Depression, specifically major depressive disorder (MDD), is a debilitating and prevalent psychiatric disorder characterized by persistent low mood, pervasive anhedonia, cognitive impairment, fatigue, and neurovegetative disturbances in sleep and appetite1. The etiology of MDD is multifactorial, involving complex, bidirectional interactions among neurochemical, neuroendocrine, and neuroimmune pathways2. Although classical monoaminergic models attribute depressive pathogenesis to functional deficits in serotonin (5-hydroxytryptamine, 5-HT), norepinephrine (NE), and dopamine (DA) neurotransmission3,4, monoaminergic dysregulation alone fails to fully account for the delayed therapeutic onset of antidepressants and the high rates of treatment resistance.

Compelling clinical and preclinical evidence indicates that chronic systemic and neuroinflammation serves as a central driver in the onset and maintenance of depressive pathology. Individuals with MDD consistently exhibit elevated systemic levels of pro-inflammatory cytokines, including interleukin-6 (IL-6), interleukin-1 beta (IL-1β), and tumor necrosis factor-alpha (TNF-α)5,6,7,8. Pro-inflammatory signaling disrupts blood-brain barrier permeability, activates microglial cells, and dysregulates the hypothalamic-pituitary-adrenal (HPA) axis via sustained glucocorticoid exposure and glucocorticoid receptor insensitivity2,8. Despite substantial progress in identifying candidate neuroimmune mediators, objective, laboratory-based molecular diagnostic tools remain unavailable in routine psychiatric practice. Clinical evaluations continue to depend almost exclusively on subjective psychometric questionnaires, such as the 9-item Patient Health Questionnaire (PHQ-9), which introduces diagnostic variability and complicates treatment stratification.

Epigenetic mechanisms, such as DNA methylation and histone modifications, mediate long-lasting alterations in gene expression in response to environmental stressors and early-life trauma. Methyl-CpG binding protein 2 (MECP2) is an essential epigenetic reader encoded on the X chromosome that binds to 5-methylcytosine (5mC) and 5-hydroxymethylcytosine (5hmC) residues in promoter and enhancer regions, orchestrating chromatin remodeling and transcriptional repression. Although originally characterized in the context of neurodevelopmental conditions such as Rett syndrome, MECP2 plays indispensable roles in adult synaptic plasticity, stress reactivity, and immunomodulation9,12,13.

At the molecular level, MECP2 directly interacts with the promoter of the IL-6 gene to suppress its transcription, thereby preventing aberrant hyperactivation of the IL-6/signal transducer and activator of transcription 3 (STAT3) inflammatory cascade9. Clinically, elevated IL-6 levels have been consistently reported in patients with depressive disorders and are thought to participate causally in mood regulation10,11. However, the precise regulatory interplay between MECP2 expression and IL-6 transcription in peripheral blood—and how their expression dynamics fluctuate with depressive disease severity—remains poorly defined in young cohorts. To bridge this knowledge gap, the present study quantified peripheral blood mRNA expression levels of MECP2 and IL-6, evaluated their correlation using nonlinear and rank-based statistical models, and examined their relationship with PHQ-9 clinical symptom dimensions in young Vietnamese individuals with depression.

A total of 34 young patients diagnosed with depressive disorders were recruited between October 2023 and October 2024. Initial screening was performed using the Vietnamese version of the PHQ-9 instrument, followed by comprehensive clinical evaluations and formal psychiatric diagnoses established by licensed psychiatrists at the Psychiatric Hospital of Ho Chi Minh City in accordance with the Diagnostic and Statistical Manual of Mental Disorders (DSM-5) criteria. Depressed participants were stratified into two clinical severity subgroups: mild-to-moderate depression (mildD group; PHQ-9 score 5–14; N = 19) and severe depression (SD group; PHQ-9 score ≥15; N = 15). All participants in the depression cohort were either antidepressant-naïve or had discontinued psychotropic medications for at least three months prior to blood collection.

An age-matched healthy control cohort consisting of 18 individuals (PHQ-9 score <4, absence of psychiatric or neurological disorders, no current medication use, and no underlying chronic medical conditions) was recruited concurrently. Due to sample volume and laboratory processing constraints, all 18 control samples were evaluated for MECP2 expression, while 8 evaluable control samples were available for IL-6 expression analysis. The study protocol was approved by the Institutional Review Board (IRB) under Decision Number 07/QĐ-IRB-VN01.017. All participants (and legal guardians for adolescent subjects under 18 years of age) provided written informed consent prior to study enrollment.

Venous blood samples (2.0 mL) were collected into EDTA-K2 anticoagulant tubes and processed within 2 h at room temperature (or within 24 h when maintained at 4–8°C). Whole blood was centrifuged at 3,000 rpm (approximately 1,000 × g) for 15 min at room temperature to fractionate blood components. Following plasma separation, the buffy coat layer (~100 μL) enriched with peripheral blood mononuclear cells (PBMCs) and leukocytes was isolated. Total RNA was extracted using the GeneJET RNA Purification Kit (Thermo Fisher Scientific, Waltham, MA, USA) under certified RNase-free conditions according to the manufacturer's protocol. Total RNA was eluted in 40 μL of sterile nuclease-free water, quantified by spectrophotometry (Eppendorf BioSpectrometer® Basic, Eppendorf, Hamburg, Germany), and stored at -30°C. Optical density ratios (A260/A280 and A260/A230) were recorded to assess RNA purity.

First-strand complementary DNA (cDNA) was synthesized from 10 μL of purified total RNA using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific) in a total reaction volume of 20 μL. Reverse transcription was carried out on a thermal cycler under the following conditions: 25°C for 10 min (primer annealing), 37°C for 120 min (cDNA synthesis), and 85°C for 5 min (enzyme inactivation), followed by rapid cooling to 4°C. Synthesized cDNA was either utilized immediately for qPCR amplification or stored in aliquots at -30°C.

Quantitative real-time PCR reactions were performed in duplicate on the Applied Biosystems™ 7500 Real-Time PCR System (Thermo Fisher Scientific). Each 10.0 μL reaction mixture contained 2.0 μL of cDNA template and 8.0 μL of reaction master mix consisting of TaqMan™ Universal Master Mix II (no UNG, Thermo Fisher Scientific) and specific TaqMan™ Gene Expression Assays: MECP2 (Assay ID: Hs05049079_g1), IL-6 (Assay ID: Hs00174131_m1), and small nuclear RNA U6 (snU6, Assay ID: 001973) as the endogenous reference gene. The amplification conditions were: initial denaturation at 95°C for 10 min, followed by 40 cycles of 95°C for 15 s and 60°C for 60 s. Baseline thresholds were set uniformly at 0.01 to ensure standardized cycle threshold (Ct) determination.

To validate assay robustness, standard curves were generated using serially diluted total RNA (two-fold serial dilutions across five concentration levels in duplicate) extracted from healthy human peripheral blood. The standard curves demonstrated excellent linearity and amplification efficiency across all targets: MECP2 (efficiency = 98.0%, R² = 0.991), IL-6 (efficiency = 95.3%, R² = 0.909), and snU6 (efficiency = 98.6%, R² = 0.991). Relative gene expression was calculated using the comparative Livak 2^-ΔΔCt method, where ΔCt = Ct_target - Ct_snU6 and ΔΔCt = ΔCt_patient - ΔCt_{control mean}.

Data distributions were evaluated for normality. Because variables demonstrated non-normal distributions, non-parametric tests were employed. Differences in relative expression levels between groups were assessed using the Kruskal-Wallis H-test. Bivariate associations were examined using Spearman’s rank correlation coefficient (ρ), Kendall’s rank correlation coefficient (τ), and distance correlation (dCor) with 1,000 bootstrap permutations to capture nonlinear associations. To model the potentially complex nonlinear relationships between MECP2 and IL-6, Generalized Additive Models (GAM) with penalized thin plate regression splines (s) were fitted. Statistical analyses and visualization were performed in R software (v4.3.0, R Foundation for Statistical Computing, Vienna, Austria). A two-tailed p-value < 0.05 was considered statistically significant.

A total of 34 young Vietnamese individuals with depression and 18 healthy controls were evaluated. Participants in the depression cohort had a mean age of 19.85 ± 2.41 years (range: 14–24 years), and the control group had a mean age of 19.61 ± 2.52 years, confirming comparable age and developmental baseline across cohorts. Based on clinical evaluation and PHQ-9 stratification, 19 patients were categorized into the mild-to-moderate depression group (mildD; PHQ-9: 5–14; including 5 inpatients and 14 outpatients/community cases) and 15 patients were categorized into the severe depression group (SD; PHQ-9 ≥15; all requiring hospitalization).

Total RNA yields ranged from 5.0 to 20.0 ng/μL, with an A260/A280 purity ratio of 1.80–2.43. The summary statistics of normalized ΔCt values for MECP2 and IL-6 in the depression cohort are detailed in Table 1. Individual demographic characteristics, PHQ-9 scores, and raw ΔCt values are compiled in Appendix 1.

Peripheral blood MECP2 mRNA expression was significantly higher in individuals with depressive disorders than in healthy controls (1.347-fold increase, Kruskal-Wallis H = 4.2357, df = 1, p = 0.0396; N = 52, comprising 34 patients and 18 controls; Figure 1A, 1B). Subgroup analysis revealed that this upregulation was driven primarily by patients with mild-to-moderate depression (mildD group), who exhibited a 1.454-fold elevation in MECP2 expression relative to controls (H = 6.9889, df = 1, p = 0.0082; N = 37).

In contrast, patients with severe depression (SD group) did not exhibit a statistically significant difference in MECP2 expression compared to healthy controls (1.22-fold, H = 0.5229, df = 1, p = 0.4696; N = 33). Direct comparison between the SD group and the mildD group showed a non-significant relative expression ratio of 0.84-fold (H = 0.8448, df = 1, p = 0.3580). Furthermore, MECP2 expression did not differ significantly between female and male depressed patients (female-to-male fold-change = 0.62, H = 2.9936, df = 1, p = 0.0836; N = 34).

Analysis of peripheral blood IL-6 mRNA expression revealed marked elevation in depressed patients compared with healthy controls (4.47-fold increase, H = 5.4675, df = 1, p = 0.0194; N = 42, comprising 34 patients and 8 controls; Figure 1C, 1D). The 95% confidence interval for the IL-6 fold-change in patients with depression ranged from 7.122 to 23.925. When stratified by clinical severity, the mildD group exhibited a 10.196-fold increase in IL-6 expression compared with controls (H = 8.5291, df = 1, p = 0.0035; N = 27).

Conversely, in patients with severe depression (SD group), IL-6 expression was not significantly elevated compared to healthy controls (1.58-fold, H = 1.3500, df = 1, p = 0.2453; N = 23). Notably, IL-6 expression in the SD group was 6.67-fold lower than that observed in the mildD group (H = 9.6364, df = 1, p = 0.0019; N = 34). A significant sex-dependent difference was also observed, with female depressed patients exhibiting 3.46-fold higher IL-6 expression than male patients (H = 4.0252, df = 1, p = 0.0448; N = 34).

To examine the functional relationship between epigenetic regulation and cytokine transcription, GAM regression and bivariate correlation analyses were conducted (Figure 2). Using ΔCt values, GAM regression revealed that IL-6 expression had a weak linear association with MECP2 (ΔCt_IL-6 ~ s(ΔCt_MECP2): deviance explained = 23.7%, adjusted R² = 0.205, p = 0.00354). Conversely, MECP2 exhibited a significant nonlinear relationship with IL-6 (ΔCt_MECP2 ~ s(ΔCt_IL-6): deviance explained = 38.9%, estimated degrees of freedom [edf] = 2.961, p = 0.0069).

When evaluating relative fold-change values, the nonlinear GAM model of IL-6 regressed on MECP2 accounted for 19.5% of the deviance (edf = 1.921, p = 0.0782), while MECP2 regressed on IL-6 accounted for 12.1% of the deviance (edf = 1.0, p = 0.0441). Correlation analyses across ΔCt metrics confirmed a statistically significant moderate positive correlation between MECP2 and IL-6 (Spearman's ρ = 0.4887, p = 0.0034; Kendall's τ = 0.3360, p = 0.0050; dCor = 0.4891). Similarly, fold-change values maintained a moderate positive correlation (ρ = 0.4877, p = 0.0034; τ = 0.3357, p = 0.0060; dCor = 0.4282, p = 0.0110), indicating concordant transcriptional co-regulation between MECP2 and IL-6.

In 32 depressed participants with complete and objective psychometric profiles (PHQ-9 scores: 5–27; 20 mild-to-moderate, 12 severe), we analyzed the association between peripheral ΔCt_IL-6 values and individual PHQ-9 items. Bivariate analyses revealed significant positive correlations between ΔCt_IL-6 (where higher ΔCt corresponds to lower relative mRNA expression) and overall depressive symptom burden (ρ = 0.439, p = 0.012; τ = 0.293, p = 0.021; dCor = 0.485, p = 0.008).

Item-level analysis demonstrated that this relationship was driven primarily by three core symptom domains:

In addition, feelings of worthlessness or guilt (Q6) displayed a significant association with ΔCt_IL-6 by distance correlation (dCor = 0.534, p = 0.004), although rank correlations did not reach statistical significance (ρ = 0.281, p = 0.119). No significant correlations were observed for fatigue (Q4), appetite changes (Q5), concentration difficulties (Q7), psychomotor agitation/retardation (Q8), or suicidal ideation (Q9).