Stem cells play a vital role in the turn over of permanently self-renewing tissues (e.g. corneal epithelium, epidermis, bone marrow). Stratified cornea epithelia serve, as ideal tissue models for studying ???stemness???, as the site of cells in the different stages of differentiation are anatomically easily identifiable. Studies targeting limbal stem cell maintenance, in-vitro gene expression during storage and differentiation will benefit future therapeutic applications (e.g. corneal transplantation). Knowledge of stem cell behaviour will help explain the pathophysiology of corneal related ocular surface disorders (e.g. idiopathic limbal stem cell deficiency, pterygium and limbal dermoids), establish new treatment modalities (e.g. allogenic cellular transplants) and help promote invivo expansion of residual stem cell populations in deficiency states. The major obstacle in the progress of research on limbal stem cells is the lack of knowledge of phenotypic markers of limbal epithelial stem cells. This project first studied the limbal expression of phenotypic markers that were discovered in other human adult and embryonic stem cell populations using microarray differential expression studies. Gene expression profiles of the relatively primitive human foetal limbus were compared with that of the central cornea. Microarray was also performed to identify the differential gene expression profile of cultured primary human limbal epithelial cells, when compared to the limbal epithelial cell population isolated after 5 serial cultures. 33 genes were upregulated in the human foetal limbus and primary cultured human limbal epithelium, when compared respectively to the central foetal cornea (first experiment) and the limbal epithelial cell population after the 5th trypsin passaged culture (second experiment). Four foetal limbal and primary limbal epithelial upregulated genes (CK15, CK14, Cdh3, and Wnt4) were confirmed to be upregulated by semi quantitative RT-PCR and immunohistochemical experiments on the human foetal cornea, adult human cadaveric cornea and cultured human limbal epithelial cells. The microarray defined phenotypic profile of both the foetal limbus and cultured adult limbal epithelial cells will help identify these cells in in-vitro and in-vivo states. The expression of these 4 selected markers in the limbal dermoid and pterygium suggests that limbal epithelial cells containing a stem cell population are involved in the pathogenesis of these two disorders.

Stem cells play a vital role in the turn over of permanently self-renewing tissues (e.g. corneal epithelium, epidermis, bone marrow). Stratified cornea epithelia serve, as ideal tissue models for studying ???stemness???, as the site of cells in the different stages of differentiation are anatomically easily identifiable. Studies targeting limbal stem cell maintenance, in-vitro gene expression during storage and differentiation will benefit future therapeutic applications (e.g. corneal transplantation). Knowledge of stem cell behaviour will help explain the pathophysiology of corneal related ocular surface disorders (e.g. idiopathic limbal stem cell deficiency, pterygium and limbal dermoids), establish new treatment modalities (e.g. allogenic cellular transplants) and help promote invivo expansion of residual stem cell populations in deficiency states. The major obstacle in the progress of research on limbal stem cells is the lack of knowledge of phenotypic markers of limbal epithelial stem cells. This project first studied the limbal expression of phenotypic markers that were discovered in other human adult and embryonic stem cell populations using microarray differential expression studies. Gene expression profiles of the relatively primitive human foetal limbus were compared with that of the central cornea. Microarray was also performed to identify the differential gene expression profile of cultured primary human limbal epithelial cells, when compared to the limbal epithelial cell population isolated after 5 serial cultures. 33 genes were upregulated in the human foetal limbus and primary cultured human limbal epithelium, when compared respectively to the central foetal cornea (first experiment) and the limbal epithelial cell population after the 5th trypsin passaged culture (second experiment). Four foetal limbal and primary limbal epithelial upregulated genes (CK15, CK14, Cdh3, and Wnt4) were confirmed to be upregulated by semi quantitative RT-PCR and immunohistochemical experiments on the human foetal cornea, adult human cadaveric cornea and cultured human limbal epithelial cells. The microarray defined phenotypic profile of both the foetal limbus and cultured adult limbal epithelial cells will help identify these cells in in-vitro and in-vivo states. The expression of these 4 selected markers in the limbal dermoid and pterygium suggests that limbal epithelial cells containing a stem cell population are involved in the pathogenesis of these two disorders.

Wong, Wai Chung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2008. / Includes bibliographical references (leaves 113-130). / Abstracts in English and Chinese. / Thesis/Assessment Committee --- p.i / Abstract --- p.ii / 中文摘要 --- p.iv / Acknowledgements --- p.vi / List of Figures --- p.vii / List of Tables --- p.ix / Table of Abbreviations --- p.x / Contents --- p.xv / Chapter Chapter I --- Introduction / Chapter 1.1 --- Stem cells --- p.1 / Chapter 1.1.1 --- Embryonic stem cells --- p.2 / Chapter 1.1.2 --- Adult stem cells --- p.2 / Chapter 1.1.3 --- Artificial stem cells --- p.5 / Chapter 1.2 --- Hair bulge stem cells (HBSC) --- p.7 / Chapter 1.2.1 --- Hair follicle --- p.7 / Chapter 1.2.2 --- The discovery of hair bulge stem cells --- p.9 / Chapter 1.2.3 --- The hair bulge niche --- p.10 / Chapter 1.2.4 --- Molecular markers --- p.12 / Chapter 1.2.5 --- Physiological roles of bulge stem cells --- p.13 / Chapter 1.2.6 --- Pathological roles of bulge stem cells --- p.15 / Chapter 1.2.7 --- Developmental plasticity of bulge stem cells --- p.16 / Chapter 1.3 --- "Regulation of adipogenic, osteogenic and cardiogenic differentiation" --- p.17 / Chapter 1.3.1 --- Regulation of adipogenic differentiation --- p.17 / Chapter 1.3.2 --- Regulation of osteogenic differentiation --- p.18 / Chapter 1.3.3 --- Regulation of cardiogenic differentiation --- p.19 / Chapter 1.4 --- Proteomics --- p.23 / Chapter 1.4.1 --- Definition of proteomics --- p.23 / Chapter 1.4.2 --- Two-dimensional gel electrophoresis (2DGE) --- p.25 / Chapter 1.4.3 --- Mass spectrometry and protein identification --- p.29 / Chapter 1.4.4 --- Other techniques associated with proteomics --- p.34 / Chapter 1.4.5 --- Proteomics and stem cells --- p.36 / Chapter 1.5 --- General summary --- p.37 / Chapter 1.6 --- Aims of my study --- p.38 / Chapter Chapter II --- Materials and Methods / Chapter 2.1 --- Animals --- p.39 / Chapter 2.2 --- Immunohistochemistry --- p.39 / Chapter 2.2.1 --- Histology --- p.39 / Chapter 2.2.2 --- Immunohistochemistry --- p.40 / Chapter 2.3 --- Establishment of hair bulge CD34+ stem cell line --- p.41 / Chapter 2.3.1 --- Isolation of hair bulge explants --- p.41 / Chapter 2.3.2 --- Establishing hair bulge primary cultured cells --- p.42 / Chapter 2.3.3 --- Purification of hair bulge stem cells --- p.43 / Chapter 2.4 --- Karyotyping --- p.45 / Chapter 2.5 --- In vitro differentiation --- p.46 / Chapter 2.5.1 --- Adipogenic differentiation --- p.46 / Chapter 2.5.2 --- Osteogenic differentiation --- p.47 / Chapter 2.5.3 --- Cardiogenic differentiation --- p.48 / Chapter 2.6 --- Proteomic analysis --- p.48 / Chapter 2.6.1 --- Sample preparation --- p.48 / Chapter 2.6.2 --- Quantification of proteins --- p.49 / Chapter 2.6.3 --- First-dimensional separation of proteins 226}0ؤ isoelectric focusing (IEF) --- p.50 / Chapter 2.6.4 --- Second-dimensional separation - sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) --- p.51 / Chapter 2.6.5 --- "Silver staining, imaging and destaining" --- p.52 / Chapter 2.6.6 --- In-gel digestion and matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry analysis --- p.53 / Chapter 2.7 --- Histochemistry --- p.54 / Chapter 2.7.1 --- Oil Red O staining --- p.54 / Chapter 2.7.2 --- Alizarin Red S staining --- p.55 / Chapter 2.8 --- Immunocytochemistry --- p.56 / Chapter 2.9 --- Semi-quantitative reverse transcription-polymerase chain reaction (RT-PCR) --- p.57 / Chapter 2.9.1 --- Isolation of total cellular RNA --- p.57 / Chapter 2.9.2 --- Complementary DNA (cDNA) synthesis --- p.58 / Chapter 2.9.3 --- Polymerase chain reaction and agarose gel electrophoresis --- p.59 / Chapter 2.10 --- Western blot analysis --- p.62 / Chapter 2.10.1 --- Sample preparation and quantification of proteins --- p.62 / Chapter 2.10.2 --- SDS-PAGE --- p.63 / Chapter 2.10.3 --- Protein transfer --- p.64 / Chapter 2.10.4 --- Immunodetection --- p.65 / Chapter 2.11 --- Cell proliferation assay --- p.66 / Chapter 2.11.1 --- Determination of growth pattern --- p.66 / Chapter 2.11.2 --- MTT assay --- p.66 / Chapter 2.12 --- Ultrastructural analysis --- p.67 / Chapter 2.12.1 --- Scanning electron microscopy (SEM) --- p.67 / Chapter 2.12.2 --- Transmission electron microscopy (TEM) --- p.68 / Chapter 2.13 --- Statistical analysis --- p.68 / Chapter Chapter III --- Results / Chapter 3.1 --- Isolation and characterization hair bulge stem cells --- p.69 / Chapter 3.2 --- Directed adipogenic differentiation --- p.70 / Chapter 3.3 --- Directed osteogenic differentiation --- p.71 / Chapter 3.4 --- Ability of cardiogenol C to induce cardiogenesis in HBSCs --- p.71 / Chapter 3.5 --- Comparative proteomic analysis of HBSC cardiogenic differentiation induced by cardiogenol C --- p.72 / Chapter 3.6 --- Role of Wnt signaling pathway in cardiogenol C-induced cardiogenesis --- p.74 / Chapter 3.7 --- Role of chromatin remodeling in cardiogenol C-induced cardiogenesis --- p.75 / Chapter 3.8 --- Legends and tables --- p.76 / Chapter Chapter IV --- Discussion --- p.98 / References --- p.113 / Appendices --- p.131 / Publication --- p.134

Thesis (DPhil (Philosophy))--University of Stellenbosch, 2007. / Should surplus embryos which are destined to be discarded be protected at all cost, to the extent that they cannot contribute to medical knowledge - knowledge which could benefit society at large? Are embryos people or merely items of property? Different moral theories address these questions in different ways. Deontologists argue that the end never justifies the means and that the right not to be killed is more fundamental than the obligation to save. Utilitarians, on the other hand, argue that certain criteria should be met before moral significance can be contributed to an entity. The question of the moral status of the embryo is, as my discussion will show, one of the most widely discussed issues in the history of bioethics. Extensive literature exists on the topic. This study holds that an Ethics of Responsibility (ER) should by applied when answering the questions posed above as it encourages one to accept responsibility for the choices or decisions made and to defend them accordingly. I have endeavoured to answer the question of the personhood and rights of the embryo within the framework of the Ethics of Responsibility. Although these concepts overlap in many ways they remain central to the debate surrounding the sanctioning or prevention of the use of human embryonic stem cells in research. After identifying the micro-issues surrounding the human embryonic stem cell debate and explaining why both the deontologist and utilitarians fail to provide any adequate answers in this respect, I turn my attention to macro-issues such as safety concerns surrounding the usages and storage of stem cells. Commercialization, power issues, accessibility and the allocation of limited resources are also examined. Living in a society such as South Africa one cannot be blind to the inequalities of our health system. On a macro level I cannot but conclude that stem cell research does not seem to be a viable exercise within the South African context. South Africa faces a health care crisis far greater than the benefits stem cell research currently has to offer. However, the need still exists for a policy to guide future lawmakers who might need to address stem cell research and to guide decisions and actions. This brings me to my final chapter, namely proposing a morally justified policy for South Africa. I propose a policy which respects and values the autonomy of the progenitors’ choices (provided they have not been coerced) and which focuses on the beneficence of the greater society. Furthermore, it is paramount that the goal of any stem cell research should be for therapeutic use ONLY. Before commencing with the extraction of the stem cells, scientists should be obligated first to present convincing evidence that they have tried alternative ways to reach the same result. Once this has been proven, a regulatory body could issue the scientist/team with a license to undertake the specific research with a specific therapy as goal in order to prevent abuse. If they are found guilty of any unethical conduct their licenses should be revoked and an investigation launched.

published_or_final_version / Biochemistry / Doctoral / Doctor of Philosophy

Cellulose binding domains (CBDs) are found in cellulolytic microorganisms as discrete domains in free cellulases or as cellulosomes, which are extracellular multi-enzyme complexes. CBDs bind to cellulose and help the catalytic domains to access cellulose substrates. CBDs are used as affinity tags for immobilizing cells, proteins or molecules on cellulose matrices. They can also be used in protein engineering to alter protein expression and solubilities. Cohesins and dockerins are domains exclusive to cellulosomes. They interact with high affinity and the interactions are Ca2+-dependent. Chelation of Ca2+ causes irreversible conformational change to dockerins thus disrupting the interactions. The first aim of this project was to validate a putative CBD from endoglucanase EngD of C. cellulovorans, to test its effect on solubility as a fusion in chimeras. The second aim was to use chimeric proteins containing CBD, cohesin, dockerin and LG to establish a system for efficient cell immobilization, expansion and harvest in hollow cellulose fibres. A putative CBD from an enzyme with its natural linker (PTCBDengD), a CBD from a scaffoldin (CBDcbpA), three cohesin domains from different strains (Cip7, Coh6 and CipC1) and an antibody binding protein (LG) were used to construct various chimeric and fusion proteins. The two CBDs were fused to different cohesins and LG respectively and the chimeras? solubility was analyzed. The results showed that fusing with CBDcbpA did not significantly help to increase the solubility of the insoluble domain Coh6 and it even greatly reduced the solubility of the soluble domain CipC1. In contrast, PTCBDengD fusion increased the solubility of Coh6 by three fold and it did not alter the solubility of soluble protein/ domains. These results suggested that PTCBDengD may be a better domain to use as a fusion tag for expression and other biotechnology applications. Cellulose binding specificity of PTCBDengD and its chimeric proteins were tested and SDS-PAGE analysis results clearly demonstrated that PTCBDengD and its chimeras specifically bound to crystalline cellulose Avicel and non-crystalline cellulose Cuprophan. The results confirmed that PTCBDengD is a true CBD. Chimeric protein pairs CBDcbpA-Cip7/ LG-Doc and Cip7-PTCBDengD/ LG-Doc were used to build the scaffold on Cuprophan hollow cellulose fibre for reversible cell immobilization studies. Cell adhesion assay results showed that the double-chimera systems efficiently immobilize cells onto Cuprophan hollow fibre. Dissociation of LG-Doc from amorphous cellulose Cuprophan-bound CBDcbpA-Cip7 by EDTA treatment resulted in decrease of cell binding by almost 90%; however, re-association of LG-Doc after EDTA dissociation only achieved 50% efficiency of cell immobilization. Dot blot and SDS-PAGE analysis showed that dissociation/ re-association of LG-Doc to Cip7-PTCBDengD could be decreased in was interfered by the presence of cellulose. Preliminary results indicated that crystalline cellulose Avicel may improve dissociation/ re-association efficiency. In conclusion, studies on recombinant proteins validated CBDengD's specific affinity to cellulose and its solubilizing effect on its fusion partner. Chimera pairs CBDcbpA-Cip7/ LG-Doc and Cip7-PTCBDengD/ LG-Doc are effective in cell immobilization. However, optimization is required to develop recyclable protein scaffolding and complexes on cellulosic matrices.

Transcription factors GATA-1, Fli-1 and FOG-1 are essential proteins for normal megakaryopoiesis, however, the detailed analyses of their functions within developmental stages of megakaryopoiesis are lacking. In my thesis, over expression of gene in target cells was adopted as the main strategy to study the biological functions of these proteins, therefore, an efficient gene delivery method was first developed by using retrovirus.This approach was then utilized to over express GATA-1, Fli-1 and FOG-1 in murine leukemia M1 cells and mouse hematopoietic stem cells (HSCs), and their effects on different developmental stages of megakaryopoiesis were investigated. In the transduced M1 cells, enforced expression of GATA-1 and Fli-1 was found to induce the megakaryocytic development, which was associated with the formation of megakaryocyte (Mk) and the increased expression of Mk specific genes c-Mpl and GPIX. In the transduced mouse HSCs, it was found that the expression of endogenous GATA-1, Fli-1 and FOG-1 was up-regulated throughout Mk differentiation; enforced expression of these transcription factors led to the significantly enhanced Mk development. Megakaryocytes over expressing GATA-1, Fli-1 and FOG-1 were characterized by the increased expression of various Mk-specific genes including GPIX, c-Mpl, platelet factor 4 (PF4), acetylcholinesterase (AChE) and NF-E2, an important transcription factor for terminal megakaryopoiesis; however, GATA-1, Fli-1 and FOG-1 displayed the different abilities in promoting the proliferation of hematopoietic cells and MK differentiation, as well as regulating other transcription factors involved in hematopoiesis. To further elucidate the role of the functional domains of Fli-1, various mutants of Fli-1 were also over expressed in mouse HSCs. The results demonstrated that first, the combination of the activation domain of Fli-1 and its Ets domain is required for early megakaryopoiesis but not sufficient for terminal megakaryopoiesis; second, DNA binding of Fli-1 was not the only requirement for early Mk enhancement, moreover, the interaction between Fli-1 and GATA-1 through the Ets domain and the resultant transcriptional synergy was the essential determinant for Fli?1 ability in Mk development. Taken together, the studies presented in this thesis provided strong in vitro evidence that GATA-1, Fli-1 and FOG-1 indeed play the critical roles in normal megakaryopoiesis.

Transcription factors GATA-1, Fli-1 and FOG-1 are essential proteins for normal megakaryopoiesis, however, the detailed analyses of their functions within developmental stages of megakaryopoiesis are lacking. In my thesis, over expression of gene in target cells was adopted as the main strategy to study the biological functions of these proteins, therefore, an efficient gene delivery method was first developed by using retrovirus.This approach was then utilized to over express GATA-1, Fli-1 and FOG-1 in murine leukemia M1 cells and mouse hematopoietic stem cells (HSCs), and their effects on different developmental stages of megakaryopoiesis were investigated. In the transduced M1 cells, enforced expression of GATA-1 and Fli-1 was found to induce the megakaryocytic development, which was associated with the formation of megakaryocyte (Mk) and the increased expression of Mk specific genes c-Mpl and GPIX. In the transduced mouse HSCs, it was found that the expression of endogenous GATA-1, Fli-1 and FOG-1 was up-regulated throughout Mk differentiation; enforced expression of these transcription factors led to the significantly enhanced Mk development. Megakaryocytes over expressing GATA-1, Fli-1 and FOG-1 were characterized by the increased expression of various Mk-specific genes including GPIX, c-Mpl, platelet factor 4 (PF4), acetylcholinesterase (AChE) and NF-E2, an important transcription factor for terminal megakaryopoiesis; however, GATA-1, Fli-1 and FOG-1 displayed the different abilities in promoting the proliferation of hematopoietic cells and MK differentiation, as well as regulating other transcription factors involved in hematopoiesis. To further elucidate the role of the functional domains of Fli-1, various mutants of Fli-1 were also over expressed in mouse HSCs. The results demonstrated that first, the combination of the activation domain of Fli-1 and its Ets domain is required for early megakaryopoiesis but not sufficient for terminal megakaryopoiesis; second, DNA binding of Fli-1 was not the only requirement for early Mk enhancement, moreover, the interaction between Fli-1 and GATA-1 through the Ets domain and the resultant transcriptional synergy was the essential determinant for Fli?1 ability in Mk development. Taken together, the studies presented in this thesis provided strong in vitro evidence that GATA-1, Fli-1 and FOG-1 indeed play the critical roles in normal megakaryopoiesis.

Cellulose binding domains (CBDs) are found in cellulolytic microorganisms as discrete domains in free cellulases or as cellulosomes, which are extracellular multi-enzyme complexes. CBDs bind to cellulose and help the catalytic domains to access cellulose substrates. CBDs are used as affinity tags for immobilizing cells, proteins or molecules on cellulose matrices. They can also be used in protein engineering to alter protein expression and solubilities. Cohesins and dockerins are domains exclusive to cellulosomes. They interact with high affinity and the interactions are Ca2+-dependent. Chelation of Ca2+ causes irreversible conformational change to dockerins thus disrupting the interactions. The first aim of this project was to validate a putative CBD from endoglucanase EngD of C. cellulovorans, to test its effect on solubility as a fusion in chimeras. The second aim was to use chimeric proteins containing CBD, cohesin, dockerin and LG to establish a system for efficient cell immobilization, expansion and harvest in hollow cellulose fibres. A putative CBD from an enzyme with its natural linker (PTCBDengD), a CBD from a scaffoldin (CBDcbpA), three cohesin domains from different strains (Cip7, Coh6 and CipC1) and an antibody binding protein (LG) were used to construct various chimeric and fusion proteins. The two CBDs were fused to different cohesins and LG respectively and the chimeras? solubility was analyzed. The results showed that fusing with CBDcbpA did not significantly help to increase the solubility of the insoluble domain Coh6 and it even greatly reduced the solubility of the soluble domain CipC1. In contrast, PTCBDengD fusion increased the solubility of Coh6 by three fold and it did not alter the solubility of soluble protein/ domains. These results suggested that PTCBDengD may be a better domain to use as a fusion tag for expression and other biotechnology applications. Cellulose binding specificity of PTCBDengD and its chimeric proteins were tested and SDS-PAGE analysis results clearly demonstrated that PTCBDengD and its chimeras specifically bound to crystalline cellulose Avicel and non-crystalline cellulose Cuprophan. The results confirmed that PTCBDengD is a true CBD. Chimeric protein pairs CBDcbpA-Cip7/ LG-Doc and Cip7-PTCBDengD/ LG-Doc were used to build the scaffold on Cuprophan hollow cellulose fibre for reversible cell immobilization studies. Cell adhesion assay results showed that the double-chimera systems efficiently immobilize cells onto Cuprophan hollow fibre. Dissociation of LG-Doc from amorphous cellulose Cuprophan-bound CBDcbpA-Cip7 by EDTA treatment resulted in decrease of cell binding by almost 90%; however, re-association of LG-Doc after EDTA dissociation only achieved 50% efficiency of cell immobilization. Dot blot and SDS-PAGE analysis showed that dissociation/ re-association of LG-Doc to Cip7-PTCBDengD could be decreased in was interfered by the presence of cellulose. Preliminary results indicated that crystalline cellulose Avicel may improve dissociation/ re-association efficiency. In conclusion, studies on recombinant proteins validated CBDengD's specific affinity to cellulose and its solubilizing effect on its fusion partner. Chimera pairs CBDcbpA-Cip7/ LG-Doc and Cip7-PTCBDengD/ LG-Doc are effective in cell immobilization. However, optimization is required to develop recyclable protein scaffolding and complexes on cellulosic matrices.

Breast cancer remains the leading cause of cancer-related death in women worldwide. Furthermore, it has been demonstrated that the treatment-resistant ER-PR-HER2/neu- sub-type is more common among women of African descent, necessitating the search for novel chemotherapies for this form of the disease. The secondary metabolites produced by marine algae represent a rich source of structurally unique compounds with chemotherapeutic potential, particularly in South Africa, whose oceans allegedly host 15 % of the total number of species in the world. Indeed, a recent study reported the isolation of a range of novel compounds from South African red and brown algae of the Plocamium, Portiera and Sargassum genera which displayed cytotoxicity against oesophageal cancer cells in vitro. The molecular mechanisms mediating this toxicity were unknown, as was the effect of these and similar compounds on metastatic ER-PR-HER2/neu- breast cancer cell lines or breast cancer stem cells. The current study aimed to address these questions by screening a library of twenty-two novel marine algal compounds for the ability to inhibit MDA-MB-231 and Hs578T breast cancer cells, while having no adverse effects on non-cancerous MCF12A breast epithelial cells. While twelve of these were toxic in the micromolar range against breast cancer cells, only the polyhalogenated monoterpenes RU004 and RU007, and the tetraprenylated quinone sargaquinoic acid (SQA) were identified as hit compounds based on the criteria that their cytotoxicity was specific to breast cancer and not healthy breast cells in vitro. On the other hand, the halogenated monoterpene RU015 was found to be highly toxic to both breast cancer and non-cancerous breast cell lines, while the halogenated monoterpene stereoisomers RU017 and RU018 were non-toxic to either of these cell lines. The mode of action of RU004, RU007, RU015 and SQA, together with the previously characterized carotenoid fucoxanthin (FXN), was assessed in terms of the type of cell death induced and the effect on cell cycle distribution of these compounds. Flow cytometric analysis of the extent of Hoescht 33342 and propidium iodide staining along with PARP cleavage studies suggested that SQA induced apoptosis in MDA-MB-231 cells. On the other hand, the highly toxic compound RU015 appeared to induce necrosis as evidenced by 50 kDa PARP cleavage product in MDA-MB-231 cells. The flow cytometry profiles of MDA-MB-231 and Hst578T cells treated with the hit compounds RU004 and RU007 were suggestive of the induction of apoptosis by these compounds. Cell cycle analysis by flow cytometry with propidium iodide staining revealed that both SQA and FXN induced G0-G1 arrest together with an increase in the apoptotic sub-G0 population, which agreed with previous reports in the literature. The molecular mechanism of action of SQA and FXN were further investigated by the identification of specific signal transducer molecules involved in mediating their anti-cancer activities. SQA was found to require the activity of numerous caspases, including caspase-3, -6, -8, -9, -10 and -13, for its cytotoxicity and was demonstrated to decrease the level of the antiapoptotic protein Bcl-2. On the other hand, FXN was shown to require caspase-1, -2, -3, -9 and - 10 for its toxicity. This, together with the ability to decrease the levels of Bcl-2, pointed to the involvement of the intrinsic pathway in particular in mediating the activity of FXN. The screening of algal compounds against non-cancerous breast epithelial cells carried out in this study, together with the investigation into their mechanisms of action, represent one of the few reports in which characterization of algal metabolites goes beyond the initial cytotoxicity assays. Finally, in order to assess the potential anti-cancer stem cell activity of the marine algal compounds, a subset of these was screened using a mammosphere assay technique developed in this study. The cancer stem cell (CSC) theory proposes that cancers arise from and are maintained by a specific subpopulation of cells able to undergo asymmetric cell division and termed CSCs. These CSCs are capable of anchorage-independent growth in serum-free culture conditions, such as those in the mammosphere assay. Using this assay, the novel halogenated monoterpene stereoisomers RU017 and RU018 were demonstrated to possess putative anti- CSC activity as evidenced by their ability to completely eliminate mammosphere formation in vitro. Furthermore, since RU017 and RU018 were non-toxic to both breast cancer and healthy breast cells, it appeared that the activity of the compounds was potentially specific to the CSCs. The results require further validation, but represent the first report of selective anti-CSC activity.