive -actinin striations in
CLCs, and can be explained with Sanger et al‟s theory that myosin insertion
into mature myofibrils occurs later in myocyte development, after the formation
of the premyofibrils which are in turn characterized by Z-bodies enriched in -
actinin29
. This observation is also consistent with Schultheiss et al‟s postulation
that thin and thick filaments assemble independently on stress fibre-like
structures during early myofibrillogenesis12
. Taken together, in vitro
characterization studies strongly suggest that CLCs resemble immature
premyofibrils28
during early myofibrillogenesis. Importantly, patient-derived
CLCs are similar to ventricular heart cells, which retain the expression of the
cardiac MHC and also do not exhibit spontaneous beatings or contractions.
However, ventricular heart cells reportedly limit negative LV remodeling and
preserve systolic function when implanted into an injured region
149. Thus, we
expect that CLCs, being a mesenchyme derivative, would confer similar if not
more enhanced improvement in post-infarct LV function.
146
4.1.6 Effect of ECM in vitro on CLCs
This thesis reports that Collagen V significantly enhanced the initial cellular
adhesion and subsequent attachment-dependent expansion of CLCs as
compared to MSCs. Cellular interaction with collagens is predominantly
mediated via the engagement of the 11, 21 and V3 heterodimeric
integrins104
. The augmented adhesion may likely be attributed to relatively
higher expression levels of the 21 and V3 integrins in CLCs as compared to
MSCs. Consistently, the single 1 integrin subunit is known to mediate cellular
adhesion and proliferation of bone marrow stromal cells104
. Similarly, the V3
integrin heterodimer has been implicated in cellular attachment of muscle
precursor cells on ECM150
. The commercial unavailability of the 11 integrin
antibody precluded a detailed examination of its role in mediating cellular
attachment to collagen -I and -V matrices in this study. Notwithstanding, the
11 integrin is believed to play relatively minor roles as compared to 21
integrin in binding the helical chains of collagen V 151
. More specifically,
11 integrin reportedly binds native collagen V, while 21 and v3 integrins
interact with collagen V via unmasked Arg-Gly-Asp (RGD) sequences of
denatured collagen V151-153
that was used in the current study.
The augmented cellular adhesion on collagen V substratum may possibly be
attributed to relatively higher expression levels of the 21 and V3 integrins in
CLCs as compared to MSCs. Consistently, the single 1 integrin subunit is
known to mediate cellular adhesion and proliferation of bone marrow stromal
cells 104
. Similarly, the V3 integrin heterodimer has been implicated in cellular
147
attachment of muscle precursor cells on ECM150
. Notwithstanding, the 11
integrin is believed to play relatively minor roles as compared to 21 integrin
in binding the helical chains of collagen V151
. The integrin inhibition assays
conducted in this study indicated that CLC adhesion on collagen -I and -V
matrices may be predominantly mediated via the engagement of the 1 integrin.
In contrast, v integrin may only play a secondary role in CLC adhesion on
collagen V as compared to collagen I. However, it remains to be determined if
differential integrin signaling events were responsible for the enhanced
adhesion and proliferation of CLCs as a consequence of their preferential
interaction with collagen V matrix.
Importantly, the v3 integrin heterodimers are implicated in cardiac
development. The expression levels of the v3 integrin heterodimers are
upregulated 1-3 weeks post myocardial infarction while the modulated spatial
distribution of v integrin and upregulation of 3 integrin is correlated with the
growth and survival of 1 integrin knock out cardiac pacemaker-like cells
during early cardiac development. However, v and 3 integrins cannot
compensate the loss of 1 integrin function during the terminal differentiation of
cardiac cells, thus implicating specific 1 integrin -induced cardiac
specialization in cardiomyocytes154
. Consistently, this study shows the
modulated expression of the GATA4, a cardiac transcription factor involved in
early cardiac development, upon functional v3 integrin inhibition.
This study shows that V3 integrin neutralization resulted in a muted response
of cardiac specific genes such as Troponin T, skeletal muscle actin and
148
ryanodine receptor 2. These observations strongly indicate that collagen V may
exert a pro-cardiogenic effect via the engagement of the V3 integrin
heterodimer in CLCs. It is important to note that the myogenic development
medium used to differentiate MSCs into CLCs contains insulin, while blocking
ligand occupancy of the v3 integrin have been shown to inhibit insulin growth
factor 1 signaling155. The v3 integrins interact with collagen V via unmasked
Arg-Gly-Asp (RGD) sequences of denatured collagen V151-153
. Thereafter, v3
integrin clustering is induced at a focal adhesion point which subsequently
activates insulin growth factor 1 signaling that may in turn be responsible for
cardiac gene expression in CLCs. Expression of GATA4 and Nkx2.5
transcription factors is associated with extracellular regulated kinase 1/2 (ERK
1/2)156, 157
, which acts downstream of the integrin-mediated signaling
pathway158
.
Moreover, the V3 integrin has been reported to enhance cardiac
differentiation of P1 embryonic carcinoma cells159
. PDGF- mediated signaling,
which reportedly influence cardiac differentiation in MSCs118, 160
is similarly
implicated with V3 integrin heterodimerization161
. More importantly, the V3
integrin reportedly blocked matrix contraction of collagen V but not collagen I
gels 128
. The results of the in vitro characterization study conducted in this study
suggest that the V3 integrin may be a dominant effector of cardiac
differentiation in CLCs as a collagen V-centric response. However, it is
important to note that functional blocking of the relevant integrins in CLCs
cultured on collagen V only demonstrated a selective attenuation of troponin T,
sarcomeric -actin and RyR2 gene expression. This strongly implicates an
149
alternative or compensatory collagen V –centric response on cardiac
differentiation in CLCs that is independent of the relevant integrins, especially
since collagen V is a multi specific ligand for other receptors such as
thrombospondin162
and heparan sulphate163
among other receptors.
Alternatively, this suggests that insulin growth factor 1 may directly influence
the expression of these cardiac genes in MDM-derived CLCs. Unexpectedly,
the results in this study showed that neutralization of either v3 did not affect
GATA4 expression in CLCs. This would imply that GATA4 expression was
independent of V3 integrin mediated signaling. However, GATA4 expression
was assessed after 7 days of the second V3 integrin inhibition. Being a
transcription factor that is involved in early cardiac development, it is possible
that the effect of functional V3 integrin neutralization may be almost
immediate or earlier than 7 days. A quantitative assessment of GATA4
expression a few seconds to a day after integrin neutralization may provide
some insights into the role of V3 integrin on GATA4 expression.
150
4.2 In vivo functional studies
Bone marrow-derived stem cell therapy presents as a promising therapeutic modality
for myocardial infarction (MI)78, 82, 164, 165
. However, there are inconsistencies in this
approach as recent clinical studies have not been able to readily reproduce the
findings of preclinical investigations and early clinical studies84, 89, 166-169
. In addition
to these discrepancies, the fate and development of the stem cells in the injured
myocardium is also controversial170-177
. Contrary to in vitro findings, which show that
MSCs can differentiate into cardiomyocytes, endothelial cells and smooth muscle
cells55, 118, 160, 178, 179
, other groups have suggested that the significant benefits of cell
transplantation therapy may be attributed to a more potent cardioprotective paracrine
effect as compared to direct myocyte replacement via extended stem cell
differentiation in vivo180-183
. However, most of these clinical trials have employed the
use of primitive, undifferentiated bone marrow stem cells, which may be less efficient
than cardiac-differentiated cells in recovering contractile performance in the infarcted
myocardium.
4.2.1 Optimal time for cell transplantation
Injecting the right cell type at the right time and in the right place is crucial in
maximizing cardiac repair. Cells that were transplanted into the compromised
myocardium 7 days following LAD-ligation as myocardial infarcts were only
clearly visible from echocardiography assessments from the 4th
day of ligation.
Furthermore, a second operative procedure proved to be too traumatic for the
immunocompromised rats in this survival model, which had been injected daily
with cyclosporine from the 3rd
day of ligation. Preliminary investigative efforts
strongly suggested that rats which had received a second operation before the
151
5th
day of ligation usually died mid-procedure or of post-operative
complications such as pneumonia; while rats that were subjected to a second
operation more than a week following LAD-ligation resulted in massive scarred
tissues that were often adhered to the chest wall. Thus, hearts were not easily
accessible via mid-or left thoracotomy for cell transplantation.
Furthermore, MSCs have been reported to exhibit poor survival rates when
transplanted into the myocardium 4 days after LAD-ligation184
, as only 0.44%
of the total MSC population injected were found to survive the transplantation
procedure. In contrast, MSC transplantation reportedly received the best
functional recovery in terms of global myocardial contractility, anti-apoptosis
and angiogenesis 1 week after myocardium infarction185
.
4.2.2 Dose dependent contribution of cell therapy
4.2.2.1 Donor cell survival
None of the MSCs were detected in frozen rat cryosections, while only trace
number of CLCs was detected via histological approaches in the low dose
therapy groups. In contrast, high-dose therapy appeared to exhibit a larger
donor cell presence in vivo. However, the fluorescent label used to track MSCs
in vivo did not colocalize with human nuclei although the myocardial or
interstitial tissues in the immediate locale of initial cell engraftment may have
retained the fluorescent label remnants of MSCs and can therefore still be
detected via immunofluorescence examination. This underscores a limitation
in current techniques in long-term cell tracking. Given that resident cells may
uptake or take a longer time to degrade the cell labels, it is important to verify
152
the identity of donor cells via the presence of human-specific markers or
genes.
It is likely that MSCs did not survive well in the myocardium following cell
transplantation. The absence of BrdU-labelled MSCs in the low-dose therapy
group as well as the trace presence of fluorescein in rat myocardium treated
with a high dose of MSCs strongly suggests a low survival rate in vivo. These
observations are consistent with current literature findings which report poor
donor MSC engraftment and survival following injection into the myocardium
regardless of dose and time of application184, 186
. Specifically, MSC presence
reportedly decreased from 34-80% to only 0.3-3.5% of the total MSC
population injected 6 weeks post transplantation186
.
In contrast, the CLC-treated myocardium showed a relatively larger presence
of CM-DiI labeled CLCs, which also stained positive for human nuclei. This
strongly implies that the CLCs that engrafted in the infarcted myocardium
may be more resilient than MSCs in surviving the traumatic process of direct
syringe injection into the myocardium and/or the harsh ischemic
microenvironment of an infarcted myocardium, thereby exhibiting a more
robust survival rate in vivo. Thus, predifferentiating MSCs into CLCs ex vivo
prior to cell transplantation therapy may enhance the donor cell survival rates
in vivo. However, these observations remain to be vindicated with a
quantitative flow cytometry analysis or fluorescence activated cell sorting
(FACS) comparing the survival of MSCs and CLCs in the infarcted
myocardium. Future experimentation should also include tracing the
fluorescence-labeled MSCs and CLCs 1, 2, 4, 6 and 8 weeks post
153
transplantation in order to gain some insight into their cell fate and
development, including survival and integration, in vivo.
4.2.2.2 Postulated role of myocardial ECM on donor cell survival
A typical type I collagen fibril comprises of a core of collagen V encapsulated
by polymerized collagen -I and -III93, 95
. The ubiquitous collagens types I-III
confers mechanical strength onto the collagen fibril, while the core collagen V
fibrils contribute towards overall fibril morphology and therefore tissue-
specificity. Besides conferring structural support and integrity, collagens
reportedly promote the cellular adhesion and expansion of as well as
differentiation in stem cells187-189
For instance, the myocardial extracellular
matrices have been reported to influence embryonic differentiation towards
cardiomyocytes190
. Collagen I represents the major components of
extracellular matrices in the myocardium while collagen V is a minor subclass
of collagens that regulates matrix assembly and possibly matrix stiffness191, 192
.
Cellular adhesion to denatured and native collagen V ECM is mediated by
distinct sets of RGD-dependent and RGD-independent receptors. Importantly,
the interstitial collagen I also interacts with collagen V in the same way, so
that when included in the triple-helical conformation of collagens, RGD
sequences in collagen V are either not accessible to cells or exhibit specific
conformations recognized by different integrins152
.
Active ECM remodeling following a myocardial infarction alters the local
collagen composition as well as the integrins that constitute the interface
between myocytes and the myocardial matrix193
. This study reports an
154
enriched collagen V deposition in the focal regions in the relatively more
viable peri-infarct borders in relation to the scar tissues of the CLC-treated
myocardium. As discussed in the preceding paragraphs, collagen V forms the
core of type I collagen fibrils. It is not surprising therefore that it is often
shielded by other more dominant collagen types in the myocardial
extracellular matrix128, 152
. It is possible that the RGD sequences usually
„hidden‟ within the triple helices of collagen fibrils may be unmasked in this
process, so that these RGD sequences are now available for interaction with
donor cells or interstitial collagen I molecules, which are significantly over-
expressed in the infarcted myocardium. Consistent with their preferential
affinity in vitro, the persistent localization of CLCs to collagen V-
encapsulated endomysial space of viable myofibers strongly suggests a
specific interaction between the engrafted CLCs and the myocardial matrix
components.
This favorable „tethering‟ of CLCs to the relatively more viable domains of
the myocardium that is characterized by elevated collagen V deposition may
provide a conducive microenvironment that is imperative for donor cell
survival and proliferation194
. This may in turn explain CLCs‟ relatively
enhanced survival in the myocardium as compared to MSCs. In contrast,
MSCs that were localized to collagen I-enriched domains within the
myocardial interstices of the infarcted myocardium were physically isolated
from the resident cardiomyocytes. Just as donor cells cannot survive, grow or
proliferate in the absence of physical tethers to functional myocardial
components or matrix elements, so the physical isolation of MSCs from
155
resident myocytes may have been responsible for diminished survival after
myocardial transplantation.
Additionally, collagen V supports a robust migration of cells belonging to a
mesenchymal lineage in vivo195, 196
. Given that CLCs are essentially derived
from the mesenchymal lineage, their distinctive spatial distribution within
myofibrillar structures may be a migratory response to collagen V. Collagen
V-mediated response may be essential to lineage commitment and functional
integration of transplanted CLCs within the cardiac syncytium since intimate
cell-cell contacts are known to induce stem cells to undergo cardiac
differentiation134, 135
. However, it remains to be elucidated if the occasional
localization of CLCs to the collagen-I enriched interstitial tissues as
demonstrated in this study, prevented their engraftment into the host
myofibrillar structure and thereby integration with resident cardiomyocytes in
the myocardium.
4.2.2.3 Cell mediated attenuation of adverse LV remodeling
Stem cells have been reported to play significant roles in alleviating the
adverse effects of ventricular remodeling197, 198
. This study shows that these
beneficial effects were indiscriminate of differentiation status as both MSCs
and CLCs significantly ameliorated ventricular dilatation and prevented
progressive wall thinning.
This cell-mediated attenuation of adverse LV remodeling was already evident
at low-dose cell therapy as exemplified by significantly smaller end-systolic
and end-diastolic volumes (Figure 32i-ii) in the cell-treated myocardium as
156
compared to control myocardium. Nonetheless, high-dose cell therapy
significantly reversed negative remodeling effects as myocardium treated with
high-dose cell therapy demonstrated significantly smaller LV dimensions,
such as significantly reduced LV internal diameters (Figure 36 i-ii) and LV
ESV and EDV (i.e ESV and EDV in Figure 36 i-ii), in relation to myocardium
treated with low-dose cell therapies. This is consistent with Menasche et al‟s
observation that LV volumes and consequently negative LV remodeling
effects were significantly reduced with a high-dose administration of skeletal
myoblasts as compared to the placebo group in the MAGIC trial62
.
ECM dysregulation in the failing heart is a primary cause of adverse LV
remodeling and myocardial fibrosis leading to impaired cardiac dysfunction.
The extracellular cardiac microenvironment in the myocardium comprises
primarily of collagen matrices. Following a myocardial infarction, cell-matrix
components are disrupted and there is extensive collagen turnover as a result
of maladaptive remodeling. Subsequently, collagen degradation in the
myocardial ECM reportedly interferes with both systolic performance199
and
relaxation properties of the ventricle200
and eventually leads to contractile
dysfunction. Villari et al also reports that the total collagen volume/mass, its
distribution as well as fiber organization collectively influence the LV
myocardial elasticity201
and therefore compliance. Thus, ECM remodeling
during the progression of heart failure is responsible for the maladaptive
reorganization of cardiac size and geometry so that collagen degradation leads
to chamber dilatation.
157
Both high- and low-dose cell-therapy demonstrated a significant amelioration
of adverse ventricular modeling effects. It is possible that physical presence of
cells may confer an added degree of compliance in collagen- enriched domains
in the infarct and peri-infarct borders of myocardium. This may in turn
increase tissue elasticity of the increasingly stiff scar or fibrotic tissues. This
may preserve regional myocardial tissue compliance in the locale of cell
engraftment. Consistently, stem cell transplantation has been reported to
improve cardiac function by modulating collagen structure in vivo that may
reduce tissue stiffness in border infarct zones of the scar tissue 202, 203
.
The insignificant number of surviving MSCs as well as CLCs in the treated
myocardium 6 weeks post transplantation in the low-dose cell therapy groups
strongly indicates that cell-mediated attenuation of adverse LV remodeling
was not dependent on sustained cell survival or integration in the myocardium.
Conversely, this underscores the likely possibility that the limitation of
negative LV remodeling may have been attributed to cell-mediated paracrine
effects. This is not surprising since bone marrow stem cell transplantation
therapy has been reported to prevent the progression of adverse LV
remodeling in the ischemic heart by secreting pro-survival cytokines with anti-
apoptotic and myoangiogenic differentiation stimulatory (discussed in
subsequent paragraphs in this chapter) properties into the myocardium even
after the disappearance of donor MSCs in vivo204
. This cardioprotective
cytokines such as the vascular endothelial growth factor (VEGF), the basic
fibroblast growth factor (bFGF), insulin-like growth factor (IGF) and the
stromal cell-derived factor (SDF), are in turn implicated in protecting the
resident cardiomyocytes in the early phase of MI205
. Both high- and low-dose
158
cell-therapy demonstrated consistent trends in a significant cell-mediated
amelioration of adverse ventricular modeling effects. Given the high death
rates in the low-dose therapy groups, it is likely that cell-mediated cardiac
repair mechanisms may be attributed to the release of paracrine factors. It is
possible that physical presence of donor cells may confer a certain degree of
compliance in the proximal regions of collagen- enriched domains in the
infarcted and peri-infarct borders of an increasingly fibrotic myocardium, thus
„softening‟ the increasingly overwhelming stiffness with their inherent
elasticity in the collagen I-enriched scar or fibrotic tissues along the anterior
wall or interventricular septum. This in turn confers a certain degrees of
compliance thus preventing progressive systolic wall thinning. Thus, donor
cell engraftment may preserve regional myocardial tissue compliance and may
prevent the progression of myocardial dyskinesia, in turn a secondary factor
that directly influences global contractility. Donor cell-induced paracrine
effects may also include the recruitment of native cardiac progenitor cells to
the site of injury. These cells may differentiate into functional cardiomyocytes
in the peri-infarct domains of the myocardium, thus decreasing the number of
„stiff‟ domains within the infarcted myocardium and enhancing overall
myocardial compliance as well as contractility.
More relevant to limiting negative LV remodeling, stem cell transplantation
has been reported to improve cardiac function by modulating collagen
architecture in vivo that may reduce tissue stiffness in border infarct zones of
the scar tissue202, 203
. Consistently, MSCs has been shown to express the
antifibrotic factor adrenomedullin, which can inhibit the post-infarction
proliferation of cardiac fibroblasts as well as attenuate the expression of
159
collagens –I and –III mRNA206
. This limits if not prevents the development of
scar tissue which would otherwise confer a degree of stiffness and result in a
noncompliant myocardium that ultimately deteriorates overall contractility.
4.2.2.4 Dose-dependent contribution of cell therapy towards myocardial
contractility
PV catheterization studies show that low-dose cell therapy improved global
myocardial contractility with respect to the control group. However, there
were no discernible differences between MSC- and CLC- therapies. At the
same time, echocardiography assessments showed that CLC-therapy
consistently improved LV dimensions and global contractilities such as EF, FS
and regional mid-anterior wall contractility, in relation to MSC therapy.
However, these trends did not achieve statistical significance. This is not
surprising given such a low incidence of cellular engraftment observed in the
low-dose therapy groups. Nevertheless, the small in vivo CLC presence in the
low-dose cell therapy group showed sign of CLC-mediated myogenic
replacement whereby engrafted CLCs demonstrated cross-striations that were
indistinguishable from host cardiomyocytes. Furthermore, CLCs that survived
in the host myocardium 6 week post-therapy engrafted as myocytes that were
electrically coupled with the resident cardiomyocytes (Figure 43-45). In
contrast, the cell fate and development of MSCs could not be studied in rats
treated with a low-dose cell therapy. Nonetheless, results from
echocardiography and pressure-volume catheterization strongly suggest that
MSCs may play a passive role towards improving overall contractile
performance.
160
The lack of discernible differences between low-dose MSC- and CLC-
therapies may be attributed to an incomplete CLC-mediated repair mechanism
in the rat myocardium 6 weeks post transplantation therapy. In contrast, the in
vivo functional data (immunofluorescent examinations of microvessels in the
MSC-treated myocardium) indicated that the MSC-therapy may mediate
cardiac repair via angiogenesis. However, these paracrine-mediated
mechanisms may decline over time with a diminishing MSC presence. In
contrast, low dose CLC-therapy showed enhanced myocardial contractility via
myogenic and neoangiogenic mechanisms. Thus, a discernible difference
between low-dose MSC- and CLC- therapies may only be observed with a
longer follow-up assessment of cardiac function.
Cell dose can exert a significant influence on cell-mediated cardiac relief.
Although no significant improvements were observed with cardiac contractile
function such as EF in the MAGIC trial, the high-dose therapy mediated a
more significant attenuation of adverse LV remodeling and chamber
dilatation62
. There are currently no reports comparing a dose-dependent effect
of cardiomyocyte transplantation on postinfarction cardiac function. However,
Menasche et al reported that both skeletal myoblast and fetal cardiomyocyte
transplantation resulted in equivalent efficacies for improving postinfarction
LV function207
. Notwithstanding, it has been established that skeletal
myoblasts do not express the gap junction protein, connexin 43149
and will not
contract synchronously with the heart, thus resulting in ventricular
arrhythmias. Importantly, cardiomyocyte transplantation offered the best
functional improvement in a functional study comparing the therapeutic
efficacies of 3 different fetal cell types such as cardiomyocytes, smooth
161
muscles and fibroblasts208
. Taken together, we expect that cardiomyocyte
transplantation would be more effective than skeletal myoblast transplantation
for the functional recovery of contractile performance in the infarcted heart.
Thus, high-dose cardiomyocyte therapy may mediate more significant
improvements in cardiac function than current transplantation treatments
involving the use of other myocyte derivatives.
BOOST trial which used 2.5 x 109 undifferentiated BM stem cells resulted in a
higher improvement in EF as compared to the ASTAMI trial which employed
7x107 BM stem cells (Table 1). Although both trials led to insignificant
improvements in global contractility with respect to their placebo groups, it is
believed that high-dose cell therapy would enhance the probability of cellular
survival and engraftment in the myocardium and may therefore provide a
clearer insight into the direct cellular effects on myocardial contractility. This
postulation was verified by a relatively larger cell presence in tissue sections
of myocardium administered with high-dose cell therapy as compared to those
who have received a low-dose of MSCs or CLCs.
In vivo functional studies showed that high-dose cell therapy was generally
more effective than low-dose cell therapy as dose-dependent assessments of
cell therapy showed that high-dose cell-therapy mediated statistically
significant improvements in LV dimensions (Figure 40i-ii), ESV, EDV and
ejection fraction (Figures 41i-ii, iv-vii). Thus, high-dose cell therapy mediated
larger improvements in restraining chamber dilatation and systolic activities as
compared to low-dose cell therapy. This may be attributed to a significantly
low cellular engraftment and integration of the respective cell types in the low-
162
dose therapy groups. Notwithstanding, some significant improvements in EDV
could already be observed in rats administered with low-dose CLC-therapy
with respect to control rats, while low-dose MSC-therapy did not mediate any
statistically significant information. Although low-dose cell therapy also
attenuated negative LV remodeling and improved global contractility, as
observed by smaller cardiac volumes (Figure 32i-ii) and enhanced minimum
dP/dt (Figures 32iii), and cardiac output (Figure 32 v) respectively, these
improvements were statistically insignificant relative to high-dose cell
therapy.
High-dose MSC therapy improved cardiac function by appearing to prevent
progressive worsening of cardiac function, which was observed with control
animals in both the high- and low-dose therapy groups. However,
echocardiography assessments showed that on the average, high-dose MSC-
therapy mediated a -0.11 to 1.6% improvement in the HR-dependent and -
independent contractility indices. These observations underlie the possibility
that MSCs contribute passively to overall contractility. In contrast, CLCs
showed ~7-13% improvement in the same parameters, and thus contributed
actively towards contractile performance. In contrast, echocardiography
analysis of global contractility low-dose therapy group did not show any
significant differences between MSC-treated and control rats, although PV
catheterization showed consistent improvements in LV function in both high-
and low- dose MSC therapy. Notwithstanding, high-dose MSC therapy was
the more effective therapy as improvements due to the low-dose MSC therapy
was statistically insignificant as compared to high-dose cell therapy (Figure
40-41).
163
More importantly, discernible functional differences in EF, FS and MPI were
also observed between high-dose MSC- and CLC-therapy (Figure 36xii-xv).
Furthermore, differences in myocardial systolic activities were uncovered with
the assessment of load-insensitive contractility measures observed with rats
treated with a high-dose cell therapy. However, among the 4 therapy groups
compared, high-dose CLC therapy mediated the best improvements in global
systolic activities as observed by the highest maximum dP/dt (Figure 41 iii)
and the best improvement in LVEF (Figure 40iii) with respect to control
animals.
164
4.3 Relative therapeutic efficacies of CLCs vs. MSCs
4.3.1 Assessment of myocardial systolic activities
4.3.1.1 Echocardiography assessments
Echocardiography assessment of LV function was derived primarily from
the parasternal short axis view, which in turn assesses the contraction of
the circumferential fibers that gives rise to radial contractility. As
discussed in the previous section, echocardiography assessments show no
significant differences in these measures of contractility between rats in
the low-dose therapy groups. Although there were no significant
differences in heart rate between the respective therapy groups, the heart
rate of rats studied ranged between 300-400bpm. It is possible that the
relative rapid heart rate in rodents may have obscured accurate analysis of
ventricular contractility between treatment groups via current diagnostic
techniques. In retrospect, sedating the rats further on a higher but safe
percentage of isoflurane to further depress the heart rate may provide
further insight into true contractility measures in rats in the respective
treatment groups. Notwithstanding, rats treated with a high-dose of CLCs
demonstrated significantly superior improvements in heart-rate dependent
global contractility indices including ejection fraction and fractional
shortening with respect to those treated with a high-dose of MSCs, in spite
of the rapid heart rate. This may be attributed to a relatively higher
frequency of CLC-derived myogenic replacement in vivo as compared to
that arising from MSCs (Figure 43).
165
HR-independent measures of cardiac contractility such as the myocardial
velocities of circumferential fiber shortening (Vcfc) were also obtained to
verify that cell-mediated enhancement in contractile function was not a
consequence of rapid heart rates. Notwithstanding, rats treated with a high
dose of CLCs showed higher improvements in Vcfc as compared to
animals administered with high-dose MSCs.
It is possible that the differences observed with these conventional
measures may be an inadvertent consequence of loading conditions209
rather than actual global systolic function. This is because derivations of
the load-sensitive contractility measures rely on the assessment of radial
or circumferential fiber contractions, which in the presence of the most
common cardiac condition of left ventricular hypertrophy (LVH) may
overestimate cardiac contractility. The contraction of excess sarcomeres
may mask impaired cardiomyocyte function by preserving normal wall
contractions. In addition, radial or circumferential fiber function is
preserved in LVH but the adverse effects of subendocardial ischemia may
lead to longitudinal fiber function that is otherwise obscured in
conventional load-sensitive contractility indicators.
Echocardiography assessments persistently showed that cell therapy
persistently attenuated left ventricular remodeling and dilatation, a benefit
that was already readily observed in rats administered with a low-dose of
cell therapy. However, high-dose cell therapy provided a clearer insight
into the direct cellular effects on myocardial contractility. In particular,
echocardiography showed significant CLC-mediated improvements in
166
systolic activities that were not confounded by the rapid heart rate, as can
be seen by the HR-independent contractility indices, such as Vcfc and the
Tei index.
4.3.1.2 Real-time pressure-volume catheterization
4.3.1.2.1 Load-sensitive hemodynamic measurements
In contrast to echocardiography measurements, PV loops derived
indices are better able to separate ventricular function into primary
systolic, diastolic and ventricular loading properties209
.
Consistently, PV measurements show similar trends in cell-
mediated attenuation of chamber dilatation and myocardial wall
thinning. However, high-dose cell therapy resulted in smaller
cardiac volumes as compared to those observed with low-dose cell
therapy. Rats treated with high-dose cell therapy also showed
superior enhancement of global contractility indices such as EF,
SW, CO and SV in relation to those administered with low-dose
cell therapy. Consistent with HR-independent echocardiography
contractility measures, high-dose CLC therapy mediated the
largest improvement in systolic activities as can be seen by
significantly elevated increase in maximum dP/dt.
4.3.1.2.2 Load-insensitive contractility measures
The biomechanical performance of cardiac muscle is influenced by
acute changes in the resting fibre length (preload) and alterations
in wall stress during ventricular ejection (afterload)132
. It is
therefore important to distinguish between haemodynamic
167
performance due to mechanical lesions from that due to depressed
myocardial function. The hemodynamic performance of the
ventricle is not an accurate measure of actual myocardial
contractile state if these 2 parameters are not held constant. A
major limitation in detecting acute contractility changes arises
when the determinants of performance other than the inotropic
state are altered by an intervention so as to change performance in
a direction similar to the presumed effect of the change in
contractile state132
.
Conventional assessments of systolic activity by echocardiography
such as ejection fraction and fractional shortening are highly
reliant on load-sensitive measures209
. In certain compensated
disease states such as LV hypertrophy (LVH), it is not surprising
that additional sarcomeres are developed so that normal cardiac
performance is sustained per unit circumference in the basal state
by maintaining seemingly normal wall shortening and thickening
values, thus obscuring actual defective cardiomyocyte function.
Furthermore, radial and circumferential fiber contractility may be
preserved in LVH, while longitudinal fibres in the
subendocardium are reportedly reduced132, 209
.
Thus, functional determinants of left ventricular (LV) contractility
should ideally be linear and load-independent while still sensitive
to changes in inotropic state. Although no measure of contractility
is absolutely load-independent, various contractility indices such
168
as PRSW, ESPVR and maximum dP/dt-EDV210
have been
reported to remain relatively unaffected by various loading
conditions in both animal and human studies211
, and have since
been widely accepted as load-insensitive indices of myocardial
contractility that are better reflections of true or absolute
contractility at the cardiomyocyte level.
Preload reduction allowed for load-insensitive systolic and
diastolic function to be examined and was achieved via a transient
depression of the inferior vena cava. The results of this study show
consistent trends in cell-mediated contractility. Both cell therapy
mediated enhanced systolic activities as compared to control rats
as can be seen by an enhanced ESPVR. However, CLC-therapy
resulted in a superior recovery of systolic performance as can be
observed by the restoration of ESPVR to baseline levels. However,
the ESPVR index is limited by its dependence of mass and heart
size209
. Consistent with other reports, we also observed that the
ESPVR curve is curvilinear with increasing IVC depression. It was
also a technical challenge to derive ESPVR slopes in the over
similar ESP ranges as suggested by Kass et al in order to reduce
the nonlinearity of this curve212
as depression of the IVC could
displace the catheter from its steady state position, sometimes
displacing the topmost pressure sensor entirely out of the chamber
as was verified visually by echocardiography. In other cases, the
displacement of the catheter often led to an interference signal
between the pressure and sometimes volume sensors and the
169
ventricular walls, causing a spike or complete distortion in the top
left corner of several PV loops so that interpretation of results
were rendered impossible. A few rats were excluded in this study
because no interpretable load-insensitive contractility data could
be derived from them.
The PRSW and dP/dt-EDV measures are 2 alternative load-
insensitive measures of systolic function. PRSW is reportedly
independent of ventricular size and mass but is sensitive to
changes in cardiac contractility, while dP-dt-EDV is another
sensitive indicator of myocardial contractility but is relatively
more pre-load sensitive as compared to PRSW209
. Both measures
demonstrated similar improvement to the ESPVR although CLC-
therapy did not lead to restoration to baseline levels. As with
ESPVR, the MSC-treated myocardium trended towards enhanced
systolic performance, but only the CLC-treated myocardium
demonstrated a superior improvement in systolic performance with
respect to control myocardium.
4.3.1.3 CLCs contribute actively towards myocardial contractile
performance via myocyte replacement
Collectively, trends in in vivo functional studies strongly indicated that
CLCs improved systolic performance without compromising end-diastolic
pressures in the infarcted myocardium. As discussed in the preceding
sections, CLCs may facilitate a synergistic recovery of cardiac function by
170
attenuating LV remodeling and limiting chamber dilatation, thereby
preserving regional tissue compliance in the collagen V-enriched peri-
infarct borders. This leads ultimately to an enhancement of regional
contractility that may contribute indirectly towards overall myocardial
contractile performance. CLCs predominantly engrafted in the collagen
V- endomysial lined viable muscle fibers and demonstrated well-
developed Z-bands and A-bands that were indistinguishable from host
cardiac myocytes. Thus, CLCs confer actual mechanical benefit via
myocyte replacement, which contributes regional and global myocardial
contractile performance in the treated myocardium.
4.3.1.4 MSCs contribute passively to myocardial systolic activities
Echocardiography assessments as well as PV catheterization show that
MSC-treated rats exhibited a relatively diminished contractile
performance as compared to CLC-treated rats. MSCs may play a passive
role towards enhancing contractile performance in the myocardium.
Consistently, immunofluorescent examinations showed that transplanted
MSCs were absent of cross striations that are consistent with
cardiomyocytes. Furthermore, MSCs appeared to be predominantly
localized to the collagen I-enriched interstitial spaces that were physically
isolated from the myocardium. This strongly suggests that most of the
engrafted MSCs remained quiescent following cell injection into the peri-
infarct borders of the myocardium. Thus, unlike CLCs, MSCs did not
integrate as functional cardiomyocytes (Figure 43) in the infarcted
myocardium.
171
Notwithstanding, both high- and low-dose MSC led to significant
improvements in global systolic activities as compared to the control.
These observations are consistent with the results of current MSC
transplantation therapies, which reported improvements in global
contractile performance as shown by enhanced EF. Given the seemingly
low survival rates of MSCs in vivo and the coincident increase in small
sized blood vessels in the MSC-treated myocardium, it is possible that
MSC-mediated improvements in systolic activities such as EF, FS and the
load-insensitive indices such as ESPVR, PRSW and maximum dP-
dt/EDV may be largely attributed to paracrine effects as MSCs may
secrete angiogenic growth factors into the microenvironment within the
locale of cell engraftment as an inadvertent consequence of paracrine
mediated effects. Alternatively, the apparent improvements in the MSC-
treated myocardium may be reflective of a possible compensatory
contractile mechanism as a passive consequence of scar stiffening in the
MSC-treated myocardium. Afterall, MSC transplantation induced systolic
wall thickening, which prevented dyskinesis, in turn a secondary factor
that directly influences systolic contractility213
. If true, high-dose MSC-
therapy may result in a deterioration of cardiac contractile function in the
long run. However, whether MSC-mediated enhancement of systolic
activities is really a compensatory mechanism can only be determined
with a longer follow-up assessment of contractile performance following
cell transplantation.
MSC transplantation has also been reported to modulate the ECM
architecture in the infarcted regions of the failing heart by decreasing the
172
levels of collagen -I and -III fractions in vivo206, 214
. However, this study
showed an extremely high death rate especially in the high-dose therapy
group, which may lead to massive inflammation at the site of cell death.
This release of inflammatory cytokines can induce subsequent myocardial
fibrosis, which may inadvertently lead to subnormal ventricular relaxation
and negate MSC’s initial positive effect on decreasing collagen levels in
vivo (seen in the low-dose therapy group). Taken together, high-dose
MSC-therapy may be counterproductive in the treatment of heart failure.
Notwithstanding, this postulation can only be affirmed with a quantitative
measure of the collagen levels in the myocardium in respective high- and
low-dose MSC therapy groups.
Alternatively, MSC-mediated improvements in systolic function may be
attributed to paracrine mechanisms more than myocyte replacement in the
MSC-treated heart. Recent studies have shown that donor MSC
engraftment and survival in the infarcted myocardium decreased rapidly
from 34-80% from the time of injection to 0.3-3.5% 6 weeks post therapy
regardless of dose and application time186
. These numbers are too low to
exert a direct influence on organ function and thus may be therapeutically
irrelevant. In addition, bone marrow stem cells reportedly confer acute
improvements in cardiac function less than 72 h after MI. This precludes
extended differentiation in vivo it is not likely that MSCs can differentiate
into cardiomyocytes within 3 days to regenerate the infarcted
myocardium. Most importantly, in vitro and in vivo animal studies have
revealed that much of the functional improvement and attenuation of
injury afforded by stem cells can be replicated by cell free, conditioned
173
media derived from stem cells. Taken together, these findings strongly
suggest that MSCs confer cardiac relief via complex paracrine
mechanisms, which ultimately preserve or enhance global contractile
performance with respect to control myocardium.
4.3.2 Assessment of myocardial diastolic activities
4.3.2.1 Echocardiography assessments
An in-depth analysis of diastolic function using tissue Doppler Imaging
techniques could not be performed as the rapid heart rates in rodents
resulted in the fusion of the E and A waves in almost every ECG cycle.
As such, echocardiography examinations could not adequately assess
diastolic activities/ dysfunction in the myocardial infarction rat models.
The Tei index however provides an insight into global myocardial
contractile performance as a summation of both systolic and diastolic
activities. It remains debatable however if this measure of contractility is
independent of heart rate209
or preload conditions215
. For this study, the
Tei index was significantly decreased in rats treated with a high-dose of
CLCs. Although not significantly different from control rats, MSC-treated
rats consistently demonstrated no nett change in this relatively HR-
independent global contractility index.
4.3.2.2 Conductance catheterization
The results of this study show that low-dose cell therapy improved
systolic performance with lower end-diastolic pressures (EDP) as
compared to control animals. However, CLC-therapy resulted in a
relatively lower EDP as compared to MSC-therapy. Furthermore,
174
increasing dose of MSCs in particular resulted in a compromised diastolic
function (Table 7). Consistently, of the 4 cell therapy groups, rats treated
with high-dose MSC-therapy demonstrated the lowest minimum dP/dt and
the slowest passive filling rate as depicted by the smallest maximum
dV/dt value.
At the same time, similar trends were observed with the loafd-insensitive
measure of isovolumetric relaxation, Tau (Weiss). High-dose MSC-
therapy consistently gave rise to the highest Tau (Weiss) value and thus
demonstrated the worst isovolumetric relaxation among of the 4 cell
therapy groups. However, high-dose MSC-therapy negative effect on
diastolic performance may have been confounded by alterations in loading
conditions. Given that these trends are not statistically significant, more
investigations are necessary to draw a firm conclusion in the effect of
high-dose MSC-therapy on possible diastolic impairment.
Taken together, these results suggests that sustenance of systolic activities
in rats treated with a high dose of MSCs may be a compensatory
mechanism in which end-diastolic pressures were correspondingly
elevated. This may not be beneficial in the long run, as chamber dilatation
may be inadvertent to offset the increased cardiac pressures within the left
ventricle. Similar to patients with heart failure, normal EF does not
exclude diastolic dysfunction216
.
It is important to stress that these mechanisms are dynamic and relatively
higher collagen levels and compromised diastolic activities can occur
175
concurrently with an apparent improvement of systolic function and
attenuation of LV remodeling in the MSC-treated myocardium at 6 weeks
post transplantation therapy. These trends may be more apparent with a
larger sample size and a longer follow-up assessment period i.e 6-12
months post therapy. More investigations are warranted before a
definitive conclusion on the effects of high-dose MSC-therapy on
myocardial diastolic activities, collagen remodeling and overall cardiac
contractile function can be derived.
Systolic function in rats appears to look more significant than that of diastolic
activities in this study. However, it is important to realize that current
techniques in assessing diastolic activities are not well optimized and need to
be further refined so as to attain a more accurate measure of diastolic indices.
One way to better study diastolic activities via echocardiography assessments
may be to decrease heart rates significantly within the 100-200 bpm range so
as to better segregate the E and A waves. However, this technique remains to
be optimized a safe and nonhazardous levels for rats to survive the otherwise
toxic effects of anesthetics at higher levels. In addition, some EDPVR data
obtained with PV catheterization were excluded as they did not yield any
discernible information. A way to overcome the large variations that existed in
this study would be to recruit more rats into this study so that the sample size
of rats approximates a normal distribution. More investigation evaluating
diastolic indices are warranted before a conclusion on relative cell-mediated
diastolic activities can be reached.
176
4.4 Cell-mediated paracrine effects
4.4.1 MSC-mediated neovascularization in the infarcted myocardium
Transplanted MSCs were localized to immature blood vessels, while CLCs
often engrafted in sites proximal to blood vessels. This strongly suggests that
MSCs may play significant roles in vasculogenesis. This observation further
reinforces the merits of committing MSCs to a cardiac-like phenotype ex vivo
before transplantation to enhance contractile performance via myocyte
replacement in the infarcted myocardium. Consistently, MSC-treated
myocardium demonstrated enhanced myocardial angiogenesis in the peri-
infarct and infarcted borders. Embryonic stem cells expressing VEGFR2 have
been reported to differentiate into endothelial and smooth muscle cells 217, 218
.
Although patient derived MSCs expressed KDR/ VEGFR2 in vitro, it
remained to be determined if MSCs underwent extended vascular
differentiation following cell transplantation. Nevertheless, intramyocardial
angiogenesis may have contributed indirectly towards overall contractile
function in the MSC transplanted animals by enhancing myocardial perfusion.
Consistently, MSCs have been reported to secrete cytokines which lead to
ultimate improvements in myocardial performance182
.
4.4.2 CLCs also elicit alternative non-myogenic mechanisms of cardiac repair
Notwithstanding, the seemingly elevated CLC presence in rats treated with a
high dose of CLCs was still disproportionate to the remarkably restoration of
myocardial systolic performance to baseline levels. Such inconsistencies
underscores the unlikely possibility of cell-mediated myocyte replacement
alone was being the primary reason for such recovery of global cardiac
function. Instead, the results of both high- and low- dose CLC therapy groups
177
consistently hint at possible cell-mediated non-myogenic cardioprotective
paracrine effects. A subset of transplanted CLCs that engrafted in the collagen
I enriched interstitial spaces between adjacent viable myofibers did not exhibit
sarcomeric cross striations. In contrast to MSCs, CLCs also did not integrate
as immature vessels. However, the CLC-treated myocardium showed a
significant increase in neoangiogenesis with respect to control myocardium.
At the same time, in vivo functional studies showed an alternate minority
population of CLCs that engrafted in the collagen I enriched interstitial spaces
between adjacent viable myofibers. These CLCs did not exhibit distinct cross
striations consistent with sarcomeric structures. However, these cells
demonstrated a coincident expression of the proliferating cell nuclear antigen
(PCNA) and the von-Willebrand factor (vWf). This strongly suggests that the
CLCs engrafted as proliferating endothelial cells in the myocardial interstices.
Additionally, resident cells in the immediate locale of CLC-engraftment in the
interstitial tissues also exhibit a coincident expression of PCNA and vWf.
Thus, CLC-therapy promoted endogenous and donor cell-derived endothelial
proliferation in the treated myocardium possibly via the secretion of relevant
cytokines such as VEGFR2. This is consistent with in vitro characterization
studies of CLCs, which show that these cardiogenic cell types simultaneously
expressed endothelial markers such as VEGFR2 and smooth muscle actin.
Thus, CLCs may also possess an inherent propensity for endothelial
differentiation in addition to myogenic differentiation.
Collectively, these results strongly suggest that CLC may also mediate
cardioprotective benefits via paracrine signaling. Thus, the combined action of
an enhanced cell-mediated regional contractility due to myogenic replacement,
178
attenuated negative LV remodeling and paracrine cardioprotection may confer
to a synergistic improvement in overall cardiac function in the CLC-treated
myocardium.
In vivo findings suggest that MSCs may contribute passively towards functional
improvement by preventing the progressive deterioration of cardiac activities.
However, MSC-mediated improvements in cardiac function as well as a
coincident increase in blood vessels in the MSC-treated myocardium despite the
absence of human nuclei in infarcted myocardium strongly suggest possible
paracrine effects that may promote angiogenesis post transplantation. This is
consistent with current literature, which reports that MSC-mediated
improvements are may be attributed to paracrine mediated205, 206
.
Similarly, the CLC-treated myocardium demonstrated enhanced endothelial cell
proliferation in vivo, which implicates CLCs in playing significant roles in
neoangigenesis, an inadvertent consequence of paracrine effects. Future
investigations should include an analysis of secretory angiogenic growth factors,
such as VEGFR2, in the microenvironment in the vicinity of cell engraftment via
an ELISA assay or flow cytometry as a quantitative measurement would provide
a more conclusive insight into the relative paracrine contribution of either cell
type.
179
4.5 Postulated role of myocardial ECM on in vivo cell fate and
development
The effect of post-infarct remodeling of myocardial extracellular matrix on in situ
stem cell differentiation remains to be determined. The developmental potential
of MSCs is reportedly regulated by underlying substrate elasticity in vitro219
.
Collagen I may impart heightened tension and mechanical stress within the cell,
leading to an overwhelming stiffness that is unfavorable towards
cardiomyogenic-like phenotype of CLCs. Consistently, MSCs have been reported
to undergo osteogenic differentiation on collagen I surfaces. In contrast, collagen
V ECM may present a more pliable environment that favors the differentiation of
CLCs. These physical forces may similarly induce appropriate cascades of
stimulatory intracellular signals that are favorable for survival, proliferation and
cardiac transdifferentiation in CLCs in vivo. Consistent with in vitro studies,
MSCs and CLCs that engrafted in the collagen I-enriched interstices in the
myocardium bordering the scar areas of the infarcted myocardium did not exhibit
expression of cardiac-specific myofibrillar proteins. In contrast, CLCs that
engrafted in the collagen V-enriched endomyosium lining viable myofibers
demonstrated well-developed Z-bands and A-bands that were indistinguishable
from host cardiac myocytes.
Transplantation of terminally differentiated fetal and adult cardiomyocytes have
reportedly led to the preservation of systolic activities and overall contractile
performance and a significant attenuation of LV remodeling and dilatation in the
infarcted myocardium. However, it is not easy to isolate cardiomyocytes from
cardiac tissue biopsies in large quantities. CLCs are partially differentiated but
180
committed to cardiac lineage. Importantly, large numbers of CLCs can be derived
in vitro before transplantation. These cells bear a resemblance to fetal
cardiomyocytes and can survive well in the myocardium. Like fetal
cardiomyocytes, CLCs also significantly enhanced contractile performance,
limited negative LV remodeling and dilatation and promoted angiogenesis in the
infarcted myocardium. However, being partially differentiated, CLCs may not
undergo extended myocyte differentiation and may thus remain quiescent in vivo,
especially if they were engrafted in a collagen I-enriched domain in the infarcted
regions of the failing heart.
Notwithstanding, the results of this study show that CLCs mediated superior
improvements in cardiac function to undifferentiated bone marrow MSC therapy
in an infarcted rat myocardium. Importantly, CLCs contribute actively towards
overall contractile performance via myocyte replacement and paracrine mediated
effects, while playing possible roles in neoangiogenesis. In contrast,
undifferentiated MSCs mediated functional improvements via a predominant
cardioprotective paracrine effect. Future directions improvements should be made
to current techniques in the delivery of viable CLCs to the injured myocardium.
Culturing of CLCs on an engineered collagen V-based scaffold and applied
directly onto the infarcted heart may enhance and maximize their survival,
engraftment and integration with host cardiomyocytes.
Although clinical trials have reported mixed outcomes, some patients have
undeniably benefited from undifferentiated bone marrow stem cell therapy. CLCs
are expected to confer equal if not superior cardiac relief in patients and present
as a promising autologous cell-based therapy for the treatment of CVD. However,
181
further preclinical investigations should be conducted before CLCs are applied in
clinical trials. For example, the beneficial effects CLC therapy should be further
investigated in large animal models with follow up assessments of cardiac
function over a period of time. It is also important to determine if CLCs
demonstrate a stable cardiomyogenic profile in vivo, and if CLC-mediated
paracrine effects excluded deleterious consequences such as fibrosis and
apoptosis in the long run.
182
Chapter 5: Conclusion
This study supports the concept of pre-differentiating bone marrow mesenchymal
stem cells into cardiomyocyte-like cells in vitro before cell transplantation therapy.
These differentiating cells exhibit distinct Z-band cross striations in vitro that is
consistent with a developing contractile phenotype. Importantly, this distinct cardiac-
like phenotype is stable in prolonged cultures. A large-scale expansion of these
cardiomyogenic cell types was further achieved via preferential interaction with
collagen V extracellular matrices.
Figure 52 summarizes the possible distinct myogenic and non-myogenic mechanisms
of cardiac repair mediated by CLCs as compared to MSCs. Importantly, cell therapy
significantly improves the efficiency of myocardial repair as human bone-marrow
derived stem cells act to enhance the elasticity of the infarct area, thus preventing
deleterious LV-remodeling and ventricular dilatation. Cell therapy also confers
cardioprotection on resident host cells via the release of paracrine factors. More
specifically, cell therapy promotes robust angiogenesis with respect to control
myocardium, as exemplified by increased vessel counts in cell-treated myocardium.
This improves myocardial perfusion, which may in turn prevent extensive apoptosis
of the surviving resident cardiomyocytes.
In particular, CLCs present as a more efficacious therapy to MSC-treatment as CLCs
contribute actively towards overall contractility via cell-derived myocyte replacement
besides promoting neoangiogenesis in the infarcted myocardium. More remarkably,
high- dose CLC therapy specifically confer added mechanical benefit by restoring
myocardial systolic performance in the treated myocardium. Thus, CLC-treated rats
183
demonstrated superior contractile performance to rats administered with
undifferentiated MSCs.
Infarcted myocardium demonstrated elevated collagen I deposition in the peri-infarct
borders and especially in the scar tissue. MSCs injected directly into the collagen I-
enriched interstices in the myocardium bordering the scar areas of the infarcted
myocardium did not exhibit expression of cardiac-specific myofibrillar proteins. In
contrast, CLCs that engrafted in the collagen V-enriched endomyosial space of viable
myofibers demonstrated well-developed Z-and A-bands indistinguishable from host
cardiac myocytes.
Predifferentiating mesenchymal stem cells into cardiomyocyte-like cells ex vivo
before transplantation is more efficacious than mesenchymal stem cells for the
treatment of chronic heart failure. CLCs may present an added value to current
clinical practice of using undifferentiated stem cells for the treatment of myocardial
infarction.
184
Figure 52. Possible MSC- and CLC- mediated repair mechanisms in the compromised myocardium.
185
References
1. Heart Disease and Stroke Statistics -- 2009 Update (At-a-Glance Version):
American Heart Association; 2009.
2. Singapore Heart Foundation Statistics. Available at:
http://www.myheart.org.sg/heart-facts/statistics/.
3. Atherosclerosis: Heart and Blood Vessel Disorders: Merck Manual Home
Edition. Available at:
http://www.merck.com/mmhe/sec03/ch032/ch032a.html.
4. Jackson G, Gibbs CR, Davies MK, Lip GY. ABC of heart failure.
Pathophysiology. BMJ. 2000;320(7228):167-170.
5. Gupta S, Markham DW, Drazner MH, Mammen PP. Fulminant myocarditis.
Nat Clin Pract Cardiovasc Med. 2008;5(11):693-706.
6. Dudac PJ. Congestive Heart Failure. United States of America: Jones &
Bartlett Publishers; 2006.
7. Buckberg GD. Form versus disease: optimizing geometry during ventricular
restoration. Eur J Cardiothorac Surg. 2006;29 Suppl 1:S238-244.
8. Ho KK, Pinsky JL, Kannel WB, Levy D. The epidemiology of heart failure:
the Framingham Study. J Am Coll Cardiol. 1993;22(4 Suppl A):6A-13A.
9. Feuerstein GZ, Weck, PK. Cardiac Remodeling: From concepts to
therapeutics. Heart Failure Reviews. 1995;4:7-19.
10. Dabiri GA, Turnacioglu KK, Sanger JM, Sanger JW. Myofibrillogenesis
visualized in living embryonic cardiomyocytes. Proc Natl Acad Sci U S A.
1997;94(17):9493-9498.
186
11. Ojima K, Lin ZX, Zhang ZQ, Hijikata T, Holtzer S, Labeit S, Sweeney HL,
Holtzer H. Initiation and maturation of I-Z-I bodies in the growth tips of
transfected myotubes. J Cell Sci. 1999;112 ( Pt 22):4101-4112.
12. Schultheiss T, Lin ZX, Lu MH, Murray J, Fischman DA, Weber K, Masaki T,
Imamura M, Holtzer H. Differential distribution of subsets of myofibrillar
proteins in cardiac nonstriated and striated myofibrils. J Cell Biol.
1990;110(4):1159-1172.
13. Gautel M, Mues A, Young P. Control of sarcomeric assembly: the flow of
information on titin. Rev Physiol Biochem Pharmacol. 1999;138:97-137.
14. Gregorio CC, Trombitas K, Centner T, Kolmerer B, Stier G, Kunke K, Suzuki
K, Obermayr F, Herrmann B, Granzier H, Sorimachi H, Labeit S. The NH2
terminus of titin spans the Z-disc: its interaction with a novel 19-kD ligand (T-
cap) is required for sarcomeric integrity. J Cell Biol. 1998;143(4):1013-1027.
15. Gregorio CC, Antin PB. To the heart of myofibril assembly. Trends Cell Biol.
2000;10(9):355-362.
16. Kontrogianni-Konstantopoulos A, Bloch RJ. The hydrophilic domain of small
ankyrin-1 interacts with the two N-terminal immunoglobulin domains of titin.
J Biol Chem. 2003;278(6):3985-3991.
17. McElhinny AS, Kazmierski ST, Labeit S, Gregorio CC. Nebulin: the
nebulous, multifunctional giant of striated muscle. Trends Cardiovasc Med.
2003;13(5):195-201.
18. Millevoi S, Trombitas K, Kolmerer B, Kostin S, Schaper J, Pelin K, Granzier
H, Labeit S. Characterization of nebulette and nebulin and emerging concepts
of their roles for vertebrate Z-discs. J Mol Biol. 1998;282(1):111-123.
187
19. Ehler E, Rothen BM, Hammerle SP, Komiyama M, Perriard JC.
Myofibrillogenesis in the developing chicken heart: assembly of Z-disk, M-
line and the thick filaments. J Cell Sci. 1999;112 ( Pt 10):1529-1539.
20. Furst DO, Obermann WM, van der Ven PF. Structure and assembly of the
sarcomeric M band. Rev Physiol Biochem Pharmacol. 1999;138:163-202.
21. van der Ven PF, Bartsch JW, Gautel M, Jockusch H, Furst DO. A functional
knock-out of titin results in defective myofibril assembly. J Cell Sci. 2000;113
( Pt 8):1405-1414.
22. Gotthardt M, Hammer RE, Hubner N, Monti J, Witt CC, McNabb M,
Richardson JA, Granzier H, Labeit S, Herz J. Conditional expression of
mutant M-line titins results in cardiomyopathy with altered sarcomere
structure. J Biol Chem. 2003;278(8):6059-6065.
23. Kontrogianni-Konstantopoulos A, Catino DH, Strong JC, Randall WR, Bloch
RJ. Obscurin regulates the organization of myosin into A bands. Am J Physiol
Cell Physiol. 2004;287(1):C209-217.
24. Wang SM, Lo MC, Shang C, Kao SC, Tseng YZ. Role of M-line proteins in
sarcomeric titin assembly during cardiac myofibrillogenesis. J Cell Biochem.
1998;71(1):82-95.
25. Yang YG, Obinata T, Shimada Y. Developmental relationship of myosin
binding proteins (myomesin, connectin and C-protein) to myosin in chicken
somites as studied by immunofluorescence microscopy. Cell Struct Funct.
2000;25(3):177-185.
26. Labeit S, Kolmerer B, Linke WA. The giant protein titin. Emerging roles in
physiology and pathophysiology. Circ Res. 1997;80(2):290-294.
188
27. van der Loop FT, van der Ven PF, Furst DO, Gautel M, van Eys GJ,
Ramaekers FC. Integration of titin into the sarcomeres of cultured
differentiating human skeletal muscle cells. Eur J Cell Biol. 1996;69(4):301-
307.
28. Rhee D, Sanger JM, Sanger JW. The premyofibril: evidence for its role in
myofibrillogenesis. Cell Motil Cytoskeleton. 1994;28(1):1-24.
29. Sanger JW, Chowrashi P, Shaner NC, Spalthoff S, Wang J, Freeman NL,
Sanger JM. Myofibrillogenesis in skeletal muscle cells. Clin Orthop Relat Res.
2002(403 Suppl):S153-162.
30. Boheler KR, Czyz J, Tweedie D, Yang HT, Anisimov SV, Wobus AM.
Differentiation of pluripotent embryonic stem cells into cardiomyocytes. Circ
Res. 2002;91(3):189-201.
31. Boheler KRaW, A.M. Myocardial Aging and stem cell biology. New York,
NY: Elsevier Science; 2002.
32. Westfall MV, Pasyk KA, Yule DI, Samuelson LC, Metzger JM. Ultrastructure
and cell-cell coupling of cardiac myocytes differentiating in embryonic stem
cell cultures. Cell Motil Cytoskeleton. 1997;36(1):43-54.
33. Oyamada Y, Komatsu K, Kimura H, Mori M, Oyamada M. Differential
regulation of gap junction protein (connexin) genes during cardiomyocytic
differentiation of mouse embryonic stem cells in vitro. Exp Cell Res.
1996;229(2):318-326.
34. Cao F, van der Bogt KE, Sadrzadeh A, Xie X, Sheikh AY, Wang H, Connolly
AJ, Robbins RC, Wu JC. Spatial and temporal kinetics of teratoma formation
189
from murine embryonic stem cell transplantation. Stem Cells Dev.
2007;16(6):883-891.
35. Shih CC, Forman SJ, Chu P, Slovak M. Human embryonic stem cells are
prone to generate primitive, undifferentiated tumors in engrafted human fetal
tissues in severe combined immunodeficient mice. Stem Cells Dev.
2007;16(6):893-902.
36. Muller R, Lengerke C. Patient-specific pluripotent stem cells: promises and
challenges. Nat Rev Endocrinol. 2009;5(4):195-203.
37. Makino S, Fukuda K, Miyoshi S, Konishi F, Kodama H, Pan J, Sano M,
Takahashi T, Hori S, Abe H, Hata J, Umezawa A, Ogawa S. Cardiomyocytes
can be generated from marrow stromal cells in vitro. J Clin Invest.
1999;103(5):697-705.
38. Pittenger MF, Mackay AM, Beck SC, Jaiswal RK, Douglas R, Mosca JD,
Moorman MA, Simonetti DW, Craig S, Marshak DR. Multilineage potential
of adult human mesenchymal stem cells. Science. 1999;284(5411):143-147.
39. Alison MR, Poulsom R, Jeffery R, Dhillon AP, Quaglia A, Jacob J, Novelli
M, Prentice G, Williamson J, Wright NA. Hepatocytes from non-hepatic adult
stem cells. Nature. 2000;406(6793):257.
40. Mezey E, Chandross KJ. Bone marrow: a possible alternative source of cells
in the adult nervous system. Eur J Pharmacol. 2000;405(1-3):297-302.
41. Krause DS, Theise ND, Collector MI, Henegariu O, Hwang S, Gardner R,
Neutzel S, Sharkis SJ. Multi-organ, multi-lineage engraftment by a single
bone marrow-derived stem cell. Cell. 2001;105(3):369-377.
190
42. Poulsom R, Alison MR, Forbes SJ, Wright NA. Adult stem cell plasticity. J
Pathol. 2002;197(4):441-456.
43. Deb A, Wang S, Skelding KA, Miller D, Simper D, Caplice NM. Bone
marrow-derived cardiomyocytes are present in adult human heart: A study of
gender-mismatched bone marrow transplantation patients. Circulation.
2003;107(9):1247-1249.
44. Ianus A, Holz GG, Theise ND, Hussain MA. In vivo derivation of glucose-
competent pancreatic endocrine cells from bone marrow without evidence of
cell fusion. J Clin Invest. 2003;111(6):843-850.
45. Murry CE, Soonpaa MH, Reinecke H, Nakajima H, Nakajima HO, Rubart M,
Pasumarthi KB, Virag JI, Bartelmez SH, Poppa V, Bradford G, Dowell JD,
Williams DA, Field LJ. Haematopoietic stem cells do not transdifferentiate
into cardiac myocytes in myocardial infarcts. Nature. 2004;428(6983):664-
668.
46. Caplan AI. Mesenchymal stem cells. J Orthop Res. 1991;9(5):641-650.
47. Krebsbach PH, Kuznetsov SA, Satomura K, Emmons RV, Rowe DW, Robey
PG. Bone formation in vivo: comparison of osteogenesis by transplanted
mouse and human marrow stromal fibroblasts. Transplantation.
1997;63(8):1059-1069.
48. Colter DC, Class R, DiGirolamo CM, Prockop DJ. Rapid expansion of
recycling stem cells in cultures of plastic-adherent cells from human bone
marrow. Proc Natl Acad Sci U S A. 2000;97(7):3213-3218.
49. Barbash IM, Chouraqui P, Baron J, Feinberg MS, Etzion S, Tessone A, Miller
L, Guetta E, Zipori D, Kedes LH, Kloner RA, Leor J. Systemic delivery of
191
bone marrow-derived mesenchymal stem cells to the infarcted myocardium:
feasibility, cell migration, and body distribution. Circulation.
2003;108(7):863-868.
50. Toma C, Pittenger MF, Cahill KS, Byrne BJ, Kessler PD. Human
mesenchymal stem cells differentiate to a cardiomyocyte phenotype in the
adult murine heart. Circulation. 2002;105(1):93-98.
51. Planat-Benard V, Menard C, Andre M, Puceat M, Perez A, Garcia-Verdugo
JM, Penicaud L, Casteilla L. Spontaneous cardiomyocyte differentiation from
adipose tissue stroma cells. Circ Res. 2004;94(2):223-229.
52. D'Angelo MG, Ausenda, C, Torrente, Y, Bordoni, A, Corti, S, Bo, R.D,
Perini, M.P, Colucci, M, Comi, H.P, Bresolin, N, Scarlato, G. 5'Azacytidine
Enhances Exogenous Gene Expression in Skeletal Muscle. Basic and Applied
Myology. 1999;9(1):5-10.
53. Hattan N, Kawaguchi H, Ando K, Kuwabara E, Fujita J, Murata M, Suematsu
M, Mori H, Fukuda K. Purified cardiomyocytes from bone marrow
mesenchymal stem cells produce stable intracardiac grafts in mice.
Cardiovasc Res. 2005;65(2):334-344.
54. Balana B, Nicoletti C, Zahanich I, Graf EM, Christ T, Boxberger S, Ravens U.
5-Azacytidine induces changes in electrophysiological properties of human
mesenchymal stem cells. Cell Res. 2006;16(12):949-960.
55. Xu W, Zhang X, Qian H, Zhu W, Sun X, Hu J, Zhou H, Chen Y.
Mesenchymal stem cells from adult human bone marrow differentiate into a
cardiomyocyte phenotype in vitro. Exp Biol Med (Maywood).
2004;229(7):623-631.
192
56. Kurita-Ochiai T, Amano S, Fukushima K, Ochiai K. Cellular events involved
in butyric acid-induced T cell apoptosis. J Immunol. 2003;171(7):3576-3584.
57. Beitnes JO, Hopp E, Lunde K, Solheim S, Arnesen H, Brinchmann JE,
Forfang K, Aakhus S. Long-term results after intracoronary injection of
autologous mononuclear bone marrow cells in acute myocardial infarction: the
ASTAMI randomised, controlled study. Heart. 2009;95(24):1983-1989.
58. Zohlnhofer D, Kastrati A, Schomig A. Stem cell mobilization by granulocyte-
colony-stimulating factor in acute myocardial infarction: lessons from the
REVIVAL-2 trial. Nat Clin Pract Cardiovasc Med. 2007;4 Suppl 1:S106-109.
59. Sussman MA, Murry CE. Bones of contention: marrow-derived cells in
myocardial regeneration. J Mol Cell Cardiol. 2008;44(6):950-953.
60. Breitbach M, Bostani T, Roell W, Xia Y, Dewald O, Nygren JM, Fries JW,
Tiemann K, Bohlen H, Hescheler J, Welz A, Bloch W, Jacobsen SE,
Fleischmann BK. Potential risks of bone marrow cell transplantation into
infarcted hearts. Blood. 2007;110(4):1362-1369.
61. Haider H, Tan AC, Aziz S, Chachques JC, Sim EK. Myoblast transplantation
for cardiac repair: a clinical perspective. Mol Ther. 2004;9(1):14-23.
62. Menasche P, Alfieri O, Janssens S, McKenna W, Reichenspurner H, Trinquart
L, Vilquin JT, Marolleau JP, Seymour B, Larghero J, Lake S, Chatellier G,
Solomon S, Desnos M, Hagege AA. The Myoblast Autologous Grafting in
Ischemic Cardiomyopathy (MAGIC) trial: first randomized placebo-controlled
study of myoblast transplantation. Circulation. 2008;117(9):1189-1200.
63. Hagege AA, Marolleau JP, Vilquin JT, Alheritiere A, Peyrard S, Duboc D,
Abergel E, Messas E, Mousseaux E, Schwartz K, Desnos M, Menasche P.
193
Skeletal myoblast transplantation in ischemic heart failure: long-term follow-
up of the first phase I cohort of patients. Circulation. 2006;114(1 Suppl):I108-
113.
64. Menasche P. Cell transplantation in myocardium. Ann Thorac Surg. 2003;75(6
Suppl):S20-28.
65. Pagani FD, DerSimonian H, Zawadzka A, Wetzel K, Edge AS, Jacoby DB,
Dinsmore JH, Wright S, Aretz TH, Eisen HJ, Aaronson KD. Autologous
skeletal myoblasts transplanted to ischemia-damaged myocardium in humans.
Histological analysis of cell survival and differentiation. J Am Coll Cardiol.
2003;41(5):879-888.
66. Menasche P. Myoblast transplantation: feasibility, safety and efficacy. Ann
Med. 2002;34(5):314-315.
67. Hodgetts SI, Beilharz MW, Scalzo AA, Grounds MD. Why do cultured
transplanted myoblasts die in vivo? DNA quantification shows enhanced
survival of donor male myoblasts in host mice depleted of CD4+ and CD8+
cells or Nk1.1+ cells. Cell Transplant. 2000;9(4):489-502.
68. Irintchev A, Zweyer M, Wernig A. Cellular and molecular reactions in mouse
muscles after myoblast implantation. J Neurocytol. 1995;24(4):319-331.
69. Pouzet B, Vilquin JT, Hagege AA, Scorsin M, Messas E, Fiszman M,
Schwartz K, Menasche P. Intramyocardial transplantation of autologous
myoblasts: can tissue processing be optimized? Circulation. 2000;102(19
Suppl 3):III210-215.
194
70. Skuk D, Caron N, Goulet M, Roy B, Espinosa F, Tremblay JP. Dynamics of
the early immune cellular reactions after myogenic cell transplantation. Cell
Transplant. 2002;11(7):671-681.
71. Orlic D, Kajstura J, Chimenti S, Jakoniuk I, Anderson SM, Li B, Pickel J,
McKay R, Nadal-Ginard B, Bodine DM, Leri A, Anversa P. Bone marrow
cells regenerate infarcted myocardium. Nature. 2001;410(6829):701-705.
72. Anversa P, Kajstura J, Leri A, Bolli R. Life and death of cardiac stem cells: a
paradigm shift in cardiac biology. Circulation. 2006;113(11):1451-1463.
73. Balsam LB, Robbins RC. Haematopoietic stem cells and repair of the
ischaemic heart. Clin Sci (Lond). 2005;109(6):483-492.
74. Fukuda K. Progress in myocardial regeneration and cell transplantation. Circ
J. 2005;69(12):1431-1446.
75. Murry CE, Field LJ, Menasche P. Cell-based cardiac repair: reflections at the
10-year point. Circulation. 2005;112(20):3174-3183.
76. Murry CE, Reinecke H, Pabon LM. Regeneration gaps: observations on stem
cells and cardiac repair. J Am Coll Cardiol. 2006;47(9):1777-1785.
77. Dawn B, Bolli R. Adult bone marrow-derived cells: regenerative potential,
plasticity, and tissue commitment. Basic Res Cardiol. 2005;100(6):494-503.
78. Strauer BE, Brehm M, Zeus T, Kostering M, Hernandez A, Sorg RV, Kogler
G, Wernet P. Repair of infarcted myocardium by autologous intracoronary
mononuclear bone marrow cell transplantation in humans. Circulation.
2002;106(15):1913-1918.
195
79. Assmus B, Schachinger V, Teupe C, Britten M, Lehmann R, Dobert N,
Grunwald F, Aicher A, Urbich C, Martin H, Hoelzer D, Dimmeler S, Zeiher
AM. Transplantation of Progenitor Cells and Regeneration Enhancement in
Acute Myocardial Infarction (TOPCARE-AMI). Circulation.
2002;106(24):3009-3017.
80. Schachinger V, Assmus B, Britten MB, Honold J, Lehmann R, Teupe C,
Abolmaali ND, Vogl TJ, Hofmann WK, Martin H, Dimmeler S, Zeiher AM.
Transplantation of progenitor cells and regeneration enhancement in acute
myocardial infarction: final one-year results of the TOPCARE-AMI Trial. J
Am Coll Cardiol. 2004;44(8):1690-1699.
81. Fernandez-Aviles F, San Roman JA, Garcia-Frade J, Fernandez ME,
Penarrubia MJ, de la Fuente L, Gomez-Bueno M, Cantalapiedra A, Fernandez
J, Gutierrez O, Sanchez PL, Hernandez C, Sanz R, Garcia-Sancho J, Sanchez
A. Experimental and clinical regenerative capability of human bone marrow
cells after myocardial infarction. Circ Res. 2004;95(7):742-748.
82. Wollert KC, Meyer GP, Lotz J, Ringes-Lichtenberg S, Lippolt P, Breidenbach
C, Fichtner S, Korte T, Hornig B, Messinger D, Arseniev L, Hertenstein B,
Ganser A, Drexler H. Intracoronary autologous bone-marrow cell transfer
after myocardial infarction: the BOOST randomised controlled clinical trial.
Lancet. 2004;364(9429):141-148.
83. Meyer GP, Wollert KC, Lotz J, Steffens J, Lippolt P, Fichtner S, Hecker H,
Schaefer A, Arseniev L, Hertenstein B, Ganser A, Drexler H. Intracoronary
bone marrow cell transfer after myocardial infarction: eighteen months'
follow-up data from the randomized, controlled BOOST (BOne marrOw
196
transfer to enhance ST-elevation infarct regeneration) trial. Circulation.
2006;113(10):1287-1294.
84. Janssens S, Dubois C, Bogaert J, Theunissen K, Deroose C, Desmet W,
Kalantzi M, Herbots L, Sinnaeve P, Dens J, Maertens J, Rademakers F,
Dymarkowski S, Gheysens O, Van Cleemput J, Bormans G, Nuyts J, Belmans
A, Mortelmans L, Boogaerts M, Van de Werf F. Autologous bone marrow-
derived stem-cell transfer in patients with ST-segment elevation myocardial
infarction: double-blind, randomised controlled trial. Lancet.
2006;367(9505):113-121.
85. Lunde K, Solheim S, Aakhus S, Arnesen H, Abdelnoor M, Egeland T,
Endresen K, Ilebekk A, Mangschau A, Fjeld JG, Smith HJ, Taraldsrud E,
Grogaard HK, Bjornerheim R, Brekke M, Muller C, Hopp E, Ragnarsson A,
Brinchmann JE, Forfang K. Intracoronary injection of mononuclear bone
marrow cells in acute myocardial infarction. N Engl J Med.
2006;355(12):1199-1209.
86. Schachinger V, Erbs S, Elsasser A, Haberbosch W, Hambrecht R,
Holschermann H, Yu J, Corti R, Mathey DG, Hamm CW, Suselbeck T,
Assmus B, Tonn T, Dimmeler S, Zeiher AM. Intracoronary bone marrow-
derived progenitor cells in acute myocardial infarction. N Engl J Med.
2006;355(12):1210-1221.
87. Hare JM, Chaparro SV. Cardiac regeneration and stem cell therapy. Curr Opin
Organ Transplant. 2008;13(5):536-542.
88. Rubart M, Field LJ. Cardiac regeneration: repopulating the heart. Annu Rev
Physiol. 2006;68:29-49.
197
89. Lunde K, Solheim S, Forfang K, Arnesen H, Brinch L, Bjornerheim R,
Ragnarsson A, Egeland T, Endresen K, Ilebekk A, Mangschau A, Aakhus S.
Anterior myocardial infarction with acute percutaneous coronary intervention
and intracoronary injection of autologous mononuclear bone marrow cells:
safety, clinical outcome, and serial changes in left ventricular function during
12-months' follow-up. J Am Coll Cardiol. 2008;51(6):674-676.
90. Aumailley M, Gayraud B. Structure and biological activity of the extracellular
matrix. J Mol Med. 1998;76(3-4):253-265.
91. Edwards DC, Sanders LC, Bokoch GM, Gill GN. Activation of LIM-kinase by
Pak1 couples Rac/Cdc42 GTPase signalling to actin cytoskeletal dynamics.
Nat Cell Biol. 1999;1(5):253-259.
92. van der Rest M, Garrone R. Collagen family of proteins. FASEB J.
1991;5(13):2814-2823.
93. Fratzl P, Misof K, Zizak I, Rapp G, Amenitsch H, Bernstorff S. Fibrillar
structure and mechanical properties of collagen. J Struct Biol. 1998;122(1-
2):119-122.
94. Eyre DR, Wu JJ, Fernandes RJ, Pietka TA, Weis MA. Recent developments in
cartilage research: matrix biology of the collagen II/IX/XI heterofibril
network. Biochem Soc Trans. 2002;30(Pt 6):893-899.
95. Birk DE. Type V collagen: heterotypic type I/V collagen interactions in the
regulation of fibril assembly. Micron. 2001;32(3):223-237.
96. Hynes RO. Integrins: bidirectional, allosteric signaling machines. Cell.
2002;110(6):673-687.
198
97. Krauss K, Altevogt P. Integrin leukocyte function-associated antigen-1-
mediated cell binding can be activated by clustering of membrane rafts. J Biol
Chem. 1999;274(52):36921-36927.
98. Li R, Mitra N, Gratkowski H, Vilaire G, Litvinov R, Nagasami C, Weisel JW,
Lear JD, DeGrado WF, Bennett JS. Activation of integrin alphaIIbbeta3 by
modulation of transmembrane helix associations. Science.
2003;300(5620):795-798.
99. Sun QH, Liu CY, Wang R, Paddock C, Newman PJ. Disruption of the long-
range GPIIIa Cys(5)-Cys(435) disulfide bond results in the production of
constitutively active GPIIb-IIIa (alpha(IIb)beta(3)) integrin complexes. Blood.
2002;100(6):2094-2101.
100. Vuori K. Integrin signaling: tyrosine phosphorylation events in focal
adhesions. J Membr Biol. 1998;165(3):191-199.
101. Damsky CH, Ilic D. Integrin signaling: it's where the action is. Curr Opin Cell
Biol. 2002;14(5):594-602.
102. Schoenwaelder SM, Burridge K. Bidirectional signaling between the
cytoskeleton and integrins. Curr Opin Cell Biol. 1999;11(2):274-286.
103. Czyz J, Wobus A. Embryonic stem cell differentiation: the role of
extracellular factors. Differentiation. 2001;68(4-5):167-174.
104. Gronthos S, Simmons PJ, Graves SE, Robey PG. Integrin-mediated
interactions between human bone marrow stromal precursor cells and the
extracellular matrix. Bone. 2001;28(2):174-181.
199
105. Roche P, Rousselle P, Lissitzky JC, Delmas PD, Malaval L. Isoform-specific
attachment of osteoprogenitors to laminins: mapping to the short arms of
laminin-1. Exp Cell Res. 1999;250(2):465-474.
106. Heino J. The collagen receptor integrins have distinct ligand recognition and
signaling functions. Matrix Biol. 2000;19(4):319-323.
107. Gullberg DE, Lundgren-Akerlund E. Collagen-binding I domain integrins--
what do they do? Prog Histochem Cytochem. 2002;37(1):3-54.
108. Ivaska J, Reunanen H, Westermarck J, Koivisto L, Kahari VM, Heino J.
Integrin alpha2beta1 mediates isoform-specific activation of p38 and
upregulation of collagen gene transcription by a mechanism involving the
alpha2 cytoplasmic tail. J Cell Biol. 1999;147(2):401-416.
109. Riikonen T, Westermarck J, Koivisto L, Broberg A, Kahari VM, Heino J.
Integrin alpha 2 beta 1 is a positive regulator of collagenase (MMP-1) and
collagen alpha 1(I) gene expression. J Biol Chem. 1995;270(22):13548-13552.
110. Pierschbacher MD, Ruoslahti E. Cell attachment activity of fibronectin can be
duplicated by small synthetic fragments of the molecule. Nature.
1984;309(5963):30-33.
111. Agrez MV, Bates RC, Boyd AW, Burns GF. Arg-Gly-Asp-containing peptides
expose novel collagen receptors on fibroblasts: implications for wound
healing. Cell Regul. 1991;2(12):1035-1044.
112. Cooke ME, Sakai T, Mosher DF. Contraction of collagen matrices mediated
by alpha2beta1A and alpha(v)beta3 integrins. J Cell Sci. 2000;113 ( Pt
13):2375-2383.
200
113. Grundstrom G, Mosher DF, Sakai T, Rubin K. Integrin alphavbeta3 mediates
platelet-derived growth factor-BB-stimulated collagen gel contraction in cells
expressing signaling deficient integrin alpha2beta1. Exp Cell Res.
2003;291(2):463-473.
114. Levinson H, Hopper JE, Ehrlich HP. Overexpression of integrin alphav
promotes human osteosarcoma cell populated collagen lattice contraction and
cell migration. J Cell Physiol. 2002;193(2):219-224.
115. Zimmermann WH, Schneiderbanger K, Schubert P, Didie M, Munzel F,
Heubach JF, Kostin S, Neuhuber WL, Eschenhagen T. Tissue engineering of a
differentiated cardiac muscle construct. Circ Res. 2002;90(2):223-230.
116. Radisic M, Park H, Shing H, Consi T, Schoen FJ, Langer R, Freed LE,
Vunjak-Novakovic G. Functional assembly of engineered myocardium by
electrical stimulation of cardiac myocytes cultured on scaffolds. Proc Natl
Acad Sci U S A. 2004;101(52):18129-18134.
117. Simpson DG, Terracio L, Terracio M, Price RL, Turner DC, Borg TK.
Modulation of cardiac myocyte phenotype in vitro by the composition and
orientation of the extracellular matrix. J Cell Physiol. 1994;161(1):89-105.
118. Shim WS, Jiang S, Wong P, Tan J, Chua YL, Tan YS, Sin YK, Lim CH, Chua
T, Teh M, Liu TC, Sim E. Ex vivo differentiation of human adult bone
marrow stem cells into cardiomyocyte-like cells. Biochem Biophys Res
Commun. 2004;324(2):481-488.
119. Keller G. Embryonic stem cell differentiation: emergence of a new era in
biology and medicine. Genes Dev. 2005;19(10):1129-1155.
201
120. Hescheler J, Fleischmann BK, Lentini S, Maltsev VA, Rohwedel J, Wobus
AM, Addicks K. Embryonic stem cells: a model to study structural and
functional properties in cardiomyogenesis. Cardiovasc Res. 1997;36(2):149-
162.
121. Molkentin JD, Kalvakolanu DV, Markham BE. Transcription factor GATA-4
regulates cardiac muscle-specific expression of the alpha-myosin heavy-chain
gene. Mol Cell Biol. 1994;14(7):4947-4957.
122. Wakimoto H, Kasahara H, Maguire CT, Moskowitz IP, Izumo S, Berul CI.
Cardiac electrophysiological phenotypes in postnatal expression of Nkx2.5
transgenic mice. Genesis. 2003;37(3):144-150.
123. Li H, S. Z, Z. H, Y. W, M. A, M. X. GATA-4 promotes Mesenchymal Stem
Cells Transdifferentiation Into Cardiomyocyte via Insulin Growth Factor
Binding Protein 4 Upregulation. Circulation. 2009;120:S1040.
124. Arminan A, Gandia C, Bartual M, Garcia-Verdugo JM, Lledo E, Mirabet V,
Llop M, Barea J, Montero JA, Sepulveda P. Cardiac differentiation is driven
by NKX2.5 and GATA4 nuclear translocation in tissue-specific mesenchymal
stem cells. Stem Cells Dev. 2009;18(6):907-918.
125. Simon JP, Shen TH, Ivanov IE, Gravotta D, Morimoto T, Adesnik M, Sabatini
DD. Coatomer, but not P200/myosin II, is required for the in vitro formation
of trans-Golgi network-derived vesicles containing the envelope glycoprotein
of vesicular stomatitis virus. Proc Natl Acad Sci U S A. 1998;95(3):1073-
1078.
126. Beltrami AP, Barlucchi L, Torella D, Baker M, Limana F, Chimenti S,
Kasahara H, Rota M, Musso E, Urbanek K, Leri A, Kajstura J, Nadal-Ginard
202
B, Anversa P. Adult cardiac stem cells are multipotent and support myocardial
regeneration. Cell. 2003;114(6):763-776.
127. Wenstrup RJ, Florer JB, Brunskill EW, Bell SM, Chervoneva I, Birk DE.
Type V collagen controls the initiation of collagen fibril assembly. J Biol
Chem. 2004;279(51):53331-53337.
128. Berendsen AD, Bronckers AL, Smit TH, Walboomers XF, Everts V. Collagen
type V enhances matrix contraction by human periodontal ligament fibroblasts
seeded in three-dimensional collagen gels. Matrix Biol. 2006;25(8):515-522.
129. Conget PA, Minguell JJ. Phenotypical and functional properties of human
bone marrow mesenchymal progenitor cells. J Cell Physiol. 1999;181(1):67-
73.
130. Phinney DG, Kopen G, Isaacson RL, Prockop DJ. Plastic adherent stromal
cells from the bone marrow of commonly used strains of inbred mice:
variations in yield, growth, and differentiation. J Cell Biochem.
1999;72(4):570-585.
131. Kunisaki Y, Takase K, Miyamoto T, Fukata M, Nonami A, Kamezaki K, Kaji
Y, Gondo H, Harada M, Nagafuji K. Marked improvement of cardiac function
early after non-myeloablative BMT in a heavily transfused patient with severe
aplastic anemia and heart failure. Bone Marrow Transplant. 2007;40(6):593-
595.
132. Ross J, Jr., Peterson KL. On the assessment of cardiac inotropic state.
Circulation. 1973;47(3):435-438.
133. Itescu S, Schuster MD, Kocher AA. New directions in strategies using cell
therapy for heart disease. J Mol Med. 2003;81(5):288-296.
203
134. Wang T, Xu Z, Jiang W, Ma A. Cell-to-cell contact induces mesenchymal
stem cell to differentiate into cardiomyocyte and smooth muscle cell. Int J
Cardiol. 2006;109(1):74-81.
135. Li X, Yu X, Lin Q, Deng C, Shan Z, Yang M, Lin S. Bone marrow
mesenchymal stem cells differentiate into functional cardiac phenotypes by
cardiac microenvironment. J Mol Cell Cardiol. 2007;42(2):295-303.
136. Kolossov E, Bostani T, Roell W, Breitbach M, Pillekamp F, Nygren JM, Sasse
P, Rubenchik O, Fries JW, Wenzel D, Geisen C, Xia Y, Lu Z, Duan Y,
Kettenhofen R, Jovinge S, Bloch W, Bohlen H, Welz A, Hescheler J, Jacobsen
SE, Fleischmann BK. Engraftment of engineered ES cell-derived
cardiomyocytes but not BM cells restores contractile function to the infarcted
myocardium. J Exp Med. 2006;203(10):2315-2327.
137. Takeda Y, Mori T, Imabayashi H, Kiyono T, Gojo S, Miyoshi S, Hida N, Ita
M, Segawa K, Ogawa S, Sakamoto M, Nakamura S, Umezawa A. Can the life
span of human marrow stromal cells be prolonged by bmi-1, E6, E7, and/or
telomerase without affecting cardiomyogenic differentiation? J Gene Med.
2004;6(8):833-845.
138. Takahashi T, Lord B, Schulze PC, Fryer RM, Sarang SS, Gullans SR, Lee RT.
Ascorbic acid enhances differentiation of embryonic stem cells into cardiac
myocytes. Circulation. 2003;107(14):1912-1916.
139. Young HE, Steele TA, Bray RA, Hudson J, Floyd JA, Hawkins K, Thomas K,
Austin T, Edwards C, Cuzzourt J, Duenzl M, Lucas PA, Black AC, Jr. Human
reserve pluripotent mesenchymal stem cells are present in the connective
204
tissues of skeletal muscle and dermis derived from fetal, adult, and geriatric
donors. Anat Rec. 2001;264(1):51-62.
140. Martin-Rendon E, Sweeney D, Lu F, Girdlestone J, Navarrete C, Watt SM. 5-
Azacytidine-treated human mesenchymal stem/progenitor cells derived from
umbilical cord, cord blood and bone marrow do not generate cardiomyocytes
in vitro at high frequencies. Vox Sang. 2008;95(2):137-148.
141. Torella D, Rota M, Nurzynska D, Musso E, Monsen A, Shiraishi I, Zias E,
Walsh K, Rosenzweig A, Sussman MA, Urbanek K, Nadal-Ginard B, Kajstura
J, Anversa P, Leri A. Cardiac stem cell and myocyte aging, heart failure, and
insulin-like growth factor-1 overexpression. Circ Res. 2004;94(4):514-524.
142. Urbich C, Aicher A, Heeschen C, Dernbach E, Hofmann WK, Zeiher AM,
Dimmeler S. Soluble factors released by endothelial progenitor cells promote
migration of endothelial cells and cardiac resident progenitor cells. J Mol Cell
Cardiol. 2005;39(5):733-742.
143. Mollenhauer HH, Morre DJ, Rowe LD. Alteration of intracellular traffic by
monensin; mechanism, specificity and relationship to toxicity. Biochim
Biophys Acta. 1990;1031(2):225-246.
144. Ikonen E, de Almeid JB, Fath KR, Burgess DR, Ashman K, Simons K, Stow
JL. Myosin II is associated with Golgi membranes: identification of p200 as
nonmuscle myosin II on Golgi-derived vesicles. J Cell Sci. 1997;110 ( Pt
18):2155-2164.
145. Percival JM, Hughes JA, Brown DL, Schevzov G, Heimann K, Vrhovski B,
Bryce N, Stow JL, Gunning PW. Targeting of a tropomyosin isoform to short
205
microfilaments associated with the Golgi complex. Mol Biol Cell.
2004;15(1):268-280.
146. Warner CL, Stewart A, Luzio JP, Steel KP, Libby RT, Kendrick-Jones J, Buss
F. Loss of myosin VI reduces secretion and the size of the Golgi in fibroblasts
from Snell's waltzer mice. EMBO J. 2003;22(3):569-579.
147. Hasegawa K, Lee SJ, Jobe SM, Markham BE, Kitsis RN. cis-Acting
sequences that mediate induction of beta-myosin heavy chain gene expression
during left ventricular hypertrophy due to aortic constriction. Circulation.
1997;96(11):3943-3953.
148. Nakanishi C, Yamagishi M, Yamahara K, Hagino I, Mori H, Sawa Y,
Yagihara T, Kitamura S, Nagaya N. Activation of cardiac progenitor cells
through paracrine effects of mesenchymal stem cells. Biochem Biophys Res
Commun. 2008;374(1):11-16.
149. Fujii T, Yau TM, Weisel RD, Ohno N, Mickle DA, Shiono N, Ozawa T,
Matsubayashi K, Li RK. Cell transplantation to prevent heart failure: a
comparison of cell types. Ann Thorac Surg. 2003;76(6):2062-2070; discussion
2070.
150. Sinanan AC, Machell JR, Wynne-Hughes GT, Hunt NP, Lewis MP.
alphavbeta3 and alphavbeta5 integrins and their role in muscle precursor cell
adhesion. Biol Cell. 2008.
151. Ruggiero F, Comte J, Cabanas C, Garrone R. Structural requirements for alpha
1 beta 1 and alpha 2 beta 1 integrin mediated cell adhesion to collagen V. J
Cell Sci. 1996;109 ( Pt 7):1865-1874.
206
152. Ruggiero F, Champliaud MF, Garrone R, Aumailley M. Interactions between
cells and collagen V molecules or single chains involve distinct mechanisms.
Exp Cell Res. 1994;210(2):215-223.
153. Fichard A, Kleman JP, Ruggiero F. Another look at collagen V and XI
molecules. Matrix Biol. 1995;14(7):515-531.
154. Guan K, Czyz J, Furst DO, Wobus AM. Expression and cellular distribution
of alpha(v)integrins in beta(1)integrin-deficient embryonic stem cell-derived
cardiac cells. J Mol Cell Cardiol. 2001;33(3):521-532.
155. Zheng B, Clemmons DR. Blocking ligand occupancy of the alphaVbeta3
integrin inhibits insulin-like growth factor I signaling in vascular smooth
muscle cells. Proc Natl Acad Sci U S A. 1998;95(19):11217-11222.
156. Liang Q, Wiese RJ, Bueno OF, Dai YS, Markham BE, Molkentin JD. The
transcription factor GATA4 is activated by extracellular signal-regulated
kinase 1- and 2-mediated phosphorylation of serine 105 in cardiomyocytes.
Mol Cell Biol. 2001;21(21):7460-7469.
157. Chen Y, Amende I, Hampton TG, Yang Y, Ke Q, Min JY, Xiao YF, Morgan
JP. Vascular endothelial growth factor promotes cardiomyocyte differentiation
of embryonic stem cells. Am J Physiol Heart Circ Physiol.
2006;291(4):H1653-1658.
158. Ross RS, Borg TK. Integrins and the myocardium. Circ Res.
2001;88(11):1112-1119.
159. Kraehenbuehl TP, Zammaretti P, Van der Vlies AJ, Schoenmakers RG, Lutolf
MP, Jaconi ME, Hubbell JA. Three-dimensional extracellular matrix-directed
207
cardioprogenitor differentiation: systematic modulation of a synthetic cell-
responsive PEG-hydrogel. Biomaterials. 2008;29(18):2757-2766.
160. Xaymardan M, Tang L, Zagreda L, Pallante B, Zheng J, Chazen JL, Chin A,
Duignan I, Nahirney P, Rafii S, Mikawa T, Edelberg JM. Platelet-derived
growth factor-AB promotes the generation of adult bone marrow-derived
cardiac myocytes. Circ Res. 2004;94(5):E39-45.
161. Schneller M, Vuori K, Ruoslahti E. Alphavbeta3 integrin associates with
activated insulin and PDGFbeta receptors and potentiates the biological
activity of PDGF. Embo J. 1997;16(18):5600-5607.
162. Takagi J, Fujisawa T, Usui T, Aoyama T, Saito Y. A single chain 19-kDa
fragment from bovine thrombospondin binds to type V collagen and heparin. J
Biol Chem. 1993;268(21):15544-15549.
163. Ricard-Blum S, Beraud M, Raynal N, Farndale RW, Ruggiero F. Structural
requirements for heparin/heparan sulfate binding to type V collagen. J Biol
Chem. 2006;281(35):25195-25204.
164. Perin EC, Dohmann HF, Borojevic R, Silva SA, Sousa AL, Mesquita CT,
Rossi MI, Carvalho AC, Dutra HS, Dohmann HJ, Silva GV, Belem L,
Vivacqua R, Rangel FO, Esporcatte R, Geng YJ, Vaughn WK, Assad JA,
Mesquita ET, Willerson JT. Transendocardial, autologous bone marrow cell
transplantation for severe, chronic ischemic heart failure. Circulation.
2003;107(18):2294-2302.
165. Chen SL, Fang WW, Ye F, Liu YH, Qian J, Shan SJ, Zhang JJ, Chunhua RZ,
Liao LM, Lin S, Sun JP. Effect on left ventricular function of intracoronary
208
transplantation of autologous bone marrow mesenchymal stem cell in patients
with acute myocardial infarction. Am J Cardiol. 2004;94(1):92-95.
166. Diederichsen AC, Moller JE, Thayssen P, Junker AB, Videbaek L, Saekmose
SG, Barington T, Kristiansen M, Kassem M. Effect of repeated intracoronary
injection of bone marrow cells in patients with ischaemic heart failure The
Danish Stem Cell study-Congestive Heart Failure trial (DanCell-CHF). Eur J
Heart Fail. 2008;10(7):661-667.
167. Penicka M, Horak J, Kobylka P, Pytlik R, Kozak T, Belohlavek O, Lang O,
Skalicka H, Simek S, Palecek T, Linhart A, Aschermann M, Widimsky P.
Intracoronary injection of autologous bone marrow-derived mononuclear cells
in patients with large anterior acute myocardial infarction: a prematurely
terminated randomized study. J Am Coll Cardiol. 2007;49(24):2373-2374.
168. Panovsky R, Meluzin J, Janousek S, Mayer J, Kaminek M, Groch L, Prasek J,
Stanicek J, Dusek L, Hlinomaz O, Kala P, Klabusay M, Koristek Z, Navratil
M. Cell Therapy in Patients with Left Ventricular Dysfunction Due to
Myocardial Infarction. Echocardiography. 2008.
169. Meluzin J, Janousek S, Mayer J, Groch L, Hornacek I, Hlinomaz O, Kala P,
Panovsky R, Prasek J, Kaminek M, Stanicek J, Klabusay M, Koristek Z,
Navratil M, Dusek L, Vinklarkova J. Three-, 6-, and 12-month results of
autologous transplantation of mononuclear bone marrow cells in patients with
acute myocardial infarction. Int J Cardiol. 2007.
170. Schierling W, Kunz-Schughart LA, Muders F, Riegger GA, Griese DP. Fates
of genetically engineered haematopoietic and mesenchymal stem cell grafts in
normal and injured rat hearts. J Tissue Eng Regen Med. 2008.
209
171. Norol F, Bonnet N, Peinnequin A, Chretien F, Legrand R, Isnard R, Herodin
F, Baillou C, Delache B, Negre D, Klatzmann D, Vernant JP, Lemoine FM.
GFP-transduced CD34+ and Lin- CD34- hematopoietic stem cells did not
adopt a cardiac phenotype in a nonhuman primate model of myocardial
infarct. Exp Hematol. 2007;35(4):653-661.
172. de Weger RA, Verbrugge I, Bruggink AH, van Oosterhout MM, de Souza Y,
van Wichen DF, Gmelig-Meyling FH, de Jonge N, Verdonck LF. Stem cell-
derived cardiomyocytes after bone marrow and heart transplantation. Bone
Marrow Transplant. 2007.
173. Jiang W, Ma A, Wang T, Han K, Liu Y, Zhang Y, Dong A, Du Y, Huang X,
Wang J, Lei X, Zheng X. Homing and differentiation of mesenchymal stem
cells delivered intravenously to ischemic myocardium in vivo: a time-series
study. Pflugers Arch. 2006.
174. Grinnemo KH, Mansson-Broberg A, Leblanc K, Corbascio M, Wardell E,
Siddiqui AJ, Hao X, Sylven C, Dellgren G. Human mesenchymal stem cells
do not differentiate into cardiomyocytes in a cardiac ischemic xenomodel. Ann
Med. 2006;38(2):144-153.
175. Thompson RB, van den Bos EJ, Davis BH, Morimoto Y, Craig D, Sutton BS,
Glower DD, Taylor DA. Intracardiac transplantation of a mixed population of
bone marrow cells improves both regional systolic contractility and diastolic
relaxation. J Heart Lung Transplant. 2005;24(2):205-214.
176. Kajstura J, Rota M, Whang B, Cascapera S, Hosoda T, Bearzi C, Nurzynska
D, Kasahara H, Zias E, Bonafe M, Nadal-Ginard B, Torella D, Nascimbene A,
Quaini F, Urbanek K, Leri A, Anversa P. Bone marrow cells differentiate in
210
cardiac cell lineages after infarction independently of cell fusion. Circ Res.
2005;96(1):127-137.
177. Olivares EL, Ribeiro VP, Werneck de Castro JP, Ribeiro KC, Mattos EC,
Goldenberg RC, Mill JG, Dohmann HF, dos Santos RR, de Carvalho AC,
Masuda MO. Bone marrow stromal cells improve cardiac performance in
healed infarcted rat hearts. Am J Physiol Heart Circ Physiol.
2004;287(2):H464-470.
178. Bayes-Genis A, Roura S, Soler-Botija C, Farre J, Hove-Madsen L, Llach A,
Cinca J. Identification of cardiomyogenic lineage markers in untreated human
bone marrow-derived mesenchymal stem cells. Transplant Proc.
2005;37(9):4077-4079.
179. Pereira WC, Khushnooma I, Madkaikar M, Ghosh K. Reproducible
methodology for the isolation of mesenchymal stem cells from human
umbilical cord and its potential for cardiomyocyte generation. J Tissue Eng
Regen Med. 2008.
180. Fazel S, Chen L, Weisel RD, Angoulvant D, Seneviratne C, Fazel A, Cheung
P, Lam J, Fedak PW, Yau TM, Li RK. Cell transplantation preserves cardiac
function after infarction by infarct stabilization: augmentation by stem cell
factor. J Thorac Cardiovasc Surg. 2005;130(5):1310.
181. Frantz S, Vallabhapurapu D, Tillmanns J, Brousos N, Wagner H, Henig K,
Ertl G, Muller AM, Bauersachs J. Impact of different bone marrow cell
preparations on left ventricular remodelling after experimental myocardial
infarction. Eur J Heart Fail. 2008;10(2):119-124.
211
182. Shyu KG, Wang BW, Hung HF, Chang CC, Shih DT. Mesenchymal stem
cells are superior to angiogenic growth factor genes for improving myocardial
performance in the mouse model of acute myocardial infarction. J Biomed Sci.
2005:1-12.
183. Gnecchi M, He H, Liang OD, Melo LG, Morello F, Mu H, Noiseux N, Zhang
L, Pratt RE, Ingwall JS, Dzau VJ. Paracrine action accounts for marked
protection of ischemic heart by Akt-modified mesenchymal stem cells. Nat
Med. 2005;11(4):367-368.
184. Jiang Y, Jahagirdar BN, Reinhardt RL, Schwartz RE, Keene CD, Ortiz-
Gonzalez XR, Reyes M, Lenvik T, Lund T, Blackstad M, Du J, Aldrich S,
Lisberg A, Low WC, Largaespada DA, Verfaillie CM. Pluripotency of
mesenchymal stem cells derived from adult marrow. Nature.
2002;418(6893):41-49.
185. Jiang CY, Gui C, He AN, Hu XY, Chen J, Jiang Y, Wang JA. Optimal time
for mesenchymal stem cell transplantation in rats with myocardial infarction. J
Zhejiang Univ Sci B. 2008;9(8):630-637.
186. Muller-Ehmsen J, Krausgrill B, Burst V, Schenk K, Neisen UC, Fries JW,
Fleischmann BK, Hescheler J, Schwinger RH. Effective engraftment but poor
mid-term persistence of mononuclear and mesenchymal bone marrow cells in
acute and chronic rat myocardial infarction. J Mol Cell Cardiol.
2006;41(5):876-884.
187. Rowlands AS, George PA, Cooper-White JJ. Directing osteogenic &
myogenic differentiation of MSCs: interplay of stiffness & adhesive ligand
presentation. Am J Physiol Cell Physiol. 2008.
212
188. Kundu AK, Putnam AJ. Vitronectin and collagen I differentially regulate
osteogenesis in mesenchymal stem cells. Biochem Biophys Res Commun.
2006;347(1):347-357.
189. Chastain SR, Kundu AK, Dhar S, Calvert JW, Putnam AJ. Adhesion of
mesenchymal stem cells to polymer scaffolds occurs via distinct ECM ligands
and controls their osteogenic differentiation. J Biomed Mater Res A.
2006;78(1):73-85.
190. Sato H, Takahashi M, Ise H, Yamada A, Hirose S, Tagawa Y, Morimoto H,
Izawa A, Ikeda U. Collagen synthesis is required for ascorbic acid-enhanced
differentiation of mouse embryonic stem cells into cardiomyocytes. Biochem
Biophys Res Commun. 2006;342(1):107-112.
191. Wenstrup RJ, Florer JB, Brunskill EW, Bell SM, Chervoneva I, Birk DE.
Type V collagen controls the initiation of collagen fibril assembly. J Biol
Chem. 2004.
192. Shimomura T, Jia F, Niyibizi C, Woo SL. Antisense oligonucleotides reduce
synthesis of procollagen alpha1 (V) chain in human patellar tendon
fibroblasts: potential application in healing ligaments and tendons. Connect
Tissue Res. 2003;44(3-4):167-172.
193. Nian M, Lee P, Khaper N, Liu P. Inflammatory cytokines and postmyocardial
infarction remodeling. Circ Res. 2004;94(12):1543-1553.
194. Fedak PW. Paracrine effects of cell transplantation: modifying ventricular
remodeling in the failing heart. Semin Thorac Cardiovasc Surg.
2008;20(2):87-93.
213
195. Murasawa Y, Hayashi T, Wang PC. The role of type V collagen fibril as an
ECM that induces the motility of glomerular endothelial cells. Exp Cell Res.
2008;314(20):3638-3653.
196. Viglio S, Zoppi N, Sangalli A, Gallanti A, Barlati S, Mottes M, Colombi M,
Valli M. Rescue of Migratory Defects of Ehlers-Danlos Syndrome Fibroblasts
In Vitro by Type V Collagen but not Insulin-Like Binding Protein-1. J Invest
Dermatol. 2008.
197. Hahn JY, Cho HJ, Kang HJ, Kim TS, Kim MH, Chung JH, Bae JW, Oh BH,
Park YB, Kim HS. Pre-treatment of mesenchymal stem cells with a
combination of growth factors enhances gap junction formation,
cytoprotective effect on cardiomyocytes, and therapeutic efficacy for
myocardial infarction. J Am Coll Cardiol. 2008;51(9):933-943.
198. Bartunek J, Croissant JD, Wijns W, Gofflot S, de Lavareille A, Vanderheyden
M, Kaluzhny Y, Mazouz N, Willemsen P, Penicka M, Mathieu M, Homsy C,
De Bruyne B, McEntee K, Lee IW, Heyndrickx GR. Pretreatment of adult
bone marrow mesenchymal stem cells with cardiomyogenic growth factors
and repair of the chronically infarcted myocardium. Am J Physiol Heart Circ
Physiol. 2006.
199. Baicu CF, Stroud JD, Livesay VA, Hapke E, Holder J, Spinale FG, Zile MR.
Changes in extracellular collagen matrix alter myocardial systolic
performance. Am J Physiol Heart Circ Physiol. 2003;284(1):H122-132.
200. Danzmann LC, Bodanese LC, Kohler I, Torres MR. Left atrioventricular
remodeling in the assessment of the left ventricle diastolic function in patients
214
with heart failure: a review of the currently studied echocardiographic
variables. Cardiovasc Ultrasound. 2008;6:56.
201. Villari B, Campbell SE, Hess OM, Mall G, Vassalli G, Weber KT,
Krayenbuehl HP. Influence of collagen network on left ventricular systolic
and diastolic function in aortic valve disease. J Am Coll Cardiol.
1993;22(5):1477-1484.
202. Berry MF, Engler AJ, Woo YJ, Pirolli TJ, Bish LT, Jayasankar V, Morine KJ,
Gardner TJ, Discher DE, Sweeney HL. Mesenchymal Stem Cell Injection
After Myocardial Infarction Improves Myocardial Compliance. Am J Physiol
Heart Circ Physiol. 2006.
203. Farahmand P, Lai TY, Weisel RD, Fazel S, Yau T, Menasche P, Li RK.
Skeletal myoblasts preserve remote matrix architecture and global function
when implanted early or late after coronary ligation into infarcted or remote
myocardium. Circulation. 2008;118(14 Suppl):S130-137.
204. Imanishi Y, Saito A, Komoda H, Kitagawa-Sakakida S, Miyagawa S, Kondoh
H, Ichikawa H, Sawa Y. Allogenic mesenchymal stem cell transplantation has
a therapeutic effect in acute myocardial infarction in rats. J Mol Cell Cardiol.
2008;44(4):662-671.
205. Uemura R, Xu M, Ahmad N, Ashraf M. Bone marrow stem cells prevent left
ventricular remodeling of ischemic heart through paracrine signaling. Circ
Res. 2006;98(11):1414-1421.
206. Li L, Zhang S, Zhang Y, Yu B, Xu Y, Guan Z. Paracrine action mediate the
antifibrotic effect of transplanted mesenchymal stem cells in a rat model of
global heart failure. Mol Biol Rep. 2009;36(4):725-731.
215
207. Scorsin M, Hagege A, Vilquin JT, Fiszman M, Marotte F, Samuel JL,
Rappaport L, Schwartz K, Menasche P. Comparison of the effects of fetal
cardiomyocyte and skeletal myoblast transplantation on postinfarction left
ventricular function. J Thorac Cardiovasc Surg. 2000;119(6):1169-1175.
208. Sakai T, Li RK, Weisel RD, Mickle DA, Jia ZQ, Tomita S, Kim EJ, Yau TM.
Fetal cell transplantation: a comparison of three cell types. J Thorac
Cardiovasc Surg. 1999;118(4):715-724.
209. Connelly KA, Prior DL, Kelly DJ, Feneley MP, Krum H, Gilbert RE. Load-
sensitive measures may overestimate global systolic function in the presence
of left ventricular hypertrophy: a comparison with load-insensitive measures.
Am J Physiol Heart Circ Physiol. 2006;290(4):H1699-1705.
210. Suga H, Sagawa K, Shoukas AA. Load independence of the instantaneous
pressure-volume ratio of the canine left ventricle and effects of epinephrine
and heart rate on the ratio. Circ Res. 1973;32(3):314-322.
211. Wallace A, Lam HW, Mangano DT. Linearity, load dependence, hysteresis,
and clinical associations of systolic and diastolic indices of left ventricular
function in man. Multicenter Study of Perioperative Ischemia (McSPI)
Research Group. J Card Surg. 1995;10(4 Suppl):460-467.
212. Kass DA, Beyar R, Lankford E, Heard M, Maughan WL, Sagawa K. Influence
of contractile state on curvilinearity of in situ end-systolic pressure-volume
relations. Circulation. 1989;79(1):167-178.
213. Muller-Ehmsen J, Peterson KL, Kedes L, Whittaker P, Dow JS, Long TI,
Laird PW, Kloner RA. Rebuilding a damaged heart: long-term survival of
216
transplanted neonatal rat cardiomyocytes after myocardial infarction and
effect on cardiac function. Circulation. 2002;105(14):1720-1726.
214. Xu X, Xu Z, Xu Y, Cui G. Effects of mesenchymal stem cell transplantation
on extracellular matrix after myocardial infarction in rats. Coron Artery Dis.
2005;16(4):245-255.
215. Yakabe K, Ikeda S, Urata J, Koga S, Chijiwa R, Ninomiya A, Fukukawa F,
Taura K, Kohno S. Significance of Tei-index alterations induced by acute
preload reduction with hemodialysis. Clin Nephrol. 2008;70(1):41-47.
216. Zile MR, Gaasch WH, Carroll JD, Feldman MD, Aurigemma GP, Schaer GL,
Ghali JK, Liebson PR. Heart failure with a normal ejection fraction: is
measurement of diastolic function necessary to make the diagnosis of diastolic
heart failure? Circulation. 2001;104(7):779-782.
217. Iida M, Heike T, Yoshimoto M, Baba S, Doi H, Nakahata T. Identification of
cardiac stem cells with FLK1, CD31, and VE-cadherin expression during
embryonic stem cell differentiation. Faseb J. 2005;19(3):371-378.
218. Yang L, Soonpaa MH, Adler ED, Roepke TK, Kattman SJ, Kennedy M,
Henckaerts E, Bonham K, Abbott GW, Linden RM, Field LJ, Keller GM.
Human cardiovascular progenitor cells develop from a KDR(+) embryonic-
stem-cell-derived population. Nature. 2008.
219. Engler AJ, Sen S, Sweeney HL, Discher DE. Matrix elasticity directs stem cell
lineage specification. Cell. 2006;126(4):677-689.