A study of natural killer cells in renal failure and in patients at cardiovascular risk

Cardiovascular disease (CVD) is the leading cause of death in the UK, accounting for more than a third of all deaths, with atherosclerotic coronary artery disease (CAD) being the commonest type of CVD. Recently, it has become recognised that in addition to traditional cardiovascular (CV) risk factors such as dyslipidaemia, hypertension, smoking, and diabetes, inflammation plays an important role in the development and progression of atherosclerosis. A concept has emerged that atherosclerosis to some extent can be viewed as a chronic inflammatory autoimmune disease in which the adaptive immune system is targeted against vascular self-antigens modified by hypercholesterolaemia, involving both the innate and adaptive immune response. Much work has been done in determining how the immune system is involved, however relatively little is known about natural killer (NK) cells – an important component of the innate immune system which acts against virally infected cells and neoplastic transformation. In addition NK cells possess cytolytic ability and provide an early source of immunoregulatory cytokines. Recently, there has been increasing evidence to support a role for NK cells in the development of atherosclerosis. The work in this thesis examines NK cell function with the aim of determining whether any changes in the function of this immune cell could have a role in the development of CVD. In order to do this, we chose two patient populations at high CV risk and compared NK cell subsets and function to healthy controls. Firstly, 66 patients with dyslipidaemia on a variety of lipid lowering treatments attending a lipid clinic, and secondly 143 patients with chronic kidney disease (CKD) including 11 with end-stage renal disease (ESRD) on hospital haemodialysis (HD). It is known that CVD is the leading cause of death in patients with CKD, and in ESRD patients have a 20-100 fold risk of premature CV death compared to age matched controls from the general population. The increased CV risk results from additional risk factors that are unique to this patient population, but in particular, these patients have an immune dysfunction that is not completely understood and a resultant inflammatory state. We determined T-cell, NKT-cell, and NK cell subsets from peripheral blood mononuclear cells (PBMCs) by flow cytometry. We then isolated NK cells from PBMCs and assessed NK cell function using a 51-Chromium release assay. These results were then correlated with clinical and laboratory results. In the patients with dyslipidaemia, we did not find any correlations between lipid levels and NK cell numbers, subsets, or cytoxicity. The presence of statin therapy or any other lipid lowering treatment did not result in a reduction in NK cell cytotoxicity. In the CKD patient group, we found a correlation between NK cell cytotoxicity and creatinine, although this did not retain significance after multivariate analysis. Interestingly, we also found a correlation between NK cell cytotoxicity and serum phosphate level, which did remain significant after multivariate regression. We are the first to report a relationship between phosphate and NK cytotoxicity. This is an interesting finding as there is increasing evidence supporting a role for hyperphosphataemia in CVD and increased mortality in both the general population and particularly in patients with ESRD. Phosphate has been shown in some studies to be an independent predictor of inflammation, and may provide the link between the high risk of CVD and CKD. The next part of this thesis was an in vitro study of membrane cholesterol in NK cells. The cell membrane supports cholesterol-rich microdomains termed “lipid rafts”, which concentrate receptors and signal transduction molecules to facilitate high efficiency signal transduction. Statins have a number of pleiotropic effects which have been explained by reduced production of isoprenoid intermediates, and depletion of cell membrane rafts. This study aimed to investigate the effects of membrane cholesterol manipulation on NK cell function, and specifically, whether the actions of statins on NK cells are due to depletion of membrane cholesterol or inhibition of isoprenylation. The NK92MI cell line was used. Cells were either cholesterol loaded, or cholesterol depleted using statins, and NK cell function assessed using a 51-Chromium release assay. Cholesterol was successfully incorporated into the membrane and rafts and was concentration dependent. The addition of cholesterol to statin treated cells restored the cholesterol content in the cell membrane and in rafts. NK cell cytotoxicity decreased with statin treatment in correspondence with raft levels, however in contrast with the increased raft levels of cells which were cholesterol loaded, NK cytotoxicity was also decreased. Measurement of active Ras (a small G-protein that is localised by isoprenylation in membrane rafts when activated), showed that statin treatment reduced Ras within the raft which was not rescued by the addition of cholesterol, suggesting that statins deplete membrane cholesterol and rafts as well as inhibiting isoprenylation. Replenishment of membrane cholesterol restores non-isoprenylated, raft-associated proteins, but does not correct the functional effects of statins. The final part of this thesis aimed to evaluate the relationship between NK cells and potential CV risk biomarkers in these two patient populations at high risk of CVD. High sensitivity C-reactive protein (hsCRP), interleukin-6 (IL-6), pentraxin-3 (PTX-3), adiponectin, and soluble ST2 (sST2) levels were determined by ELISA, and correlated with NK cell numbers, phenotype, and function, as well as other routine biochemical and haematological parameters. We were not able to determine any definite relationships between the biomarkers studied and NK cell function, although there was an association between PTX-3 and NK cytotoxicity that was only found in inflamed (hsCRP>2mg/L) patients. In conclusion, these studies have provided further insight into the role of NK cells in a group of patients that have not previously been studied: patients with a range of CKD, in addition to patients with dyslipidaemia. This study is the first to associate NK cell cytotoxicity with serum phosphate levels which may have clinical implications. Further studies are needed to clarify whether other immune abnormalities occur in the context of hyperphosphataemia, and the causes and consequences of this. We have also demonstrated that statins deplete membrane lipid rafts as well as inhibit isoprenylation, suggesting a novel dual mechanism of action which merits further investigation