1. Vitamin D and high-density lipoprotein cholesterol (HDL-C) levels were lower in patients with depression than in healthy controls.

2. Vitamin D levels were negatively correlated with the severity of depressive symptoms.

3. Vitamin D and HDL-C may be inversely associated with depression.

Serum vitamin D exerts a broad range of effects by binding to the vitamin D receptor, which is expressed in nearly all tissues and cells, including brain regions implicated in neuropsychiatric disorders [1, 2]. Several studies have reported an association between vitamin D deficiency and psychiatric or mood disorders [3, 4, 5, 6]. However, findings from randomized controlled trials (RCTs) on the efficacy of vitamin D supplementation in treating depression remain inconsistent [7, 8, 9, 10]. This discrepancy may arise from overlooked factors, such as vitamin D dosage, individual age and sex, recurrent depressive episodes, the influence of antidepressant medications, and comorbid conditions affecting vitamin D absorption and metabolism. Consequently, depression in relation to circulating vitamin D levels requires further validation.

Vitamin D is a fat-soluble vitamin whose active form, 1,25-dihydroxyvitamin D [1,25(OH)2D], regulates cellular differentiation and biosynthetic pathways, including lipid biosynthesis, via mitochondrial vitamin D receptors [11, 12]. Observational studies suggest that vitamin D deficiency is associated with unfavourable blood lipid profiles, with inverse correlations observed between vitamin D levels and total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and triglyceride (TG) [13]. Lipids are highly abundant in neural tissues and play a crucial role in neurodevelopment [14]. Research has established connections between lipids and depression [15, 16, 17]. Notably, a Mendelian randomization study investigating lipids and depression indicated a potential causal relationship between triglycerides and depressive symptoms [18]. Lipid homeostasis contributes significantly to multiple interconnected processes governing mood regulation and suicidal behaviours, including serotonin neurotransmission [19, 20], neurogenesis [21], and neuroprotection against both excitotoxicity and systemic inflammation [22]. Consequently, disturbances in lipid metabolism are increasingly recognised as potential biomarkers for depression [23].

While the individual associations of vitamin D and lipid profiles with depression have been widely studied, their relative strength, comparative importance, and potential co-occurrence patterns within a homogeneous cohort of first-episode, drug-naïve patients remain unclear. Systematic evaluation of both biomarkers in this well-defined population helps to delineate a more comprehensive physiological profile of depression. In this study, we recruited a cohort of relatively young, drug-naïve, first-episode patients with depression without major physical comorbidities, along with carefully matched healthy controls. This design enables a clearer investigation of the relationship between vitamin D, lipids, and depression. Our work aims to clarify their association with the disorder and explore their potential role in its pathogenesis, thereby contributing novel insights for clinical strategy development.

Of the 686 participants initially recruited (198 healthy controls and 488 drug-naïve patients with first-episode depression), 94 were excluded for the following reasons: 22 could not provide informed consent, 46 did not meet the inclusion criteria, and 26 failed to complete all scale assessments and blood draws. Consequently, a final cohort of 180 healthy controls and 412 depressed patients was included in the study (see Fig. 2 for details).

Fig. 2.

Participants recruitment flowchart.

Demographic characteristics for both groups are detailed in Table 1.

The two groups did not differ significantly regarding sex, ethnicity, residence, age, or BMI. The overall sample medians were as follows: age, 22 years; BMI, 20.31 kg/m². Compared to the depression group, the healthy control group exhibited significantly lower PHQ-9 and HAMD-17 scores (p 0.001), and notably higher HDL-C and vitamin D levels (p = 0.001 and p = 0.021, respectively). However, no significant between-group differences were detected in LDL-C, TC, or TG levels (p $>$ 0.05).

Following stratification by sex, between-group comparisons were conducted. As demonstrated in Fig. 3A, male participants in the depression group exhibited significantly lower vitamin D levels compared to the healthy control group (p = 0.013). Fig. 3B reveals that female participants in the depression group showed significantly reduced HDL-C levels relative to the healthy control group (p 0.001). No other significant differences in lipid levels were observed between groups (p $>$ 0.05).

Fig. 3.

Violin plots of intergroup differences in vitamin D and lipid profiles by sex. (A) illustrates the differences in vitamin D and lipid levels between the healthy control group and the depression group in males, whereas (B) presents the corresponding comparisons in females. *p 0.05, **p 0.001.

As presented in Table 2, Spearman’s correlation analysis demonstrated that vitamin D levels in depressed patients exhibited significant negative correlations with both HAMD-17 scores (r = –0.179, p 0.001) and PHQ-9 scores (r = –0.180, p 0.001). Similarly, TG levels showed inverse relationships with depression severity scores (HAMD-17: r = –0.141, p = 0.004; PHQ-9: r = –0.120, p = 0.015). Conversely, no statistically significant associations were found between depression scores and HDL-C, LDL-C, or TC levels (p $>$ 0.05).

As presented in Table 3, following adjustment for sex, age, and BMI, the analysis revealed two significant protective associations: participants with elevated HDL-C levels exhibited a substantially reduced likelihood of depression compared to those with lower levels (adjusted OR = 0.317, 95% CI: 0.173–0.583, p 0.001). Higher vitamin D levels were similarly associated with decreased depression likelihood (adjusted OR = 0.950, 95% CI: 0.920–0.982, p = 0.002). These findings carry the implication that both HDL-C and vitamin D may serve as factors inversely associated with depression.

Our study comprised 412 patients with first-episode, drug-naïve depression and 180 healthy controls, with rigorous exclusion criteria for physical comorbidities, recurrent episodes, and any history of medication or treatments. The analysis revealed significantly higher levels of both HDL-C (which facilitates reverse cholesterol transport and clearance) and vitamin D in healthy controls compared to depressed patients. When stratified by sex, male patients in the depression group exhibited significantly lower vitamin D levels than healthy male controls. In contrast, female patients showed significantly reduced HDL-C levels compared to healthy female controls. Notably, vitamin D levels in females were consistently lower than in males. This disparity may be attributed to women’s greater use of cosmetics or sun protection measures, which can limit sunlight exposure, as well as their relatively lower engagement in outdoor activities compared to males.

Sex, age, and BMI, known determinants of lipid metabolism [24, 25], were included as covariates in our modelling analysis. The results suggested that HDL-C and vitamin D may confer a reduced risk of depression, implying their potential role as protective factors against the disorder. This observation is consistent with prior studies demonstrating that elevated lipid levels, particularly TC and LDL-C (which facilitate cholesterol delivery to peripheral tissues), correlate with more severe depressive symptoms [17, 18]. Conversely, a specific correlation exists between decreased HDL-C levels and symptoms of depression [26, 27] and may function as a predictive biomarker for depression severity [28]. Similarly, in patients with obsessive-compulsive disorder (OCD), lower HDL-C levels have been linked to increased suicidal ideation [29].

A bidirectional relationship exists between vitamin D and lipids. Firstly, as a fat-soluble vitamin, vitamin D relies on dietary lipids for dissolution and intestinal absorption. Secondly, through binding to its receptor and modulating gene expression [11, 12], vitamin D influences cholesterol homeostasis and fatty acid metabolism, thereby regulating blood lipid concentrations. Epidemiological studies have demonstrated that individuals with vitamin D deficiency frequently present with elevated TC and LDL levels, with this deficiency being particularly prevalent in obese populations [30]. Furthermore, vitamin D shows an inverse association with circulating lipid levels, and supplementation has been shown to exert beneficial effects on lipid profiles [31]. Interestingly, weight reduction has also been associated with improved vitamin D status [32]. In summary, the vitamin D-lipid interaction is reciprocal: while vitamin D requires lipids for absorption and storage, it simultaneously plays a crucial role in lipid metabolism. However, confounding factors must be considered. For example, decreased physical activity and reduced sunlight exposure, both common among patients with depression, may independently contribute to lower vitamin D levels alongside elevated TC and TG concentrations.

The roles of vitamin D and lipids in depression, however, involve greater complexity. Our study identified an inverse association between vitamin D levels and depressive symptoms, consistent with previous research findings [5]. The immune-inflammation hypothesis, linking dysregulated inflammation to neural circuit and neurotransmitter alterations in depression, has gained substantial support [33]. Notably, depression, dyslipidaemia, cardiovascular disease, and insulin resistance share common immune-inflammatory alteration [33, 34]. Vitamin D regulates inflammatory cytokine production and suppresses pro-inflammatory cell proliferation [35]. Through vitamin D receptor (VDR) signalling, it suppresses NLRP3-mediated immune responses [36]. Chronic inflammation is often associated with elevated cholesterol and triglyceride levels [37]. Studies indicate that in human monocytes, lipids activate the NLRP3 inflammasome. This activation promotes caspase-8 maturation through a cascade involving Lyn/Syk-dependent calcium influx and the generation of reactive oxygen species.

In addition to its anti-inflammatory effects, vitamin D modulates brain neuroplasticity, neurotransmitter biosynthesis, neuroprotection, and synaptic transmission by regulating neurotrophic factors and redox signalling pathways [38, 39, 40]. A study in pregnant rats demonstrated that vitamin D deficiency led to offspring with reduced cortical thickness and enlarged lateral ventricles [41, 42]. Furthermore, vitamin D deficiency alters neuronal morphology, impairing neurite outgrowth, branching, and periaqueductal length, which may contribute to speech and cognitive dysfunction [43, 44]. Research has also shown that a high-fat diet (HFD) disrupts hippocampal synaptic plasticity, reducing dendritic spine density and impairing long-term potentiation (LTP) [45, 46]. Chronic HFD exposure in adult mice leads to a reduction in key hippocampal neurotrophic factors, such as brain-derived neurotrophic factor (BDNF) [47, 48, 49]. Notably, these neurobiological alterations are frequently observed in depression [50, 51, 52, 53].

In summary, lipids are not only essential components of cell membranes but also play a pivotal role in nervous system function. Dysregulated lipid metabolism has been linked to neurodegenerative and psychiatric disorders, whereas vitamin D may indirectly support neural health and plasticity through its modulation of lipid metabolism. Therefore, maintaining an equilibrium between these factors is critical for optimal neurological function.

In our study, discrepancies were observed between correlation and regression analyses. Several factors may explain this inconsistency: First, correlation analysis exclusively measures linear relationships; when variables exhibit nonlinear associations (e.g., curvilinear patterns), the correlation coefficient may approach zero. Second, outliers may disproportionately affect correlation results, whereas regression analysis typically reduces their influence through least squares estimation or robust methods. Additionally, while bivariate correlation examines pairwise relationships, regression analysis simultaneously evaluates the effects of multiple independent variables on the dependent variable. Importantly, some variables may demonstrate modified influence patterns after covariate adjustment. In our study, regression models controlled for sex, age, and BMI, key covariates known to influence lipid metabolism, thereby strengthening the validity of our results.

Serum vitamin D and HDL-C levels were significantly lower in patients with depression compared to healthy controls. Moreover, a negative correlation was observed between serum vitamin D concentrations and the severity of depressive symptoms, indicating that lower vitamin D levels were associated with more severe clinical manifestations of depression. These findings collectively suggest that both vitamin D and HDL-C may be inversely associated with depression, likely involved in the underlying biological pathways related to depression.

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Study conception and design: LY, ZL. Data acquisition: LY, SM, ZWN, WW, LK, DX, ZLN, YW. Analysis and interpretation of data: LY, HL, WY, HC. Drafting: LY, ZL, HC. Revised the manuscript for intellectual content: LY, HC and ZL. All authors contributed to editorial changes in the manuscript. All authors reviewed and approved the final version. All authors have participated sufficiently in the work and agreed to be accountable for all aspects of the work.

This study was carried out in accordance with the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Renmin Hospital of Wuhan University (approval no.WDRY2020-K191). All patients were enrolled from the outpatient clinic via convenience sampling and provided written informed consent.

We gratefully acknowledge the participants and their families for participation in this study.

This study supported by grants from the Brain Science and Brain-like Intelligence Technology-National Science and Technology Major Project (grant number: 2021ZD0202000) and the National Key Research and Development Project of China (grant number 2024YFC3308400).

The authors declare no conflict of interest.