Effects of SSADH gene variants on enzyme stability and activity involvement in neuronal diseases and oxidative stress

Effects of SSADH Gene Variants on Enzyme Stability and activity Involvement in Neuronal Diseases and Oxidative Stress Gamma-aminobutyric acid (GABA) is the main inhibitory neurotransmitter in the central nervous system and disturbance in the GABAergic system can cause neuronal diseases, neurodegeneration, psychiatric disorders and intellectual disability (Govindpani et al., 2017). Succinic semialdehyde dehydrogenase (SSADH) is a mitochondrial homotetrameric enzyme, catalyzing a reaction in the GABA catabolic pathway. Indeed, it oxidizes succinic semialdehyde (SSA) to succinate, thus supplying Krebs cycle, in the so called “GABA shunt” (Malaspina et al., 2016). SSA could also be converted, in a minor cytosolic pathway, into γhydroxybutyric acid (GHB) by the enzyme AKR7A2. Furthermore, as a secondary function, SSADH oxidizes and therefore detoxifies, 4-2- hydroxynonenal (4-HNE), a highly reactive and toxic degradation product of peroxidized polyunsaturated lipids. Indeed, the SSADH deficiency (SSADHD) is characterized by oxidative stress and mitochondrial damage. Different mutations in the ALDH5A1 gene (coding for SSADH) have been found, which lead to enzyme failure and causing SSADHD which is a rare autosomal recessive disorder and the best characterized inherited metabolic disorder of GABA catabolism (Malaspina et al., 2016). Several common polymorphisms (SNPs) of the ALDH5A1 gene also occur; some of them, when overexpressed in vitro, negatively affect SSADH activity (Akaboshi et al., 2003; Malaspina et al., 2016). Interesting correlations between GABA catabolism impairment and the progression of Alzheimer's disease (AD) have been suggested in recent years. In particular, accumulation of GABA metabolites (2-4-DHBA and GHB) in the AD patients’ serum has been shown, probably due to defects in the activities of the SSADH or AKR7A2 enzymes (Salminen et al., 2016). In addition, one of the most frequent SNPs of the ALDH5A1 gene, the c.538C>T, enhances cognitive decline and affects survival in the elderly (De Rango et al., 2008). In the present PhD project, studies have been conducted on the effects of ALDH5A1 gene variants on the activity and stability of the enzyme and on their involvement in neuronal damage and oxidative stress, by three different approaches. First of all, in two Italian patients affected by SSADHD, mutations and SNPs and their correlation with decrease in enzyme activity and in tetramer stability was demonstrated, by in vitro overexpression and in silico analysis (in collaboration with Prof. M. Falconi). Taken together, the results obtained in the first part of my project show that the mechanisms underlying SSADH activity loss could rely on amino acids changes, which modify monomer structure and affects tetramer stability (Leo et al. 2017, Menduti et al., 2018). Sequence analysis of ALDH5A1 gene showed that one of the two patients was a compound heterozygote for c.1226G>A (p.G409D) and the novel missense mutation, c.1498G>C (p.V500L). The p.V500L mutation, despite being conservative, produces a complete loss of enzyme activity, because it induces destabilizing energies in the tetramer. Furthermore, we studied a further clinical case of SSADHD, carrying a double mutant allele with the already reported pathogenic mutation c.526G>A (p.G176R) together with the common SNP c.538C>T (rs2760118) (p.H180Y). The other double mutant allele harbors another novel missense mutation c.1267A>T (p.T423S) together with the SNP c.709G>T (rs62621664) (p.A237S). When overexpressed in cells, the activity of the mutant SSADH carrying the c.1267A>T substitution is comparable to that of the wt enzyme. However, when the c.1267A>T substitution is together with the SNP c.709G>T, the SSADH activity drops to about 30%. The c.526G>A mutation, alone or in combination with the SNP c.538C>T, produces an almost inactive enzyme. By the observation of decreased protein levels, and the calculated instability of the tetramer, we suggest that this mutant SSADH might undergo increased degradation in the cell, thus resulting in the loss of enzyme activity. Thus, besides discovering to novel mutations associated with SSADHD, we demonstrated that common SNPs and pathological mutations may act in synergy, thus offering new perspectives to the genetics of SSADHD (Menduti et al., 2018). As a second goal, studies were conducted on the effects of the transiently overexpressed SSADH gene variants, associated with a lower enzymatic activity, in the cell response to oxidative stress, with respect to the wt form. In particular, the CTT haplotype (i.e. the variants c.106G>C, c.538C>T and c.545C>T in linkage disequilibrium) was studied. Interestingly, we observed that protein level produced by CTT variant in U87 neuronal cells is lower with respect to the wt protein, suggesting that it may undergo degradation, which does not seem to be mediated by proteasomal activity. Treatment of transiently transfected U87 cells with paraquat, a molecule known to produce reactive oxygen species in cells and oxidative stress, was performed to evaluate the effects of CCT or wt on cell viability, lipid peroxidation, 4-HNE-proteinadducts formation and mitochondrial damage. The results obtained suggest that CTT SSADH overexpression negatively influence cell viability, lipid peroxidation and mitochondrial transmembrane potential. CTT overexpression appears to be noxious to untreated cells. Furthermore, the data obtained seem to confirm the hypothesis that SSADH defends cells from oxidative stress induced by paraquat, since effect was observed in terms of protection from lipid peroxidation and mitochondrial transmembrane potential. Finally, a case-control study on 300 Italian patients affected by Alzheimer’s disease and 300 healthy individuals was conducted, in order to assess whether the presence of the more frequent SNPs of the ALDH5A1 gene, as well as of AKR7A2 gene could be associated with Alzheimer's disease. The DNA samples of both patients (300 individuals) and controls (300 individuals) with the same range of age (58-88 years old), were obtained from the IRCCS Fatebenefratelli (Brescia) Biobank. Three SNPs in ALDH5A1 coding sequence (c.106G>C, c.538C>T and c.545C>T) and one SNP in the AKR7A2 coding sequence (c.446G>A) were analyzed. For all the considered SNPs no statistically significant difference was obtained between the two groups. We found that the T allele of c.538C>T ALH5A1 SNP, as well as the haplotype c.538C>T-c.545C>T (TC), is slightly increased in individuals affected by Alzheimer’s. So far, no correlation was found between disease condition and ALDH5A1 and AKR7A2 genotypes for the examined SNPs. Possibly, it would be interesting to enlarge the sample of AD and CTRL, in order to highlight genetic factors that contribute with a light effect on the risk of occurrence of a complex pathology such as AD.