👤 Ancha Baranova

Major depressive disorder (MDD) in adolescents is a critical public health concern, yet objective diagnostic biomarkers remain lacking. We conducted an integrative lipidomics study across human cohorts and a chronic unpredictable mild stress (CUMS) rat model. Targeted UPLC-MS/MS profiling was applied to a training cohort (95 MDD, 40 controls), and untargeted UPLC-HRMS profiling to an independent cohort (56 MDD, 37 controls). Candidate biomarkers were identified using univariate tests, partial least squares discriminant analysis, and three feature-selection methods (Boruta, LASSO, RFE), with predictive performance evaluated by cross-validation and external replication. Translational relevance was examined in CUMS rats through behavioral assays and lipidomic profiling of serum and brain tissues. Pathway enrichment and regression models explored metabolic context and clinical associations. In the training cohort, we found that 244 lipids were significantly altered, highlighting altered glycerophospholipid, glycerolipid, and sphingolipid metabolism. A 29-lipid panel achieved 90.4% cross-validation accuracy, while a reduced 7-lipid subset reached 94.8%. In the validation cohort, an 8-lipid panel achieved 71.2% accuracy, and a minimal 2-lipid set-LPA(18:2) and SPH(d16:1)-reached 72.1%. Cross-species analysis confirmed consistent downregulation of SPH(d16:1) in serum of both humans and rats, and of LPC(0:0/16:0) specifically in the rat prefrontal cortex. Regression analyses linked sex, age, and anxiety severity to lipid alterations. This cross-platform, cross-species study identifies reproducible lipid signatures of adolescent MDD, highlights SPH(d16:1) and LPC(0:0/16:0) as translational biomarkers, and implicates glycerophospholipid metabolism in MDD pathophysiology, providing a foundation for biomarker-guided diagnostics and therapeutics. Show less

Altered levels of human plasma metabolites have been implicated in the etiology of bipolar disorder (BD). However, the causality between metabolites and the disease was not well described. We performed a bidirectional metabolome-wide Mendelian randomization (MR) analysis to evaluate the potential causal relationships between 871 plasma metabolites and BD. We used DrugBank and ChEMBL to evaluate whether related metabolites are potential therapeutic targets. Finally, Bayesian colocalization analysis was performed to identify shared genomic loci BD and identified metabolites. Our MR results showed that six metabolites were significantly associated with a reduced risk of BD, including arachidonate (20:4n6) (OR: 0.90, 95% CI: 0.84-0.95) and sphingomyelin (d18:2/24:1, d18:1/24:2) (OR: 0.92, 95% CI: 0.87-0.96), while five metabolites were significantly associated with an increased risk of BD, including 1-palmitoyl-2-linoleoyl-GPE (16:0/18:2) (OR: 1.09, 95% CI: 1.05-1.13). However, our reverse MR analysis showed that BD was not associated with the levels of any metabolite. Additionally, the leave-one-out analysis revealed SNPs within chromosome 11 loci harboring MYRF, FADS1, and FADS2 as ones with the potential to influence partial causal effects. Druggability evaluation showed that 10 of the BD-related metabolites, such as sphingomyelin and cytidine, have been targeted by pharmacologic intervention. Colocalization analysis highlighted one colocalized region (chromosome 11q12) shared by 11 metabolites and BD and pointed to some genes as possible players, including FADS1, FADS2, FADS3, and SYT7. Our study supported a causal role of plasma metabolites in the susceptibility to BD, and the identified metabolites may provide a new avenue for the prevention and treatment of BD. Show less

Type 2 diabetes (T2D) is a chronic metabolic disorder that has high comorbidity with mental disorders. The genetic relationships between T2D and depression are far from being well understood. We performed genetic correlation, polygenic overlap, Mendelian randomization (MR) analyses, cross-trait meta-analysis, and Bayesian colocalization analysis to assess genetic relationships between T2D and depression, in the forms of major depressive disorder (MDD) and depressed affect (DAF). Then, the summary data-based MR (SMR) analysis was performed to prioritize genes contributing to MDD and to T2D from functional perspective. MDD-driven signaling pathways were constructed to understand the influence of MDD on T2D at the molecular level. T2D has positive genetic correlations both with MDD (r Our study points out that depression, in the forms of MDD and DAF, may increase the risk of T2D. Analysis of underlying genetic variation and the molecular pathways, connecting depression and T2D, indicate that the pathophysiological foundations of these two conditions have a notable overlap. Show less

Lipoprotein (a) [Lp(a)] is a highly heterogeneous lipoprotein particle promoting panvascular disease. Structurally, it consists of an LDL-like core covalently bound to apolipoprotein (a) [apo(a)]. Molecular determinants linking various genetic variants of apo(a) constituent of Lp(a) to vascular pathology remain incompletely defined. We have built a model allowing dissection which variations in LPA gene are functional, and which are mere associates of these functional variations. Copy number changes in kringle IV type 2 (KIV-2), together with a spectrum of single nucleotide polymorphisms (SNPs), regulate apo(a) size, expression, and function. These variants can be broadly categorized into Lp(a)-increasing, Lp(a)-lowering, and null alleles, with distinct prevalence across populations. Notably, risk alleles such as rs10455872 and rs3798220 account for substantial variance in circulating Lp(a) and confer elevated susceptibility to coronary artery disease, whereas splice-altering and nonsense alleles markedly reduce Lp(a) concentrations. The therapeutic implications of modifying circulating Lp(a) levels are profound. While conventional lipid-lowering therapies exert little influence on Lp(a), antisense oligonucleotides (pelacarsen) and small interfering RNA agents (olpasiran, SLN360) achieve robust Lp(a) reductions. Integrating genetic insights with structural modeling provides a framework to disentangle functional from proxy associations within LPA and neutralize the cardiovascular hazard conferred by elevated levels of Lp(a). Show less

MicroRNA-9-5p (miR-9-5p) is highly expressed in the brain and has been implicated in the risk of schizophrenia. We compared the expression levels of miR-9-5p in schizophrenia cases and healthy controls and evaluated whether regulatory targets of miR-9-5p are enriched in schizophrenia genome-wide risk genes. Literature-based analysis was conducted to construct molecular pathways connecting miR-9-5p and schizophrenia. We found that the expression levels of miR-9-5p were down-regulated in the peripheral blood of schizophrenia patients compared with those in healthy controls. miR-9-5p can regulate 24 out of the 1136 genome-wide risk genes of schizophrenia, which was higher than by chance (hypergeometric test P = 4.09E-06). The literature-based analysis showed that quantitative genetic changes driven by miR-9 exert more inhibitory (the IL1B, ABCB1, FGFR1 genes) than promoting (the INS gene) effects on schizophrenia, suggesting that miR-9 may protect against schizophrenia. Our results suggest that miR-9-5p deficiency may contribute to the development of schizophrenia. Show less

High risk of macrovascular complications in patients with type 2 diabetes mellitus (T2DM) is caused by insulin resistance and atherogenic dyslipidemia that may be genetically determined. The aim of this study was to assess the association of polymorphic genetic variants APOA5 (S19W/rs3135506), CETP (Taq1B/rs708272), PON1 (Q192R /rs662) and PPARG (Pro12Ala /rs1801282) with T2DM and macrovascular complications in patients with T2DM resident in Northwestern Russia. We examined 386 patients with T2DM and 199 healthy controls. Genotyping was performed by polymerase chain reaction followed by restriction analysis. The study revealed the protective role of allele 12Ala of PPARG gene against T2DM development (odds ratio [OR]=0.58; 95% confidence interval [CI] 0.39-0.85). B1B1 genotype of CETP was associated with increased risk of stroke in T2DM patients (OR=1.85; 95%CI1.07-3.21). RR genotype of PON1 was associated with increased risk of T2DM with stroke (OR=2.98; 95%CI1.01-8.84). According to study results Pro12Ala (rs1801282) variant of PPARG affected the risk of T2DM; polymorphic variants of CETP (Taq1B/rs708272) and PON1 (Q192R/rs662) contributed to the risk of macrovascular complications of T2DM. Show less