The left-lateralisation of citrate synthase activity in the anterior cingulate cortex of male violent suicide victims

Prefrontal parts of both hemispheres of 24 suicide victims (21 males/3 females) with unknown data both on psychiatric comorbidity and on possible psychotropic medication preceding death (typical for most of suicide cases autopsied in the Department of Forensic Medicine at the Medical University of Gdańsk) and 24 (20 males/4 females) controls were obtained during routine forensic autopsies in accordance with existing EU law regulations [17, 18]. The study has been approved by the local ethics committee of the Medical University of Gdańsk as performed in accordance with the ethical standards laid down in the Declaration of Helsinki of 1989.

Detailed diagnostic and demographic data of investigated cases are present in the Supplementary Table. The suicide cohort included violent suicides, which prevail in our autopsy material. Control cases of natural manner of death were more numerous than those of unnatural manner (20 and 4, respectively). Only sudden death cases were investigated in suicide and control cohorts. All brains were free of gross neuropathology suggestive of vascular, traumatic, inflammatory, neoplastic and neurodegenerative processes. Macroscopic evaluation of brains was confirmed by histopathological investigation in cases, where the cause of death was unclear at autopsy and the routine histopathological evaluation of internal organs was necessary for the forensic diagnosis (i.e. in 20 control cases of natural manner of death). Neither chronic nor acute pathological processes were observed microscopically in these cases in neocortical areas and other brain regions in hematoxylin–eosin stained sections. Among others, neuronal necrosis as a consequence of protracted antemortem hypoxia was excluded by histopathological investigation. Neurodegenerative changes such as amyloid plaques, perivascular amyloid deposits and neurofibrillary tangles were not observed microscopically in the AgNOR silver staining in prefrontal regions in those cases, which were included in cohorts investigated previously [17]. Blood was tested for the presence of ethanol at each autopsy. The majority of investigated cases (12 suicide victims and 15 controls) revealed the blood alcohol concentration (BAC) below the limit of quantification (LOQ), i.e. < 0.2 g/l according to internationally accepted analytical guidelines. The remaining 12 suicide victims and 9 controls revealed BAC in the range of 0.24–2.8 g/l (the highest value in one of hanging cases) and 0.3–2.8 g/l (the highest value in the victim of transport accident), respectively.

Prefrontal parts of the brains were separated at forensic autopsies from both hemispheres by coronal sections at the level of temporal poles. Immediately after the separation, cortical samples for CS assays were isolated bilaterally from the rostral (pregenual) part of the AC located closely to the genu of corpus callosum. Each sample was approximately 10 mm in length, 5 mm in width, and 2–3 mm thick, i.e. cortical samples were isolated by an experienced forensic pathologist (KK) under visual control at the clearly visible border with the subcortical white matter (thus each cortical sample contained all cortical layers). Immediately after the isolation, cortical samples were transferred to the deep-freezing refrigerator and stored at − 80 °C. After sampling procedure the remaining prefrontal parts were preserved for the morphological and molecular investigations, which were presented previously [17, 18].

Approximately 0.1 g AC sample was placed in 3 ml of 20 mmol/L Tris chloride buffer pH 7.8 containing 0.2% Triton X-100. The tissue was thawed, minced finely with scissors, homogenized manually with a Teflon-pestle homogenizer (small size), and centrifuged at 30,000 g for 20 min. The resulting supernatant was decanted, and the pellet was resuspended in 2 mL of isolation medium, rehomogenised and centrifuged as above. The supernatant was combined with that obtained after the first centrifugation step and used for enzyme assay.

CS activity was measured by following the formation of 5-thio-2-nitrobenzoic acid (TNB) during the reaction: CoA-SH + 5,5’dithio-bis-2-nitro-benzoic acid (DTNB) → CoA-S–S-TNB (yellow product), coupled with the reaction catalyzed by citrate synthase: oxaloacetate + acetyl-CoA + H₂O → citrate + CoA-SH, as described previously [28].

Briefly, the assay medium (final volume 1 mL) contained: 100 mmol/L Tris–HCL pH 8.1, 0.1 mmol/L DTNB, 0.5 mmol/L acetyl CoA, and 0.5 mmol/L oxaloacetate (OAA). The reaction was started by adding OAA. The assay was performed in duplicate at 37 °C. The yellow product CoA-S–S-TNB was quantified by measuring absorbance at 412 nm (molar absorption coefficient 13.6 × mM⁻¹ × cm⁻¹) using a Beckman DU68 spectrophotometer (Beckman Instruments, Fullerton, CA, USA). Absorbance changes were linear against both time and enzyme concentration. Enzyme activity was expressed as nmol × min⁻¹ × mg⁻¹ protein. Protein assays were performed according to the Peterson’s method [29].

Statistical analyses were performed with the data analysis software system STATISTICA version 10 (StatSoft®, Inc. 2011, www.​statsoft.​com). As normal distribution was not given for analysed data (i.e. significant values of Kolmogorov–Smirnov and Lilliefors tests were obtained), non-parametric statistical procedures were used in hierarchic mode.

Furthermore, the laterality index of CS activity in the AC (100 × [left − right]/[left + right]) was calculated in each case to compare the lateralisation effect between study groups. Age, postmortem interval, brain weight and BAC (values below LOQ were accounted null values in statistical analysis) were considered as numerical confounding variables. Therefore, the subsequent GCD procedure was applied to analyse associations between these variables and dependent variables, i.e. CS activity bilaterally and laterality index. Supplementary to GCD analyses, Spearman’s correlation coefficients were calculated to determine the impact of these variables which might confound the dependent variables.

Following the GCD analysis, unadjusted two-way post hoc comparisons with Mann–Whitney U-test and the χ²-test were used to detect possible differences between the studied groups with respect to the variables mentioned above (i.e. CS activity, laterality index, and confounders). All statistical tests were two-tailed. In general, P values of < 0.05 were accepted as statistically significant.

Kruskal–Wallis analysis of the variance of ranks (H-test) with subsequent U-tests were performed for the evaluation of differences in CS activity and laterality index related to sex between suicides and controls; in this procedure U-test P-values were adjusted to multiple comparisons according to the Bonferroni correction. The differences in investigated parameters related to BAC levels were analysed in a comparable manner (i.e. in cases with BAC values higher than LOQ versus remaining cases).

Further analyses by U-tests revealed an increased laterality index in suicides compared to controls due to the left-lateralised CS activity in the AC in the former study group (U-test P = 0.00009), which was driven mainly by male subjects (see next paragraph). However, the inter-group difference in CS activity observed in the right AC (i.e. a decrease in suicides compared to controls) was insignificant, which could be related to the accentuated variability of results (see Tables 1, 2, 3 and Supplementary Table).

Suicidal and control groups were matched by sex (non-significant χ²-test P value, see Tables 1, 2, 3 and Supplementary Table). According to the effect of sex suggested by the initial GCD procedure, female subjects revealed higher CS activity in both groups bilaterally and the difference was significant in the left AC in controls (median values in females and males: 202.04 and 158.5 nmol × min⁻¹ × mg⁻¹ protein, respectively; U-test P = 0.036, corrected for multiple comparisons). However, very small numbers of female subjects in compared groups prevent from far-reaching conclusions regarding sex-specific differences in CS activity. Further analysis revealed that the laterality index was significantly increased only in male suicide victims compared to male controls (U-test P = 0.0003, corrected for multiple comparisons, see Tables 1, 2, 3). Therefore, the observed phenomenon of left-lateralised CS activity in the AC in suicide was specific for males.

Initial analyses by the GCD procedure revealed no associated impact of any of numerical confounders (i.e. age, PMI, BAC, and brain weight) and forensic diagnosis (i.e. suicides vs. controls) on both CS activity bilaterally and laterality index (non-significant Wald statistic P values). In the subsequent analysis by U-tests, age, PMI, and BAC revealed no significant differences between suicides and controls, whereas the brain weight was significantly higher in the former group (see Tables 1, 2, 3 and Supplementary Table). However, further Spearman’s correlations analysis did not suggest that either brain weight or other numerical confounders influenced the results of comparisons between CS activity or its laterality index in study groups (see Tables 1, 2, 3).

Moreover, no significant differences in investigated parameters were found between inebriated and remaining cases in the entire pool of results as well as in compared groups, also in hemisphere-specific statistical analyses (insignificant H-tests P-values followed by insignificant U-tests P-values corrected for multiple comparisons).