Putative neuroprotective and neurotoxic kynurenine pathway metabolites are associated with hippocampal and amygdalar volumes in subjects with major depressive disorder

Abstract

Inflammation-related changes in the concentrations of kynurenine pathway metabolites occur in depression secondary to medical conditions but are not firmly established in primary mood disorders. Reductions in hippocampal and amygdalar volume that putatively reflect dendritic atrophy are widely reported in major depressive disorder (MDD). Here we tested whether the relative serum concentrations of putatively neuroprotective (kynurenic acid (KA)) and neurotoxic (3-hydroxykynurenine (3HK) and quinolinic acid (QA)) kynurenine pathway metabolites were altered in primary MDD and whether these metabolites were associated with hippocampal and amygdalar volume. A total of 29 moderately to severely depressed unmedicated subjects who met DSM-IV criteria for MDD and 20 healthy controls (HCs) completed a structural MRI scan and provided blood sample for kynurenine metabolite analysis, performed using high-performance liquid chromatography with tandem mass spectrometry. Cytokine concentrations were measured with ELISA and gray matter volumes were measured with the automated segmentation software, FreeSurfer. An a priori defined variable of interest, the KA/QA ratio, a putative neuroprotective index, trended lower in the MDD versus the HC group and correlated negatively with anhedonia but positively with the total hippocampal and amygdala volume in the MDD subjects. The post hoc data reduction methods yielded three principal components. Component 1 (interleukin-1 receptor antagonist, QA, and kynurenine) was significantly elevated in MDD participants versus the HCs, whereas component 2 (KA, tryptophan, and kynurenine) was positively correlated with hippocampal and amygdala volume within the MDD group. Our results raise the possibility that an immune-related imbalance in the relative metabolism of KA and QA predisposes to depression-associated dendritic atrophy and anhedonia.

Figure 1

Main branches of the kynurenine pathway. Each box represents a metabolite resulting from the oxidation of tryptophan. Putative neurotoxic metabolites are colored red whereas KA, which is putatively neuroprotective, is colored green. The black italicized text shows the enzymes that catalyze each step in the metabolic pathway. The blue and red stars indicate that the metabolite is able to cross the blood–brain barrier. The effects on NMDA receptor activity listed for some metabolites have been established in vitro, but the extent to which the metabolite concentrations achieve sufficiently high levels in vivo to impact the function of this receptor system remains unclear. NMDA, N-methyl-D-aspartate glutamate receptor. A full text version of this figure is available at Neuropsychopharmacology Journal online.

Figure 2

Scatterplots showing the distributions of kynurenine pathway metabolites and the principal components in MDD subjects and HCs. (a) Kynurenic acid. (b) 3-hydroxykynurenine. (c) Quinolinic acid. (d) Principal component 1 (IL-1RA, QA, and KYN). After controlling for sex, age, and BMI, the MDD group showed higher levels of component 1 relative to the healthy controls (F=4.5, p=0.039). (e) Principal component 2 (KA, TRP, KYN). (f) Principal component 3 (3HK, BDNF, and TRP). The MDD subjects are represented by blue squares and the HC subjects by pink circles. The black lines show the mean concentration and the error bars show the SEM. A full text version of this figure is available at Neuropsychopharmacology Journal online.

Figure 3

(a) Scatterplot showing the correlation between total hippocampal GM volume (X axis) and principal component 2 in subjects with MDD (β-weight=0.43, t=2.4, p=0.024). Hippocampal volumes are provided in cubic millimeters. The linear trendline is displayed in black. (b) Scatterplot showing the correlation between total amygdala GM volume (X axis) and principal component 2 in subjects with MDD (β-weight=0.38, t=2.2, p=0.038). Amygdala volumes are provided in cubic millimeters. (c) Scatterplot showing the correlation between total hippocampal GM volume (X axis) and the serum concentration of the ratio of kynurenic acid (KA) to quinolinic acid (QA) (Y axis) in subjects with MDD (β-weight=0.42, t=2.5, p=0.022). (d) Scatterplot showing the relationship between total amygdalar GM volume (X axis) and the ratio of serum KA to QA (Y axis) in subjects with MDD (β-weight=0.37, t=2.2, p=0.041).