Nutritional and pharmacological regulation of cerebral capillary function

Abstract

Alzheimer’s disease (AD) is the most common cause of dementia pathologically characterised by neurovascular inflammation, extracellular proteinaceous deposits enriched in amyloid-β (Aβ) and formation of neurofibrillar tangles. The cerebrovasculature in subjects with AD is also significantly altered and is first indicated by the progressive dysfunction of the cerebral capillary vessels. Despite evidence of an underlying vascular contribution to onset and progression of AD, most research focuses on factors regulating the more advanced pathological processes.Clinical and experimental evidence suggest vascular risk factors influence the onset and progression of AD. Several studies have demonstrated that atherosclerosis, cardiovascular disease, hypertension, dyslipidemia and insulin resistance are positively associated with AD risk. Furthermore, population studies have found that chronic ingestion of pro-inflammatory diets enriched in saturated fatty acids (SFA), trans-fatty acids and cholesterol are positive risk factors for AD and markedly exacerbate cerebral pathology in animal models. A putative mechanism for the dietary-fat/amyloidogenic pathway was demonstrated in recent animal studies where chronic consumption of diets enriched in SFA and cholesterol significantly compromised blood-brain barrier (BBB) function in wild-type (WT) mice, resulting in substantial blood-to-brain delivery of circulating Aβ and potential exacerbation of cerebral amyloid load. Significant plasma Aβ is derived from liver and small intestine, secreted into circulation associated with triglyceride-rich lipoproteins (TRLs).Dietary lipid regulation of TRL-Aβ was equivocally demonstrated in WT mice fed an SFA enriched diet, where intestinal Aβ biogenesis and secretion was enhanced. Given that humans are predominantly in an absorptive state, the cumulative cerebrovascular effects of transient exposure to dietary stimulated plasma Aβ may be AD-risk relevant. However, a dietary-lipid/Aβ axis for cerebrovascular function could also provide therapeutic opportunities to reduce AD risk.Observational studies suggest that calorie restriction and regular consumption of diets which supress inflammation may delay development and progression of AD. Recent findings have demonstrated that consumption of diets rich in monounsaturated fatty acids (MUFAs) and ω-3 polyunsaturated fatty acids (PUFAs), in particular docosahexaenoic acid (DHA), may reduce risk for AD via suppression of inflammatory pathways. Furthermore, beneficial effects of unsaturated fatty acids on prevention of cerebrovascular dysfunction have been demonstrated in vivo. The association between dietary fat intake, hypercholesterolemia and increased AD prevalence suggests the possibility that lipid-modulating agents might also delay onset and progression of AD. Some epidemiological studies have provided evidence of a lower prevalence of diagnosed AD in patients with hypercholesterolaemia treated with statins (3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase inhibitors). In one clinical study, an older generation cholesterol lowering agent probucol was found to stabilise cognitive impairment in elderly AD subjects. The mechanisms for the pharmacological effects of lipid lowering agents in prevention of AD are unclear, but may include vascular benefits associated with lower plasma lipids, regulation of Aβ biogenesis and kinetics or indicative of anti-oxidant and antiinflammatory properties of the agents.At present, there is lack of effective strategies for the treatment of AD. Given that cerebral capillary disturbances is amongst the earliest indicators of disease onset, it seems a reasonable proposition to suggest that recognition of and timely intervention to restore capillary integrity would be therapeutically beneficial. This thesis explores the novel hypothesis that “dietary fat induced cerebral capillary dysfunction can be corrected by timely provision of an anti-inflammatory diet enriched in ω-3 fatty acids, or lipid lowering agents that attenuate cerebrovascular exposure to peripheral Aβ and inflammation”. The hypothesis presented is supported by a substantial literature review in Chapter 1 of this thesis.As a ‘proof of concept’ approach to the indicated hypothesis, the effects of lipid lowering agents, atorvastatin, pravastatin and probucol, for ‘prevention’ of dietary fat induced BBB dysfunction was determined (Chapter 1: article 5). Wildtype mice were chronically fed an SFA diet supplemented with atorvastatin and pravastatin. In addition, the effects of probucol on BBB integrity were determined in mice fed a cholesterol supplemented diet. All mice were maintained on their respective diets for 12 weeks. Cerebral extravasation of plasma protein immunoglobulin-G (IgG) was assessed utilising 3-dimensional (3-D) immunofluorescent microscopy. Immunoglobulin-G was used as a surrogate marker of non-specific cerebrovascular permeability. Significant BBB dysfunction and enhanced cerebral extravasation of IgG was evident in mice fed SFA and cholesterol diets. Atorvastatin, a lipid soluble HMG-CoA reductase inhibitor prevented SFA induced parenchymal extravasation of IgG at 12 weeks when incorporated into the diet. In contrast, hydrophilic pravastatin had no effect on BBB integrity. In cholesterol supplemented mice, probucol maintained BBB function and extravasation of IgG was not evident. These findings suggest that some lipid lowering agents may effectively prevent dietary fat induced BBB dysfunction and their efficacy may be dependent on solubility. The drug effects on BBB integrity were independent of significant modifications on plasma lipid homeostasis. These findings support the hypothesis that pleiotropic properties of lipid lowering agents maintain BBB integrity.Evidence of protective effect of lipid modulating agents on BBB function was concomitant with an attenuation in TRL-Aβ biogenesis is reported in Chapter 2. The article presented in this chapter investigated in vivo the putative effects of probucol on in TRL-Aβ homeostasis in absorptive epithelial cells of the small intestine in WT mice. Intestinal Aβ abundance was stimulated in mice fed a high-fat (HF) diet enriched in SFA and cholesterol for 4 weeks. Drug effects were determined in mice given low-fat (LF) and HF diets supplemented with 1% probucol. Quantitative immunofluorescent microscopy determined intestinal Aβ and apolipoprotein B (apo B) abundance. Apo B is an obligatory structural component of TRL originating from liver and absorptive epithelial cells of the small intestine. Apo B staining was detected in both the perinuclear region of the enterocytes and the lacteals in all groups. However, HF feeding and probucol treatment increased secretion of apo B into the lacteals without any change in net villi abundance. On the other hand, HF induced enterocytic perinuclear Aβ was significantly attenuated by probucol. No significant changes in Aβ were observed within the lacteals. These findings support the notion that probucol normalised the HF induced intestinal Aβ biogenesis and availability of TRL-Aβ for secretion and may confer protection against AD by reducing exposure to plasma TRL-Aβ.Chronic ingestion of SFAs compromise BBB integrity, leading to cerebral extravasation of apo B lipoproteins enriched in Aβ. In contrast, diets enriched in PUFA oils had no detrimental effect. Rather, ω-3 and ω-6 fatty acids generally confer protection via suppression of inflammation. Chapter 3 investigated in WT mice if a PUFA diet enriched in DHA restored BBB integrity and attenuated parenchymal apo B abundance induced by chronic ingestion of SFA. Cerebrovascular leakage of apo B was quantitated utilising immunofluorescent staining. The plasma concentration of brain-derived S100B was measured as a measure of brain-to blood leakage and complimentary marker of BBB function. In mice fed SFA for 12 weeks, provision thereafter of a DHA enriched diet exacerbated parenchymal apo B retention, concomitant with a significant increase in plasma cholesterol and S100B. In contrast, provision of a LF diet following chronic SFA feeding had no effect on SFA induced parenchymal apo B abundance. In the established mouse model of BBB dysfunction with a heightened state of cerebrovascular inflammation, the provision of unsaturated fatty acids may be detrimental, possibly as a consequence of a greater susceptibility for lipid peroxidation.The ‘proof of concept’ study presented in Chapter 1 (article 5) reported that the SFA induced parenchymal accumulation of plasma proteins could be prevented by co-administration of some lipid lowering agents. Restoration of BBB function is clinically relevant in a therapeutic context. Chapter 4 explored whether lipidlowering agents could reverse BBB disturbances induced by chronic SFA feeding. Wild-type mice were fed an SFA diet for 12 weeks to induce BBB dysfunction and then randomised to receive atorvastatin, pravastatin or ibuprofen in combination with the SFA-rich diet for an additional 2 or 8 weeks. Abundance of plasma-derived IgG and apo B lipoproteins within brain parenchyme were quantified utilising immunofluorescence microscopy. Atorvastatin treatment for 2 and 8 weeks restored BBB integrity, indicated by a substantial reduction of IgG and apo B, particularly within the hippocampus. Pravastatin, a water-soluble statin was less effective than atorvastatin (lipid-soluble). Statin effects were independent of changes in plasma lipid homeostasis. Ibuprofen, a lipid-soluble cyclooxygenase (COX) inhibitor attenuated cerebral accumulation of IgG and apo B as effectively as atorvastatin. These findings are consistent with the drug effects being independent of plasma lipid homeostasis.The outcomes presented in this thesis provide novel insight into the positive effects of lipid lowering agents and non-steroidal anti-inflammatory drugs (NSAIDs) on regression of SFA induced BBB dysfunction. The effects of lipid lowering statins on BBB dysfunction were similar to that of ibuprofen. In addition, statin effects were independent of significant modifications in plasma lipid homeostasis, suggesting pleiotropic anti-inflammatory anti-oxidative effects. Moreover, statins enhance endothelial nitric oxide (NO) bioavailability that is essential for regulation of cerebral perfusion and improved endothelial function. In the established mouse model of BBB dysfunction induced by SFA feeding, provision of diet enriched in ω- 3 fatty acid DHA exaggerated BBB dysfunction. At high concentrations, ω-3 fatty acids are subject to lipid peroxidation and could notionally promote oxidative stress under certain conditions. At a heightened state of inflammation, provision of ω-3 fatty acid may exacerbate BBB dysfunction. Given the importance of cerebral capillary vessels in AD pathophysiology, it is our contention that animal model studies investigating the putative role of anti-inflammatory agents on restoration of BBB integrity will be exceedingly informative for clinical research studies.