Major depressive disorder (MDD) is a common mental illness and one of the leading causes of disability worldwide (McIntyre and Lee, 2016; Otte et al., 2016). At least one-third of patients experience recurrent episodes, contributing to its chronic and relapsing course (Kessler and Bromet, 2013). The etiology of MDD involves a complex interplay of genetic and environmental factors, including common and rare genetic variants, as well as environmental stressors (Otte et al., 2016; Carvalho Silva et al., 2024; Kendall et al., 2021; Krishnan and Nestler, 2008; Schramm et al., 2020). Beyond these well-established contributors, growing evidence points to immune dysregulation as a key pathophysiological mechanism in MDD. Elevated levels of pro-inflammatory cytokines and chemokines, together with altered frequencies and increased activation of immune cells in peripheral blood (PB), have been reported in individuals with depression (Beurel et al., 2020; Miller, 2025; Miller et al., 2009). Within this inflammatory context, dendritic cells (DCs) are of particular interest due to their central role in linking innate and adaptive immune responses (Guilliams et al., 2016; Steinman, 2012). However, their cytokine-producing capacity and relationship with depressive symptoms and severity have not yet been investigated.

DCs are a heterogeneous group of peripheral immune cells specialized in antigen presentation (Segura, 2016). They recognize pathogen-associated (PAMPs) and damage-associated molecular patterns (DAMPs) that signal infection or cellular stress, thereby initiating adaptive immune responses by presenting antigenic peptides to naïve T cells (Guilliams et al., 2016; Mildner and Jung, 2014). Upon antigen uptake within a specific cytokine milieu, DCs upregulate MHC class II, costimulatory molecules, and chemokine receptors, enabling migration to lymph nodes to prime naïve T cells (Segura and Villadangos, 2009). In humans, conventional DCs (cDCs) comprise CD1c⁺ cDCs (DC2), the most abundant subset in PB, which primarily activate CD4⁺ T cells, and CD141⁺ cDCs (DC1), a smaller subset specialized in cross-presentation to CD8⁺ T cells and antiviral or antitumor immunity (Balan et al., 2019; Collin and Ginhoux, 2019). Furthermore, type 3 DCs (DC3s) were discovered in humans and mice, as an independent lineage of DCs arising from Ly6C⁺ monocyte-DC progenitors. DC3 have been attributed a high capacity for the in vitro induction of Th17 cells (Liu et al., 2023; Villani et al., 2017).

DCs produce a broad range of cytokines, including TNF, IL-1β, IL-6, IL-10, IL-12, and IL-23, many of which have been reported to be elevated in the PB of individuals with depression (Collin and Bigley, 2018; Farooq et al., 2017; Haapakoski et al., 2015; Kohler et al., 2017; Leite Dantas et al., 2021). Additionally, DCs exert anti-inflammatory functions by secreting IL-10 and transforming growth factor (TGF)-β, promoting regulatory T-cell (Treg) expansion, and upregulating indoleamine 2,3-dioxygenase (IDO), a key enzyme in tryptophan metabolism implicated in MDD (Leite Dantas et al., 2021; Mellor and Munn, 2004). Plasmacytoid DCs (pDCs) represent a minor subset of circulating DCs, and a small subset of those cells produces high amounts of type I interferons (IFNs) during antiviral responses (Ali et al., 2019; Reizis, 2019; Mann-Nuttel et al., 2021).

Sex has a significant impact on immune function, with males and females differing in specific innate and adaptive immune responses, which may contribute to their differential susceptibility to autoimmune diseases and viral infections (Klein and Flanagan, 2016). Growing evidence indicates that sex-specific biological mechanisms influence the vulnerability, course, and treatment outcomes of depression. Women show a higher prevalence of MDD and experience greater rates of residual symptoms and treatment resistance, potentially reflecting differences in genetic regulation of neural circuitry and plasticity, neurotransmitter systems, hormonal milieu, and immune modulation (Di Benedetto et al., 2024; Kuehner, 2003). Supporting this, a recent meta-analysis of inflammatory profiles in adults with MDD reported that women with depression exhibited significantly higher circulating levels of CRP and IL-6 compared to healthy women, whereas these associations were absent in men (Jarkas et al., 2024). These findings underscore the importance of accounting for biological sex in studies investigating cytokine production in depression.

Few studies have examined DCs in humans or rodent models of depressive disorder. A recent human study using immunophenotyping of circulating immune cells in MDD found that effector memory CD4⁺ T cells negatively predicted improvement of depressive symptoms after 3 months of treatment, whereas pDCs predicted a reduction in hopelessness (Stolfi et al., 2025). Another human study identified DCs as mediators linking social stress to reduced efficacy of immunogenic chemotherapy (Yang et al., 2019). This effect was driven by glucocorticoid-induced upregulation of TSC22D3 (GILZ), which impairs DC antigen presentation (Cohen et al., 2006; Vetillard and Schlecht-Louf, 2018). Elevated TSC22D3 expression correlated with higher cortisol, negative mood, and poor treatment response, while in mice, stress-induced glucocorticoid signaling increased TSC22D3 in tumor-infiltrating DCs suppressing type I IFN and T cell-derived IFN-γ responses (Yang et al., 2019). Depression-associated behaviors have also been shown to alter DC phenotype and function. Six days of social disruption stress induced a mature, pro-inflammatory profile in splenic DCs of subordinate mice, characterized by increased MHC class I, CD80, CD44, TNF, and IL-6 expression and glucocorticoid resistance (Powell et al., 2009). Adoptive transfer of these DCs enhanced antiviral T-cell responses and tumor immunity in naïve recipients (Powell et al., 2011). Our previous work demonstrated stress-induced alterations in DC phenotype and cytokine profiles in a mouse model of depression-like behavior. We found that chronic social defeat stress reduced splenic DC frequencies, while stress-susceptible mice exhibited enhanced cDC maturation characterized by elevated MHC class II and CD80 expression, increased Th17 cells, and higher corticosterone levels (Leite Dantas et al., 2021; Ambree et al., 2018).

In the present study, we characterized circulating DCs in individuals with MDD and HC, stratified by depression severity and sex, and examined their associations with depressive symptoms and their predictive value for disease severity. Our findings reveal alterations in DC populations and their effector cytokines in MDD, suggesting a contributory role in the immunopathophysiology of depression.

In this exploratory study, we characterized circulating DC subsets and their cytokine-producing profiles in individuals with MDD compared with HC, with a particular focus on sex-specific immune alterations. Although overall DC subset frequencies were largely comparable between groups, several notable cellular patterns emerged. We observed a reduction of circulating pDC numbers, particularly in severe MDD, and a selective increase in cDC and IL-23⁺ cDC percentages in MDD. Moreover, we identified a distinct sex-specific DC profile in this study. Females with MDD exhibited elevated circulating IL-23⁺ cDCs that correlated with depressive symptom items and severity, whereas males displayed higher proportions of TNF⁺, IL-1β⁺, and IL-6⁺ cDCs without significant associations with symptom items and severity. These findings suggest a sex-dependent peripheral DC pattern in MDD, with women showing enhanced circulating IL-23⁺ DCs and men exhibiting a bias toward a more proinflammatory cDC cytokine profile.

pDCs can promote immunity via type I IFN production or regulate immune responses by inducing Treg cells. Depending on the context, these functions may confer host protection or contribute to pathology (Bauer et al., 2016). The observed reduction in pDCs is consistent with stress hormone-related suppression of type I IFN-producing pDCs in PB in humans (Lepelletier et al., 2010; Shodell et al., 2003; Shodell and Siegal, 2001). Human pDCs are highly glucocorticoid-sensitive and undergo glucocorticoid-induced apoptosis. Likewise, hyperinflammatory states (e.g., severe viral infection or critical illness) show lower circulating pDCs and dampened type I IFN activity despite elevated pro-inflammatory cytokines (Grimaldi et al., 2011; Hadjadj et al., 2020; Lu et al., 2025). It is thus possible that chronic stress in MDD may contribute to a contraction of the pDC/type I IFN compartment, as seen in stress-exposed human settings. In excluding a link between pDCs and depressive items and severity, our findings differ from a recent study showing that pDCs predicted improvement in hopelessness following 3 months of treatment of acute depression (Stolfi et al., 2025). Thus, pDC alterations may be state-dependent and modulated by inflammation or pharmacological treatment. The expansion of IL-23⁺ cDCs, on the other hand, represents a more specific immune profile in MDD. Since cytokine expression on a single cell level was equivalent in all groups, our finding reflects an increase in the proportion of IL-23 producing cDCs rather than enhanced cytokine production capacity on a single cell level. This selective increase of peripheral IL-23⁺ cDCs in depression points to an altered differentiation or activation of cDCs in the context of chronic stress and low-grade inflammation.

Sex-stratified analyses revealed that these immune alterations occurred primarily in depressed females. After controlling for age, BMI, smoking, and depression severity, we found that pDCs were reduced and IL-23⁺ cDCs were elevated in the PB only in female patients, compared to sex-matched controls. In contrast, males with MDD showed increased frequencies of TNF⁺, IL-1β⁺, and IL-6⁺ cDCs compared with female patients, but no difference in circulating IL-23⁺ cDCs. Cluster and discriminant analyses confirmed that DC profiles segregated primarily by sex. The lower IL-23⁺ cDC frequencies in healthy women compared with the other groups suggest that depression induces a shift from a homeostatic state comprising a low IL-23 producing DC profile to a more IL-23-skewed phenotype in cDCs in MDD that is specific to females.

Several underlying immune mechanisms may explain altered IL-23⁺ cDC frequencies in depressed females compared to sex-matched controls, such as increased expansion, enhanced cell survival and/or altered redistribution of functional DC subsets. Accordingly, sex hormones profoundly influence DC differentiation and responsiveness. Estrogens enhance NF-κB-dependent cytokine transcription and TLR signaling, whereas progesterone and androgens have been shown to suppress DC activation (Butts et al., 2007; Laffont et al., 2014; Mackern-Oberti et al., 2017). Under chronic stress and elevated glucocorticoids, as observed in MDD, this can lead to a paradoxical immune state where some DC functions are suppressed while others, such as IL-23 production, are amplified. Depressed women may therefore exhibit an exaggerated IL-23 response due to a convergence of estrogen signaling, glucocorticoid exposure, and stress-induced inflammatory cues. Furthermore, the IL-23/Th17 axis is biased toward females in autoimmune disorders such as multiple sclerosis and atopic dermatitis, both of which involve DC-derived IL-23 as a key pathogenic driver (Li et al., 2019; Vaknin-Dembinsky et al., 2006). Activation of IL-23/Th17 axis as well as elevated IL-23 blood levels have been described in depression (Daray et al., 2024; Moulton et al., 2025; Galecka et al., 2021), albeit the cellular source of IL-23 has remained unclear. Our data suggest that circulating cDCs are a cellular source of this cytokine. The interaction between DCs and DAMPs may also contribute to this effect. Chronic psychological stress and neuroinflammation elevate DAMPs such as high-mobility group box 1 (HMGB1), S100 proteins, and heat shock proteins, which signal through TLR2, TLR4, or RAGE on cDCs (Ma et al., 2024). Although the role of DAMPs in MDD is still being elucidated, growing evidence supports their involvement in its pathophysiology (Franklin et al., 2018; Peng et al., 2025; Serna-Rodriguez et al., 2022). These signals, especially when combined with cytokines such as GM-CSF, can induce IL-23 expression in DCs (Poppensieker et al., 2012). Our data are consistent with the idea that in depressed women, cDCs are more readily activated by DAMP-mediated pathways, whereas in healthy women, these cells remain in a more quiescent state. This hypothesis is supported by our previous work demonstrating that GM-CSF and CCR4 signaling promote IL-23 production in DCs in a murine model of CNS autoimmunity (Poppensieker et al., 2012; Ruland et al., 2017; Scheu et al., 2017). Given that some, though not all, studies have reported elevated GM-CSF levels in MDD (Annam et al., 2024; Hemmati et al., 2019), this pathway represents a plausible mechanism. Also, osteoprotegerin (OPG), a soluble member of the tumor necrosis factor receptor superfamily, has been shown to affect DC survival and cytokine production. LPS treatment of murine BM-derived OPG knock out DCs resulted in increased survival and elevated production of cytokines (TNF, IL-12, IL-23) of these cells compared to wildtype DCs, while adding OPG to the culture reduced their survival and cytokine production (Chino et al., 2009). Albeit, elevated OPG levels related to calcification and inflammation have been associated with severe mental disorders (Hope et al., 2010), postmenopausal women with MDD had decreased OPG levels compared to HC (Atteritano et al., 2013). Considering the mean age of 55 years in the female study group presented here, and the fact that around 55% of the MDD patients were either undergoing menopause or had reached the postmenopausal phase, it is possible that reduced OPG levels in female patients may contribute to the observed increased frequency of IL-23⁺ cDCs. Taken together, these findings suggest that the increase in IL-23⁺ cDCs in women with MDD reflects a biologically specific consequence of chronic stress, sex hormone modulation, and innate immune receptor activation.

The correlation analyses further strengthen the clinical relevance of this immune phenotype. IL-23⁺ cDC frequencies were positively associated with multiple depressive symptom domains and overall severity scores on both the BDI-II and IDS-C scales in females, whereas no significant associations were seen in males. Multivariate modeling further revealed that IL-23⁺, TNF⁺, and IL-10⁺ cDCs, together with BMI and age, predicted over 51% of the variance in MDD severity, as measured by the IDS-C sum score, in women, but not in men. This sex-specific linkage between IL-23⁺ cDCs and MDD severity suggests that these cells may not merely be markers but functional contributors to mood dysregulation. IL-23 and its downstream effector, IL-17, can influence the brain through microglial activation, increased blood–brain barrier permeability, and altered monoaminergic signaling, all mechanisms implicated in major depression (Gaffen et al., 2014). The pattern observed here suggests that in women with MDD, IL-23⁺ cDCs may reflect or promote a Th17-driven inflammatory milieu linked to affective dysregulation. The established association between Th17 activation and depressive symptom severity (Moulton et al., 2025; Beurel et al., 2013; Beurel and Lowell, 2018; Kim et al., 2021) further supports this interpretation. Moreover, the increase in CD11c⁺ DCs observed in severe MDD may influence chemokine signaling relevant to disease mechanisms. Consistent with this, we recently reported elevated plasma CCL17 levels, a chemokine primarily secreted by cDCs, in individuals with MDD (Freff et al., 2022). Peripheral blood IL-23⁺ cDCs may therefore reflect central immune processes driving depressive symptomatology in women or rather promote peripheral chronic inflammation with detrimental effects in MDD.

Our results have several translational implications. They add cytokine-producing DC phenotypes to the expanding list of immune alterations implicated in MDD. They also underscore that sex is not simply a covariate but a key biological stratifier. From a biomarker perspective, IL-23⁺ cDC frequencies could help define inflammatory subtypes of depression, potentially guiding immune-targeted interventions. Also from a public health perspective, implementing IL-23⁺ cDC screening could support assessment of severity of depression in females in the future. The findings also align with the concept of immuno-metabolic depression, a subtype characterized by low-grade inflammation and metabolic dysregulation with obesity, dyslipidaemia, and insulin resistance. This recently proposed concept of immuno-metabolic depression affects about 20–30% of patients (Penninx et al., 2025). Increased body weight in immuno-metabolic depression may influence IL-12 and IL-23 production in DCs; indeed, Sumarac-Dumanovic et al. reported upregulation of the IL-23/IL-17 axis in obese women, while weight loss reduced leptin and inflammatory cytokines, particularly IL-23 (Landgren et al., 2023; Sumarac-Dumanovic et al., 2009). Although IL-23⁺ cDCs did not correlate with BMI in our cohort, increased body weight and metabolic stress could amplify this inflammatory pathway, particularly in women, providing a link between metabolic state and immune activation in depression.

In this study, we focused on patients with MDD, however, other psychiatric disorders such as bipolar disorder (BD) and schizophrenia, are also frequently associated with immune alterations. A meta-analysis has shown that these mental disorders share common immune features, including altered peripheral blood cytokine profiles. In particular, schizophrenia and BD have been associated with elevated peripheral IL-23 blood levels (Debnath and Berk, 2017; Li et al., 2015), suggesting that DCs may contribute to immune dysregulation not only in MDD but also in other psychiatric disorders.

This study has several limitations. The sample size was moderate, and its cross-sectional design limits causal inference. Genetic risk factors for the development of MDD, such as SLC6A4, BDNF, and CACNA1c, the ladder coding for the pore-forming α1C subunit of the L-type calcium channel (LTCC) Ca_V1.2, were not available for this study group. This is relevant because expression of Ca_V1.2 channels by DCs has been shown to regulate, e.g., cytokine production and MHC-class II expression (Boller et al., 2025). Moreover, variants in SLC6A4 and BDNF have been associated with altered IL-6 and BDNF levels in MDD, highlighting the potential contribution of genetic risk factors to immune alterations in depression (Colle et al., 2017; Su et al., 2009). In addition, circulating DCs were analyzed rather than tissue-resident or CNS-infiltrating cells, limiting conclusions about central mechanisms. Pharmacological treatments varied across participants, and potential drug effects on DC function cannot be fully excluded. Moreover, although cluster analyses indicated distinct sex-specific immune profiles, overall separation between MDD and HC was moderate, suggesting that DC alterations represent one component within a broader immune landscape in depression.

Future research should replicate these findings in larger, medication-free cohorts and incorporate longitudinal sampling to examine whether IL-23⁺ cDC numbers fluctuate with symptom changes or treatment response. In addition, our study highlights the importance of investigating DCs in the context of other mental disorders, both as potential biomarkers and potential future treatment targets. It further underscores the need for sex-stratified research in this field. Ultimately, prospective studies in larger cohorts within translational frameworks, such as the DFG funded Collaborative Research Centre 393 (SFB/TRR 393), could identify immune and metabolic predictors of illness trajectories and inform the development of sex-specific immunomodulatory strategies for affective disorders (Kircher et al., 2025).

In conclusion, our findings demonstrate that cytokine-producing DC subsets are altered in MDD in a sex-specific manner. These results identify IL-23-producing DCs as a potential immunological bridge between female sex, immune dysregulation, and depressive symptomatology, supporting the notion that sex-tailored immunomodulatory approaches may hold promise in the treatment of depression.