Part I of this thesis deals with general methodology and the experimental work can be clearly divided into two sections. Part II deals with studies carried out chiefly to determine further the epidemiological factors influencing serum lipid (and in particular, serum triglyceride) levels in the population groups of Southern Africa. The original objectives are described on page 70 and the main conclusions summarised on page 114. Part III describes three studies which were conducted in an attempt to fill some of the gaps in the considerable literature on the relationship between dietary carbohydrate and serum lipids in man. Both in the review of the literature at the beginning of this section and in the interpretation of the results of each of the studies, discussion has been chiefly limited to experiments conducted in man. There is a great deal of information available on studies carried out in experimental animals which show marked species differences from man with regards kinetic behaviour of serum and liver triglycerides(l). Where relevant, of course, reference has been made to these studies. The significance of each of the three studies has been discussed separately, but the principal objectives are mentioned on page 138 and the general conclusions are summarised on page 205.

An investigation was carried out to study whether long-chain fatty acids (LCFAs) have an effect on digestion of waste sludge under anaerobic conditions. Four different kinds of LCFAs were used in this study. The 18 carbon series with 0, 1, 2 and 3 double bonds were studied to evaluate the degree of saturation on fatty acid degradation. Due to their molecular structure, unsaturated LCFAs are more soluble than saturated LCFAs. Oleic, linoleic, linolenic acid with an ascending number of double bonds were tested as representatives for three different degrees of saturation. In addition, stearic acid, a saturated fatty acid was also tested. LCFAs were added to sewage sludge at concentrations ranging from 5% to 20% on a weight basis and the pH, solids reduction and COD reduction were determined. The results suggested that in addition to degrading in the digesters, all unsaturated acids contributed additional solids removal, compared to the control group. In contrast, stearic acid did not affect the solids removal. The COD reduction was similar to solids reduction in that additional COD was destroyed when unsaturated LCFAs were added to the sludge. The mechanism for additional solids reduction is not known. / Master of Science

INTRODUCTION: Specialized pro-resolving lipid mediators (SPM) such as resolvin-D1 (RvD1) act to resolve vascular inflammation and may guard against the progression of restenosis following cardiovascular interventions. Stimulating synthesis of these mediators directly in vascular cells may increase their local availability, and thus, protect against restenotic injury. However, the ability of endothelial (EC) and vascular smooth muscle cells (VSMC) to produce SPMs from their polyunsaturated fatty acid precursor decosahexaenoic acid (DHA) via lipoxygenase (LO) enzymatic transformation remains unknown. We sought to determine whether vascular cells produce SPMs from DHA and, if they do, how inflammation and mechanical injury of the vasculature alter biosynthesis. METHODS: Primary cultures of human saphenous vein endothelial and smooth muscle cells were treated with DHA in cell culture media (+ 10% serum) for 4h-24h. Freshly dissected rabbit aorta was incubated intact or following gentle endothelial denudation in cell culture media (+10% serum) with or without DHA for 48h. SPM levels in media were quantified by LC-MS/MS and ELISA and lipoxygenase expression and localization were assessed by western blotting and immunofluorescence staining, respectively. RESULTS: EC and SMC receiving media without DHA did not synthesize SPMs within the detection limits of the assay, whereas DHA treatment produced 17-HDHA, 14-HDHA, Mar1, RvD5, RvD2, and a dose and time-dependent increase in RvD1 production in EC (10.1 ±1.0 pg for 1000nM at 24h) and SMC (7.4 ± 0.2 pg for 1000nM at 24h). Intact rabbit aorta incubated in DHA+ media produced 0.24 ± 0.05 pg RvD1/mg tissue whereas aorta incubated in DHA− media produced 0.13 ± 0.007 pg RvD1/mg tissue. Moreover, EC-denuded aortas produced less RvD1/mg tissue than intact aortas. 5-LO was expressed in both cell types, however DHA induced 5-LO expression in EC (1.3 fold -DHA) but not in SMC. DHA promoted a nuclear to cytoplasmic shift of 5-LO in both EC and SMC. Finally, TNF-α stimulated an increase in RvD1 production in EC. CONCLUSIONS: Human vascular cells and rabbit vascular tissue can biosynthesize SPMs de novo from their precursor DHA, signifying a new source of SPMs in the vasculature.

Chitin is a polysaccharide derived from the shell of crustaceans. Chitosan is a deacylated chitin derivative and ChitoRich™ is a chitosan-based formulation. Chitosan is known to inhibit dietary fat absorption. ChitoRich™ and chitosan were evaluated for their ability to control fat absorption in growing rats. Forty-eight rats (about 150 g) were fed for 4 weeks on 12 different diets containing two levels of vegetable shortening (15 and 5%) and fiber (5 and 2.5%) with cellulose, chitosan, or ChitoRich™ as the dietary fiber sources. Rats fed ChitoRich™ ate less and gained less body weight than rats fed the other fiber sources. Rats fed ChitoRich™ adapted over time; thus, the weight control effect was not significant during the third and fourth weeks. Feed efficiency of the rats fed ChitoRich™ was lower than that seen in rats fed the other fibers, indicating that calories from a higher percent of absorbed macronutrients were required to maintain their bodies. Apparent fat digest (AFD) was lower in rats fed the ChitoRich™ than in the rats fed cellulose. The reduction of AFD was not as great as that seen by other researchers, possibly due to the low solubility and high viscosity of shortening, compared to corn oil. AFD of the rats fed ChitoRich™ increased over time, suggesting an adaptation to ChitoRich&trad;. Apparent protein digestibility (APD), apparent Ca digestibility (ACaD), and weight gain per unit of protein intake were less in the rats fed ChitoRich™ than in the other fiber-fed groups. ChitoRich™-fed rats had retarded body weight gain, reduced levels of serum albumin, total liver lipids, and epididymal fat weight when compared to the other fiber groups, possibly due to the reduced caloric intake, FD, and APD. Serum cholesterol level was not affected by the fiber source. Liver iron and zinc concentrations were not different among fiber sources, indicating that chitosan and ChitoRich™ may not inhibit the absorption of trace minerals. Liver retinol concentration in the ChitoRich™-fed rats was higher than in the cellulose- or chitosan-fed groups, possibly due to the antioxidant effect of ascorbic acid. All rats were free of microscopic lesions, suggesting that chitosan and ChitoRich™ are safe sources of dietary fiber under these dietary conditions. ChitoRich™ is effective for reducing dietary fat absorption and body weight gain.

Effects of Fe and Se status on GSHPx activity and lipid peroxidation in liver and intestinal mucosa were studied. Rats were Se and Fe supplemented ( +Se+Fe), Sedeficient and Fe-supplemented (-Se+Fe), Se supplemented and Fe overloaded (+Se++Fe) by intramuscular injection, or Se deficient and Fe overloaded (-Se++Fe) for 20d. Fe overloaded tissues had more Fe, but hemoglobin was unaffected. Liver and mucosal GSHPx activity was low in Se-deficient rats. Thiobarbituric acid reactive substances (TBARS) were higher in Fe overloaded and -Se++Fe vs +Se++Fe tissues. Mucosal TBARS was higher in -Se++Fe rats gavaged with CBrCl3. In experiment 2, Fe overload was induced by a 2% carbonyl iron, low-Se diet fed for 2mo, the +Se++Fe and -Se++Fe groups. Low Se-, low Fe-diet was fed to rats supplemented with Fe or Fe and Se in the water, the -Se+Fe and +Se+Fe controls. Iron overloaded tissues had more Fe. Liver and mucosal GSHPx activity was lower in Se-deficient and +Se++Fe vs +Se+Fe rats. TBARS was higher in Fe overloaded, -Se++Fe vs +Se++Fe, and CBrCl3 tissues. Hemoglobin and serum Fe were lower in the -Se++Fe group. In experiment 3, low-Se, low-Fe diet was fed for 20d, the -Se-Fe, +Se-Fe, -Se+Fe and +Se+Fe groups. Mucosal Fe was lower in the Fe-deficient rats. Cytochrome P-450 and GSHPx activities were lower and TBARS was higher in Se-deficient tissues. In experiment 4, the +Se+Fe, +SeFe, -Se+Fe, -Se-Fe, +Se++Fe, and -Se++Fe groups treated as in experiment 3. Iron overloaded tissues had more Fe and TBARS, but hemoglobin and serum Fe were unaffected. GSHPx and Cytochrome P-450 activities were lower in Se-deficient and in +Se++Fe vs +Se+Fe rats. CBrCl3 did not affect TBARS. High TBARS occurred in liver and mucosa of Fe overloaded rats. Chronic Fe overload was required to reduce liver and mucosal GSHPx activity. The combination of Se deficiency and Fe overload caused very high TBARS. Low oral CBrCl3 doses elevated mucosal TBARS, a first report of extrahepatic action of CBrCl3 in vivo. Iron deficiency did not affect GSHPx activity or CBrCl3 induced lipid peroxidation.

The effects of pentachlorophenol (PCP), a widely used pesticide, on the electrical characteristics of lipid bilayer membranes has been studied. When a small amount of PCP (even at a concentration of a few micromoles per liter) is present in the electrolytic solution surrounding the membrane, the electrical conductivity of the membrane significantly increases. The present work was concerned with detailed measurements of the changes in the conductivity caused by PCP under chemically controlled conditions. The experimental results were analyzed to determine the permanent species in the membrane, and an attempt was made to correlate the data with existing models of transport.

Pentachlorophenol (PCP) is one of the prominent environmental pollutant that has penetrated into food chain and is present in humans. Health concerns have been raised since daily intake of PCP by the US population is estimated to be 16-19 µg. PCP facilitates dissipation of electrochemical potential gradients of hydrogen ions across energy transducing membranes, which are the energy sources for the conversion of adenosine diphosphate into adenosine triphosphate. Closely linked to these dissipative effects is the development of electrical conductivity in lipid membranes, induced by the presence of PCP. Three modes of PCP - induced membrane electrical conductivity were theoretically analyzed and experimentally verifiable formulations of each models were developed. Experimental studies using the charge - pulse method involved characterization of the time dependent transmembrane voltage over a wide pH range, from 1.8 to 9.5, for 30 µM concentrations of PCP. Lipid membranes were prepared from dioleoyl phosphatidylcholine. It was shown that three PCP molecular species were determining the transmembrane transfer of hydrogen ions: electrically neutral PCP molecules (HA), negatively charged pentachlorophenolate ions (A⁻) and negatively charged heterodimers (AHA⁻). It was found that at pH>9 the membrane electrical conductivity was determined by the transmembrane movement of A⁻ ions, whenever pHAHA⁻ species. Two new membrane surface reactions were proposed as supplementary mechanisms for the generation of AHA⁻ in addition to the formation of AHA⁻ by the recombination of HA and A⁻, HA + A⁻→ AHA⁻. These new reactions are, (i) 2HA → H⁺ + AHA⁻, and (ii) H20 + 2A⁻ → OH' + AHA⁻. Reaction (i) provides formation of membrane permeable heterodimers AHA⁻ at pH < < 5.5 and reaction (ii) at pH> > 5.5. The maximum surface density of AHA" heterodimers was 0.09 pmol/cm² • The rate constant of formation of AHA' by recombination, HA + A⁻ → AHA' was estimated to be k[subscript f] = 2.6xl0⁹ cm² mol⁻¹ s⁻¹ and the dissociation rate constant for AHA⁻ Further, it was possible to determine the rate constants of transmembrane translocation for A' and AHA⁻ ions to be k[subscript a] = 6.6x10⁻⁵ s⁻¹ and k[subscript aha] = 1200 S⁻¹, respectively.

It has been found that the herbicide 2,4-D has the ability to increase the rate of transport of positive ions of several kinds and inhibit the transport of negatively charged tetraphenylborate ions in lipid bilayer membranes. Only the neutral molecules of 2,4-D are transport active. The ionized 2,4-D molecules do not modify the transport of ions, and do not by themselves permeate through lipid membranes. The results suggest that the enhancement of transport of positively charged ions is dominated by the increase of the ion translocation rate constant. It has been shown that membrane transport of negatively charged tetraphenylborate ions is suppressed by 2,4-D. The effect is dominated by the suppression of translocation of these ions across membrane interior, rather than by the decrease of their adsorption at the membrane surface. It has been shown that the enhancement of nonactin-mediated transport of potassium ions by 2,4-D can be accounted for by a simple carrier model. From the changes of kinetic parameters of nonactin-K('+) transport, membrane conductance due to positively charged tetraphenylarsonium ions and also from the changes of membrane conductance and relaxation time constant due to transport of negatively charged tetraphenylborate ions, the changes of the electric potential of the membrane interior have been estimated. The potential of the membrane interior becomes more negative in the presence of 2,4-D and its change is proportional to the aqueous concentration of 2,4-D. The effect of 2,4-D on ion transport was explained by the hypothesis that a layer of 2,4-D molecules is absorbed within the membrane/water interfacial region, and that the 2,4-D molecules are oriented in such a way that their dipole moment is directed toward the aqueous medium. The results suggest that this layer is located in the hydrocarbon side of the interface. The hypothesis has been confirmed by the measurements of changes of electric potential difference across air/water and air/lipid monolayer/water interfaces. It has been found that the electric potential of the nonpolar side of the interface decreases in the presence of neutral molecules of 2,4-D, which is in agreement with the conclusions drawn from the results of membrane experiments. The effect of the other auxin-type phenoxy herbicides, 2,4,5-T and 2,4-DB on lipid bilayer membranes has been found to be similar to that of 2,4-D. In contrast, the phenoxy acid 2,4,6-T, has very little or no herbicidal activity, and at the same time has small effect on ion transport in membranes. Biologically active 2,4-D derivatives, amino acid conjugates of 2,4-D (isoleucine, leucine and valine conjugates) have been found to be also transport active in a manner similar to 2,4-D. Similar conclusions have been drawn from experiments with natural auxin indole acetic acid. The results obtained in this work suggest the existence of correlation between the biological activity of herbicides acting as plant growth regulators and their ability to enhance transport of positively charged ions across lipid membranes. This work provides insight into the physical origin of such activity.

OBJECTIVE: The close relationship between obesity and type 2 diabetes (T2D) highlights the fact that most diabetes patients are overweight or obese. We propose that elevated glucose and free fatty acid levels in those patients cause beta-cell dysfunction. Chronic exposure to excess nutrients (glucose and free fatty acid) leads to glucolipotoxicity, characterized by basal insulin hypersecretion, a left-shift in the glucose dose-dependent insulin secretion curve, and blunted glucose-stimulated insulin secretion. One of the suggested reasons for this defect is elevated intracellular lipid. In this study, our objective was to investigate whether reducing beta-cell lipid levels can reverse basal insulin hypersecretion. METHODS: INS-1 (823/13) cells were cultured in 4 or 11 mM glucose media. Elevated glucose and KCl doses were added to cells in the insulin secretion experiments. In the KCl-induced insulin secretion experiment, cells were treated with a combination of 12 mM glucose and 250 μM diazoxide, then assigned to different test concentrations with elevated KCl doses. Insulin release and content were measured by the insulin ultra-sensitive homogenous time-resolved fluorescence (HTRF) kit (Cisbio). Following that, we monitored intracellular Ca2+ activity of KCl-induced insulin secretion on a fluorescence spectrophotometer F-2000 (Hitachi). Additionally, we acutely added Adipo C (20 µM) or fatty acid-free BSA to cells to reduce the lipids levels in the ß-cells. We also stained with Nile Red (Sigma) to examine the intrinsic lipid droplets in those cells. RESULTS: ß-cells cultured in excess nutrients (11 mM glucose) exhibited a left shift in the glucose dose-dependent response curve. The hypersecretion at low glucose could be blocked by the KATP channel activator, diazoxide, indicating that Ca2+ influx drives the increase in secretion at glucose concentrations normally considered basal. Here we extend this left shift to include KCl-induced insulin secretion, supporting a role for Ca2+ in the observed hypersensitivity. KCl-induced Ca2+ influx was also left-shifted. Interestingly, we found acute exposure to Adipo C or fatty acid-free BSA reversed the basal hypersecretion in cells cultured in excess nutrients. CONCLUSION: The work presented in this study provided supporting evidence that ß-cells cultured in excess nutrients were hypersensitive to glucose while extending these studies to KCl-induced insulin release. The excess nutrient-induced left shift in both glucose and KCl-stimulated insulin secretion was mediated by increased Ca2+. Thus, we postulate that excess nutrient exposure increases ß-cell plasma membrane lipids that alter Ca2+ handling to allow increased Ca2+ influx at inappropriate low glucose concentrations. Our results demonstrated that cells acutely exposed to the putative long-chain acyl-CoA synthetase inhibitor Adipo C or fatty acid-free BSA reversed basal insulin hypersecretion and supports a role for lipids mediating the adverse effect. T2D patients with obesity have a similar physiologically elevated fasting blood glucose and lipid. Thus, our findings suggest lowering lipid levels in ß-cells may have therapeutic potential in treating hyperinsulinemia leading to T2D.

Background: In the pregnant population, being physical active and meeting gestational weight gain (GWG) guidelines have numerous health benefits for both mother and infant. Markers of lipid metabolism are known to be influenced by these two variables in the non-pregnant population. However, the relationship between physical activity (PA) and GWG on lipid markers has yet to be assessed during pregnancy. My thesis aims to address this gap in the literature. Methods: The first objective of my thesis was to examine the relationship between maternal PA and GWG on gross measurements of fetal and placental development (n=40). Specifically, three markers of placental efficiency (Pl-E) were examined (birthweight [BW], BW-to-placenta weight ratio, and residual BW). The second objective of my thesis was to analyze maternal serum lipid and glucose markers (n=40), in mid (24-28 weeks) and late (34-38 weeks) gestation as well as from the umbilical cord (UC) as they relate to both PA and GWG. The third objective of my thesis was to explore how PA level and GWG status affect markers of lipid metabolism in term placenta (n=31). Markers of placental lipid transport (FATP1, FABP4, FAT/CD36) were assessed at the protein level, and enzymatic activity of placental lipoprotein lipase was also measured. Lastly, placental lipid storage was assessed by examining triglyceride content, paired with lipid droplet staining. Results: There was no relationship between PA independently or in combination with GWG on any Pl-E markers. A significant association was found between GWG and BW in women who gained weight excessively compared to insufficiently. Neither PA nor GWG categorization was associated with maternal lipid and glucose markers. Total cholesterol levels measured in UC serum were significantly lower in women categorized as active throughout pregnancy (p<0.0001) or whose activity dropped in late gestation (p<0.0001) compared to those who were inactive v throughout gestation. Glucose levels were lower in UC blood of women who gained weight appropriately in mid-gestation compared to those who gained insufficient (p=0.040) or excessive (p=0.021) weight. In terms of placental fatty acid transport, there was a significant interaction between PA status and GWG categorization and placental FATP1 protein expression (F=14.62, p<0.0001). Finally, while no differences were found in placental lipid droplet staining, the droplets were more likely to be clustered within the syncytiotrophoblast border. Conclusion: In conclusion, maternal PA had no association with Pl-E, while GWG was only associated with BW. My thesis work found that while maternal serum lipid markers were not associated with PA and GWG, both maternal PA and GWG status were related to changes in UC and placental lipid markers throughout pregnancy. In combination with previous research from our lab, it is suggested that women who are physically active during pregnancy, and gain weight appropriately may be transporting fewer nutrients (i.e. fatty acid, glucose, cholesterol) to the placenta than those who are inactive, yet simultaneously increasing metabolization. Future research should further investigate these findings by performing functional experiments.