24s-Hydroxycholesterol. Studies on regulatory mechanisms behind its formation in the brain and its potential use as a marker for neurodegeneration

Abstract

Cholesterol 24-hydroxylase (CYP46A1) belongs to the cytochrome P450 super family and is responsible for conversion of cholesterol to the oxysterol 24S- hydroxycholesterol (24S-OHC). This structural modification allows 24S-OHC to traverse the blood brain barrier and this pathway is the major one for elimination of cholesterol from the mammalian brain. CYP46A1 is almost exclusively located to neurons in the brain and retina. The promoter region of CYP46A1 shows classical hallmarks of a gene with putative housekeeping function. Oxidative stress was the only factor initially shown to cause signicant increase in CYP46A1 reporter activity.

The aim of this thesis was to obtain a more profound knowledge about regulation of CYP46A1 and the regulatory importance of its product. In addition the possibility was investigated that the levels of 24S-OHC in cerebrospinal fluid may be used diagnostically.

In paper I we tested the possibility that epigenetic factors are important for CYP46A1 expression. We could demonstrate both in vivo and in vitro that the histone deacetylase inhibitors Valproate and Trichostatin A induce the expression of CYP46A1, essentially reprogramming non-neuronal cells to express CYP46A1 to the same level as that found in adult neurons.

In paper II we investigated the inhibitory effect of the antifungal drug Voriconazole on CYP46A1 in mice and we hypothesised that inhibition of CYP46A1 may have a role in the reported side effects in connection with Voriconazole (neurological and visual disturbances). A decrease in concentration of 24S-OHC in the mouse brain and a reduction in the lathosterol:cholesterol ratio, an index of brain cholesterol synthesis, was demonstrated.

In paper III we investigated the effect of omega-3 polyunsaturated fatty acids in Syrian Hamsters. Enrichment of diet with omega 3-fatty acids resulted in increased CYP46A1 mRNA levels in the brain.

In paper IV we investigated the consequences of an overexpression of CYP46A1. A transgenic mouse model overexpressesing human CYP46A1 was developed and characterized. Significant expression of the human CYP46A1 protein was found in brain, testis and eye and the brain in which contained more than 10-fold higher levels than the other organs. Circulating levels of 24S-OHC were increased by 4-6 folds and the fecal excretion of this steroid in free form was increased more than 20 fold. In the brain of the transgenic mice the total amount of CYP46A1 protein was increased 2-4 fold and had similar cellular distribution to the endogenous enzyme. The level of 24S-OHC in the brain was about double compared to the controls. It is known that 24S-OHC is an efficient ligand for the liver X receptor (LXR) in vitro. The overproduction of 24S-OHC did not significantly activate any LXR target genes in the brain, while in the liver some of these genes were affected but not in the direction expected in connection with LXR activation.

Based on in vitro studies a regulatory link between 24S-OHC and APOE has been suggested and addition of 24S-OHC stimulates secretion of APOE from cultured astrocytes. In paper V we investigated if there is a correlation between the concentration of APOE and 24S-OHC in cerebrospinal fluid in patients. A significant correlation was found in patients with Alzheimer s disease (AD) and mild cognitive impairment (MCI), but not in the control group. In paper VI we investigated the possibility that the level of 24S-OHC in cerebrospinal fluid may be an early biomarker for AD. We could show that 24S-OHC is as sensitive as the standard diagnostic biomarkers for AD (tau-protein, phospho-tau, and beta amyloid). Interestingly 24S-OHC appeared to be a more sensitive marker for MCI than the standard biomarkers. According to current concepts the levels of cholesterol in critical neuronal membranes is of importance for the balance the between α- and β- secretase pathway. An increased consumption of this cholesterol by increased activity of CYP46A1 would be expected to favour the α- secretase pathway and reduce amyloid formation. Under in vitro conditions 24S-OHC seems to have a direct inhibitory effect on amyloid formation. The mouse model with overexpressed CYP46A1 presented in paper IV will be ideal to test the hypothesis that an upregulation of CYP46A1 is neuroprotective and reduces amyloid formation. As shown in paper I histone deacetylase inhibitors are potential drugs to reach this goal. High intake of omega-3 fatty acids seems to have some neuroprotective effects in humans and the results shown in paper III is consistent with the possibility that part of this may be mediated by an effect on CYP46A1. Inhibition of CYP46A1 would be expected to have a negative effect on the function of the central nervous system and possibly also retina. The results of paper II demonstrated that the drug Voriconazole, with known negative side effects on CNS and visual function, inhibits the flux in the mevalonate pathway in the brain. In part this may be the consequence of inhibition of CYP46A1. Results are presented to indicate that the levels of 24S-OHC in cerebrospinal fluid may be used as diagnostic tool in connection with neurodegeneration (paper V and VI)