Search

Background: Systemic lupus erythematosus (SLE) is a multi-system autoimmune disease characterised by autoantibody production and variable clinical features, ranging from mild to severe disease. Patients with SLE are at increased risk of developing accelerated atherosclerosis. Biomarkers have potential utility in SLE as markers of disease or predictors of future clinical events and mortality. Objective The aim of this thesis was to identify serological biomarkers predictive for erosive arthritis (EA), cardiovascular events (CVEs), mortality and subclinical atherosclerosis in SLE. Methods: In chapters 2 to 4, study subjects were SLE patients from Bath. Anti-cyclic citrullinated peptide antibodies (ACPA) and HLA-DR and -DQ were studied for markers of EA, and anticardiolipin (aCL) and lipoprotein profiles for markers of CVEs and mortality. In chapters 5 and 6, study subjects were women with SLE from Manchester. B-mode ultrasound scans of subjects' carotid arteries were performed at baseline and follow-up time-points to detect atherosclerotic plaque. Baseline IgG and IgM antiphospholipid (aPL) antibodies and CV risk factors were studied for markers of subclinical atherosclerosis. Clinical data collected for all studies included SLE features and auto-antibody profiles. Results: ACPA was identified as a marker of a SLE phenotype with EA - "rhupus". Patients with major erosive arthritis were HLA-DQB1*0302 carriers. Increased aCL GPL levels and total cholesterol : high density lipoprotein-C (TC : HDL-C) ratio were markers for future CVEs, and increased TC : HDL ratio, aCL GPL and lipoprotein(a) concentrations were markers for increased mortality. Lower HDL-C concentrations and anti-annexin A5 (anti-AnxA5) GPL were markers of carotid plaque progression. Conclusion: This thesis identified new markers for EA, subclinical atherosclerosis and future CVE and mortality risk in SLE. Strategies to incorporate these new CV markers into clinical CV risk assessments may assist in distinguishing the subset of SLE patients most at risk of developing accelerated atherosclerosis.

Antibiotic-associated diarrhoea (AAD) and colitis, with the causative agent being the Gram-positive anaerobe, Clostridium difficile, are some of the most important hospital-acquired infections and significant burdens to healthcare services worldwide. Treatment of the infection is often ineffective and currently no vaccine is available against C. difficile infection (CDI). Research to identify novel virulence factors potentially leads to the development of new therapeutic and prophylactic drugs. As lipoproteins have been shown to play key roles in the virulence of several pathogens, the aim of this project was to investigate whether lipoproteins are involved in the virulence of C. difficile. Lipoproteins are anchored to the extracellular side of the cytoplasmic membrane in Gram-positive bacteria. Two enzymes are involved in the biosynthesis of lipoproteins: lipoprotein diacylglycerol transferase (Lgt) attaches lipoproteins to the membrane, and lipoprotein signal peptidase (Lsp) cleaves the signal peptide from the amino-terminus of lipoproteins. In order to study lipoprotein processing in C. difficile, lgt and lsp mutants of the C. difficie 630Δerm strain were generated using the ClosTron system. Antibody reactivity of 14 C. difficile lipoproteins was also investigated. It was shown in this study that lgt mutation caused changes in the lipoproteome of C. difficile. Therefore, inactivation of the lgt gene allowed investigation of the global contribution of lipoproteins to bacterial processes. The physiology and virulence of the lgt mutant was studied in vitro and in vivo. Surprisingly, many of the assayed phenotypes were not significantly affected by disruption of the lgt gene. Nevertheless, the ability of the lgt mutant to adhere to Caco-2 cells was markedly reduced. In addition, the phenotype of the lgt mutant observed in mice suggests that the faecal shedding of C. difficile is affected by Lgt inactivation. In further studies, the CD0873 lipoprotein as a potential adhesin of C. difficile was identified by in silico approach. Contribution of the CD0873 lipoprotein to the adherence of C. difficle was investigated by several different assays and the results strongly suggest that the CD0873 lipoprotein is directly involved in adhesion

Exaggerated postprandial lipaemia has been implicated in the development of atherosclerosis. Thus, by reducing postprandial TAG concentrations, exercise may play a role in delaying atherogenic progression. This thesis sought to explore the qualitative nature of, and the mechanisms behind, the moderate exercise-induced attenuation to postprandial lipaemia. Before the experimental studies commenced, a reproducibility study was undertaken. This showed that in a group of eight middle-aged men, the postprandial plasma TAG response differed by only 1.9 ± 5.1 % (mean ± standard error) on a testretest basis, indicating that the oral fat tolerance test had enough precision to detect the effect of exercise on TAG metabolism. Previous work suggested that the exercise-induced reduction to lipaemia was linked to the energy expended by exercise. As the attenuation may have been mediated by energy deficit, rather than exercise per se, a study comparing the effect of a 90-minute moderate exercise session with an equivalent dietary-induced energy deficit on postprandial lipid metabolism was undertaken, in a group of eleven postmenopausal women. This showed that the reduction in postprandial lipaemia elicited by exercise was far greater than that elicited by intake-restriction (20 % vs. 7 %). The second experimental study aimed to establish the effect of a 90-minute moderate exercise session on postprandial chylomicron- and very-Iow-density lipoprotein (VLDL)-TAG concentrations, and its effect on exogenous (through use of a l3e-Iabelled lipid) and endogenous fat oxidation, in a group oftwelve middle-aged men. Exercise reduced postprandial lipaemia by 23 %, and over three-quarters of this reduction was due to lower VLDL-TAG concentrations. Increases in endogenous fat oxidation accounted for over half of the increase in postprandial fat oxidation. In the third experimental study, the effect of a 90-minutes moderate exercise session on Intralipid clearance, and postprandial lipaemia, was determined in a group of eight middle-aged men. Exercise attenuated postprandial lipaemia by 18 %, but did not increase Intralipid clearance. Taken together, these data imply that moderate exercise predominantly reduced postprandial TAG concentrations by reducing hepatic VLDL secretion, rather than increasing TAG clearance, and this effect is not mediated by whole-body energy deficit. In addition, this work has shown that moderate exercise is effective at attenuating postprandial lipaemia in middle-aged men and postmenopausal women.

Lipid metabolism plays a role in the pathogenesis of atherosclerotic cardiovascular disease (ASCVD) through aberrant lipid biosynthesis and trafficking via lipoproteins. The dysregulation of these pathways occurs across various tissues and understanding tissue-specific lipid biosynthetic flux can unravel the metabolic contribution to physiological responses. Furthermore, understanding how these fluxes are regulated at the genetic level may uncover new mechanisms to be exploited as clinical targets. Sphingolipids constitute a pool of bioactive lipids that are critical in maintaining cell function and lipid homeostasis. Owing to their abundance in circulation and enrichment in lipoproteins, sphingolipids have been emerging as biomarkers of several diseases including ASCVD. In this dissertation, we study the intersection of sphingolipid metabolism and dietary fat in the progression of ASCVD and other co-morbidities of metabolic syndrome. First, we applied mass spectrometry to quantify the metabolic flux of cis (CFAs) and trans monounsaturated fatty acids (TFAs) into sphingolipids by the initial, rate-limiting enzyme serine palmitoyltransferase (SPT). We identified novel sphingolipids synthesized from TFAs that exhibit high affinity for the sphingolipid pathway through their preferential incorporation into the long-chain base of sphingolipids and secretion from hepatocarcinoma cells. Next, we designed custom high-fat diets (HFDs) with identical macronutrient composition enriched in either CFAs or TFAs to characterize the role of sphingolipid metabolism in the progression of hepatic steatosis and atherosclerosis induced by dietary fat. Administration of a HFD enriched in TFAs to low-density lipoprotein receptor deficient (Ldlr-/-) mice accelerated liver steatosis, hepatic very-low-density lipoprotein (VLDL) secretion, and atherogenesis compared to a HFD enriched in CFAs. Mechanistically, we demonstrate that in vivo SPT inhibition mitigated these phenomena while significantly diminishing atherogenic VLDL consisting of TFA-derived polyunsaturated sphingolipids. SPT inhibition additionally remodeled the lipidome by shifting hepatic fatty acids towards phospholipids, indicating that diverting fatty acids away from SPT and downstream sphingolipids may be beneficial on these HFDs. Collectively, we elucidate a novel mechanism in which SPT flux synergizes with hepatic lipoprotein secretion to deliver atherogenic sphingolipids into circulation. Thus, we uncovered a critical metabolic link between sphingolipids and lipoprotein metabolism that can be therapeutically targeted in ASCVD in response to specific dietary fats.

Atherosclerosis is a condition caused by cholesterol accumulating in arterial intimal cells and is a disease that kills more people in the United States and globally than any other disease. Atherosclerosis is commonly recognized to arise from arterial intimal macrophage cholesterol accumulation, but cell lineage tracing technology has shown that a large majority of cholesterol-laden intimal cells found in atherosclerotic arteries are actually vascular smooth muscle cells that have switched phenotypes to a macrophage-like cell. This vascular smooth muscle cell to macrophage-like cell phenotypic switch is known as transdifferentiation and can be triggered by vascular smooth muscle cell cholesterol accumulation. While vascular smooth muscle cell to macrophage-like cell transdifferentiation is considered pro-atherogenic, mechanisms that govern this process are not entirely elucidated. Furthermore, lipid-laden macrophage-like cells of vascular smooth muscle cell origin have demonstrated impaired ABCA1-dependent cholesterol efflux, but it is poorly understood why this occurs in these cholesterol-filled cells. A possible mechanism, which may at least partially attribute to lipid-filled macrophage-like cells demonstrating attenuated ABCA1-dependent cholesterol efflux, is miR-33a expression, as a primary function of this microRNA is to silence ABCA1 expression, but this has yet to be rigorously investigated. For me to directly test this, I would need to identify an immortalized vascular smooth muscle cell line that modulates cholesterol homeostasis and transdifferentiates into macrophage-like cells when exposed to high cholesterol levels, similar to primary vascular smooth muscle cells. I characterized the vascular smooth muscle cell line MOVAS cells to determine that this cell line regulates cholesterol metabolism and responds to hypercholesterolemic conditions, similar to primary vascular smooth muscle cells. Thus, the MOVAS cell line was used to generate miR-33a knockout MOVAS cells to compare to miR-33a endogenously expressing wild-type MOVAS cells to delineate any possible pro-atherogenic role of miR-33a expression in cultured vascular smooth muscle cells. When these two types of MOVAS cells were cholesterol-loaded to convert into macrophage-like cells, this resulted in the wild-type MOVAS cells exhibiting impaired ABCA1-dependent cholesterol efflux. In the cholesterol-loaded wild-type MOVAS macrophage-like cells, I also observed a delayed restoration of vascular smooth muscle cell phenotype when these cells were exposed to the ABCA1 cholesterol acceptor, apoAI. My results imply that miR-33a expression in vascular smooth muscle cells drives atherosclerosis by triggering macrophage-like cell transdifferentiation via attenuated ABCA1-dependent cholesterol efflux. Based on my findings, testing the potential pro-atherogenic impact of vascular smooth muscle cell miR-33a expression in atherogenic animal models is warranted.

Elevated levels of low-density lipoprotein (LDL) and its derivative lipoprotein(a) (Lp(a)) in the bloodstream are directly implicated in the development and progression of atherosclerotic cardiovascular diseases (ASCVD). LDL homeostasis is mainly regulated by the LDL receptor (LDLR), which is expressed predominantly on the cell surface of hepatocytes where it binds and removes free LDL particles from circulation. Endocytosed LDLR may then either be recycled back to the cell surface or diverted to lysosomes for its degradation. LDLR recycling is therefore critical for efficient LDL uptake. Endocytic recycling is a complex process involving several trafficking steps and many potential molecular mediators. Here I characterized the role of a small GTPase, RAB10, in the trafficking of LDLR. I found that RAB10 depletion inhibits LDLR recycling from endocytic recycling compartment (ERC), thereby reducing surface LDLR abundance and cellular LDL uptake. Further investigation showed that RAB10 also promotes recycling of another cell surface receptor, the transferrin receptor (TFR), but from a different endocytic compartment. Our findings suggest that RAB10 impacts recycling of LDLR and TFR by promoting vesicle trafficking from different intracellular compartments. Although Lp(a) is another casual risk factor for ASCVD, there are no specific Lp(a)-lowering treatments, and their development is limited by a gap in knowledge regarding Lp(a) regulation. Unlike the well-known receptor mediated pathway for LDL homeostasis, the mechanism(s) for cellular uptake of Lp(a) are poorly understood, and several receptors, including LDLR, have been individually studied with conflicting data. To address this gap, we applied a high-throughput whole-genome CRISPR screen to interrogate the regulators of Lp(a) endocytosis in HuH7 cells. Surprisingly, LDLR and other known regulators of LDLR, including SCAP, MBTPS2, MYLIP were among the top hits in the screen. No other receptors, including those previously proposed to serve as Lp(a) receptors, exhibited a functional influence on Lp(a) uptake in this screen. Our results suggest that the LDL receptor is the primary mediator of Lp(a) uptake in hepatocytes. Collectively, the work done in this thesis increases our understanding of receptor mediated trafficking of LDL and Lp(a).

Atherosclerosis is a chronic inflammatory disease that results from the accumulation of cholesterol-containing LDL particles in the vessel wall, and is the underlying pathology for many cardiovascular diseases. Current clinical guidelines focus on LDL-cholesterol lowering as a means of cardiovascular disease prevention. The immune system also plays a major role in the pathogenesis of this disease, and emerging evidence points to the therapeutic potential of targeting inflammation to combat cardiovascular diseases. Studies have identified dual roles of apolipoprotein E (apoE) and apoE receptors in modulating plasma lipids and inflammation, and have implicated these proteins as key players in atherogenesis. Moreover, genome-wide association studies have linked polymorphisms in the genes encoding these proteins with altered cardiovascular disease risk in humans. ApoE was first identified as a component of circulating lipoproteins, and the importance for ligand-receptor interactions between lipoprotein-associated apoE and hepatic endocytic receptors for maintaining healthy cholesterol levels has been since well established. However, the roles of apoE and apoE receptors expressed in other tissues and their involvement in the pathogenesis of atherosclerosis have not been fully elucidated. The goals of this dissertation research are to (1) investigate the impact of human APOE polymorphisms on macrophage-driven mechanisms of atherosclerosis, and (2) evaluate the effect of global and immune-cell specific expression of a novel mutation of the apoE receptor, LRP1, on development of metabolic disease and atherosclerosis.The functions of macrophage expression of the different human apoE isoforms – apoE2, apoE3, and apoE4 – in modulating atherogenesis was assessed by transplantation of donor bone marrow expressing the apoE isoforms into recipient ApoE-/- mice. Despite the similar plasma cholesterol levels between groups, we observed significant reduction in atherosclerosis in the mice receiving apoE3-expressing bone marrow, but not apoE2- or apoE4-expressing bone marrow, indicating that apoE3 is anti-atherogenic, whereas apoE2 and apoE4 are pro-atherogenic in immune cells. Further experiments revealed differential mechanisms by which apoE2 and apoE4 promote macrophage inflammation. ApoE2 increased intracellular cholesterol accumulation in myeloid cells, which was associated with enhanced myelopoiesis and inflammasome-mediated inflammation. Conversely, apoE4 increased inflammatory signaling pathways by promoting oxidative stress. Taken together, these results showed that APOE polymorphisms can influence atherosclerosis risk independently of plasma cholesterol levels by increasing macrophage inflammation. This research also investigated the impact of mutation of the proximal dileucine motif of the cytoplasmic tail of LRP1 on development of high-fat diet-induced cardiometabolic disease. This mutation was hypothesized to cause LRP1 dysfunction in various tissues by impairing its association with endosomal recycling adaptor protein, SNX17. We showed that Ldlr-/- mice expressing the LRP1 dileucine mutation displayed increased hypercholesterolemia that resulted in accelerated atherosclerosis progression. Interestingly, the same phenotype was observed in Ldlr-/- mice transplanted with LRP1 dileucine mutant-expressing bone marrow, suggesting significant contribution of LRP1 dysfunction in immune cells to increasing atherosclerosis risk. This study was the first to demonstrate the functional role of the proximal dileucine motif for LRP1 anti-atherogenic functions in vivo, and may indicate the therapeutic potential of improving LRP1 function by targeting pathways that enhance its intracellular trafficking.

Plasma triacylglycerols (TG) are elevated in diabetes, metabolic syndrome, obesity, and dyslipidemia. Very-low density lipoprotein (VLDL) is the main plasma carrier of TG and the direct metabolic precursor of low-density lipoprotein (LDL), the main carrier of plasma cholesterol and the major causative risk factor for atherosclerosis. Binding of LDL to heparan sulfate on the arterial wall initiates retention and modifications of LDL in the arterial intima, triggering atherosclerosis. Studies presented in this dissertation show that variations in TG levels and lipoprotein binding to heparin, a model for heparan sulfate, alter the structural and biochemical stability of VLDL and LDL, and increase their atherogenic potential. The molecular consequences of variations in the lipoprotein TG content and LDL-heparin binding were determined by combining heparin affinity chromatography with biochemical, spectroscopic and electron microscopic techniques. Remodeling of human VLDL and LDL by thermal denaturation was used to mimic key aspects of lipoprotein remodeling in vivo. Our studies revealed that increasing the TG content in VLDL promotes changes in the lipoprotein size and release of the exchangeable apolipoproteins. Similarly, increased TG content in LDL promotes lipoprotein remodeling and fusion. Additionally, an increase in TG content increases lipoprotein susceptibility to oxidation and lipolysis, thereby promoting the generation of free fatty acids that augment fusion. Consequently, TG-induced destabilization may be a general property of plasma lipoproteins. Our studies showed that binding to heparin initiates irreversible pro-atherogenic remodeling of human LDL. As a result of heparin binding, LDL showed decreased structural stability and increased susceptibility to hydrolysis and fusion. Further, phospholipid hydrolysis and/or glycation of LDL (as occurs in diabetes) increased the proteolytic susceptibility of apolipoprotein (apo)B (the major apolipoprotein of VLDL and LDL) and its heparin binding affinity. LDL derived from hyperglycemic patients with type-2 diabetes, became particularly destabilized following heparin binding causing apoB fragmentation and LDL fusion. In summary, binding to heparin alters apoB conformation and triggers pro-atherogenic LDL modifications including proteolysis, lipolysis and structural destabilization. Furthermore, phospholipid lipolysis and glycation of LDL in vitro strengthen its binding to heparin. Together, these findings help establish a mechanistic link between diabetes and atherosclerosis.

Abstract: Phosphatidylcholine (PC), the most abundant phospholipid in eukaryotic cells, is an important component of cellular membranes and lipoprotein particles. The enzyme CTP: phosphocholine cytidylyltransferase (CT) regulates de novo PC synthesis in response to changes in membrane lipid composition in all nucleated mammalian cells. The aim of this thesis was to determine the role that CTα plays in metabolic function and immune function in the murine intestinal epithelium. Mice with intestinal epithelial cell-specific deletion of CTα (CTαIKO mice) were generated. When fed a chow diet, CTαIKO mice showed normal lipid absorption after an oil gavage despite a ~30% decrease in small intestinal PC concentrations relative to control mice. These data suggest that biliary PC can fully support chylomicron output under these conditions. However, when acutely fed a high-fat diet, CTαIKO mice showed impaired intestinal fatty acid and cholesterol uptake from the intestinal lumen into enterocytes, resulting in lower postprandial plasma triglyceride concentrations. Impaired intestinal fatty acid uptake in CTαIKO mice was linked to disruption of intestinal membrane lipid transporters (Cd36, Slc27a4 and Npc1l1) and higher postprandial plasma Glucagon-like Peptide 1 and Peptide YY. Unexpectedly, there was a shift in expression of bile acid transporters to the proximal small intestine of CTαIKO mice, which was associated with enhanced biliary bile acid, PC and cholesterol output relative to control mice. Gene expression profiling of small intestinal epithelial cells showed induction of transcripts linked to cellular proliferation and inflammation in CTαIKO mice relative to control mice. Colonic inflammation after loss of intestinal epithelial cell CTα was linked to increased intestinal permeability (as assessed by Fluorescein Isothiocyanate-Dextran gavage), invasion of the intestinal epithelium by microbes, and enhanced inflammatory cytokine secretion. Impaired intestinal barrier function in CTαIKO mice was mechanistically linked to induction of endoplasmic reticulum stress and depletion of goblet cells. Antibiotics and 4-phenylbutyric acid both partially ameliorated inflammatory cytokine secretion in CTαIKO mice, suggesting that microbes and endoplasmic reticulum stress are key drivers of the inflammatory phenotype in CTαIKO mice. In further support of a role for de novo PC synthesis in colonic barrier function, feeding C57BL/6J mice a choline-deficient diet increased their susceptibility to Citrobacter rodentium-induced colitis relative to mice fed sufficient dietary choline.In conclusion, de novo PC synthesis in the small intestinal epithelium is required for dietary lipid absorption under certain dietary conditions, and the re-acylation of biliary lyso-PC cannot compensate for loss of CTα under these conditions. Furthermore, CTα activity prevents invasion of the intestinal epithelium by microbes. Accordingly, disruption of CTα in intestinal epithelial cells induces spontaneous colitis in mice. Finally, an adequate supply of dietary choline, the essential nutritional substrate for CTα, is an important factor in protecting the colon from Citrobacter rodentium -induced inflammation.

The glycoprotein CD36 is expressed in the plasma membrane (PM) of many cell types that surround or contact arteries, including macrophages, myocytes, and endothelial cells. CD36 binds oxidized low density lipoprotein (oxLDL), which promotes atherosclerosis, and fatty acids (FA), which promotes their cellular uptake. To gain insights into the molecular mechanisms of uptake, HEK293 cells expressing CD36 were studied by cell biological and fluorescence methods. To test our hypothesis that the PM is not an impermeable barrier to FA and that FA move into cells by diffusion via their uncharged form, we first applied biophysical fluorescence spectroscopy to directly measure transmembrane FA movement and membrane fluidity. Expression of CD36 in HEK293 cells did not increase either transport across the PM or the fluidity of the PM compared to HEK293 cells without CD36; however, CD36 enhanced intracellular FA esterification. Furthermore, the widely used “inhibitors” of FA transport did not alter either the rapid FA transmembrane diffusion in HEK293 cells or diffusion in control experiments with protein-free phospholipid bilayers. To gain new insights into the physiological relevance of FA binding to CD36, we applied surface plasmon resonance (SPR) to quantify FA and oxLDL binding to the ectodomain of CD36. Structurally distinct FA [saturated, monounsaturated (cis and trans), polyunsaturated, ω-3, ω-6, and oxidized FA] were pulsed in a solubilized form (bound to methyl-β-cyclodextrin) across SPR channels, generating real-time association and dissociation binding curves. With the exception of the oxidized FA hydroxyoctadecadienoic acid (HODE), all FA tested bound to CD36 with rapid association and dissociation kinetics similar to human serum albumin. In addition, FA increased oxLDL binding to CD36. To investigate whether FA affect CD36-mediated oxLDL uptake in live cells, we monitored fluorescent oxLDL (Dii-oxLDL) uptake using confocal microscopy. Addition of exogenous FA to serum-free media enhanced dose-dependent oxLDL uptake. Exceptions were ω-3 FA, which bound to CD36, and HODE, which did not bind to CD36, demonstrating FA structure-specific effects on a major function of CD36 and a new mechanistic link between atherosclerosis and high levels of FA in obese and Type-II diabetic individuals.

Successful development of effective therapeutics for Alzheimer’s disease (AD) requires a deep understanding of the mechanisms of disease pathogenesis. Recent human induced pluripotent stem cell (hiPSC)-derived models have shown it is possible to recapitulate the complex genetic diversity of a patient and more completely model AD pathology. While most work has focused on neuronal AD phenotypes, there is mounting evidence that failure to regulate cholesterol homeostasis by neighboring nonneuronal support cells may be involved in AD pathogenesis. Amyloid precursor protein (APP) or multiple fragments generated by APP proteolytic processing have previously been implicated in the regulation of cholesterol metabolism. However, the physiological function of APP in regulating lipoprotein homeostasis in astrocytes, which are responsible for the brain’s de novo cholesterol biosynthesis and regulation, remains unclear. To address this, we independently validated and characterized a novel in vitro differentiation protocol to generate human astrocytes from hiPSC. We found that cells derived by this method recapitulated native astrocytes in their form and function. We also utilized CRISPR/Cas9 genome editing to generate isogenic APP knockout (KO) hiPSCs. We found that APP KO astrocytes have reduced cholesterol, elevated levels of sterol regulatory element-binding protein (SREBP)-target gene transcripts, and increased low-density lipoprotein (LDL) receptor protein, all downstream consequences of reduced lipoprotein endocytosis. To elucidate which APP fragments regulate cholesterol homeostasis and examine whether familial AD mutations in APP affect lipoprotein metabolism, we analyzed an isogenic allelic series using the APP Swedish and APP V717F mutations. Only astrocytes homozygous for the APP Swedish (APPSwe/Swe) mutation, which had reduced full-length APP (FL APP) due to aggressive beta-secretase cleavage, recapitulated the APP KO phenotypes. Further, astrocytic internalization of amyloid-beta (A-beta), another ligand of LDL receptors, was impaired in APP KO and APPSwe/Swe astrocytes. Given that FL APP is known to bind to multiple LDL receptors, we propose that FL APP acts as a co-receptor for known LDL receptor ligands and is required for proper cholesterol homeostasis and A-beta clearance in human astrocytes.

Mycobacterium tuberculosis, the cause of tuberculosis, survives for long periods in a latent state in infected individuals, and the immune system is typically able to control but not eliminate the bacteria. Latency is a complex phenomenon but involves, in part, interactions of the bacteria and its unique lipoproteins and lipoglycans with macrophages, the main cells that become infected. The purpose of this dissertation was to expand understanding of how M. tuberculosis engages with macrophages. In the first part, we characterized the lipoglycan binding function of M. tuberculosis lipoprotein LprG. We determined the binding properties of these M. tuberculosis lipoglycans to LprG using surface plasmon resonance. We also verified the presence of a non-acyl chain dependent binding mode to LprG, and determined that LprG also binds mannan. Finally, we determined that one function of LprG is to facilitate exposure of LAM on the bacterial cell surface for interaction with macrophages. LprG-deficient M. tuberculosis had reduced surface-exposed lipoarabinomannan, and had reduced ability to block phagolysosome maturation, a known immune evasion mechanism that requires lipoarabinomannan. These studies contribute to understanding of LprG, and develop increased knowledge of how M. tuberculosis lipoarabinomannan is exposed to macrophages to block phagolysosome fusion, a process involved in bacterial persistence and intracellular survival.In the second part, we studied the TLR2 signaling response of macrophages to M. tuberculosis. We determined that TLR2 was required for M. tuberculosis to trigger NF-κB and ERK, and that TLR2 signaling results in balanced downstream effects. NF-κB is required for expression of pro-inflammatory IL-12, and M. tuberculosis-stimulated Tpl2-ERK signaling suppressed IL-12 while inducing anti-inflammatory IL-10. These effects reduced CD4+ T cell responses against M. tuberculosis. Tpl2-deficient macrophages expressed IL-12 in response to M. tuberculosis, and were more potent at stimulating antigen-specific T cells, upon initial stimulation and recall. These findings contribute to understanding of the signaling triggered by M. tuberculosis, and the role of the macrophage-intrinsic ERK cascade in inhibiting T cell-mediated host defense. Together, these studies expand understanding of the regulation of macrophages by M. tuberculosis in ways that promote long-term survival of the bacteria, and may potentiate latent infection.

Atherosclerosis, hardening and narrowing of the arteries, is the principal underlying cause of heart attacks, strokes, and peripheral vascular disease, which kills more than 600,000 Americans each year. High plasma levels of low-density lipoproteins (LDL) are linked to the formation of atherosclerotic plaques in arteries. LDL is the last metabolic product of very low-density lipoprotein (VLDL), which is secreted from the liver along with one molecule of apolipoprotein B (apoB). Current therapies to control levels of LDL include: cholesterol synthesis inhibitors or statins, low-fat diets and antisense oligonucleotides to reduce cholesterol levels. Recent studies recommend lower clinical levels of plasma LDL to maintain an individual’s health, especially of those who have already developed atheroscle- rotic plaques. However, existing therapies are often unable to achieve these aggressive limits. Furthermore, patients have shown various levels of intolerance to these treatments. In order to develop new, targeted drugs, that can control LDL levels with minimal side effects, it is imperative to understand, in detail, the process of apoB-containing lipoprotein formation. ApoB is one of the largest human proteins known (4563 residues) and previous attempts to solve the structure have been unsuccessful, mainly due to analyzing the protein as a whole or by large sections. To advance the field we will go by a different approach. I present here a construct that represents roughly 8% of the whole protein, apoB9-17 (residues 430 to 782). This section of the protein is believed to play a pivotal role in the assembly process of LDL. My hypothesis is that this construct will be well-behaved and suitable for structural and functional analysis. The study shows that apoB9-17 can be produced in considerable quantities from bacterial cells and can be purified by means of a 6-histidine tag with a good yield. Furthermore, circular dichroism analysis shows the construct contains the expected secondary structure at room temperature and is stable at a wide temperature range (50 to 70 ◦C) at low concentrations. The construct here described will be useful to test the effect of mutations such as the one found in patients with Familial hypobetalipoproteinemia (FHBL). Furthermore, this construct contains two regions believed to be of vital importance for LDL particle formation: the alpha-helical region (residues 430 to 570) is believed to associate with MTP at the initial stages of LDL formation and the c-sheet (residues 614 to 782), which may form part of the lipid recruiting process. Both essential aspects to ultimately develop therapies that can modulate VLDL particle formation.

Fusobacterium nucleatum is a Gram-negative bacterium that colonizes the oral cavity as well as other sites in the human body. It is known for its ability to adhere to a large number of different bacterial species and serving as a "bridging" organism. It plays an important role in the formation and maintenance of mature oral biofilms. At the same time it has been implicated in a number of important oral and systemic diseases and is considered an opportunistic pathogen. Very little is known about the outer membrane proteins of F. nucleatum and the molecular mechanisms of its adhesion. There is a large number of predicted membrane proteins that are expected to localize to either of the two membranes surrounding the cell. A number of lipoproteins are predicted from sequence searches, yet none are currently characterized on a molecular level. We have identified and characterized a novel hypothetical protein in F. nucleatum encoded by the ORF FN1253. The gene was shown to play a role in interspecies interactions with oral streptococci and was named aid1 (Adhesion Inducing Determinant 1). Aid1 is important in modulation of adhesion to oral streptococci, as aberrant expression of the aid1 gene leads to an uncharacteristic development of the dual species biofilm, which ultimately could lead to changes in the overall multispecies biofilm formation. The functional domains of the protein were also analyzed, and we showed that this small lipoprotein localizes to the inner leaflet of the outer membrane and requires the presence of its signal sequence as well as the anchoring cysteine residue for proper processing and maturation of the final protein product. Aberrant processing of the protein led to rapid turnover inside the cytosol, as the accumulation of the unprocessed protein appeared to be toxic to the cells. Expression of the aid1 gene and the Aid1 protein production is tightly regulated, particularly in the presence of oral streptococci. We showed that the gene can be transcribed as two messenger RNA species, and only the longer polycistronic messenger is used for transcription. To the best of our knowledge, this is the first molecular characterization of a fusobacterial lipoprotein, and the first hypothetical protein that is implicated in interspecies interactions within the oral cavity. This work also describes an interesting translational and transcriptional regulatory mechanism utilized by F. nucleatum, and is the first study to describe the regulation of a lipoprotein in this organism.

Phospholipids (PLs) are important structural components of biological membranes and precursors of numerous signaling molecules. The fatty acyl composition of PLs determines the biophysical characteristics of membranes. Multiple lines of evidence demonstrated that changes in fatty acyl composition could potentially affect the properties of proteins associated with membranes and influence the biological processes that occur on membranes. However, there is little understanding of how regulatory pathways control PL fatty acyl composition in vivo or how such regulation dictates physiological responses. In this work, we investigated the regulation of membrane fatty acyl composition by the Liver X Receptor (LXR)-Lysophosphatidylcholine Acyltranferase 3 (Lpcat3) pathway and its physiological or pathological relevance in lipid homeostasis and metabolic diseases.In chapter 2, we define a nuclear receptor pathway for the dynamic modulation of membrane composition in response to changes in cellular lipid metabolism. Ligand activation of LXR preferentially drives the incorporation of polyunsaturated fatty acids into PLs through induction of the remodeling enzyme Lpcat3. Promotion of Lpcat3 activity decreases endoplasmic reticulum (ER) stress induced by saturated free fatty acids in vitro or by obesity and hepatic lipid accumulation in vivo. Conversely, Lpcat3 knockdown in liver exacerbates ER stress and inflammation. Mechanistically, Lpcat3 modulates inflammation both by regulating c-Src (Proto-oncogene tyrosine-protein kinase Src) and JNK (c-Jun N-terminal kinases) activation through changes in membrane composition and by affecting substrate availability for inflammatory mediator production. These results outline an endogenous mechanism for the preservation of membrane homeostasis during lipid stress and identify Lpcat3 as an important mediator of LXRs effects on metabolism.In chapter 3, we show that Lpcat3 is a critical determinant of triglyceride secretion due to its unique ability to catalyze the incorporation of arachidonate into membranes. Mice lacking Lpcat3 in the intestine fail to thrive during weaning and exhibit massive enterocyte lipid accumulation and reduced plasma triglycerides. Mice lacking Lpcat3 in the liver show reduced plasma triglycerides, hepatosteatosis, and secrete lipid-poor very low density lipoprotein (VLDL) lacking arachidonoyl PLs. Mechanistic studies indicate that Lpcat3 activity controls membrane lipid mobility in living cells, suggesting a biophysical basis for the requirement of arachidonoyl PLs in lipidating lipoprotein particles. These data identify Lpcat3 as a key factor in lipoprotein production and illustrate how manipulation of membrane composition can be used as a regulatory mechanism to control metabolic pathways.

High density lipoprotein (HDL), low density lipoprotein (LDL), and very low density lipoprotein (VLDL) are lipoproteins previously shown to bind many basic and neutral hydrophobic drugs in serum. These interactions impact the distribution, delivery, metabolism, and excretion of drugs and are important in determining drug activity, pharmacokinetics, and toxicity in the human body. Information about drug-lipoprotein interactions and the strength of these interactions can be useful in determining the distribution of drugs following administration. The research presented in this dissertation uses high performance affinity chromatography (HPAC) and packed columns to study binding of the drug propranolol to immobilized lipoproteins such as HDL, LDL, and VLDL. Through these studies, two types of interactions were identified between the lipoproteins and propranolol and verapamil. The first interaction has a relatively high affinity and likely involves binding of the drug by surface apolipoproteins. This high-affinity saturable interaction was stereoselective for LDL. HDL and VLDL did not exhibit stereoselectivity. The second type of interaction observed in each lipoprotein had a lower affinity involved partitioning of the drug into the non-polar core of lipoproteins. Additional work analyzing the theory and experimental conditions needed for the detection of multiple binding mechanisms in HPAC columns when using frontal analysis is also presented. This work focuses on the evaluation of binding models that incorporated both a saturable type of binding and a non-saturable interaction. These evaluations make it possible to determine the experimental conditions that would be required for detection of this type of multi-mode interaction. These studies demonstrate that HPAC is a useful tool in characterizing mixed-mode interactions, as can occur with complex particles like lipoproteins. The affinity columns containing immobilized lipoproteins allowed these studies to be conducted using the same column for hundreds of experiments with short analysis times. The combined result of these advantages was the ability to quickly obtain precise data over a variety of drug concentrations. The results of these experiments indicate that similar columns prepared with other lipoproteins or biological membranes can be used in similar HPAC binding studies. Advisor: David S. Hage

To overcome the membrane permeability barrier and low nutrient availability in the environment, Gram-negative organisms have evolved many mechanisms dedicated to the intake of nutrients. One such mechanism is the long-chain fatty acid uptake pathway. This pathway involves machinery that transports fatty acids across the outer membrane and into the cell, where the lipid can be used for either nutrition or for remodeling the structure of the membrane. Interestingly, the fatty acid species that can be recognized by this machinery differ between organisms; the aquatic pathogen Vibrio cholerae demonstrates a much wider substrate recognition profile than other Gram-negative species. In this work we elaborate on the lipid nutrients accessible to V. cholerae, demonstrating that it can use lysophosphatidylcholine as both a carbon source as well as a source of fatty acids for remodeling its outer membrane. We identify the enzyme responsible for the breakdown of lysophosphatidylcholine, VolA (Vibrio outer membrane lysophospholipase A). VolA is conserved in many Vibrio species as well as other aquatic Gram-negatives, annotated as a putative lipase. We show VolA is co-expressed with the outer membrane fatty acid transporter FadL; FadL serves to transport across the outer membrane those long-chain fatty acids freed from lysophosphatidylcholine by VolA. VolA is expressed on the surface of the cell as a membrane anchored lipoprotein; this is novel as VolA is the first identified example of a surface-localized lipoprotein phospholipase. Biochemical characterization of VolA shows that it acts as a canonical lysophospholipase in vitro, suggesting that it works in tandem with the FadL transporter, freeing fatty acids from lysophosphatidylcholine at the surface of the cell to be brought in via the fatty acid uptake pathway. This work expands on the currently understood lipid uptake abilities of Vibrio cholerae, demonstrating a novel mechanism for utilizing a nutrient not previously thought to metabolized. VolA is an important to our understanding of the larger picture of lipid uptake and how it contributes to the survival of Gram-negative organisms.

The outer membrane (OM) is the first line of defense for Gram-negative bacteria against the exterior environment, as it forms an impermeable lipid barrier to protect the cell from antibiotics and other stresses. This impermeability is due to the fact that the OM is an asymmetric lipid bilayer, consisting of outer leaflet lipopolysaccharides and inner leaflet phospholipids. Additionally, there are many proteins unique to the OM, mainly the Outer Membrane Porins (OMPs) and OM lipoproteins. The cell utilizes dedicated machines to assemble each of the proteins and lipids into this complex folding environment. As OM integrity is vital, the cell needs to have complex surveillance systems to monitor folding and assembly of all OM components. In E. coli and related bacteria, homeostasis of the OM is monitored by the essential transcription factor, σE, which regulates OM assembly by transcribing a suite of ~100 protein encoding genes and two small, regulatory RNAs (sRNAs).σE activity is controlled by the degradation rate of its anti–σ, RseA, which holds σE inactive in the inner membrane. σE is further regulated by RseB, which binds to RseA and protects RseA from proteolytic cleavage. While mechanisms that can cause RseA degradation are known, the in vivo signal that relieves the inhibitory effect of RseB was not understood.In this thesis, I set out to understand the role of RseB in the σE signaling system, and to identify the signal that inactivated RseB in vivo. I found that the presence of RseB significantly inhibited the dynamic range of σE activation, such that σE is activated only when signals to activate RseA cleavage and inhibit RseB were both present in the cell. I next discovered that RseB is inactivated by lipopolysaccharde, thus only in the presence of two concomittant sources of envelope stress is σE maximally activated. As the σE, RseA, and RseB operon is conserved in genomes with σE, we hoped that our findings regarding the role of RseB in Escherichia coli would be a broadly applicable paradigm for how Gram-negative bacteria regulate OM homeostasis.I found it curious that the assembly machines for the lipoproteins of the OM were not represented in the σE regulon, even though this regulon encodes genes involved in every step of lipopolysaccharide and OMP transport and assembly. As lipoproteins are the major proteins of the OM, I set out to determine whether σE senses lipoprotein folding stresses and whether σE had additional mechanisms to regulate lipoprotein folding. I discovered a new σE–dependent sRNA (MicL) that decreases translation of the lipoprotein Lpp, the most abundant protein in the cell. By regulating Lpp translation through MicL, σE is able to regulate a major fraction of cellular translation and control the majority of protein flux destined to the OM. Additionally, misaccumulation of Lpp induces the σE response and proper regulation of Lpp plays an important role in envelope homeostasis.

As obesity rates increase in developed nations, the prevalence of other obesity-related metabolic disorders such as hyperlipidemia, atherosclerosis, and non-alcoholic fatty liver disease (NAFLD) has also increased drastically. A greater understanding of the pathobiology and molecular mechanisms of these disorders becomes increasingly relevant. Recent studies demonstrated that certain alleles of the pancreatic group 1B phospholipase A2 (PLA2G1B) are associated with increased obesity and that Pla2g1b-deficient (Pla2g1b-/-) mice are protected against diet-induced obesity and diabetes. Furthermore, supplying mice with lysophosphatidylcholine (LPC), the enzymatic product of Pla2g1b, causes hyperglycemia and insulin resistance, which are risk factors for hyperlipidemia, atherosclerosis, and NAFLD. Thus, Pla2g1b may play a role in the development of these obesity-related diseases and inhibiting Pla2g1b and/or LPC may change the progression of these metabolic disorders. The goal of the first part of this dissertation was to identify a Pla2g1b-mediated protection against hyperlipidemia. Findings from these studies indicate that Pla2g1b-/- mice are protected against high fat diet-induced hyperlipidemia and gain less weight than Pla2g1b+/+ controls due to decreased very-low-density lipoprotein (VLDL) production and increased postprandial triglyceride-rich lipoprotein clearance. Further results indicate that supplying LPC stimulates VLDL production in both Pla2g1b-/- and Pla2g1b+/+ mice and that LPC is directly used as a substrate to make VLDL-triglyceride (TG). Taken together, these findings indicate that inhibition of luminal Pla2g1b may be a viable strategy to decrease hyperlipidemia in vivo.The goal of the second part of this dissertation was to identify a mechanism of LPC-mediated inhibition of hepatic oxidative function. Pla2g1b-/- mice have decreased plasma LPC and increased hepatic oxidation that is inhibited by intraperitoneal injection of LPC by an unknown mechanism. Findings indicate that though low micromolar concentrations of LPC decrease the mitochondrial membrane potential, oxidation rate remains equal to controls up to incubation with 80 µM LPC. However, when isolated mitochondria or hepatocytes are supplied with 100 µM LPC, decreased substrate-stimulated oxidation and increased mitochondrial permeability are observed. These findings indicate that LPC plays a major role in the maintenance of hepatic mitochondrial integrity and function and that low micromolar changes in intra- and extracellular LPC concentration can greatly affect hepatic oxidation rate.The goal of the third part of this dissertation was to identify a Pla2g1b-mediated protection against atherosclerosis and NAFLD. Findings indicate decreased obesity, VLDL-TG, VLDL-cholesterol, low-density lipoprotein-cholesterol, and hepatomegaly in Pla2g1b-/-/Ldlr-/- mice compared to Pla2g1b+/+/Ldlr-/- controls. Findings also suggested decreased atherosclerotic lesion size in Pla2g1b-/-/Ldlr-/- mice.Taken together, the findings from this dissertation have three implications in the field of metabolic research. First, these studies indicate a vital role for Pla2g1b and LPC in modulating hepatic function. Regardless of the Pla2g1b status of experimental animals, the mechanisms of LPC action appear intact. Secondly, these studies indicate that inhibition of Pla2g1b provides protection against diet-induced obesity in two diverse murine models. Thirdly, decreased VLDL-TG is described as a hallmark of inhibition of Pla2g1b. The development of a single drug to inhibit the action of PLA2G1B would likely lead to improvements in obesity, diabetes, hyperlipidemia, and potentially atherosclerosis.

Amongst its numerous functions, the liver is responsible for the synthesis and secretion of very low-density lipoprotein (VLDL). VLDL particles play the important role of facilitating the transport of lipids within the aqueous environment of the plasma; yet high plasma concentrations of these particles result in the pathogenesis of atherosclerosis, while low VLDL secretion from the liver results in hepatic steatosis. VLDL synthesis in the hepatocyte is completed in the Golgi apparatus, which serves as the final site of VLDL maturation prior to its secretion to the bloodstream. The mechanism by which VLDL’s targeted transport to the plasma membrane is facilitated has yet to be identified. Our lab has identified this entity. Our findings suggest that upon maturation, VLDL is directed to the plasma membrane through a novel trafficking vesicle, the Post-Golgi VLDL Transport Vesicle (PG-VTV). PG-VTVs containing [3H] radiolabeled VLDL were generated in a cell-free in vitro budding assay for study. First, the fusogenic capabilities of PG-VTVs were established. Vesicles were capable of fusing with the plasma membrane and delivering the VLDL cargo for secretion in a vectorial manner. The next goal of our study is to characterize key regulatory molecular entities necessary for PG-VTV biosynthesis. A detailed analysis was undertaken to determine the PG-VTV proteome via western blot and two-dimensional difference in gel electrophoresis. The identification of key molecular regulators will potentially offer therapeutic targets to control VLDL secretion to the bloodstream.

The TonB system couples cytoplasmic membrane-derived proton motive force energy to drive ferric siderophore transport across the outer membrane of Gram-negative bacteria. While much effort has focused on this process, how energy is harnessed to provide for transport of ligands remains unknown. Several bacterial viruses ("phage") are known to require the TonB system to irreversibly adsorb (i.e., establish infection) in the model organism Escherichia coli. One such phage is φ80, a "cousin" of the model temperate phage λ. Determining how φ80 is using the TonB system for infection should provide novel insights to the mechanisms of TonB-dependent processes. It had long been known that recombination between λ and φ80 results in a λ-like phage for whom TonB is now required; and this recombination involved the λ J gene, which encodes the tail-spike protein required for irreversible adsorption of λ to E. coli. Thus, we suspected that a φ80 homologue of the λ J gene product was responsible for the TonB dependence of φ80. While φ80 has long served as a tool for assaying TonB activity, it has not received the scrutiny afforded λ. Indeed, only a small portion of its genome (not including a J gene homolog) was known. To facilitate the use of φ80 as a molecular tool, we determined and annotated the full genome of the φ80vir strain commonly used for TonB studies. The 46,285-bp φ80vir genome contained 71 predicted open reading frames, in which the structural genes showed strong synteny with other known lambdoid viruses, with an overall degree of mosaicism indicative of multiple genetic exchanges with other viruses. There was also evidence of unique gene acquisition in the form of morons, one of which, termed cor, when cloned and expressed in E. coli, blocked TonB-dependent transport of ferric siderophore and irreversible adsorption of φ80vir to the TonB-dependent outer membrane protein FhuA. These and other findings reported in this dissertation provide a foundation for the use of φ80vir as a tool to dissect the mechanisms of TonB-dependent energy transduction.

High density lipoproteins (HDL) are complexes of phospholipid, cholesterol and protein that circulate in the blood. Epidemiological studies have demonstrated a strong inverse correlation between plasma levels of HDL associated cholesterol (HDL-C) and the incidence of cardiovascular disease (CVD). Clinically, HDL-C is often measured and used in combination with low density lipoprotein cholesterol (LDL-C) to assess overall cardiovascular health. HDL have been shown to possess a wide variety of functional attributes which likely contribute to this protection including anti-inflammatory and anti-oxidative properties and the ability to remove excess cholesterol from peripheral tissues and deliver it to the liver for excretion, a process known as reverse cholesterol transport. This functional diversity might be explained by the complexity of HDL composition. Recent studies have taken advantage of advances in mass spectrometry technologies to characterize the proteome of total HDL finding that over 50 different proteins can associate with these particles. This adds to a growing body of evidence that supports the global hypothesis of this thesis which is that the total pool of HDL in an individual is composed of numerous subspecies with distinct protein and lipid compositions and therefore will have distinct functional properties. Additionally, we believe that the composition of HDL is dynamic and can change in response to changes in the environment of the blood, as can occur in disease. To test these hypotheses we devised an approach based on three aims. Aim 1: Identify and characterize HDL subspecies based on protein composition. Aim 2: Analyze functional heterogeneity across separated plasma HDL fractions. Aim 3: Examine the effect of type 2 diabetes on HDL subspecies distribution in young adults. To accomplish these goals we have developed novel methods for the separation and fractionation of HDL subspecies from human plasma and their subsequent proteomic and functional analysis. Our findings support the existence of a diverse array of HDL species with distinct functionalities which correlate strongly with specific lipid or protein components of the particle. Additionally, we found that the composition of specific subfractions of HDL are altered in type 2 diabetes and that these fractions can predict vascular health better than LDL-C or HDL-C, suggesting that these may prove to be better biomarkers for cardiovascular risk than the current clinical standard. A detailed understanding of HDL subspeciation will be invaluable in the further development of HDL as a biomarker and therapeutic target not only for cardiovascular disease but a variety of disease states.

Toll-like receptor 2 (TLR2) agonists in the mycobacterial cell wall are important determinants of immune responses of Mycobacterium tuberculosis (Mtb). The Mtb lipoprotein LprG was found to associate with glycolipids, contributing to its TLR2 agonist activity. Isolation of LprG from mycobacteria suggests that LprG is able to bind multiple cell wall glycolipids, including lipomannan (LM), and lipoarabinomannan (LAM). We have some evidence that LprG prefers binding to LAM over LM. Experiments are ongoing to investigate the mechanism and kinetics of this preferential binding.The glycolipid chaperone role of LprG in the context of the whole bacterium TLR2 agonism was examined when individual lipoproteins were overexpressed. Overexpression of acylated LprG increased the TLR2 agonist activity of the whole bacterium; however, a non-acylated form of LprG does not increase TLR2 activity of the bacteria. We have found that LprG and its glycolipid cargo are major contributors to whole Mtb TLR2 agonist activity.

GfcC is a small (26 kDa) periplasmic protein encoded within the seven gene gfc operon comprised of gfcABCDE, etp and etk of E. coli. This operon in enteropathogenic strains is essential for the expression of a group 4 polysaccharide capsule whose oligosaccharide repeat units are identical to those in lipopolysaccharide . The GfcC structure shows striking similarities with the outer membrane lipoprotein Wza from group 1 polysaccharide export. Both have two β-grasp domains for each monomer and an amphipathic C-terminal helical region. GfcC and its homologs also have a 40-residue long helical hairpin inserted into the first β-grasp domain not found in Wza. The C-terminal helices of Wza form a novel octameric alpha-helical barrel in the outer membrane. This barrel most likely exports the polysaccharide from the periplasm where it is synthesized. Although GfcC has a similar helix, it is shorter and folded back onto a β-grasp domain and hence GfcC exists as a soluble monomer in solution. Since the group 4 operon also has a wza homolog, gfcE, the function of GfcC is not clear. The following gene encodes GfcD, predicted to be a large β-barrel outer membrane lipoprotein. Homologs of gfcC and gfcD are fused in several species, notably in Burkholderia sp., suggesting an interaction in vivo between the two encoded proteins. YraM, a 575-residue outer membrane lipoprotein essential for growth in H. influenzae, has two distinct domains, an N-terminal domain with tetratricopeptide repeats and a C-terminal periplasmic binding protein-like domain. The C-domain has conserved residues clustered in the cleft where other binding proteins bind small molecule ligands. The N-domain resembles other proteins that bind protein or peptidoglycan ligands. The two domains are arranged adjacent connected by a 6-residue linker. Normal Mode analysis (specifically Elastic network modeling) suggests flexibility in the overall conformation between the two domains with the linker acting as a hinge. The function of YraM is unknown but it represents a novel fusion of domains with two distinct binding capabilities –one a small molecule ligand (to the C-domain) and the other possibly a protein (to the N-domain), both binding partners are yet to be identified.

Coronary heart disease is the leading cause of death in the United States, and epidemiological studies have shown that an elevated plasma triglyceride level is an independent risk factor for cardiovascular disease. In 2001, a new member of the exchangeable apolipoprotein (apo) family, apoA-V, was discovered using comparative genomics. Studies with transgenic mice over-expressing human apoA-V and apoA-V knock-out mice have shown that there exists a negative correlation between apoA-V levels and plasma triglyceride concentrations. Such findings have also been observed in humans, where single nucleotide polymorphisms within the APOA5 gene have been associated with hypertriglyceridemia. These physiological findings suggest that apoA-V plays a major role in the regulation of triglyceride metabolism, and elucidating the mechanism by which apoA-V functions could yield therapeutic effects.Several features of apoA-V, including an extremely low plasma concentration, a lack of correlation with cholesterol levels despite its association with high density lipoprotein, and its insolubility at neutral pH in the absence of lipid, are unlike other exchangeable apolipoproteins. Findings indicate apoA-V is comprised of two independently folded structural domains, with the amino-terminal domain forming an amphipathic alpha-helix bundle in the absence of lipid. In the presence of lipid, the helix bundle motif is postulated to unfurl, exposing hydrophobic residues for lipid interaction. The carboxy-terminal region plays a key function in apoA-V lipid binding, consistent with its known association with plasma lipoproteins.

Abstract: Overproduction of intestinal chylomicrons (CM) has been proposed to contribute to fasting and post-prandial (PP) dyslipidemia and may accelerate the development of cardiovascular disease (CVD) during obesity, insulin resistance (IR) and diabetes. However, the impact of morphological changes in intestinal mucosa structure have not been investigated during IR and intestinal dyslipidemia. The first objective of this thesis was to characterize intestinal villi morphology and to determine whether a morphological relationship exists with enterocytic apoB48 (a marker of CM), and intestinal lymph secretion of apoB48 in the obese and IR JCR:LA-cp rat. The second objective was to assess the impact of n-3 PUFA supplementation on PP dyslipidemia in the JCR:LA-cp rat. Intestinal hypertrophy was observed in IR rats, corresponding to an increase in intestinal and lymphatic apoB48 expression. Further, a dietary intervention of n-3 PUFA showed lower PP plasma concentrations of apoB48 and PP plasma inflammatory markers. We conclude that intestinal hypertrophy may contribute to intestinal CM overproduction during obesity and IR. Additionally, dietary n-3 PUFA improves PP lipemia and the associated PP inflammatory response in the JCR:LA-cp rat model.

Mycobacterium tuberculosis has established infection in one third of the world population. Toll‐like receptor 2 is important for control of mycobacterial infection, and is important for numerous immunological processes. Toll‐like receptor 2 has an unusually diverse ligand repertoire, including lipoproteins, glycolipids, proteins, and glycans. The purpose of the studies described in this thesis was to better characterize the mechanisms of TLR2 activation by mycobacterial lipoproteins. In the first part, we examined co‐receptor and accessory receptor contribution to TLR2‐mediated detection of the mycobacterial lipoproteins LpqH, LprA, LprG, and PhoS1. We found that while most of the lipoproteins required TLR2, TLR1, and CD14, some did not require CD14, and LprA did not require TLR1. These results suggest that the protein component of bacterial lipoproteins contributes significantly to their ability to activate TLR2. Furthermore, in an effort to determine whether different TLR2 co‐ and accessory receptor requirements might impact mycobacterial infection, we analyzed murine lung antigen presenting cell subsets and found that expression of TLR2, CD14, and CD36 varied amongst them. This impacted their ability to respond to the mycobacterial lipoprotein LpqH. Lung macrophages expressed less TLR2 than alveolar macrophages or dendritic cells, and were significantly less responsive to LpqH than alveolar macrophages. A second focus of this thesis was to describe the structural motifs responsible for the ability of LprG to signal independent of its covalently‐attached acyl chains. Toward this end, we found that LprG carried glycolipids in a hydrophobic pocket and delivered them to TLR2 for recognition. A point mutation in the hydrophobic pocket abrogated glycolipid binding and TLR2 activation by the non‐acylated LprG. We found that while LprA could carry diacylated glycolipids, only LprG carried triacylated glycolipids, and that these glycolipids signaled in a TLR2 and TLR1 dependent manner. Furthermore, we found that LprG increased the potency of glycolipids for TLR2 activation by several orders of magnitude, and that the glycolipids carried in the pocket of acylated LprG explain the difference between LprA and LprG. We propose that LprG functions to carry mycobacterial glycolipids for the purpose of envelope construction. Together, the studies described in this thesis demonstrate that TLR2 detects its agonists via multiple mechanisms, and include the description of a novel recognition mechanism by which TLR2 co‐opts a bacterial glycolipid carrier as a ligand delivery system.

Ehrlichia chaffeensis, an obligatory intracellular bacterium of human monocytes-macrophages, causes human monocytic ehrlichiosis (HME). The bacterium has a biphasic developmental cycle in mammalian cells that alternates between a small ‘dense-cored cell’ (DC) and a large ‘reticulate cell’ (RC), as defined by morphological features, differential surface protein expression, and infectivity. However, how E. chaffeensis intracellular development is regulated is poorly understood. The type IV secretion (T4S) system is critical for the virulence of several pathogens. E. chaffeensis T4S apparatus, virB/D genes are split into two operons: virB3–virB6 (preceded by sodB) and virB8–virD4. Between these two operons, there are duplications of virB4, virB8, and virB9. We found that the transcription of all five loci was down-regulated prior to the release of E. chaffeensis from host THP-1 cells and up-regulated at the initiation of the exponential growth. An E. chaffeensis 12.3-kDa hypothetical protein, EcxR, specifically bound to the promoter regions upstream of virB/D loci. EcxR also activated transcription of all five virB/D loci in lacZ reporter constructs. The expression of ecxR was positively auto-regulated by EcxR. These results suggest that the five virB/D loci are coordinately regulated by EcxR to allow developmental stage–specific expression of the T4S system in E. chaffeensis. The two-component regulatory system (TCS) composed of a sensor histidine kinase and a response regulator, allows bacteria to sense signals and respond to changes in their environment through the activation or repression of specific genes. The genomes of E. chaffeensis and A. phagocytophilum were each predicted to encode three pairs of TCSs. All six genes encoding three histidine kinases and three response regulators were expressed in both E. chaffeensis and A. phagocytophilum cultured in human leukocytes. Pretreatment of host-cell-free E. chaffeensis or A. phagocytophilum with closantel, an inhibitor of histidine kinases, completely blocked the infection of host cells. Treatment of infected cells one day post infection with closantel cleared the infection in dose-dependent manner. Autokinase activities of the three recombinant histidine kinases from E. chaffeensis were inhibited by closantel in vitro. A number of E. chaffeensis genes, including the six TCS genes, were down-regulated within 5-60 min post closantel treatment. These results suggest that these TCSs play an essential role in the infection and survival of E. chaffeensis and A. phagocytophilum in human leukocytes. Caulobacter crescentus CtrA (cell cycle transcription regulator A) is a transcriptional regulator that allows the coordination of cell cycle progression and morphogenesis by controlling the expression levels of ~100 genes. In E. chaffeensis, CtrA mRNA and protein levels were down-regulated within 6 h post infection in synchronously infected cell culture, and significantly up-regulated at the late stage of infection prior to the bacterial release. C. crescentus CtrA-binding motif (CtrA box) was predicted in the upstream regions of more than 30 E. chaffeensis genes. Temporal mRNA expressions of these genes were divided into four patterns: 1) up-regulated at the late exponential phase, similar to CtrA, 2) up-regulated at the lag phase and down-regulated during the exponential phase, 3) up-regulated at the lag phase but not down-regulated during the exponential phase, 4) constitutively expressed. E. chaffeensis CtrA specifically bound to the promoter regions of E. chaffeensis ctrA and surE, and rCtrA transactivated surE promoter-lacZ reporter constructs. These results suggest that CtrA is a global regulator controlling the intracellular development of E. chaffeensis. In E. chaffeensis, lipoproteins are required for bacterial infection to the host cells. OmpA (outer membrane protein A, also known as a peptidoglycan-associated lipoprotein) was E. chaffeensis surface exposed. E. chaffeensis binding and infection to the host cells were inhibited by rabbit anti-OmpA IgG. OmpA mRNA and protein expressions were down-regulated during the lag phase and the early exponential phase, but highly up-regulated at the late exponential phase. CtrA-binding site was predicted in the promoter region of ompA gene. CtrA bound to this region and activated OmpA expression. The results suggest that CtrA regulates ompA expression, consequently, E. chaffeensis infection of host cells.

Mycobacterium tuberculosis (Mtb) infects one third of the global population, resulting in two million deaths annually. In most infected individuals the immune response is sufficient to control infection, though sterile immunity is rarely achieved and latent infection persists indefinitely due to mycobacterial ability to subvert the immune response on several levels. If the host immune system becomes weakened the infection will reactivate. In vitro studies have indicated that Mtb infection in macrophages down-regulates interferon-gamma (IFN-gamma)-induced major histocompatibility complex class II (MHC-II) antigen presentation, possibly accounting for one mechanism by which mycobacteria can survive in the face of a T-cell response. In this work, these studies have been extended to in vivo analysis of lung antigen presenting cells (APCs). Using a strain of BCG which expresses GFP in a murine model of disease, we have identified infected macrophages and dendritic cells and determined their surface expression of MHC-II. We found that MHC-II levels are significantly lower on infected than on uninfected cells from the same lung, providing in vivo support for the ability of Mtb to down-regulate MHC-II. This phenomenon has been observed with both whole bacilli and purified cell wall Toll-like receptor (TLR) ligands, such as the lipoproteins LpqH and LprG. Although it is generally supposed that lipoproteins signal though TLR2 equivalently in strength and in an acyl-chain-dependent fashion, this has only been confirmed with a few proteins. In this work, we assess and compare the ability of a select group of immunodominant lipoproteins from mycobacteria to signal through TLR2 and also determine whether they require acyl chains or if they have other structural determinants of signaling. The data discussed here mostly concerns the closely related lipoproteins LprA and LprG. We found that while both LprA and LprG are TLR2 agonists, LprG is stronger and can signal through TLR2 without its acyl chains while LprA requires them. The TLR2 signaling ability of non-acylated LprG may be due to another molecule within a hydrophobic pocket in the protein. These data characterize novel mycobacterial TLR2 ligands and provide insight into the possible function of LprG, a protein that is both important to the bacterium and a major modulator of the host immune response.

Escherichia coli assembles functional amyloid fibers called curli. Produced by many Enterobacteriaceae spp., curli fibers are associated with biofilm formation, host cell adhesion and invasion, and immune system activation. CsgB nucleates the major curli subunit protein, CsgA, into a self-propagating amyloid fiber on the cell surface. CsgA and CsgB cell surface association and subsequent fiber polymerization are efficient processes, yet little is known about how the subunits reach and become positioned on the surface. I have characterized the role of CsgG in curli subunit secretion across the outer membrane, and found that CsgG is the central molecule of the curli secretion and assembly complex. CsgG formed an oligomeric complex in the outer membrane that interacted with at least three other csg encoded proteins. I found that the CsgG lipoprotein spanned the outer membrane, was sufficient for curlin transport across the outer membrane, and that a specific signal peptide on the N-terminus of the major fiber subunit directed protein secretion via the CsgG secretion apparatus. I also discovered that CsgG was non-uniformly distributed on the surface of curli-producing cells, and that the assembly, spatial organization, and secretion activity of CsgG was modified by other components of the curli assembly machine. One of the other curli assembly proteins, CsgE, was found to specifically function in the secretion pathway and to modulate both the secretion activity of CsgG and to act directly on CsgA to prevent self-polymerization. These results suggest a model where secretion and fiber assembly are tightly coupled and highly ordered processes.

There is abundant evidence that the subclasses within a given lipoprotein class differ in their associations with coronary heart disease. Since subclass distributions can vary widely from person to person, individuals with the same levels of LDL cholesterol and HDL cholesterol may be at different cardiovascular risk and respond differently to dietary and drug therapy. Unfortunately, existing laboratory methods of subclass measurement are too time-consuming and expensive to be used in routine clinical practice. Using a new approach to lipoprotein analysis that exploits the natural proton NMR spectroscopic differences exhibited by lipoprotein particles of different size, we have developed a new quantitative NMR technology for use in clinical laboratory medicine. The newly developed NMR LipoProfile assay rapidly and simultaneously quantifies the lipoprotein subclass particle concentrations of 10 lipoprotein species (3 VLDL, IDL, 3 LDL and 3 HDL) with good intraassay and interassay precision. Extensive validation studies were conducted that established robustness of the NMR lipoprotein particle assay. The average particle sizes of the major lipoprotein classes determined by NMR correlate very well with those estimated by gradient gel electrophoresis. Emerging clinical data from several coronary disease outcome studies indicate that NMR-derived lipoprotein particle parameters are superior predictors of cardiovascular disease risk compared to traditional cholesterol risk factors. The speed and efficiency of NMR lipoprotein subclass profiling make it a potentially valuable research tool and cost-effective means of assessing and managing heart disease risk in the general population.

The effects of vitamin E supplementation during late gestation were evaluated in dairy or beef cattle in three experiments. Dairy cows were injected with vitamin E and selenium (Se) in one study; beef cows were offered oral vitamin E supplementation via a free-choice vitamin-mineral mix in two studies. Breed-related effects were also evaluated. Jerseys had higher blood Se and lower serum vitamin E concentrations than Holsteins at dry-off and higher blood Se concentrations than Holsteins 3-4 weeks pre-calving and at calving. Selenium supplementation increased blood Se concentrations at calving. Treatment did not affect serum vitamin E concentrations at calving or post-calving, nor blood Se concentrations post-calving. Beef cattle consuming supplemental vitamin E (treatment) had greater responses to treatment when calving in late winter than when calving in late summer. Treated multiparous cows calving in winter had increased serum and colostral vitamin E concentrations but treatment did not affect colostral or serum immunoglobulin G (IgG) concentrations of their calves. Calves from treated, multiparous cows calving in winter had increased 205-day adjusted weaning weights (AWWs). Treatment did not affect colostral vitamin E or IgG concentrations of nulliparous heifers calving in winter, nor serum vitamin E or IgG concentrations, or AWWs of their calves. Six hundred IU supplemental vitamin E/head/day did not affect serum vitamin E concentrations of dams, colostral vitamin E or IgG concentrations, serum vitamin E or IgG concentrations, or growth of calves in the summer-calving herd. Breed-related differences in vitamin E concentrations and AWW occurred in both herds. Consumption of 600-1000 IU vitamin E/cow/day (treatment) during late gestation via a free-choice vitamin-mineral mix increased vitamin E concentrations in serum and the lipoprotein fraction containing no apolipoprotein B (non-ApoB) from 1-2 weeks pre-calving to calving. Treatment group calves had higher vitamin E concentrations in serum and non-ApoB than control group calves. Treatment did not affect cholesterol or phospholipid concentrations in serum or the lipoprotein fractions of either cows or calves. Treatment increased vitamin E cholesterol (VEC) and vitamin E phospholipid (VEPL) ratios in the serum and non-ApoB of cows and calves. Various breed-related differences also occurred.

Analytical ultracentrifugation has been employed to investigate the lipoprotein spectra of rabbits fed cholesterol and humans with type III hyperlipoproteinaemia. The results show considerable similarities between the lipoprotein spectra of the two, whether studied individually or after being averaged. Despite this similarity in spectra, the rabbit fed cholesterol was not considered to be a suitable experimental model for human type III hyperlipoproteinaemia as the lipoproteins probably have a different metabolic origin. On feeding cholesterol to rabbits, two peaks of lipoproteins were identified within the very low density lipoprotein (VLDL) region of Sf 12 to 400. The peak of smaller lipoproteins floated between Sf 12 and 60 and are a well known component of the serum of rabbits fed cholesterol. The larger lipoproteins, floating at Sf 270, have not been reported before. The greater part of this thesis is concerned with the study of the similarities of, and differences between, these two populations of lipoproteins. Rabbits were fed 0.8% cholesterol in their diet for two weeks and the two peaks of VLDL isolated from serum by preparative ultracentrifugation. The smaller lipoproteins had a peak Sf averaging 37, mean diameter of 36 nm, density of 1.00 g/ml, and their chemical composition agreed closely with previous reports. The larger particles had a peak Sf averaging 270, mean diameter of 80 nm, density of 0.97 g/ml and a high (80%) cholesterol ester content and low (4%) triglyceride content. The fatty acid composition of the cholesterol esters, phospholipids and triglycerides was similar in both fractions. It was proposed that the two particles were both remnant chylomicrons as all the differences seen between the two particles could be accounted for by their difference in size. They were named small (SCR) and large chylomicron remnants (LCR). Metabolic studies on the LCR and SCR using radioactive tracers confirmed that they both originated from the intestines, and indicated that the main site of catabolism of these particles was the liver. On injecting LCR and SCR, labelled in their cholesterol moiety, into recipient rabbits fed normal or cholesterol diets, it was observed that LCR disappeared from the plasma faster in the normal animals, but SCR was removed more rapidly in the rabbits fed cholesterol. An hypothesis was proposed to explain this difference in catabolism. The large number of samples that were subjected to analytical ultracentrifugation necessitated the efficient calculation of the results from this machine. As the manual method first set up for the job was found to be rather tedious and clumsy, a computer program was written. This meant that all necessary corrections could be incorporated and output could be in the form of a convenient and useful spectrum. Continuous development of this program throughout the duration of this work has produced a simple but useful and versatile program.

This thesis describes two approaches for studying of lipoprotein metabolism in sheep. The first approach involves the assay of lipoprotein lipase (LPL) activity to determine the role of lipoprotein-triacylglycerol fatty acids in fat deposition in sheep. This enzyme is the rate limiting enzyme in the hydrolysis of fatty acids from lipoprotein-triacylglycerol. The second approach was to characterize and quantify in vivo lipoprotein metabolism using iodinated very low density lipoprotein (¹²⁵I-VLDL) and low density lipoprotein (¹³¹I-LDL). Cross-bred lambs were divided into two treatment groups and either weaned early at 5 weeks of age or remained suckling. Lambs were slaughtered at 12 or 23 weeks at which time the body composition and adipose tissue LPL activity were determined. The differences in rearing led to differences in body composition. The suckled lambs were larger and fatter than weaned lambs. The increased fatness in the suckled lambs was associated with increased LPL activity (U/mg protein) in subcutaneous adipose tissue and was reflected in higher LPL activity in post-heparin plasma (PHP) taken 2 days prior to slaughter. The role of insulin in the regulation of LPL activity was investigated by either infusing a subset of the weaned and suckled lambs with insulin for 7 or 18 weeks or using the euglycemic clamp technique to study the effect of short insulin infusions. The long term infusion of insulin had no significant effect on PHP LPL or on adipose tissue LPL (U/g tissue). However, after infusing insulin for 6h at 6.3 mU.kg⁻·⁷⁵.h⁻¹ during the euglycemic clamps, a two fold increase in LPL activity in biopsied subcutaneous adipose tissue was observed. In the second approach, in vivo lipoprotein metabolism was investigated in 4 lambs using apolipoprotein B as a marker. Following the simultaneous injection of ¹²⁵I VLDL and ¹³¹I VLDL, the specific activities of apoB in VLDL, IDL and LDL fractions were determined. ApoB specific activity curves demonstrated that VLDL is metabolised to IDL and subsequently to LDL. The turnover of VLDL-B (3.45mg.d⁻¹.kg⁻¹) and LDL-B (4.8mg.d⁻¹.kg⁻¹) was calculated by fitting the VLDL-¹²⁵I-B and LDL-¹³¹I-B specific activity data to a mono-exponential equation. The metabolism of lipoproteins, inferred from the study of apoB, was shown to be similar in sheep to that reported in other animals although the amount of lipoprotein synthesised was low. A model to describe the kinetics of apoB metabolism in sheep was developed using SAAM. The proposed model features a three pool delipidation chain for VLDL, and subsystems containing two pools for IDL and LDL. IDL may be catabolised to LDL or cleared directly from the plasma. The developed model can now be used to compare the metabolism of lipoproteins in different physiological states and to design new experiments to study lipoprotein metabolism further.