HIV-infected individuals are at increased risk of age-related non-AIDS morbidity and mortality compared with HIV-uninfected persons. It is speculated that HIV-infected individuals may not only be aging chronologically, but also undergoing accelerated biological aging. This is supported by clinical reports of conditions classically associated with the normal aging process appearing at an earlier age in HIV-infected persons compared to age-matched controls. Chronological age is an imprecise measure of biological aging due to inter-individual differences in rates of aging and therefore ‘biomarkers of aging’ may be used to assess biological age. The eye may be a uniquely useful site as a model of aging. It is easily accessible for examination and several components can be measured and assessed objectively e.g. lens density, retinal vascular calibre, corneal endothelial cell counts and the retinal nerve fibre layer thickness. This case-control study of 504 adults recruited from one district in Cape Town, South Africa assessed whether HIV-infected individuals have more advanced ocular aging, systemic frailty and cellular senescence than an HIV-uninfected group of similar age. Accelerated biological aging was demonstrated in HIV-infected individuals compared to their uninfected counterparts. HIV infection was also associated with frailty. Ocular parameters provided evidence of greater aging within the HIV-infected group, particularly objective measurement of retinal vascular calibre and lens density. These data suggest that as well as increased biological aging at a cellular and systemic level, ocular aging occurs as part of the accelerated aging phenotype in HIV infection. This study provides novel data about accelerated biological aging in sub- Saharan Africa and a platform for addressing future research questions relating to accelerated aging trajectories in HIV infection, the relative contributions of the infection and antiretroviral therapy, and whether biological age is dependent upon the duration of untreated disease or nadir CD4 count. As the HIV-infected population continues to age and expand, accelerated biological aging may have wideranging implications for the burden and management of HIV-related morbidity.

The antiphospholipid syndrome (APS) is characterised by the presence of pathogenic antiphospholipid antibodies (aPL) that bind phospholipid via the N-terminal domain I (DI) of the protein cofactor beta2-glycoprotein I (beta2GPI). This thesis aims to investigate and identify the nature of this interaction at the molecular level. In order to study DI, a system for expressing and purifying sufficient quantities of this protein was developed using Escherichia coli (E.coli) as the expression host. This was achieved primarily by synthesising a gene with all codons optimised for E.coli, tight regulation of expression and recovery of DI from the periplasm to facilitate folding. Recombinant purified DI bound clinically relevant human monoclonal and polyclonal purified IgG aPL in both solid an fluid phase assays. Hypotheses were generated that identified candidate epitopes for aPL binding as potentially involving residues D8, D9, E23, E26, E27, R39, G40, R43 and the DI-II interlinker region. In total, 15 mutants of DI targeting these areas were created and computer modelling employed to predict the likely structural effects of these mutations upon DI. Expressed DI mutants were then tested in both solid and fluid phase assays for binding to polyclonal IgG derived from patients with APS and compared to wild-type DI. Some mutations, such as those targeting R39, caused loss of binding to aPL in the fluid phase whilst others caused enhanced binding over wild-type DI. In conclusion, this thesis demonstrates that DI of 62GPI may be expressed using E.coli and binds clinically relevant IgG aPL. Detailed mutational studies support the concept that aPL bind discontinuous epitopes on DI involving regions D8 and D9, R39 to R43 and the DI-II interlinker which are in close proximity to each other in the tertiary structure. The ability to produce a mutant of DI with enhanced binding over wild-type holds therapeutic potential.

Autontibodies to a wide variety of antigens are associated with systemic lupus erythematosus (SLE) and the Antiphospholipid Syndrome (APS). Previous studies have demonstrated the importance of somatic mutations and arginine residues in the complementarity determining regions (CDRs) of pathogenic anti-dsDNA and antiphospholipid antibodies. This thesis describes the study of two human monoclonal IgG antibodies, B3 (anti-DNA) and IS4 (antiphospholipid) that were derived from a patient with active SLE and primary APS respectively. I have demonstrated in-vitro expression and mutagenesis of B3 and IS4 and used this expression system to investigate the importance of the arginine residues in B3VH and IS4VH. The mutant heavy chains, as well as the wild-type VH were expressed with different light chains and the resulting antibodies assessed for binding to nucleosomes, alpha-actinin, cardiolipin (CL), phosphatidylserine (PS), beta-2-glycoprotein I (foGPI), and the N-terminal domain of p2GPI (Domain I) using direct binding assays. The results obtained have shown that the presence of arginine at position 53 in B3VH was essential but not sufficient for binding to dsDNA and nucleosomes. Conversely, the presence of this arginine reduced binding to alpha-actinin, pzGPI and Domain I of P2GPI. The fact that the arginine to serine substitution at position 53 in B3VH significantly alters binding of B3 to different clinically relevant antigens, but in opposite directions implies that this arginine residue plays a critical role in the affinity maturation of the antibody B3. Furthermore, of four arginine residues in IS4VH CDR3 substituted to serine, two at positions 100 and 100g reduced binding to all antigens, while two at positions 96 and 97 reduced binding to fcGPI but increased or decreased binding to CL and PS. Only one H/L chain combination bound neutral phospholipid and none bound dsDNA hence, these effects are particularly relevant to antigens important in APS. Therefore, my findings suggest that these four arginine residues have developed as a result of somatic mutations driven by an antigen containing both phospholipid and frGPI. These results extend our knowledge of the structure-function relationship of human anti-DNA and anti phospholipid antibodies and aid in our understanding of how these antibodies lead to pathogenicity and what we need to target in the future for possible therapies.

The human immunodeficiency virus type I (HIV-1) is a retrovirus that damages the human immune system, which is suppressed by a cellular factor APOBEC3G in non-permissive cells. The viral infectivity factor (Vif) can induce a poly-ubiquitination degradation of APOBEC3G to counteract the immune response by forming an E3 ubiquitin complex composed of cellular proteins Elongin B (EloB), Elongin C (EloC), Cullin 5 and Ring-Box protein. In this project, we solved the first structure for the Vif SOCS-box and EloBC complex in solution by Nuclear Magnetic Resonance, which shows that the proline-rich motif in the SOCS-box binds to the EloB carboxyl terminus by backbone interaction, based on weak van der Waals forces. Based on the results from cell assays it shows that the residues on the Vif proline-rich motif play an important role in viral infectivity; the proline-rich motif induced structure of the C-terminal tail has been demonstrated to be critical for EloB to perform further biological function, by a mechanism which allows HIV-1 to evade the immune response. In addition, expression and purification on full-length Vif in the presence of EloBC and core binding factor β (CBFβ) has been performed. These studies showed that soluble Vif in the tetramer is achieved when it is expressed at a low temperature with a low IPTG concentration. Solubility tests indicate that this complex must be kept in high salt concentration in order to prevent self-association. Comparison with previous studies on Vif expression and purification. This protocol enables one to obtain purified Vif from E. coli and opens the way to solve the structure by X-ray crystallography and extend our understanding on the Vif-induced APOBEC3G degradation mechanism.

Aside from the typical retroviral gag, pol, and env genes, HIV-1 encodes a set of accessory proteins. Amongst those, Vpu modulates the expression of several cell surface receptors within the infected cell to promote HIV-1 replication. Particularly, Vpu enhances virus release by overcoming the antiviral action of tetherin. HIV-1 is not the only virus to have evolved countermeasures to inactivate tetherin. Here we have investigated the strategy employed by Kaposi’s sarcoma-associated herpesvirus (KSHV) to escape from tetherin-mediated restriction of virus particle release. We have found that KSHV encodes a ubiquitin ligase, K5, that ubiquitinates tetherin on its cytoplasmic tail to target it for endolysosomal degradation. We then assessed the determinants in the Vpu transmembrane domain required to antagonize tetherin. Three amino acid residues that form one face of the transmembrane region of Vpu were found to be important for the interaction with tetherin. In contrast to Vpu from HIV-1 Group M, those encoded by Group O have not acquired the capacity to counteract tetherin. The defects map to the transmembrane domain and the membrane-proximal hinge region of the first alpha helix of the cytoplasmic tail; both are respectively defective for tetherin binding and trans-Golgi network-associated subcellular localization. As the panel of Vpu’s targets is increasing, we have collaborated with Paul Lehner’s group to identify new cellular substrates of Vpu. The glutamine transporter, SNAT-1, was identified by proteomic screening (SILAC). SNAT-1 is downregulated and degraded in HIV-1 infected CD4+ T cells. We are currently investigating the effects of Vpu-mediated SNAT-1 degradation and glutamine limitation on virus growth and cell proliferation/survival in primary CD4+ infected T cells. In parallel, we have investigated the molecular mechanism by which Vpu achieves SNAT-1 depletion.

Following the identification of a highly resistant protease containing a novel insertion (35QN), together with a series of additional gag mutations. We wished to explore the gag-protease interrelationships, and associated replication capacity deficit/compensation. A replication competent NL4-3 based HIV-1 vector system was developed allowing ligation of full length gag with or without protease derived from patient plasma virus. Drug susceptibility assays were undertaken using the MTT assay, and replication capacity assessed by a GHOST indicator cell line. Recombinant viruses containing the protease alone or protease with full length gag demonstrated reduced susceptibility to Pis. The recombinants containing the gag from the patient isolated virus were shown to have a much reduced replication capacity compared to wild-type, even when cloned with the cognate protease. However, the virus containing the patient virus derived gag and protease, demonstrated a 2-fold increase in resistance to some of the Pis tested. The production of high titred stocks was shown to require multiple reversions in gag. This occurred in both viruses containing this region derived from the original patient virus. The reversion pattern was shown to be dependant on whether or not the recombinant contained the highly resistant protease. The lack of reversions from the input genotype in the protease only virus isolate, suggest that this virus is not growth impaired, which was also confirmed by a growth comparison study. The interaction of protease with the gag demonstrates a co-dependence regarding the evolution under drug selection pressure. The removal of the pressure causes reversion of some, but not all of the gag mutations. The lack of reversion in the protease gene suggests that the virus mutates gag to accommodate the protease. The reversion pattern in gag was particularly evident in the p2 protein and or the p2/p7 cleavage site.

Despite the magnitude of the HIV pandemic, the events involved in the initial HIV-1 entry into the body are not yet fully understood. Although the principle mode of HIV-1 transmission is through mucosal surfaces, the oral epithelium appears to be less susceptible to HIV-1 infection than vaginal epithelium. In addition, infections with co-pathogens that elicit immune activation, such as Candida albicans, may also promote HIV-1 infection. The project objectives are to determine whether (i) HIV-1 is able to bind and integrate into oral and vaginal epithelial cell lines, (ii) epithelial cell lines are able to transfer viable virus from their surface to permissive cells, (iii) epithelial cells are responding to HIV-1 with changes in intracellular signalling or gene expression profiling, and (iv) Candida albicans can affect epithelial susceptibility to HIV-1 or whether it can bind and/or transfer HIV-1 to permissive cells. We demonstrate that oral, oro-pharyngeal and vaginal epithelial cell lines do not express canonical receptors for HIV-1 but they do express other receptors known to promote HIV-1 binding, including GalCer and syndecan-1. Oral and vaginal epithelial cell models can capture HIV-1, which subsequently does not appear to integrate into the epithelial genome. Therefore, viral replication is not supported. Notably, HIV-1 captured on the epithelial surface remains infectious and can be transferred to permissive cells. Furthermore, like epithelial cell lines, C. albicans can also directly bind and transfer HIV-1 to permissive cells. The carbohydrate moieties chitin and β-glucan appear to play a role in mediating viral binding. Notably, transfer of HIV-1 to permissive cells occurs from chitin but minimally from p-glucan. This indicates that fungal-viral interactions may occur at mucosal surfaces that potentially promote HIV-1 infection.

Antigen-specific Th2 cells are central to the pathogenesis of asthma and other allergic diseases. Th2 cells secrete the cytokines IL-4, IL-5 and IL-13, responsible for many of the features of allergic inflammation such as eosinophilia, IgE production and mucus hypersecretion. Cysteinyl leukotrienes (CysLTs) are known potent lipid mediators involved in mucus production, bronchoconstriction and leukocyte migration, and act via the receptors CysLTiR and CysLT2R. Recently, CysLTs have been implicated in Th2 responses in mouse models, although the exact mechanisms are unclear. Preliminary microarray studies on polarised human Thl and Th2 cells suggested a highly selective expression of CYSLTR1 by human Th2 cells. Subsequent RT-PCR analysis confirmed that Th2 cells selectively express CYSLTR1 mRNA at high levels. Calcium signalling experiments revealed that Th2 cells respond selectively to leukotrienes compared to Thl cells, with a rank order of potency similar to that reported for CysL R (LTD4>LTC4>LTE4). This profile of leukotrienes responsiveness was blocked using the known selective CysLTiR antagonists MK571, Montelukast and Zafirlukast. Additionally, the LTD4-induced signalling in Th2 cells is mediated via CysLTiR coupled to Gaq and Gaj proteins, which are blocked when using selective CysLTiR antagonists. Finally, LTD4 is a potent chemoattractant for Th2 cells, which migrate to LTD4 at low nanomolar concentrations in a dose-dependent manner. Migration of human Th2 cells towards LTD4 was inhibited by selective CysLTiR antagonist MK571. Interestingly, LTD4 had no significant effect on the proliferation or differentiation of Th2 cells, or on the expression of Th2-specific cytokines. Microarray studies on LTD4-treated Th2 cells identified key immediate early genes that were upregulated after 30 minutes of treatment, and that were differentially expressed compared to untreated cells. This is the first report that human T cell subsets express functional CysLTiR and important implications for our understanding of the anti-inflammatory effect mechanism of CysLTiR antagonists.

Here presented is a detailed account of two novel findings, as well as further clarification of another known effect of the heterodimeric cytokine IL-27 on T cell biology. The first novel finding describes a potent anti-inflammatory mechanism through the suppression of T cell derived GM-CSF production via IL-27 stimulation. While IL-27 suppressed GM-CSF production by Th1 cells in vitro, it did not influence the production of GM-CSF in committed Th17 cells. il2T'- T cells produced more GM-CSF in response to Toxoplasma gondii infection, suggesting IL-27 signalling regulates T cell derived GM-CSF production in vivo. Furthermore, IL-27 suppressed GM-CSF production by human Thl, but not Th17-polarised cells in vitro, suggesting a conserved mechanism of suppression between species. While IL-27 is a known inducer of IL-10 production in C04+ T cells, the phenotype of these cells remained unknown, with three possible phenotypes proposed (Th1, Tr1cells displayed a Thl phenotype with expression of Thl markers while lacking factors associated with Tr1 and Treg cells.

No description available.