A predominant species of heparan sulfate proteoglycan that consisted of at least two subunits linked by disulfide bonding was isolated from cell layers of normal ("cobblestone") bovine vascular endothelial cells in culture. Treatment of the parent molecules with dithiothreitol caused their complete cleavage and permitted the subsequent separation of the larger and smaller subunits on Sepharose CL4B columns. Removal of dithiothreitol by dialysis resulted in the reformation of large disulfide-bonded molecules but such recombination of the subunits was prevented by prior reductive alkylation using iodoacetamide. Buoyant density gradient analysis as well as gel chromatography on Sepharose CL6B columns, following alkaline borohydride and nitrous ac i d treatment of individual carbohydrate-rich subunits, showed that the latter consisted of core proteins associated solely with heparan sulfate glycosaminoglycans. The sizes of the latter were estimated by chromatographic techniques to be approximately 50 000 and 14 000 daltons in the case of the larger and smaller subunits, respectively. This is the first description of disulfide-bonded proteoheparan sulfates in bovine aortic endothelial cells. Studies of the effects of various extracellular matrices on the proliferative behaviour of bovine aortic endothelial cells in culture revealed that extracellular matrix material from rat smooth muscle cells stimulated proliferation more than did other matrices. Bovine aortic endothelial cells also changed their morphology and cell-surface proteoglycan profiles in response to particular extracellular matrices. Enzymic modifications of matrices did not, however, cause noticeable changes in the cell surface proteoglycans synthesized by bovine aortic endothelial cells. This discrepancy suggested that the observed differences in cell-surface proteoglycan profiles cannot be ascribed to any specific single constituent of the extracellular matrix but that its overall architecture may be the sole determinant of such differences. When the turnover of endothelial cell proteoglycans was assessed, degradation of both intracellular and pericellular proteoglycans was inhibited by lysosomotropic agents. This indicated that these macromolecules may be degraded within the lysosomes; the cell layer proteoglycans are apparently internalized prior to their degradation in this location. Failure by both NH₄Cl and chloroquine completely to block the degradation of intracellular as well as pericellular proteoglycans suggested that other mechanisms of degradation also exist. The results extend biochemical data on endothelial cell surface proteoglycans.

Type I collagen is the predominant collagen within the stroma and plays an important role in the processes of tumour cell invasion and metastasis during which the collagens within the stroma is degraded. Total RNA was extracted from different stages of breast cancer and adjacent normal tissue for analysis of collagen gene expression by Northern blot hybridisation. Stage I breast tumours had increased α1(I) and α2(I) collagen mRNA, whereas stages II and III tumours had decreased mRNA levels when compared to the adjacent normal tissue. This stage-specific change in collagen gene expression was confirmed by non-radioactive in situ hybridisation and the results indicated that α1(I) and α2(I) collagen mRNA was produced by the stromal fibroblasts and not the tumour cells. To determine whether this altered collagen gene expression was manifested in other cancers, α1(I) and α2(I) collagen mRNA levels were analysed in colorectal carcinoma samples by in situ hybridisation. Colon cancer as in the case of breast cancer, also showed stage specific changes in collagen gene expression. Dukes C and D colon cancer samples had decreased collagen mRNA levels compared to Dukes A and B. Mutated Ras has been shown to affect collagen mRNA levels in vitro (Slack et al, 1992), therefore the colon samples were analysed for Ras mutations in an attempt to correlate Ras mutations with the decreased levels of α1(I) and α2(I) collagen mRNA. Colorectal DNA samples were screened for Ras mutations by SSCP and direct sequence analysis. No possible association was found between the presence of Ras mutations and the decreased collagen gene expression. To gain greater insight into exactly how tumour cells modulate the collagen produced by normal fibroblasts, primary breast fibroblasts (prepared from breast tissue) were cocultured with various breast tumour cell lines. The fibroblasts were also incubated with conditioned media prepared from the tumour cells. Collagen production was analysed using the collagenase assay and the results showed that co-cultured tumour cells, as well as growth in the presence of tumour cell conditioned media, resulted in decreased type I collagen production by the fibroblasts. Type III collagen is often produced in conjunction with type I collagen and we have found that the breast tumour cells modulated type III collagen in the same way as type I collagen. These results demonstrated that a factor(s) was secreted by the tumour cells which affected collagen production. This factor was further shown to stimulate the fibroblasts to produce type I collagenase as analysis of the medium from co-cultured fibroblasts and tumour cells indicated the presence of collagenases. The tumour cell conditioned media was subsequently shown by Western blot analysis to contain a protein of similar molecular weight to the tumour cell derived collagenase stimulatory factor (known as EMMPRIN or extracellular matrix metalloproteinase inducer) which stimulates fibroblasts to secrete collagenases and has been shown to play a crucial role in tumour invasion (Biswas 1982, 1984 and Biswas et al, 1995). In order to determine whether fibroblasts of different origins reacted similarity when cocultured with breast tumour cell lines, WI-38 lung fibroblasts and FGo skin fibroblasts were co-cultured with breast tumour cells. WI-38 fibroblasts responded in the same way as breast fibroblasts (having decreased collagen production), FGo fibroblasts had no effect or slightly elevated collagen production, depending on the tumour cell line. These results suggested that the response to tumour cells is tissue specific. The decrease in type I collagen produced by the fibroblasts when incubated with the tumour cell conditioned media was not due to a decrease in α1(I) and α2(I) collagen mRNA as shown by Northern hybridisation. Type III collagen mRNA was affected differently, the levels were either decreased or increased depending on the tumour cell line being used. We postulate that the fibroblasts and tumour cells required contact for type I collagen mRNA to be decreased. Northern hybridisation showed that types I and III collagen mRNA levels were decreased when tumour cells were co-cultured with the fibroblasts. To demonstrate that specific contact was in fact required, the tumour cells were separated from the fibroblasts by a diffusible membrane and the levels of collagen mRNA were not adversely affected. Tumour cells, therefore can modulate collagen production by normal fibroblasts in two ways I); cause the fibroblasts to secrete collagenases which will degrade the collagen and 2); decrease collagen mRNA. Both of these mechanisms would aid the tumour in invasion and metastasis.

Smooth muscle cells isolated from neonatal rat hearts synthesize and secrete radioactively labelled proteoglycans into two distinct extracellular compartments, the pericellular (cell surface/matrix layer) and the culture medium (extracellular). Cultures grown in the presence of ascorbic acid synthesize proteoglycans that are more highly sulphated than those produced in the absence of ascorbate. The glycosaminoglycan chains associated with the proteoglycans synthesized by rat smooth muscle cells were heparan sulphate, chondroitin sulphate and dermatan sulphate. There was no evidence for the synthesis of hyaluronic acid by these cells. Most of the heparan sulphate was found to be associated with the pericellular and intracellular compartments, whereas the extracellular compartment contained the bulk of the chondroitin sulphate. In the presence of ascorbate there was an increase in dermatan sulphate content of the pericellular compartment at the expense of heparan sulphate, whilst in the absence of ascorbate the heparan sulphate content of this compartment was significantly increased. Hyaluronic acid and the antibiotic Tunicamycin had no effect on the biosynthesis of sulphated macromolecules produced by the rat smooth muscle cells. However, p-nitrophenyl-β-D-xyloside increased by 10-fold the amount of radioactive sulphate incorporation into macromolecules in the extracellular compartment. This increase was due to increased sulphation of glycosaminoglycan chains synthesized in the presence of the exogenous acceptor, as evidenced by the sulphate/ uronate ratio of these sulphated macromolecules. Furthermore, heparan sulphate secretion into the extracellular compartment was decreased whilst dermatan sulphate increased in the presence of xyloside. Pulse-chase experiments with radioactive sulphate were used to study the pathways and kinetics of secretion in the rat smooth muscle cell system. The data from these studies are consistent with a very rapid intracellular sulphation mechanism followed by rapid secretion to the pericellular compartment of macromolecular sulphated proteoglycans. Subsequently some of these molecules then travel to the extracellular compartment. The time that different proteoglycan species remain associated with the pericellular compartment is influenced by the different matrix connective tissue proteins found in this compartment as a result of ascorbate supplementation or deprivation. During the course of these investigations, it was observed that the pericellular compartment contributed to catabolism of sulphated macromolecules. The sulphated proteoglycans associated with this compartment are acted upon by a sulphatase or sulphatases to give rise to free radioactive inorganic sulphate and macromolecules which have been desulphated. That this process occurs in the pericellular compartment only was proven by the use of intracellular lysomotrophic inhibitors and by the continuous exposure of sulphate labelled macromolecules to the extracellular extract. Neither resulted in the release of radiolabelled inorganic sulphate from sulphated macromolecules.

The chemical relationship which exists between haemoglobin and bile pigments was first established in 1933, when, after monumental studies covering a period of 24 years, Hans Fischer and his colleagues succeeded in synthesizing bile pigments and protoporphyrin. He thereby demonstrated the fundamental tetrapyrrolic structure which is common to both haem and bilirubin. Prior to this, in vivo experiments on dogs performed by Mann, Sheard, Bollman and Blades showed that haemoglobin, injected into the arterial blood of the spleen or bone-marrow, produced a rise in the bilirubin content of the venous blood. Haematin was also evident in both arterial and venous blood.

The dysregulation of numerous genes has been associated with the development and progression of cancer, many of which are being investigated as potential therapeutic targets. Previous work in our laboratory and others reported the elevated expression of the nuclear import protein Karyopherin Beta 1 (Kpnβ1) in various cancers. The inhibition of Kpnβ1 by siRNA silencing inhibited the proliferation of cancer cells and induced cell death via apoptosis while having little effect on non-cancer cells. These findings suggested that Kpnβ1 has potential as an anti-cancer therapeutic target. Using an in silico screening approach to identify small molecule inhibitors of Kpnβ1 with anti-cancer activity, a number of compounds were selected for further investigation in our laboratory. The aim of this study was to investigate a novel small molecule, Compound 60 (9-[(1-methyl-3-piperidinyl)methoxy]-4-[(6-methyl-2-pyridinyl)methyl]-7-(5-methyl-2-thienyl)-2,3,4,5-tetrahydro-1,4-benzoxazepine); by monitoring (i) its effect on cancer cell biology using cervical and oesophageal cancer cell lines, (ii) its effect on nuclear import activities associated with Kpnβ1, (iii) its in vitro ADME pharmacokinetics and in vivo anti-cancer properties and (iv) performing biophysical analysis of Kpnβ1:C60 interactions. Cervical and oesophageal cancer cells were found to be more sensitive to C60 treatment compared to non-cancer epithelial cell. C60 treatment resulted in the inhibition of cancer cell proliferation, colony formation, migration and invasion. G1/S cell cycle arrest and a reduced expression of cell cycle regulatory proteins such as Cyclins D1, B1 and A as well as CDK4 was observed on treatment with C60. C60 induced cell death via apoptosis as observed PARP cleavage. These results suggest that C60 has an inhibitory effect on cancer cell biology. Immunofluorescent analysis and nucleo-cytoplasmic western blot analysis showed that C60 treatment resulted in the cytoplasmic retention of Kpnβ1. Immunofluorescent analysis and luciferase reporter assays showed that C60 inhibited the nuclear entry and transcriptional activity of a Kpnβ1 cargo, NFκB. Similarly, the transcriptional activity of cargo proteins NFAT and AP-1 were also inhibited. This suggests that C60 inhibits the nuclear entry of Kpnβ1 and thus its function as a nuclear importer of cargo proteins. In vitro ADME pharmacokinetics analysis found C60 to have high solubility, permeability and plasma protein binding properties and a short half-life. These findings suggest that C60 may have good oral absorption but rapid clearance in living systems. In vivo toxicology studies showed that C60 is tolerable, allowing for its testing in a xenograft nude mouse model. Intraperitoneal injection of C60 selectively inhibited the growth of oesophageal tumour cells with a significant effect observed on KYSE 30 oesophageal tumours. To investigate Kpnβ1:C60 binding interactions, Kpnβ1 was purified using the GST-tagged purification technique. Purified Kpnβ1:C60 interactions were monitored using the Bio-layer interferometry technique. Our preliminary data suggest an interaction between C60 and Kpnβ1 with Kd values in the micromolar range. We obtained varying Kd values, hence further optimisation is required to arrive at a conclusive Kd value. In conclusion, this study is a first to show that Compound 60 (9-[(1-methyl-3-piperidinyl)methoxy]-4-[(6-methyl-2-pyridinyl)methyl]-7-(5-methyl-2-thienyl)-2,3,4,5-tetrahydro-1,4-benzoxazepine), a small molecule identified to bind Kpnβ1 in an in silico screening approach interferes with the subcellular localisation of Kpnβ1, inhibits the localisation and transcriptional activity of Kpnβ1 cargo proteins and inhibits cancer cell biology in vitro and in vivo. In vitro ADME pharmacokinetics shows that C60 has tolerable drug properties. Biophysical analysis shows that C60 appears to bind Kpnβ1 in the micromolar range. Together, these results provide evidence for C60 as a compound that warrants further investigation as an anti-cancer agent.

<div>Utilization of nucleic acids to manipulate genetic information within a cell is known as gene therapy. It has provided researchers with unprecedented opportunities in treatment and mitigation of several life-threatening diseases. Gene therapy is an attractive alternative to conventional chemotherapy or radiation therapy due to its high efficiency, minimal side effects, and potential to evade drug resistance. The versatility of gene therapy makes it useful for the treatment of diseases dangerous disease like cancer. However, delivery of nucleic acid payloads to the intended target has been the bottleneck in clinical translation of such therapies. Here, we have developed and evaluated three different delivery systems (lipid based, polymer based and lipid-polymer hybrid) which can complex nucleic acid payloads, able to target specific cell types and get dissembled on cellular internalization to release the therapeutic payload. Our lipid and lipid-polymer hybrid delivery systems utilize a novel bacterial peptide sequence which enables these vectors to “stick” to fibronectin present in tumor extracellular matrix making them attractive for intravesical administration in bladder cancer management. Additionally, these systems have pH responsive modalities which aids in vector dissemble under acidic endosomal pH conditions for efficient release of therapeutic cargos after internalization into target cells.</div><div><br></div><div>In our efforts to develop an ideal delivery system with a tunable the assembly/disassembly properties, we synthesized a library of pendent polymer with biodegradable polycarbonate backbone. The ability of the pendent groups to form host-gest interaction with hydrophobic core of cationic cyclodextrins determined the stability of the delivery system. We demonstrate the capability of such polymer systems to form nano-dimensinal complexes with nucleic acid and transfect cancer cells. The above-mentioned property of cyclodextrins to form host-guest interaction with hydrophobic molecules also forms the basis of its utilization in the treatment of a rare metabolic disorder called Niemann Pick type C disease where there the cells loses the ability to remove stored cholesterol from endo-lysosomal compartments. Cyclodextrin forms host-gest complexes with aberrantly stored cholesterol and helps normalization of cellular cholesterol level. However, the soluble nature and small size of the cyclodextrin causes very rapid clearance for the body necessitating dosage levels as high as 9000mg/kg for therapeutic benefit. We have also developed a library of polyrotaxanes, where multiple cyclodextrins are threaded onto linear polymer chains and end-capped with bulky groups to prevent slippage. This kind to assembly drastically increases the systemic circulation time by preventing rapid renal clearance. We evaluated the ability of these constructs to serve as a long circulation delivery system for delivery of β-cyclodextrins for potential treatment of Niemann Pick type C disease. Finally, we have studied the structure-function relationships of these supramolecular assemblies to aid in rational design of therapeutic polyrotaxanes.</div>

Glucose uptake occurs in skeletal muscle under basal conditions, and increases in response to stimuli such as insulin and exercise. Exercise is known to increase blood flow, and it appears that insulin has similar hemodynamic effects, including increased blood flow and capillary recruitment, which can modify the amount of glucose uptake occurring under each condition. Here we study factors affecting both basal and stimulated myocyte glucose uptake, with a particular focus on vasoactive agents. Insulin stimulates the release of endothelin-1 (ET-1), a potent vasoconstrictor, from endothelial cells in culture. As yet it is unknown whether ET-1 is a type A (causing nutritive perfusion) or a type B (non-nutritive) vasoconstrictor, so here we use the pump-perfused rat hindlimb to characterize the distribution effects of ET-1. We show that ET-1 causes a type A vasoconstriction, stimulating basal metabolism at low doses, while at high doses the distribution of flow changes to become non-nutritive, inhibitory to metabolism. As a general vasodilator prevents both metabolic and hemodynamic effects, the effects on metabolism are due to the redistribution of flow. These redistribution effects are confirmed by the ability of high dose ET-1 to decrease aerobic tension development in the contracting hindlimb, and by the ability of low dose ET-1 to increase the interstitial glucose concentration. Given this understanding of the effects of ET-1 alone, we can investigate the interactions between ET-1 and insulin. In the perfused rat hindlimb, insulin has not been observed to have any vasodilatory effect, whereas here for the first time insulin appears to have vasodilator-like actions against ET-1 mediated vasoconstriction. Also, the redistribution of flow by ET-1 does not appear to alter the metabolic effect of insulin to cause glucose uptake at either dose of ET-1 used. Nitric Oxide (NO) is thought to be the mechanism by which insulin causes vasodilation in muscle. A previous study has shown that methacholine (MC), by increasing NO, was able to augment insulin-mediated glucose uptake and capillary recruitment, while other NO donors were unable to do so. Here we show that, at the dose used to increase glucose uptake in the previous study, MC has only a vasodilatory effect, and no direct effect on glucose uptake, in the perfused rat hindlimb. At higher doses, an effect on glucose uptake can be observed. This means that the increase in capillary recruitment by MC was responsible for the elevated insulin-mediated glucose uptake, and there was no direct effect of MC on glucose uptake. A recent publication suggested that the Na+-D-glucose cotransporter (SGLT1) was essential for insulin-mediated glucose uptake, although not required for basal glucose uptake. The implications of this detract from our proposed role of blood flow redistribution in insulin action. In attempting to reproduce these results in the perfused rat hindlimb we found that SGLT1 is not required for insulin-mediated glucose uptake, and confirmed this using a low sodium buffer, which would also inhibit the transporter. We conclude that SGLT1 is not required for insulin-mediated glucose uptake. Our results therefore suggest that complex interactions are involved in insulin action, some of which involve hemodynamic actions that are capable of altering insulin-mediated glucose uptake, and others in which insulin itself can limit the action of other vasomodulators, such as ET-1. It is apparent, however that SGLT1 in the endothelium may not be necessary for the metabolic effects of insulin, and that blood flow distribution, or capillary recruitment is therefore of great importance in delivering glucose to myocytes.

Virulence determinants of M. tuberculosis remain largely unknown. Of key interest has been the ability of the bacterium to survive intracellularly within its host cell, the macrophage, and its ability to cause extensive tissue necrosis. Exported proteases are commonly associated with virulence in bacterial pathogens, yet their role in Mycobacterium tuberculosis has virtually not been studied. Preliminary experiments showed M. tuberculosis culture filtrates contained a proteolytic activity inhibited by mixed serine/cysteine protease inhibitors and activated by Ca²⁺, features typical of some serine proteases, notably subtilisins, and possibly metalloproteases. Purification attempts were unsuccessful. A family of five genes that encode putative, secreted, serine proteases has recently been described in M. tuberculosis. These proteases share 36-47% sequence identity and are all encoded with putative signal peptides, suggesting that they are translocated across the cytoplasmic membrane. One member, mycP1, was selected for further study. The gene product, mycosin-1, was 30-35% identical to bacterial subtilisin-like serine proteases and contained the classic catalytic triad and oxyanion hole. Mycosin-1 also contained a typical signal peptide, a likely propeptide, and a Cterminal hydrophobic sequence with a high transmembrane potential. Topology analyses predicted mycosin-1 to be a type I ectoprotein. Consistent with this, expression of mycosin-1 in M. tuberculosis and in Mycobacterium smegmatis transformed with mycP1 (M. smegmatis-P1) was limited strictly to the cell envelope, as seen by Western blotting, and immunogold electron microscopy. Only full-length, 50-kDa mycosin-1 was observed by Western blotting in broth-grown M. tuberculosis and M. smegmatis-P1 lysates, whereas a 40-kDa species was detected in 6-week M. tuberculosis culture filtrates. A similar 40-kDa immunoreactive band was also observed in lysates of macrophages infected with M. tuberculosis, consistent with robust transcription of the mycP 1 gene during growth in macrophages. Since putative mature mycosin-1 has a molecular weight of 38.6 kDa, the 40-kDa protein may represent activated mycosin-1 after propeptide cleavage. In conclusion, mycosin-1 is an exported, cell envelopeassociated subtilisin homolog that is expressed during growth of M. tuberculosis in vitro and in macrophages.

Alzheimer's disease (AD) is commonly characterized by a loss of cognitive function due to the deterioration of neuronal synapses from the presence of senile amyloid beta-42 (Aß42) plaques. Evaluating cognitive deficits caused by Aß42 using human cortical neurons poses a challenge due to sourcing difficulties, and the use of animal models to assess drug efficacy creates biological hurdles from lack of species translatability. Recent advances in induced-pluripotent stem cell technology have enabled the development of mature, human-based cortical neuron models. The development of an hiPSC-cortical neuron differentiation protocol facilitates the exploration of disease onset and functional analysis from a patient-derived cell source, and further investigation of potential therapeutic treatments, while eliminating biological efficacy concerns. Long-term potentiation (LTP) was utilized as an in vitro correlate for memory and learning to quantify cognitive deficits in sporadic AD (SAD) and familial AD (FAD) systems and assess drug treatments for the prevention of Aß42-induced neurotoxicity. Synaptic connectivity and LTP induction through high-frequency stimulation was simulated through cortical neurons cultured on microelectrode arrays (MEAs), such that the functional activity of the neuronal population could be assessed overtime. AD therapeutic treatments were shown to block the Aß42-induced neurotoxic loss of synaptic plasticity and maintain persistent LTP in a model for SAD. Subsequently, FAD was assessed through the differentiation of patient-derived AD iPSCs, where LTP proficiency could be evaluated to relate to clinical cognitive evaluations. This study established a serum-free, in vitro human-derived iPSC-cortical neuron protocol that could be adapted to validate disease mechanisms and drug efficacy in patient-derived neural networks as a potential platform for precision medicine.

Preeclampsia is the leading cause of foetal and maternal mortality and morbidity in developing countries. In South Africa, maize is a dietary staple for most black African populations and is susceptible to contamination by mycotoxins such as fumonisin B1 (FB1).Fumonisin B1 is a ubiquitous secondary metabolite of Fusarium fungi produced predominantly by Fusarium verticillioides. This mycotoxin shares structural similarities with the backbone of sphingoid bases (sphinganine and sphingosine) which are substrates for the biosynthesis of complex sphingolipids. The mechanism of FB1 toxicity therefore is centred on the disruption of this process. The aim of the present study was to elucidate the possible causal link between FB1 and preeclampsia. Following ethical approval, 20 normotensive and 20 preeclamptic patients were recruited into the study. Blood and placental tissue were collected and processed for further analysis. The presence of FB1 was verified using standard immunohistochemical and electrophoretic techniques. The levels of FB1 and sphingolipids were quantified using high performance liquid chromatography (HPLC). Western blotting was conducted to confirm the presence of FB1 in the serum. Placental tissue apoptosis was evaluated using Hoechst staining and other markers. Lipid peroxidation was measured in serum and placental tissue of both groups. Fumonisin B1 was immunolocalised within the endothelial cells and mesenchymal cells of placentas from both groups, while FB1 was present in cytotrophoblastic cells of preeclamptic patients only. In addition, FB1 concentrations were significantly higher in preeclamptic compared to normotensive serum samples. Sphinganine was significantly elevated in preeclamptic serum samples whilst there was no statistical difference in the sphingosine levels between the groups. Chromatin condensation was higher in the preeclamptic patients. Caspase 3 and Fas were present with greater intensity in preeclamptic samples. The levels of lipid peroxidation were significantly higher in both serum and placental tissue of preeclamptic patients. This study has demonstrated not only the presence of FB1 in the serum and placental tissues of pregnant women but also the potential effects of this mycotoxin in the humans. / Thesis (M.Med.Sc.)-University of KwaZulu-Natal, Durban, 2012.