On Parkinson's disease and schizophrenia : case control studies, cellular localization and modelling of candidate genes

Abstract

The causes of many major neurodegenerative, neurodevelopmental and psychiatric diseases are unknown. This thesis has aimed at applying knowledge about the dopamine (DA) neuron system and other key systems to clinical material (DNA and post mortem brain tissue) collected from patients with Parkinson s disease (PD), schizophrenia (SZ) and controls, in order to decipher genetic risk factors. A further aim has been to develop animal models of the disorders. Revelation of genetic risk factors should allow insight into pathogenic mechanisms and development of novel classes of drugs able to prevent and/or counteract disease. The candidate gene approach was chosen, because linkage studies have proven difficult in complex diseases in which multiple genetic risk factors are implicated.

Key genetic findings were modelled in cellular systems to determine if identified genetic variants influence gene and/or protein function. In situ hybridization of human post mortem tissues from tissue banks was used to study the cellular expression patterns and mRNA levels of candidate genes in SZ, PD and control brains NURR1 is one good candidate gene for disorders with an involvement of the DA neuron system since knockout mice do not develop DA neurons in mesencephalon. We sequenced NURR1 in PD, SZ and control material and found three unique missense mutations within exon 3 in psychiatric patients. These three mutations all disturbed NURR1-mediated transcription when modelled in cell lines. We also found single nucleotide polymorphisms (SNP) in the NURR1 promoter and analyzed them with regard to endophenotypes of SZ and personality traits without identifying any strong association. Because of the role of Nurr1 for midbrain DA neurons it has been hypothesized that it has a role in rewarding behaviour. Nurr1 is also located within an alcohol preference quantitative trait locus. The impact of heterozygous deletion of Nurr1 on ethanol consumption as a model for drug-induced reward and on wheel running as a model for natural reward was therefore investigated. The Nurr1+/- did not develop high ethanol consumption nor did they develop as much running as wild-type (WT) mice. Nurr1 therefore seems to be important in mediating the reinforcing properties of ethanol and running. Nurr1 was sequenced in three different mice strains (C57Bl/6, DBA/2 and 129/Sv) with different preferences for ethanol. Two dinucleotide repeats were found in the Nurr1 promoter that were longer in mice with low preference for ethanol (DBA/2 and 129/Sv) compared to those with high preference for ethanol (C57Bl/6).

A second project focused on the toxic effects of aldehydes on the DA systems and possible genetic defects in defence systems against such toxic agents. ADH4 codes for an alcohol dehydrogenase which can convert retinol to retinal and also act upon a wealth of other alcohols and aldehydes. We found a significant association between one SNP in the promoter region of ADH4 and idiopathic PD. In-vitro transfection studies showed that this SNP reduced transcriptional activity 25-30%. To determine if Adh4 knockout mice (Adh4 -/-) show any changes in DA system-related activity, we investigated locomotion responses to drugs acting on the DA system, as well as biochemical parameters. In an attempt to challenge the DA system from early development and throughout life groups of Adh4 -/- and WT mice where fed a lipid-rich Western diet, expected to induce higher levels of toxic aldehydes through lipid peroxidation. Two corresponding groups were fed a normal diet.

Spontaneous locomotor activity was decreased in Adh4 -/- compared to WT, whereas administration of apomorphine, an unselective DA agonist, led to a much higher locomotor response in Adh4 -/-. d-Amphetamine, acting on the DA transporter, induced increased horizontal activity in Adh4 -/-. Groups on the two different diets showed similar effects of lacking Adh4. DA levels in striatum were significantly reduced in Adh4 -/-. Our data suggest that absence of Adh4 disturbs the function of the DA system. ADH1C is the closest relative to ADH4 and therefore another candidate gene. We found the occurrence of a nonsense mutation in ADH1C in PD patients. This truncating mutation occurs early in the protein (after 78 of 350 amino acids) and individuals carrying the mutation are thus functionally equivalent to heterozygous knockouts. The distribution of ADH1, 3 and 4 mRNAs was investigated in rat, mouse and human brain. Only ADH3 mRNA was found to be expressed in brain tissue of all three species. The cytosolic aldehyde dehydrogenase (ALDH1) is involved in DA degradation. ALDH1 was found to be highly expressed in control DA cells of both substantia nigra (SN) and the ventral tegmental area (VTA). A reduced expression was seen in SN but not VTA in PD. In SZ we found ALDH1 expression at normal levels in the DA cells in SN but significantly reduced levels in those of the VTA. We conclude that low levels of ALDH1 expression correlate with DA neuron dysfunction in both PD and SZ.

Paraoxonase 1 (PON1) is a potent enzyme for the detoxification of such environmental toxins that have been implicated in the pathogenesis of PD. Its physiological function is likely to be that of an aldehyde dehydrogenase associated with high-density lipoproteins (HDLs) in blood. We studied nonsynonymous PON1 mutation in PD and could confirm association with PD as reported by others (Ahkmedova et al, J Neurol Sci 184:179-82, 2001). We also made further studies of polymorphisms in the PON gene cluster which also include the related PON2 and PON3 genes, but no further associations were found. NOTCH4 lies within the MHC region at 6p21.3, a region that has been linked to SZ by several groups. However we did not detect any association of the reported NOTCH4 polymorphisms with SZ in our material. We are also unable to detect any association of reported SNPs to endophenotypes of SZ or personality traits. To summarize, multiple genetic risk factors seem to be involved in PD an SZ, some of them have been detected and/or further characterized in the present study.