Formation of an intact, continuous and biological interface with proper zonal organization between mechanically dissimilar tissues is a key challenge in complex tissue engineering. The presence of a stable interface between soft and hard tissues is important. In particular, the presence of the osteochondral interface can prevent mechanical failure by reducing the shear stress across it. It also prevents vascularization and subsequent mineralization of the uncalcified cartilage, thus maintaining the normal tissue function. In this study, we demonstrated that with the use of mesenchymal stem cells, the collagen scaffold and the microencapsulation technology, an osteochondral interface with a zone of calcified cartilage could be generated in vitro in 5 weeks. Specifically, by placing an undifferentiated mesenchymal stem cell-collagen gel between an upper cartilage-like part and a lower bone-like part, cells in the middle layer were able to remodel the collagen gel into an interface similar to that found in vivo. Hypertrophic chondrocytes populated this in vitro generated interface, secreting GAGs, collagen type II and X, and calcium phosphates. Vertically running collagen fibers were found in this interface as well. We also demonstrated the importance of culture medium together with an appropriate configuration for interface formation. In particular, only with the use of both the chondrogenic medium and the three-layer configuration could we generate the osteochondral interface in vitro. Finally we conducted a pilot animal study on the efficacy of cartilage repair using constructs with a pre-formed osteochondral interface and demonstrated that cartilage re-surfacing was successful in only one month. Hyaline-like cartilage with a continuous tidemark was regenerated. This observed phenomenon could be maintained up to 3 months. Results of this study contribute to the development of better cartilage repair in future. / published_or_final_version / Mechanical Engineering / Doctoral / Doctor of Philosophy

Neural stem cells (NSCs) exist in both embryonic and adult neural tissues and are characterized by their self-renewal capacity and multipotency that contribute to the generation of three major cell types in the vertebrate central nervous system (CNS):neurons, oligodendrocytes and astrocytes. The tremendous therapeutic potential of NSCs to treat the neurodegenerative diseases and repair brain injuries has provoked intensive study in the molecular regulation of their induction, maintenance and differentiation. Previous study reported that Sox9, a member of high-mobility-group(HMG) containing SoxE transcription factors family, plays important roles in regulating the formation and maintenance of NSCs in both mouse and chick CNS, as well as the cell fate switch between neuronal and glial. Whether it plays similar roles in human NSCs (hNSCs)is still unknown. My RT-qPCR analysis showed that SOX9is expressed at a basal level in human embryonic stem cells (hESCs) and up-regulated upon commitment into neural lineage and maintained at a high level in hESCs-derived hNSCs. I therefore hypothesized that SOX9 might also be involved in the induction, maintenance and differentiation of hNSCs. To test this, two stable hESC lines(HES2)were generated with each constitutively expressing short hairpin RNA (shRNA) against SOX9andGL2 Luciferase (Luc, as control) respectively. Upon neural induction, SOX9-knock-down(KD) hESCs were able to commit neural lineage and differentiate into NSCs/neurospheres (NSPs), however, these NSCs exhibited reduced multipotency and glial marker (GALC, CD44) expressions but enhanced self-renewal compared to the shLuc NSCs. Hence, SOX9 is required for both the induction and maintenance of multipotent hNSCs. Strikingly, extensive TUJ1+ neurites and advance groupings of these neurites into bundles were observed in SOX9-KD NSPs after three days and seven days neuronal differentiation respectively, suggesting premature neurogenesis as a result of SOX9 ablation. In addition, RT-qPCR analysis revealed down-regulated expression of NSC marker HES1but induced proneural basic helix-loop-helix transcription factor MASH1in shSOX9-1208 NSCs. The inhibitory role of HES1 on the expression and functions of MASH1 has been reported to be essential for the timely generation of neurons. Hence, ablation of SOX9 is likely to relieve the inhibition on MASH1activity via down-regulated HES1expression and leads to early neuronal differentiation. Expression of the potent neurite blocker NG2 was also found to be reduced in SOX9-KD NSCs which may explain the extensive neurite network observed. Altogether, similar to previous studies in mouse NSCs, SOX9 is also required for the induction and maintenance of hNSCs. However, this study further reveals a putative novel role of SOX9 in preventing premature neuronal differentiation by regulating the expressions of HES1 to counteract MASH1 function and NG2 to control neurite outgrowth. / published_or_final_version / Biochemistry / Master / Master of Philosophy

Thrombocytopenia is one of the most common hematologic presentations of active human cytomegalovirus (HCMV) infection, especially in recipients of allogeneic hematopoietic stem cell transplantations and newborns of congenital HCMV infection. However, mechanisms of HCMV-induced thrombocytopenia have not been well understood. The precursor of circulating platelets – megakaryocyte, is derived from hematopoietic stem/progenitor cell in bone marrow. We postulate that inhibition to megakaryocytic development is the major pathogenesis of HCMV-induced thrombocytopenia. Megakaryocytic cells as well as supportive microenvironment in bone marrow are major targets of HCMV infection. Presented study mainly focused on the impacts of HCMV to megakaryocytic cells and multipotent mesenchymal stromal cells (MSCs) - the precursor of bone marrow stromal cells. Based on a megakaryocytic cell model challenged by HCMV in vitro, inhibited megakaryocytic endomitosis, proliferation, and cellular expression were respectively demonstrated as decreased polyploidy population, decreased colony formation, and reduced c-Mpl (thrombopoietin receptor) expressing cells. Evoked apoptosis of megakaryocytic cells was also evidenced with increased phosphatidylserine exposure on cell surface and intracellular caspase-3 activation after HCMV infection. Involvement of mitochondrial-mediated intrinsic apoptosis was further shown as losing JC-1 fluorescent signal in infected megakaryocytic cells. These results suggest that inhibition induced by HCMV is exerted through multiple processes directly affecting the megakaryopoietic development. Functional failure of bone marrow microenvironment was demonstrated in bone marrow derived MSCs infected by HCMV in vitro. Suppressed cytokine production, impaired cellular migration, and hindered differentiation of HCMV-infected MSCs were respectively demonstrated by lowered level of stromal cell-derived factor 1 in culture medium, decreased number of cells passed through a porous membrane in a transwell culture, and reduced differentiated cells in either adipogenic or osteogenic induction cultures. Alongside with these changes, HCMV-induced programmed cell death further contributed to the supportive failure. Autophagic cell death in infected MSCs was demonstrated as massive accumulation of vacuoles with double membrane structure and LC-3b II molecules followed by viability loss. De novo apoptosis was also observed as another process of programmed cell death, shown as increased phosphatidylserine exposure on cell surface and intracellular caspase-3 activation of infected MSCs. Increased programmed cell death appeared to be associated with extensive HCMV replication in MSCs, which was featured with typical cytopathic morphology, expression of viral tegument protein pp65, and massive accumulation of various viral particles including mature virions. Sustained activation of extracellular signal-regulated kinases likely represented a signal transduction network connecting viral expression or replication with programmed cell death. In a “MSCs-dependent” megakaryopoiesis model, HCMV-infected MSCs failed to support survival and maintenance of megakaryocytic cells. Taken together, these results suggest that active HCMV expression or replication inhibits multiple cellular functions and induces multiple processes of programmed cell death of MSCs. Such inhibition compromises supportive functions of bone marrow microenvironment, and subsequently reduces platelet production in an indirect manner. In summary, HCMV suppresses cellular function and induced apoptosis on both megakaryocytic cells and their supportive cells, MSCs. Therefore, the inhibitory effects of HCMV on megakaryopoiesis are operated via both direct and indirect mechanisms. / published_or_final_version / Paediatrics and Adolescent Medicine / Doctoral / Doctor of Philosophy

The potential applications of mesenchymal stem cells (MSCs) have been widely advocated, however, many barriers hinder their clinical utilization. Enhancement of the homing of human MSCs (hMSCs) to the target tissues remains a clinical challenge. To overcome this hurdle, the mechanisms responsible for migration and engraftment of hMSCs have to be defined. My study aimed to explore both the underlying mechanisms and means of enhancing the migration of hMSCs. A graft versus host disease (GvHD) injury model and a novel orthotopic neuroblastoma model were established to delineate the distinct property of hMSCs homing towards either injured or cancerous tissues. This highly specific homing process was further revealed to be in a CXCR4-dependent manner. Notably, a novel gene, exchange protein directly activated by cAMP (Epac), was demonstrated to be actively involved in the hMSCs homing process. hMSCs expressed functional Epac and its activation significantly enhanced the migration and adhesion of hMSCs. Furthermore, Epac activation directly contributed to the chemotactic response of hMSCs to SDF-1, suggesting that Epac is linked to the stromal cell derived factor-1 (SDF-1) signaling cascades. Importantly, the homing of hMSCs towards injured tissues in vivo could be dramatically increased by Epac activation. hMSCs are adherent cells and their migration to distant tissues thus requires detachment into a suspension state. This disruption of cell-extracellular matrix interaction, known as anoikis stress, triggers programmed cell death, leading to a marked decrease in the efficiency of cell trafficking and engraftment. Anoikis stress induced massive cell death has emerged as the major challenge in the application of hMSCs. How some of the hMSCs can overcome this adversity and migrate towards distant destinations remains largely unexplored. It was observed that the surviving hMSCs circumvented anoikis stress by forming self-supporting cellular aggregates. Compared to adherent hMSCs, aggregated-hMSCs had better migratory response to both SDF-1α and SDF-1α analogue (CTCE-0214). Such enhanced migratory effect was proven to be CXCR4-dependent both in vitro and in vivo by using a CXCR4 specific antagonist (AMD3100). Although the viability of hMSCs under anoikis stress dramatically decreased, CTCE-0214 could promote cell survival and facilitate the migration of hMSCs towards injured targets. This phenomenon could be partially explained by the increase in anti-apoptosis effect via up-regulated Bcl-2 expression and autophagy activation under CTCE-0214 treatment. The exact effects of hMSCs on tumor growth and progression have long been controversial. Significant fasten growth and promoted metastasis of neuroblastoma in vivo was observed in hMSCs co-transplanted mice in this study. Reciprocally, hMSCs could not only be recruited by primary tumor, but also be selectively attracted by metastatic loci. This recruitment was significantly reduced when hMSCs were pre-treated with AMD3100, suggesting that the SDF-1/CXCR4 axis was a prime mover in this process. In summary, my study demonstrated that the migratory property of hMSCs could be enhanced by novel intrinsic and extrinsic factors using both in vitro and in vivo models. This study provides a new prospective on MSCs biology during the ex vivo manipulation process and I proposed means to overcome some of these hindrance so we can maximize the efficacy of clinical MSCs application in the future. / published_or_final_version / Paediatrics and Adolescent Medicine / Doctoral / Doctor of Philosophy

Myelin is essential for neuronal survival and maintenance of normal functions of the nervous system. Demyelinating disorders are debilitating and are often associated with failure of resident oligodendrocyte precursor cells (OPCs) to differentiate into mature, myelinating oligodendrocytes. Derivation of OPCs, from a safe source that evades ethical issues offers a solution to remyelination therapy. We therefore hypothesized that bone marrow stromal cells (BMSCs) harbour neural progenitor cells at a pre-commitment stage and that in vitro conditions can be exploited to direct differentiation of these cells along the oligodendroglial lineage. For the current study, adult rat BMSCs used were >90% immunopositive for CD90, CD73, STRO-1 (stromal cell markers), 10% for nestin (neural progenitor marker) but negligible for CD45 (haematopoietic cell marker) as measured by flow cytometry. Transfer of the culture from a highly adhesive substratum to a moderately adhesive substratum resulted in increase in proportion of p75-positive cells but CD49b-positive cells remained at 97% and Sox 10-positive cells remained negligible. Transfer of the culture to a non-adherent substratum fostered the generation of neurospheres comprising cells that were positive for the neural stem/progenitor cell (NP) marker, nestin, and for the neural crest markers CD49b, p75 and Sox10. Prior to this stage, the BMSCs were not yet committed to the neural lineage even though transient upregulation of occasional marker may suggest a bias towards the neural crest cell lineage. The BM-NPs were then maintained in adherent culture supplemented with beta-Heregulin (β-Her), basic fibroblast growth factor (bFGF) and platelet-derived growth factor-AA (PDGF-AA) to direct differentiation along the oligodendroglial lineage. Within two weeks of glial induction, cells expressing the OPC markers - NG2, Olig2, PDGFRa and Sox10, were detectable and these could be expanded in culture for up to 3 months with no observable decline in marker expression. These BM-OPCs matured into myelinating oligodendrocytes after 2 weeks in co-culture with either dorsal root ganglion neurons or cortical neurons. In vivo myelination by BM-OPCs was demonstrated by exploitation of the non-myelinated axons of retinal ganglion cells of adult rats. By 8 weeks post-injection of BM-OPCs into the retina, myelin basic protein-positive processes were also observable along the retinal axons. The results not only suppport our hypothesis, but also provide pointers to the adult bone marrow as a safe and accessible source for the derivation of OPCs towards transplantation therapy in acute demyelinating disorders. / published_or_final_version / Biochemistry / Doctoral / Doctor of Philosophy

Human embryonic stem cells (hESCs) are characterized by unlimited proliferation (self-renewal), capability to differentiate into derivatives of all three germ layers (pluripotency), and abbreviated cell cycle structure. Despite tremendous efforts in identification of important regulators, the complicated molecular mechanisms and essential effectors underlying the distinctive features of hESCs have not yet been fully elucidated. Forkhead box transcription factor M1 (FOXM1) has been demonstrated to be critical for the maintenance of pluripotency in mouse embryonic stem cells (mESCs) and mouse embryonal carcinoma cells (mECCs). The present study hypothesized that FOXM1 is important to the self-renewing capacity and pluripotency of hESCs. The objectives of this study were to characterize FOXM1 expression in undifferentiated and differentiating hESCs, and to study the effect of perturbing FOXM1 expression on pluripotency and proliferation. Undifferentiated VAL3 analyzed by bivariate flow cytometric analysis revealed that FOXM1 expression was regulated in a cell cycle phase-dependent manner, with expression level increased from G1 through S phase, and eventually reached peak levels in G2/M phase. To study the subcellular localization of FOXM1 with respect to cell cycle progression, VAL3 cells were synchronized by nocodazole-mediated cell cycle block, followed by immunocytochemical analysis. The result indicated that FOXM1 underwent nuclear translocation at late-S and early-G2 phase of the cell cycle. When VAL3 spontaneously differentiated as embryoid bodies (EBs), the mRNA expression of FOXM1 displayed profound fluctuation over the differentiation process. Retinoic acid (RA) treatment induced rapid differentiation of VAL3, yet differential expression pattern of FOXM1 was observed for cells grown in different culture media. FOXM1 mRNA expression persisted in differentiating VAL3 cultured in mTeSR. By contrast, RA-driven differentiation of VAL3 cultured in conditioned medium was accompanied by transient depletion and resurgence of FOXM1 protein expression. Differentiation of VAL3 driven by Definitive Endoderm kit did not alter FOXM1 expression, whereas induced differentiation by Bone Morphogenic Protein 4 (BMP4) led to repression of FOXM1. The functional role of FOXM1 in hESCs was investigated with the use of siRNA. Transient knockdown of FOXM1in VAL3 did not induce substantial repression of pluripotent marker (OCT4, SOX2, NANOG) expression nor significant morphological change of colonies, despite upregulation of early differentiation marker SSEA-1. Intriguingly, FOXM1 depletion led to altered cell cycle progression and delay in G2 phase progression, possibly attributed to the downregulation of Cyclin B1 and Cdc25B. Also FOXM1 knockdown impaired VAL3 proliferation, yet no prominent defect in mitosis was observed. In conclusion, the present study reported for the first time the expression and functions of FOXM1 in undifferentiated hESCs. Upon differentiation, FOXM1 expression varied in cells committing to different lineages. Depletion of FOXM1 did not interfere with hESCs pluripotency, but hindered cell cycle progression and cell proliferation, suggesting that FOXM1 is mainly involved in promoting rapid proliferation of hESCs. The functional role and regulatory mechanics of FOXM1 in hESCs cell cycle control and differentiation warrant further investigation. / published_or_final_version / Biochemistry / Master / Master of Philosophy

Mesenchymal stem cell (MSC) transplantation has achieved only modest success in the treatment of ischemic heart disease due to poor cell viability in the diseased microenvironment. Genetic manipulation on the MSCs holds promising prospects in enhancing cell tolerance against adverse environmental conditions. Recent studies demonstrate that the activation of the NRG1 (neuregulin 1) - ERBB4 (v-erb-b2 avian erythroblastic leukemia viral oncogene homolog 4) pathway can enhance pro-survival signaling, stimulate mature cardiomyocyte cell cycle re-entry and cell division. In this study, I aimed to determine whether activating NRG1-ERBB4 in MSCs can enhance their cardioprotective effects following myocardial infarction. In chapter 3, I determined that MSC endogenously expresses NRG1, but not ERBB4. Considering the absence of ERBB4 in the MSCs might lead to mute response to its ligand NRG1, I exogenously manipulated ERBB4 into MSCs. In chapter 4, MSCs, with or without ERBB4 overexpression were transplanted into mice following myocardial infarction. The transplantation of MSCs with ERBB4 expression considerably improved left ventricular ejection fraction and reduced infarctsize, compared to unmodified MSCs and direct NRG1 injection. ERBB4 overexpression induced greater MSC survival following infarction. The transduction of ERBB4 in MSCs increased cell mobility and apoptotic resistance via a PI3K/Akt pathway under hypoxic conditions in the presence of NRG1. The transplantation of MSCs with ERBB4 expression induced cardiomyocyte division and protected them against apoptosis during early phase of infarction. In chapter 5, a novel autocrine loop regarding to NRG1-ERBB4-NRG1 signaling was identified. MSCs with ERBB4 overexpression in turn increased NRG1 synthesis and secretion. Conditioned medium of ERBB4-expressing MSCs containing elevated NRG1, promoted cardiomyocyte growth, division and anti-senescence, whereas neutralization of NRG1 blunted these effects. Injecting ERBB4-expressing MSCs restored NRG1 in the infarcted myocardium to a level comparable with that of the normal myocardium. These findings collectively suggest overexpressing ERBB4 in MSCs enhances the effectiveness of MSCtherapy following myocardial in farction through potentiating MSC survival and revitalizing endogenous repair and regeneration. The combination of ERBB4 and MSC is more efficient than naïve MSC or solely recombinant NRG1 injection, emerging as potential target for developing novel strategy in treating myocardial diseases. / published_or_final_version / Medicine / Doctoral / Doctor of Philosophy

Hirschsprung’s (HSCR) disease is a congenital disorder in which some enteric ganglion cells are absent in the colon due to incomplete colonization of neural crest cells (NCCs) in the hindgut, causing chronic constipation. A significant number of HSCR patients also clinically present with other NC- associated disorders, such as ventricular and atrial septal defects (VSD/ASD). A hypomorphic allele or SNP of a major gene, RET, causes or imparts susceptibility to HSCR. In particular, SNP (rs2435357) residing in the intron 1 of RET gene was found to be highly associated with HSCR and lead to reduced RET expression. However, the molecular basis of syndromic HSCR with VSD/ASD is largely unclear. In our project, with the use of the induced pluripotent stem cell (iPSC) technology, we aim to establish a patient-specific model unravel the etiology of HSCR and the associated disorders. To this end, 3 iPSC clones from a syndromic HSCR patient with VSD/ASD, carrying the RET risk allele in rs2435357 were generated. We attempted to use different protocols to directly differentiate iPSCs into NCCs with unique HOX expression patterns, corresponding to anterior cranial/vagal or posterior vagal/trunk NCCs. Consistently, the patient iPSCs displayed similar capacities in generating NCCs at all axial levels, marked by HNK-1 and 〖p75〗^NTR. Nevertheless, the patient NCCs and their derivatives exhibited severe migration and/ or differentiation defects in making neurons and smooth muscle cells. In particular, HNK-1+〖p75〗^NTR+ HOX+ (vagal/trunk) NCCs derived from patient-iPSCs were less migratory compared to the control NCCs, while no obvious migration defect was observed in their cranial counterpart, indicating that the migration defect was only restricted to the more posterior NCCs. In addition, these patient NCCs were less capable in generating neurons and readily biased toward generating glial cells. Intriguingly, the neural differentiation defects were restricted to NC lineage. The capacity of patient iPSCs to make various types of CNS progenitors and neurons was comparable to that of the control iPSCs, nicely recapitulating the patient’s phenotype where only enteric neurons, but not CNS progenitors were affected. Subsequent expression analysis revealed that patient NCCs express lower level of RET which is known to be regulating enteric NCC migration and differentiation. Whole transcriptome RNA sequencing analysis also revealed an enhanced expression of genes associated with gliogenesis and a reduced expression in genes associated with neurogenesis and migration. Moreover, the expression of a new candidate gene ALDH3B1 was shown to be significantly reduced in the HSCR-iPSC-derived NCCs that might contribute to the disease pathogenesis. In summary, these data suggests that reduced RET expression in HSCR patient NCCs may at least partly account for the disease phenotypes. / published_or_final_version / Surgery / Doctoral / Doctor of Philosophy

The pulp organ is retained in the pulp chamber of teeth and maintains the biological and physiological vitality of the surrounding dentin. It works as a biosensor and generates secondary dentine and tertiary dentine to resist tooth abrasion and pathogenic stimuli (Zhang and Yelick, 2010). However, dental pulp is vulnerable to injury (Smulson and Sieraski, 1989). Most people experience some irreversible pulpal diseases during their lifetime. Hence, pulp regeneration is one of the research tasks in dentistry that attracts much attention. Stem cell transplantation is a plausible strategy for the regeneration of dental pulp organ. Dental pulp stem cells (DPSCs)derived from heavy or inflamed dental pulps have the natural advantage in pulp regeneration due to its dentinogenic potentiality (Huang et al., 2009). DPSCs are delivered into prepared root canal, which then differentiate into odontoblasts, fibroblasts, and other kinds of cells. It was shown that these transplanted DPSCs were able to produce dentin and formed a dentin-pulp like tissue both in vitro and in vivo(Huang, 2009).However, low survival rate of the transplanted cells is a common problem in pulp regeneration. Overexpression of Bcl-2 could enhance cell anti-apoptotic ability. Studies of many kinds of cell transplantation showed that a large number of cells died upon grafting and a large proportion of cell death seemed to have occurred due to apoptosis (Liu et al., 2013; Zhang et al., 2001).The aim of this study was to improve cell survival through making DPSCs overexpress lymphoma 2 (Bcl-2) protein.Bcl-2 is a proto-oncogene which playsa significant role in (anti) apoptosis. Former studies in the literature have provided evidences that overexpressing Bcl-2 could reduce cell apoptosis. However, this strategy has not been studied in the modification of DPSCs. In this study, DPSCs were isolated from discarded third molars of adults and manipulated to overexpressing Bcl-2. Proliferation of modified DPSCs was analyzed by static batch culture, CCK-8 test and BrdU based proliferation test. Apoptosis of modified DPSCs was analyzed by measuring DNA fragments in the cells. Modified DPSCs generated a higher maximum cell population during static batch culture and showed higher viability (the ratio of live cells to total cells). CCK-8 test showed that the population of modified DPSCs increased faster than control group cells and wild type cells. Modified DPSCs were not better than the other cells in proliferative ability, but had lower apoptosis level when culturing in serum free medium. Hence, overexpressing Bcl-2 could increase cell population, the mechanism is to help DPSCs survival rather than promote the proliferative ability of cells. / published_or_final_version / Dentistry / Master / Master of Philosophy

No description available.