Electroconvulsive therapy suppresses the neurotoxic branch of the kynurenine pathway in treatment-resistant depressed patients.

BACKGROUND: Neuroinflammation is increasingly recognized as contributing to the pathogenesis of depression. Key inflammatory markers as well as kynurenic acid (KYNA) and quinolinic acid (QUIN), both tryptophan metabolites, have been associated with depressive symptoms and suicidality. The aim of the present study is to investigate the peripheral concentration of cytokines and tryptophan and kynurenine metabolites in patients with unipolar treatment-resistant depression before and after electroconvulsive therapy (ECT), the most effective treatment for depression. METHODS: Cytokines in plasma from patients with major depressive disorder (MDD; n = 19) and healthy volunteers (n = 14) were analyzed with electrochemiluminescence detection. Tryptophan and kynurenine metabolites were detected with high-performance liquid chromatography (HPLC) and LC/MS. KYNA was analyzed in a second healthy control cohort (n = 22). RESULTS: Patients with MDD had increased plasma levels of interleukin (IL)-6 compared to healthy volunteers (P < 0.05). We also found an altered kynurenine metabolism in these patients displayed by decreased plasma levels of KYNA (P < 0.0001) as well as a significantly increased QUIN/KYNA ratio (P < 0.001). Plasma levels of tryptophan, kynurenine, and QUIN did not differ between patients and controls. Treatment with ECT was associated with a significant decrease in the plasma levels of tryptophan (P < 0.05), kynurenine (P < 0.01), and QUIN (P < 0.001), whereas plasma levels of KYNA did not change. The QUIN/KYNA ratio was found to significantly decrease in ECT-treated patients (P < 0.05). There was a significant inverse correlation between symptom severity and kynurenine levels at baseline (r = -0.67, P = 0.002). CONCLUSIONS: This study confirms an imbalanced kynurenine pathway in MDD supporting the hypothesis of a netstimulation of N-methyl-D-aspartic acid (NMDA) receptors in the disorder. Treatment with ECT profoundly decreased QUIN, an NMDA-receptor agonist previously suggested to be implicated in the pathogenesis of depression, an effect that might have bearing for the good clinical outcome of ECT.

Polymorphisms in AKR1C4 and HSD3B2 and differences in serum DHEAS and progesterone are associated with paranoid ideation during mania or hypomania in bipolar disorder.

Paranoia is commonly a mood-incongruent psychotic symptom of mania which may be related to dopamine dysregulation. Progesterone and its metabolite allopregnanolone (ALLO) have been found in animals to antagonize the effects of dopamine. We therefore examined serum progesterone, its endogenous antagonist DHEAS and polymorphisms of the genes coding for certain steroidogenetic enzymes (AKR1C4, HSD3B2, and SRD5A1) in 64 males and 96 females with bipolar 1 or 2 disorder with or without paranoid ideation during mood elevation. Euthymic morning serum progesterone, DHEAS and cortisol concentrations were measured in males and in premenopausal women who were in follicular phase and not taking oral contraceptives. In women only, SNPs in AKR1C4 reduced the likelihood of having exhibited paranoid ideation by circa 60%. The haplotype of all 4 SNPs in the AKR1C4 gene reduced the risk of exhibiting paranoia by 80% (OR 0.19, 95% CI 0.06-0.61, p=0.05). A history of paranoid ideation was not, however, related to progesterone or DHEAS concentration. Serum DHEAS and progesterone concentrations were lower in men who had shown paranoid ideation during mania/hypomania compared with those who had not (F=7.30, p=0.006) however this was not coupled to polymorphisms in the selected genes. The ancestral G in rs4659174 in HSD3B2 was in men associated with a lower risk of paranoid ideation (likelihood ratio χ(2) 3.97, p=0.046, OR 0.31 (95% CI 0.10-0.96)) but did not correlate with hormone concentrations. Hence, gene variants in the steroidogenetic pathway and steroids concentration differences may be involved in the susceptibility to paranoia during mood elevation.

AKR1C4 gene variant associated with low euthymic serum progesterone and a history of mood irritability in males with bipolar disorder.

BACKGROUND: Irritable mood during mood elevation is common in bipolar disorder. The progesterone metabolite allopregnanolone (ALLO) has been implicated in other disorders presenting with irritability. This study aimed to test whether a history of manic/hypomanic irritability is associated with low serum progesterone levels; and whether single nucleotide polymorphisms (SNPs) in gene coding for steroidogenetic enzymes (HSD3B2, SRD5A1 and AKR1C4 were coupled to previous manic irritability and/or with serum progesterone concentrations. METHODS: Morning serum progesterone concentrations during euthymic phase of bipolar illness types 1 and 2 were assessed in 71 males and 107 females. Previous manic/hypomanic irritability was assessed using the Affective Disorders Evaluation. Selected SNPs were analyzed: i) aldoketoreductase-type-4 (AKR1C4 - rs17306779, rs3829125, rs10904440, rs12762017, and rs11253048), ii) 3-ß-hydroxysteroid-dehydrogenase (HSD3B2 - rs4659174, rs2854964, and rs3765948), iii) steroid-5-α-reductase (SRD5A1 - rs8192139, rs181807, rs3822430, and rs3736316). RESULTS: In males, progesterone concentrations were lower in those who had shown manic/hypomanic irritability compared with nonirritable (F=7.05, p=0.0099). SNPs rs17306779, rs3829125, and rs10904440 were associated with manic/hypomanic irritability. A cystine to serine change at position 145 in AKR1C4 (rs3829125) was associated with lower serum progesterone (F=6.34, p=0.014). There were no associations in females. LIMITATIONS: Relatively small sample sizes. CONCLUSION: Low progesterone levels and a cystine to serine change at position 145 in AKR1C4 gene are associated with manic/hypomanic irritability in males. Given that the enzyme AKR1C4 has both dehydrogenating and reductive activities in the steroidogenetic pathway, a missense variation in the gene may predispose to manic/hypomanic irritability by altering the relationship between progesterone and ALLO concentrations in the brain.