Polymorphisms in genes in phospholipase A2 signalling pathway and niacin skin flush response in patients with schizophrenia and schizoaffective disorder Objectives: To test phospholipid membrane hypothesis we investigated a possible role of gene polymorphisms involved in PLA2 signaling pathway on the risk of developing schizophrenia or schizoaffective disorder and clinical expression of illness. We included five gene polymorphisms into genetic analysis: rs10798059 (A/G polymorphism - BanI variant) of the PLA2G4A gene for cPLA2, rs4375 (C/T polymorphism) of the PLA2G6 gene for iPLA2, rs1549637 (A/T polymorphism) of the PLA2G4C gene for the cPLA2, rs689466 (A/G polymorphism) of the PTGS2 gene for COX-2 and rs1800206 (C/G polymorphism - L162V variant) of the PPARα gene. We further tested the possible impact of polymorphisms in genes of the cPLA2/COX-2 metabolic cascade (BanI of the PLA2G4A and rs689466 of the PTGS2) and fatty acid (FA) content of red blood cell (RBC) membranes on attenuated niacin skin flush response in patients. Patients and methods: The study included 203 patients with schizophrenia or schizoaffective disorder and 291 healthy controls. Identification of investigated polymorphisms was performed by PCR-RFLP analysis. For niacin sensitivity we used four patches that contained different concentrations of niacin solution (0.1 M, 0.01 M, 0.001 M and 0.0001 M) which were kept on the forearm skin. After 5 minutes, patches were removed, and visual evaluation of flushing, rated from 0 to 3 (0 = no erythema, 1 = incomplete erythema, 2 = complete erythema within definite patch area, 3 = erythema plus edema beyond the patch area) was done in 5-minute intervals for 15 minutes by two independent raters. According to the method of volumetric niacin response (VNR), visual scores for each individual, across three time intervals at each of the four different niacin concentrations were summed, and regarding concentrations, total VNR and VNR1-4 were yielded. To test FA VIII content, total RBC lipids were extracted, converted into FA methyl esters, and further analyzed by gas chromatography. We determined the relative amount of different saturated FAs (SFAs), monounsaturated FAs (MUFAs) and polyunsaturated FAs (PUFAs) from n-3 and n-6 series. We further introduced several indices and ratios into analysis. Results: Allele and genotype frequencies of investigated gene polymorphisms between patients and controls were not significantly different (P > 0.05). While investigating interactions between different genotypes of the PLA2 and PTGS2 genes, we detected significantly higher frequency of the CT genotype (PLA2G6) and AT genotype (PLA2G4C) combination in the patient group (OR = 2.2, CI = 1.1 - 4.3, z = 2.30, P = 0.02). However, distributions of other genotype combinations did not differ between groups (P > 0.05). Furthermore, there was no synergy between alleles of tested polymorphic variants of the PLA2 and PTGS2 genes (P > 0.05). Neither meta-analysis of the case control studies, nor meta-analysis of the family based studies, revealed statistically significant association between BanI polymorphism of the PLA2G4A gene and etiology of illness. We found no evidence for association between age at first hospital admission and investigated polymorphisms, but we detected small, although significant impact of polymorphic variations of the PLA2G6, PLA2G4C and PPARα genes on PANSS psychopathology. Polymorphic variations accounted for 3.2 – 12% symptoms' variability. General and total symptoms showed positive correlation with the presence of the C allele of the C/T polymorphism (PLA2G6) in all patients (P < 0.05). In males, negative symptoms correlated positively with the presence of the T allele of the PLA2G4C genotype (P < 0.01), while general and total symptoms were positively associated with combination of the C allele (PLA2G6) and T allele (PLA2G4C) (P < 0.05, P < 0.01). Furthermore, females heterozygous for the PPARα genoytpe (L162V), manifested significantly lower negative symptoms than homozygous (L162L) (P < 0.05). The mean values of total VNR (21.4 ± 3.8 vs. 12.7 ± 6,0) and VNR1-4 IX were significantly higher in controls (P < 0.001). PLA2G4A and PTGS2 gene polymorphisms affected niacin flushing to a small, although significant extent, accounting for approximately 3.4% of the VNR variability (each polymorphism accounted for 1.7%, P < 0.01) in the patient group. Patients lacking the A allele in their PLA2G4A genotype (genotype GG) or those carrying G allele in their PTGS2 genotype (AG heterozygous or GG homozygous) had higher values of total VNR. In patients, the total SFAs were significantly increased (P < 0.01), while both n-3 and n-6 PUFAs (P < 0.01), and total PUFAs were significantly reduced (P < 0.001). Lower contents of linoleic acid (LA) (P < 0.01), eicosapentaenoic acid (EPA) (P < 0.001) and docosahexaenoic acid (DHA) (P < 0.01) contributed the most to decreased PUFA levels. In patients, significant decreases were also detected for double bond index (DBI) (P < 0.001), peroxidizability index (PI) (P < 0.01) and several ratios, such as EPA/AA (P < 0.001), total PUFAs/SFAs (P < 0.001) and PUFAs/MUFAs (P < 0.01). However, although both n-3 and n-6 PUFAs were reduced in patients, an association with niacin sensitivity was not detected. Conclusion: Investigated gene polymorphisms, individually, do not contribute to the development of illness. CT genotype (PLA2G6) and AT genotype (PLA2G4C) combination is associated with higher risk for schizophrenia and schizoaffective disorder. Investigated polymorphic variants, alone, or in combination, show no impact on age at first admission. Interaction between PLA2G6, PLA2G4C and PPARα gene polymorphisms and gender, in a smaller extent, modulate expression of clinical symptoms of illness. BanI (PLA2G4A) and rs689466 (PTGS2) polymorphisms affect niacin flush response to a negligible extent, while deficit of n-3 and n-6 PUFAs is not associated with attenuated skin flush in the patient group.