Introduction

Psoriasis is a chronic inflammatory skin disease characterized by excessive keratinocyte proliferation and abnormal differentiation, and immune cell infiltration into the epidermis^1,2. The interplay between keratinocytes and immune cells, including dendritic cells (DCs) and T cells, drives disease pathogenesis, with the IL-23/IL-17A axis playing a central role^3–5. IL-23, primarily secreted by DCs, promotes the differentiation of IL-17-producing γδT cells and Th17 cells^3,4,6,7. IL-17, in turn, binds keratinocyte receptors, triggering epidermal hyperproliferation and the release of chemokines and antimicrobial peptides (AMPs)^5,8,9. These factors recruit immune cells to psoriatic lesions, amplifying IL-23 production and establishing a self-perpetuating inflammatory loop^4,9–11. FDA-approved biologics, such as secukinumab, ixekizumab, and guselkumab, target IL-23 or IL-17A to mitigate symptoms^1,10,12. Additionally, IL-1β from DCs enhances Th17 activation^13,14, while plasmacytoid DCs (pDCs) infiltrate lesions and secrete type I interferons and proinflammatory cytokines¹⁵. Notably, AMPs like LL37 form complexes with self-DNA/RNA from damaged cells, activating pDCs via TLR7 or TLR9 and exacerbating inflammation^16–18.

Imiquimod (IMQ), also known as R837, is an imidazoquinoline derivative with antiviral and immunomodulatory properties. IMQ-containing cream is clinically used to treat actinic keratoses, superficial basal cell carcinoma, and genital warts. However, repeated application of IMQ to mouse skin induces psoriasis-like dermatitis, providing a widely used experimental model of psoriasis¹⁰. IMQ was originally characterized as a Toll-like receptor 7 (TLR7) agonist that activates MyD88-dependent NF-κB pathway, leading to proinflammatory cytokine production¹⁹. Beyond its canonical TLR7 signaling, IMQ also exerts TLR-independent effects^20–22. It directly binds to mitochondrial complex I, promoting mitochondrial reactive oxygen species (ROS) production in DCs. These mitochondrial perturbations drive NLRP3 inflammasome activation, which mediates IL-1β maturation and pyroptotic inflammation. Nevertheless, IMQ-induced skin inflammation remains only partially attenuated in TLR7- or NLRP3-deficient mice, suggesting the existence of additional, unexplored pathways contributing to psoriatic inflammation^23–25.

The endoplasmic reticulum (ER) is essential for maintaining protein quality control, lipid metabolism, and intracellular calcium homeostasis²⁶. In response to environmental insults such as oxidative stress, infection, or UV irradiation, ER function can be perturbed, triggering the unfolded protein response (UPR) through IRE1α, PERK, and ATF6 signaling. While primarily cytoprotective, sustained or dysregulated UPR signaling contributes to inflammation and has been implicated in chronic inflammatory diseases^27,28. The ER also forms physical contact sites with mitochondria, referred to as mitochondria-associated ER membranes (MAMs), which play a critical role in calcium exchange and mitochondrial integrity²⁹. Excessive or dysregulated ER-mitochondria tethering has been associated with mitochondrial calcium overload, ROS generation, and organelle dysfunction. Previous studies have suggested a role for ER stress in concert with TLR in worsening psoriasis²². However, the upstream events linking IMQ to organelle stress responses remained unclear.

Gelsolin is a multifunctional, actin-binding protein that exists in two major isoforms: cytoplasmic (cGelsolin) and secreted plasma (pGelsolin)^30,31. It regulates actin filament assembly and disassembly in a calcium-dependent manner and plays critical roles in cell morphology, migration, and signal transduction. Recent studies have highlighted its importance in immune regulation. Gelsolin facilitates antigen presentation by modulating cytoskeletal dynamics in DCs³². Furthermore, it has been shown to suppress NLRP3 inflammasome activation by stabilizing cytosolic ion homeostasis and reducing mitochondrial stress³³. Gelsolin also binds intracellular calcium, potentially buffering calcium overload during stress responses^30,31,34. Despite these regulatory functions, the role of Gelsolin in psoriasis and its interaction with ER stress pathways remain poorly characterized.

Here, we demonstrate that IMQ induces ER stress and NLRP3 inflammasome activation, thereby driving psoriasis-like inflammation. Moreover, we identify Gelsolin as a novel IMQ-binding protein that suppresses UPRs. Loss of Gelsolin exacerbates mitochondrial dysfunction, and its reduced expression correlates with psoriasis severity, highlighting both Gelsolin and UPRs as potential therapeutic targets.

Results

IMQ Induces Cytosolic and Mitochondrial Ca²⁺ flux, leading to Mitochondrial Dysfunction and Inflammasome Activation in DCs

To elucidate how IMQ activates innate immune signaling, we first examined intracellular Ca²⁺ dynamics in LPS-primed bone marrow-derived dendritic cells (BMDCs). Live-cell imaging with Fluo-8 revealed that IMQ, but not the related TLR7 agonist resiquimod (RSQ), induced a rapid increase in cytosolic Ca²⁺ concentration (Figure S1). This response was retained in MyD88-deficient BMDCs (Figure 1A), indicating a MyD88-independent mechanism. Flow cytometric analysis using the mitochondrial Ca²⁺ indicator Rhod-2 showed that IMQ also promoted Ca²⁺ transfer toward mitochondria (Figure 1B), suggesting that cytosolic Ca²⁺ elevation is accompanied by enhanced ER–mitochondria communication. IMQ stimulation led to an increase in intracellular reactive oxygen species (ROS) levels, whereas treatment with 2-APB, an inhibitor for IP₃ receptor, significantly reduced ROS production (Figure 1C). Consistently, 2-APB also suppressed IL-1β release, cell death, and Caspase-1 cleavage (Figure 1D, 1E, and 1F), indicating that intracellular Ca²⁺ mobilization contributes to inflammasome activation and pyroptosis. In addition, IMQ-induced IL-1β production and cell death induction was abrogated by NLPR3 deficiency (Figure S2), demonstrating IMQ promotes NLRP3 inflammasome activation. To assess mitochondrial responses, we monitored mitochondrial membrane potential and mitochondrial ROS generation. IMQ, but not RSQ, caused depolarization and increased mitochondrial ROS levels (Figure 1G, and 1H). Treatment with the antioxidants glutathione ethyl ester (GSHEE), NAC, or PDTC significantly reduced IMQ-induced IL-1β release and cell death (Figure 1I, and 1J), and the mitochondrial-targeted antioxidant MitoQ showed a similar inhibitory effect (Figure 1K). Together, these results demonstrate that IMQ elevates intracellular and mitochondrial Ca²⁺ and ROS levels, both of which are closely associated with NLRP3 inflammasome activation in dendritic cells.

Figure 1.

IMQ Promotes UPR and Facilitates IL-23 Expression in DCs

Given that IMQ increased cytosolic and mitochondrial Ca²⁺ and promoted ER–mitochondria communication, we next examined whether ER homeostasis is affected by IMQ stimulation. A proximity ligation assay revealed that IMQ enhanced the formation of MAMs, indicating strengthened ER–mitochondria contacts (Figure 2A). To elucidate the role of IMQ in modulating ER bioprocesses, we investigated its impact on unfolded protein response (UPR), a key cellular mechanism activated under ER stress, by quantifying the expression of UPR-related markers. We first examined the spliced form of Xbp1 mRNA (Xbp1s mRNA), a critical indicator of ER stress generated through IRE1α. IMQ stimulation significantly upregulated Xbp1s mRNA expression in both WT and MyD88-deficient BMDCs, demonstrating robust induction of ER stress through a MyD88-independent pathway (Figure 2B). This finding suggests that IMQ triggers ER stress independently of the canonical TLR signaling. By contrast, RSQ stimulation did not induce Xbp1s mRNA expression in BMDCs, indicating that the ER stress response is specific to IMQ and not a general feature of TLR7/8 agonists (Figure S3A). To confirm the involvement of IRE1α in IMQ-induced Xbp1s expression, we pretreated BMDCs with 4μ8c, a selective inhibitor of IRE1α RNase activity. This pretreatment significantly suppressed IMQ-induced Xbp1s mRNA expression, underscoring the critical role of IRE1α in mediating this ER stress response (Figure S3B).

Figure 2.

In addition to the IRE1α-Xbp1 axis, we explored other branches of the UPR to gain a comprehensive understanding of effects of IMQ on ER stress signaling. IMQ stimulation markedly increased the phosphorylation of PERK and its downstream target, eIF2α, in BMDCs, but not RSQ (Figure 2C, and S3C). Furthermore, IMQ treatment upregulated the expression of UPR-associated genes, including Dnajb9 (also known as Erdj4) and Ddit3 (also known as Chop), which are hallmarks of ER stress responses, in a Myd88-independent manner (Figure 2D). Collectively, these results demonstrate that IMQ potently induces multiple UPR pathways through ER stress, including IRE1α and PERK, independently of MyD88 signaling.

Previous studies have reported that ROS induces ER stress³⁵. To investigate whether mitochondria are required for ER stress induction upon IMQ stimulation, we examined the effects of mitochondrial depletion. Treatment with the mitochondrial uncoupler carbonyl cyanide m-chlorophenyl hydrazone (CCCP) combined with ectopic expression of the ubiquitin E3 ligase Parkin induces robust mitophagy, leading to effective mitochondrial clearance from cells. We generated Parkin-expressing RAW264.7 cells and cultured them with CCCP to induce mitochondrial depletion (Figure S4A and S4B). In both WT and mitochondria-depleted cells, IMQ stimulation increased the expression of ER stress-related genes (Figure S4C). These findings suggest that IMQ activates ER stress through a mechanism that is distinct from its effects on mitochondria.

Since IL-23 is a critical inflammatory cytokine in the pathogenesis of psoriasis^10,36, we next evaluated the impact of IMQ treatment on Il23a expression in BMDCs. IMQ treatment markedly induced Il23a expression (Figure 2D), whereas this induction was completely abolished in MyD88-deficient cells. Interestingly, pretreatment with 4μ8c significantly suppressed IMQ-induced Il23a transcription but not RSQ-induced Il23a transcription (Figure 2E), suggesting that UPR enhances IMQ-mediated Il23a expression. A luciferase assay using an Il23a promoter construct (−2,030 to +20 bp) revealed that Il23a promoter activity was increased by XBP1s or MyD88 overexpression and synergistically enhanced by their co-expression (Figure 2F, and S5). Additionally, RSQ-induced Il23a expression, which was induced in an IRE1α-independent manner, was enhanced by thapsigargin (Tg), an ER stress inducer, whereas Il6 expression was unaffected (Figure 2G). These results suggest that IMQ activates TLR and ER stress pathways via distinct mechanisms, which converge to cooperatively enhance Il23a expression.

IMQ Drives Psoriasis-Associated Gene Expression in Keratinocytes via the IRE1α-XBP1s Axis of the UPR

In psoriatic lesions, keratinocytes exhibit hyperproliferation and aberrant differentiation, contributing to the inflammatory milieu of psoriasis^1,2. To investigate whether IMQ induces UPRs in keratinocytes, we stimulated primary mouse keratinocytes with IMQ and assessed key UPR markers. RT-qPCR analysis revealed that IMQ significantly upregulated mRNA expression of Xbp1s, Hspa5, and Ddit3, indicating robust activation of the UPRs (Figure 3A). Concurrently, IMQ stimulation increased cytoplasmic Ca²⁺ influx, as measured by Fluo-8 staining, and enhanced phosphorylation of PERK and eIF2α, further confirming UPR activation (Figure 3B, and 3C).

Figure 3.

To explore the broader impact of IMQ-induced UPRs on keratinocyte gene expression, we performed RNA-sequencing (RNA-seq) analysis on IMQ-stimulated primary mouse keratinocytes, with or without pretreatment with 4μ8c. Differential expression analysis identified 286 upregulated and 459 downregulated genes in IMQ-treated keratinocytes compared to controls (Figure 3D, and Tables S1, and S2). Notably, 37 (18%) of the upregulated genes overlapped with those previously reported in human psoriatic lesions, suggesting that IMQ recapitulates key molecular features of psoriasis in keratinocytes (Figure 3E, and Tables S3, and S4). Gene ontology (GO) analysis of the upregulated genes revealed significant enrichment of pathways related to cellular metabolism and proliferation, consistent with the hyperproliferative phenotype of psoriatic keratinocytes (Figure 3F, and Table S5). Pretreatment with 4μ8c significantly reversed IMQ-induced gene expression changes, indicating that the IRE1α-XBP1s axis is critical for these transcriptional responses (Figure 3G, and Table S6).

To validate these findings, we performed RT-qPCR on selected psoriasis-associated genes, including Ccl20, Nr4a3, and Defb14 (the mouse ortholog of human β-defensin 3, hBD3). IMQ significantly induced the expression of these genes, and 4μ8c pretreatment effectively reversed this induction, confirming the dependency on IRE1α-mediated UPR signaling (Figure 3H). These results collectively demonstrate that IMQ activates the UPRs in keratinocytes, particularly through the IRE1α-XBP1s axis, to drive the expression of psoriasis-associated genes, thereby contributing to the inflammatory and hyperproliferative features of psoriatic lesions.

UPRs Exacerbate IMQ-Induced Dermatitis, and Ca²⁺ Flux Inhibition Ameliorates It

To investigate the in vivo role of the UPR in IMQ-induced psoriasis-like dermatitis, we evaluated the expression of ER stress-related genes in a mouse model. We quantified mRNA levels of key UPR markers in ear tissue following topical IMQ application. Our results revealed significant upregulation of Xbp1s, Hspa5 (also known as BiP or Grp78), and Hsp90b1 (also known as Grp94) mRNA in IMQ-treated ears compared to untreated controls, indicating robust induction of ER stress in psoriatic lesions (Figure 4A). To explore the downstream consequences of IMQ-induced ER stress, we measured mitochondrial DNA (mtDNA) levels, a marker of mitochondrial damage, in both lesional tissue and systemic circulation. Notably, mtDNA levels were significantly elevated in psoriatic lesions, consistent with mitochondrial dysfunction, but remained unchanged in serum, suggesting that mtDNA release is localized to the site of inflammation (Figure 4B).

Figure 4.

To further elucidate the functional role of UPR in IMQ-induced dermatitis, we pretreated mice with tunicamycin, a pharmacological inducer of ER stress, prior to IMQ application. Tunicamycin pretreatment markedly exacerbated IMQ-induced ear thickening and epidermal hyperplasia, as assessed by histological analysis (Figure 4C, 4D, and 4E). Additionally, tunicamycin significantly enhanced the expression of psoriasis-associated genes, including Defb14 (encoding mouse BD14 [mBD14]), in IMQ-treated ears, indicating that heightened ER stress amplifies the inflammatory and psoriatic phenotype (Figure 4F). To examine the role of calcium signaling in this context, we treated mice with carbachol, an inositol 1,4,5-trisphosphate receptor (IP₃R) activator that promotes Ca²⁺ release from the ER. Carbachol treatment significantly worsened IMQ-induced dermatitis, as evidenced by increased ear thickening, epidermal hyperplasia and enhanced expression of psoriasis-related inflammatory genes, further supporting the role of Ca²⁺-mediated ER stress in disease exacerbation (Figure S6).

Conversely, to assess whether inhibiting Ca²⁺ flux could reduce IMQ-induced pathology, we pretreated mice with 2-APB. Strikingly, 2-APB pretreatment significantly reduced IMQ-induced ear thickening and epidermal hyperplasia, as determined by histological analyses (Figure 5A, 5B, and 5C). Moreover, 2-APB treatment markedly decreased the expression of inflammatory and psoriasis-related genes in IMQ-treated ears (Figure 5D). Collectively, these findings demonstrate that Ca²⁺-triggered ER stress, coupled with mitochondrial damage, plays a pivotal role in driving IMQ-induced psoriasis-like dermatitis. Furthermore, inhibition of Ca²⁺ flux via IP₃R blockade ameliorates disease severity, highlighting a potential therapeutic strategy for targeting ER stress in psoriasis.

Figure 5.

Mitochondrial DNA and mBD14 Complex Activate pDC via TLR9

Mitochondrial damage promotes the release of mtDNA into the cytoplasm, where it eventually activates the NLRP3 inflammasome^37,38. We found that IMQ stimulation significantly increased mtDNA levels in both mitochondrial and cytoplasmic fractions, while cytosolic nuclear DNA (nDNA) remained unchanged (Figure 6A). Furthermore, 2-APB treatment significantly suppressed increment of cytoplasmic mtDNA induced by IMQ (Figure 6B). IMQ also increased the amount of oxidized DNA in cytosolic fraction, as measured by dot blot analysis (Figure 6C), and enhanced the presence of mitochondria in the extracellular milieu (Figure 6D). Furthermore, extracellular mtDNA levels were elevated in an NLRP3-dependent manner (Figure 6E). These results suggest that IMQ increases cytoplasmic mtDNA levels, which contribute to NLRP3 inflammasome-dependent pyroptosis and its subsequent release into the extracellular milieu.

Figure 6.

AMPs such as human LL37 and BD3 exacerbate psoriasis by binding to extracellular self-derived nucleic acids and activating pDCs via TLR9^16–18. We hypothesized that DNA released from IMQ-stimulated DCs forms a complex with mBD14. Indeed, pDCs incubated with DNA from IMQ-stimulated BMDCs and recombinant mBD14 exhibited significantly higher Il6 mRNA expression than those treated with DNA or mBD14 alone (Figure 6F). In contrast, DNA from unstimulated BMDCs failed to induce Il6 expression, even in the presence of mBD14 (Figure 6F).

Since oxidized mtDNA activates NLRP3 in the cytoplasm^37,38, we further hypothesized that it also enters the extracellular space and, together with mBD14 secreted by keratinocytes, activates pDCs. To test this, we treated pDCs with synthetic mtDNA (syn-mtDNA) with or without recombinant mBD14 and measured cytokine expression. Syn-mtDNA alone, even with mBD14, did not induce TNF or IL-6 expression (Figure 6G). However, co-treatment with the oxidized form (syn-ox-mtDNA) and mBD14 significantly upregulated TNF and IL-6 at both mRNA and protein levels (Figure 6G). This induction was abolished in TLR9- and MyD88-deficient pDCs, confirming the essential role of TLR9 in this response (Figure 6H). To determine whether DNA sequence specificity affects pDC activation, we tested synthetic oxidized double-stranded DNA from genomic DNA (syn-ox-Gapdh). Syn-ox-Gapdh induced Il6 mRNA expression at levels comparable to syn-ox-mtDNA, suggesting that DNA oxidation, rather than sequence specificity, drives pDC activation (Figure 6I). Lastly, to assess mBD14-DNA interactions under physiological conditions, we overexpressed FLAG-tagged mBD14 or hBD3 in HEK293T cells and performed DNA immunoprecipitation using an anti-FLAG antibody. mtDNA was efficiently immunoprecipitated with both mBD14 and hBD3 (Figure 6J).

IMQ Interacts with Gelsolin and Gelsolin Deficiency Increases UPR

To uncover the molecular mechanisms by which IMQ induces ER stress and activates UPRs, we sought to identify proteins that interact with IMQ (Figure 7A). IMQ was conjugated to magnetic beads and incubated with whole cell lysates from BMDCs. We next performed two rounds of mass spectrometric analysis and compiled a list of proteins that were specifically detected in the IMQ-beads, as well as those that were more abundantly detected in the IMQ-beads compared to the control beads (Table S7, and S8). Among these, we focused on Gelsolin, which was consistently detected in both independent experiments (Table S9). To validate this interaction, we overexpressed FLAG-tagged Gelsolin in HEK293T cells and incubated cell lysates with IMQ-conjugated beads. Immunoprecipitation assays confirmed that FLAG-Gelsolin was co-precipitated with IMQ-conjugated beads, but not with control beads (Figure 7B). Surface plasmon resonance (SPR) analysis further revealed a binding affinity (K_D) of approximately 1.8 μM, supporting the specificity of this interaction (Figure S7). These findings establish Gelsolin as a novel molecular interactor of IMQ, potentially regulating its proinflammatory effects.

Figure 7.

To investigate the subcellular localization of Gelsolin and its potential dynamics upon IMQ stimulation, we expressed FLAG-tagged Gelsolin in mouse embryonic fibroblast (MEF) and performed immunostaining before and after IMQ treatment. Under basal conditions, FLAG-Gelsolin did not colocalize with both ER and mitochondria (Figure S8). Following IMQ stimulation, however, we observed a portion of FLAG-Gelsolin colocalized with both ER and mitochondria (Figure 7C). These findings suggest that Gelsolin relocalizes to ER and mitochondria in response to IMQ and may function at ER to regulate stress responses.

To investigate the role of Gelsolin in IMQ-induced ER stress, we generated Gelsolin-deficient MEFs using the CRISPR/Cas9 system (Figure S9), achieving complete knockout as confirmed by western blot analysis (Figure 7D). Gelsolin-deficient MEFs and WT control cells were stimulated with IMQ, and the UPR was assessed. RT-qPCR revealed significantly elevated mRNA expression of Xbp1s and Ddit3 in Gelsolin-deficient MEFs compared to WT cells following IMQ stimulation, indicating enhanced UPR activation (Figure 7E, and 7F). Given the interplay between ER stress and mitochondrial function, we examined mitochondrial parameters in IMQ-treated MEFs. MitoSOX staining revealed significant increase in mitochondrial ROS production in Gelsolin-deficient MEFs compared to WT cells (Figure 7G). Moreover, PLA assay revealed that IMQ stimulation increased ER-mitochondria contact sites in Gelsolin-deficient MEFs compared to WT cells (Figure 7H), indicating enhanced ER-mitochondria tethering. These findings suggest that Gelsolin suppresses IMQ-induced ER stress and mitochondrial dysfunction by modulating ER-mitochondria interactions, highlighting its protective role in psoriasis pathogenesis.

Human and Mouse Psoriatic Lesions Exhibit Reduced Gelsolin and Enhanced ER Stress

To further explore the physiological link between Gelsolin, psoriasis pathogenesis, and the UPR, we re-analyzed RNA-seq datasets derived from non-lesional and lesional skin in a mouse model of IMQ-induced psoriasiform dermatitis, as well as from normal and lesional skin of human psoriasis patients. Re-analysis of the RNA-seq datasets revealed a set of transcriptional alterations characteristic of psoriatic lesions. Both in mice and humans, the expression of psoriasis-associated genes, including Il23a and S100a9, was markedly elevated in lesional skin (Figure 8A and 8B). In human lesions, Defb103b, a gene previously reported to be downregulated in psoriasis, showed decreased expression, while its mouse ortholog Defb14 was similarly downregulated in IMQ-induced lesions. Moreover, the UPR-related gene Hspa5 was consistently induced in lesional skin across both species. Notably, the expression of Gelsolin was significantly reduced in psoriatic lesions of both mice and humans. Pearson correlation analysis showed a negative correlation between Gelsolin and Hspa5 (r = -0.86, p = 0.00626) in lesional skin (Figure 8C). Together, these results not only recapitulate known transcriptional features of psoriatic lesions but also highlight a striking and conserved downregulation of Gelsolin, which is closely linked to UPR activation and may play a key role in disease pathogenesis.

Figure 8.

Discussion

Our study demonstrates that the UPR and the NLRP3 inflammasome exacerbate IMQ-induced psoriasis (Figure 8D). We found that IMQ disrupts mitochondrial homeostasis, leading to excessive ROS production, mitochondrial damage, and the release of mtDNA into the cytoplasm. This is supported by our observation that IMQ-induced IL-1β release from BMDCs was significantly suppressed by the mitochondrial ROS-specific inhibitor mitoquinone (MitoQ), as well as by antioxidants such as GSHEE, NAC, and PDTC. These findings highlight mitochondrial dysfunction as a central driver of inflammasome activation and inflammation in IMQ-induced dermatitis. In addition to its effects on mitochondrial ROS, IMQ increased ER–mitochondria contact sites, a condition commonly associated with facilitated Ca²⁺ dynamics between these organelles (Figure 8D). Although Ca²⁺ transfer can promote mitochondrial ROS production^39–42, our study does not distinguish which specific Ca²⁺ flux pathways are responsible for the observed effects. Notably, inhibition with 2-APB significantly reduced IMQ-induced ROS generation and IL-1β release; however, because 2-APB modulates multiple Ca²⁺-related processes beyond ER-derived flux, these results should be interpreted as indicating that Ca²⁺ dysregulation in general, rather than a single defined pathway, contributes to IMQ-driven mitochondrial stress and inflammasome activation. While previous studies have shown that IMQ directly targets mitochondrial complex I and NQO2, leading to ROS production and NLRP3 activation²⁰, our findings suggest that additional mechanisms, specifically, Ca²⁺ dysregulation and ER-mitochondria interactions, contribute to mitochondrial ROS production.

IMQ also triggered the UPR, characterized by the induction of Xbp1s in both DCs and keratinocytes (Figure 8D). Given that ER Ca²⁺ depletion disrupts protein folding and induces ER stress⁴³, it is likely that Ca²⁺ release from the ER contributes to UPR activation. Additionally, mitochondrial ROS have been reported to act as upstream signals for ER stress and UPR activation⁴⁴, further supporting the link between mitochondrial dysfunction and ER stress responses in IMQ-induced skin inflammation. Our study highlights the critical role of the IRE1α-XBP1s axis in regulating IL-23 expression in DCs. IMQ-induced Il23a mRNA expression was partially suppressed by 4μ8c, an IRE1α inhibitor, while overexpression of XBP1s and MyD88 synergistically enhanced the promoter activity of Il23a gene, which contains binding sites for both XBP1s and NF-κB. These findings suggest that IL-23 expression in DCs is driven by the convergence of the IRE1α-XBP1s and TLR7-MyD88-NF-κB pathways. Notably, Il23a mRNA induction was abolished in MyD88-deficient BMDCs, confirming the indispensable role of TLR7-MyD88 signaling, with XBP1s enhancing its expression. Furthermore, RNA-seq analysis revealed that IMQ failed to induce IL-23 expression in keratinocytes where TLR7 expression is rare but activated the Ca²⁺-IRE1α-XBP1s axis to drive the expression of psoriasis-associated genes. These results suggest that in keratinocytes, IMQ primarily regulates gene expression through the UPR rather than TLR7.

Our data may suggest that IMQ stimulation also promoted the release of oxidized mtDNA into the extracellular environment via NLRP3 inflammasome-dependent pyroptosis (Figure 8D). Previous studies have established that mtDNA acts as a DAMP, activating pathways such as TLR9 and other cytosolic DNA sensors such as cGAS, NLRP3, and AIM2 to drive inflammatory and autoimmune diseases^{37,38,45–47}. Our data further revealed that mtDNA enhanced IL-6 and TNF-α production by pDCs in the presence of mBD14. Additionally, synthetic DNA experiments confirmed that oxidized mtDNA complexed with mBD14 activated pDCs via TLR9, suggesting that mBD14 may protect extracellular mtDNA from DNase-mediated degradation or facilitate its endosomal uptake. Given that Defb14 expression in keratinocytes is induced via a UPR-dependent mechanism, targeting the UPR or its downstream effectors, such as AMPs and chemokines, may provide a novel therapeutic strategy to prevent mtDNA-protein complex formation and mitigate psoriasis pathology.

Beyond immune regulation, metabolic alterations also contribute to psoriasis pathogenesis²². High-fat diets have been shown to exacerbate psoriasis via an XBP1-dependent mechanism, promoting mitochondrial ROS production and UPR activation in TLR-stimulated DCs. In our study, IMQ treatment induced UPR activation in psoriatic lesions, and administration of tunicamycin, an ER stress inducer, significantly increased ear thickness and upregulated psoriasis-related gene expression in IMQ-treated mice. Furthermore, carbachol, an IP₃R activator known to trigger UPR and NLRP3 inflammasome activation, exacerbated IMQ-induced dermatitis and enhanced the expression of psoriasis-related genes. These findings reinforce the role of the Ca²⁺-UPR axis in psoriasis pathology. Importantly, 2-APB significantly suppressed caspase-1 activation in BMDCs and reduced IL-17 expression in skin lesions, ultimately ameliorating psoriasis symptoms in vivo. Collectively, these results suggest that Ca²⁺ transfer from the ER to mitochondria, followed by ER stress and NLRP3 inflammasome activation, represents a key pathogenic mechanism driving psoriasis development.

We identified Gelsolin as a critical regulator that suppresses IMQ-induced ER stress responses (Figure 8D). Gelsolin-deficient cells exhibited enhanced activation of UPR upon IMQ stimulation. These findings are consistent with a recent study in macrophage-specific Gelsolin-deficient mice, which demonstrated increased mitochondria damage and NLRP3 inflammasome-mediated IL-1β secretion and aggravated arthritis, indicating a broader role for Gelsolin in modulating inflammation³³. Given its established function in actin remodeling^30,31, Gelsolin may influence the structural integrity of ER-mitochondria contact sites, which are essential for Ca²⁺ exchange between the two organelles. The loss of Gelsolin could increase these contact sites, resulting in excessive Ca²⁺ influx into mitochondria and subsequent mitochondrial dysfunction, including ROS generation. Additionally, Gelsolin’s ability to bind cytosolic Ca²⁺ may contribute to its regulatory function. By buffering excess Ca²⁺, Gelsolin may prevent aberrant Ca²⁺ signaling that would otherwise promote ER stress and inflammasome activation. These findings highlight Gelsolin as a key modulator of intracellular Ca²⁺ dynamics, ER stress responses, and inflammatory signaling pathways, offering new insights into its potential as a therapeutic target in inflammatory skin diseases such as psoriasis.

Importantly, our analysis of human psoriatic lesions revealed significant Gelsolin downregulation alongside upregulation of ER stress genes such as Hspa5 (Figure 9B). Furthermore, Pearson correlation analysis showed a negative correlation between Gelsolin and Hspa5, suggesting that reduced Gelsolin expression exacerbates ER stress and inflammation in human psoriasis. These findings in human psoriasis patients validate Gelsolin’s protective role, suggesting that its downregulation in psoriatic lesions may contribute to disease severity, potentially correlating with clinical metrics like the Psoriasis Area and Severity Index (PASI). Gelsolin supplementation or strategies to enhance its expression could thus represent novel therapeutic approaches.

Our findings reveal a previously unrecognized interplay between the NLRP3 inflammasome, ER stress, and Ca²⁺ homeostasis in IMQ-induced psoriasis (Figure 8D). Furthermore, IMQ-induced mtDNA release amplifies inflammation through pDC activation in the presence of antimicrobial peptides, such as mBD14, while Gelsolin serves as a protective factor by regulating ER stress and mitochondrial function. These findings suggest that targeting ER-mitochondria interactions, and UPR-associated molecules could offer novel therapeutic strategies for psoriasis.

Material and Methods

Mice

C57BL/6J mice were purchased from CLEA Japan (Tokyo, Japan) and maintained in accordance with the guidelines of the Committee on Animal Research at the Nara Institute of Science and Technology. MyD88-deficient mice were purchased from Oriental Bio Service.

Cells

HeLa cells, HEK293T cells, and MEFs were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37°C in a humidified 5% CO₂ incubator. Mouse BMDCs and pDCs were obtained by culturing mouse bone marrow cells in RPMI 1640 medium (Nacalai Tesque) supplemented with 10% FBS, 1% antibiotic-antimycotic mixed stock solution (Nacalai Tesque), 0.1 mM 2-mercaptoethanol, and 10 ng/mL recombinant mouse GM-CSF (PeproTech) or 100 ng/mL mouse Flt-3 ligand (BioLegend) for 6–8 days.

Isolation and culture of primary mouse keratinocytes

Primary mouse keratinocytes were isolated from neonatal mouse skin. Briefly, peeled mouse skin was washed with PBS and then incubated with CnT-Prime medium (CELLnTEC) containing 2.5 mg/mL dispase II and 1% penicillin-streptomycin mixed solution (Nacalai Tesque) at 4°C overnight. The next day, the skin was washed with PBS and the epidermis was isolated from the dermis. The epidermis was incubated with accutase (Nacalai Tesque) for 20 minutes at room temperature to isolate keratinocytes. Keratinocytes were passed through a 100-μm filter and pelleted. Keratinocytes were resuspended in CnT-Prime medium containing 1% penicillin-streptomycin mixed solution and seeded onto culture dishes precoated with Cellmatrix Type I-C (Nitta Gelatin).

Reagents and plasmids

All ligands for pattern recognition receptors were purchased from Invivogen unless otherwise stated. LPS and ATP were purchased from Sigma-Aldrich. Working concentrations of each innate immune ligand were as follows: LPS (20–1000 ng/mL), imiquimod (5–50 μg/mL), resiquimod (50 μg/mL), ATP (5 mM), nigericin (5 μM), and D19 (100 nM). Recombinant mouse beta-defensin 14 was purchased from Prospec and used at 5 μg/mL. Each inhibitor was used at the following concentrations: PDTC (10 μM, Sigma-Aldrich), GSHEE (10 mM, Sigma-Aldrich), NAC (10 or 20 mM, Thermo Fisher Scientific), 2-APB (50 μM, Wako), 4μ8c (25 or 32 μM, Cayman), mitoquinone (10 μM, Wako). Inhibitors were added directly to the cell culture medium 0.5–1 h before stimulation. Tunicamycin (LKT-Labs) and thapsigargin (Wako) were used to induce ER stress.

To overexpress Xbp1s, the mouse Xbp1s coding sequence (CDS) was inserted into a pFLAG-CMV2 expression vector (Sigma-Aldrich). For the reporter assay, the mouse Il23a genomic fragment (−2,030 to +20) was amplified by PCR and inserted into the pGL3-promoter (Promega).

Synthesis of oxidized DNA fragment

Oxidized DNAs were generated by PCR in the presence of 200 μM 8-Oxo-2′-dGTP (TriLink) and then purified using a QIAquick Nucleotide Removal Kit (Qiagen). Synthesized oxidized DNA fragments and mBD14 were mixed in FBS-free RPMI 1640 and incubated for 15 min at room temperature prior to stimulation. The primers used for oxidized DNA synthesis are described in Supplementary Table 10.

Flow cytometry-based cell death assay

LPS primed or unprimed BMDCs were seeded onto uncoated 96-well plates at a density of 1.0 × 10⁵ cells per well. After 3 or 18 h of stimulation, cells were harvested and stained with 200 ng/mL propidium iodide (PI, Nacalai Tesque) and Annexin V-APC (BD Bioscience) in Annexin V binding buffer (10 mM Hepes-NaOH [pH 7.2], 140 mM NaCl, 2.5 mM CaCl₂). Data were collected by FACS Accuri (BD Bioscience).

Measurement of intracellular ROS

BMDCs were seeded onto untreated 96-well plates at a density of 1.0 × 10⁵ cells per well. After 30 min of stimulation with 50 μg/ml IMQ, CellROX™ Deep Red (Thermo Fisher Scientific) was added directly to the culture medium at a final concentration of 1 μM and incubated at 37°C for 15 min. The cells were then washed with PBS and the fluorescence intensity was analyzed by FACS Accuri. To measure mitochondrial ROS, 5 × 10⁵ BMDCs were seeded onto untreated 24-well plates and incubated with 2.5 μM MitoSOX™ Red Mitochondrial Superoxide Indicator (Thermo Fisher Scientific) in RPMI 1640 medium for 30 min. The cells were then washed with RPMI 1640 medium and seeded onto untreated 24-well plates followed by stimulation with 50 μg/ml IMQ. After 30 min of IMQ stimulation, the cells were washed with PBS and analyzed by flow cytometry.

PLA

BMDCs were seeded at a density of 1.5 × 10⁵ onto 24-well plates with poly-L-lysin-coated coverslips and stimulated with LPS for 3 h followed by stimulation with 50 μg/mL IMQ. The cells were then washed with PBS, fixed in 10% formaldehyde, and incubated at room temperature. After 10 min, fixation was stopped using 1 M glycine and the cells were washed with PBS. The coverslips containing the cells were stored in 100 mM glycine. The cells were permeabilized in PBS containing 0.1% Triton-X100 for 20 min at room temperature and then washed twice with PBS. Blocking solution was then added and the cells were incubated at 37°C for 30 min. The cells were then treated with PBS containing anti-VDAC1/3 antibody (ab14734, Abcam) and anti-IP₃R1 antibody (an264281, Abcam) at 4°C overnight and washed three times with TBS-T. Subsequent experiments were performed using kits (Duolink In Situ PLA Probe Anti-Mouse MINUS [DUO92004, Sigma-Aldrich] and Duolink In Situ PLA Probe Anti-Rabbit PLUS [DUO92002, Sigma-Aldrich]) according to the manufacturer’s instructions. Nuclei were stained with Hoechst33342. PLA signals were visualized by fluorescence microscopy and the number of PLA dots per cell was counted.

Mitochondrial Ca²⁺ measurement

First, 5 × 10⁵ BMDCs were seeded onto untreated 24-well plates and incubated with 5 μM Rhod-2-AM (Dojindo) in RPMI 1640 for 30 min. The cells were then washed three times with HBSS and seeded onto untreated 24-well plates, followed by stimulation with 50 μg/ml IMQ. After 30 min of stimulation, the cells were washed with HBSS and analyzed by FACS Accuri.

Mitochondrial membrane potential analysis

BMDCs were seeded onto untreated 96-well plates at a density of 1.0 × 10⁵ cells per well. After 1 h of stimulation with 50 μg/ml IMQ, the cells were stained with 2 μM JC-1 (Dojindo) for 30 min at 37°C in RPMI 1640 medium. The cells were washed with PBS and resuspended with imaging buffer (Dojindo) for analysis by FACS Accuri.

Cellular fractionation and measurement of mtDNA

Stimulated cells were washed with PBS and lysed with digitonin lysis buffer (50 mM HEPES-NaOH [pH 7.4], 150 mM NaCl, 20 μg/ml digitonin). The cell lysate was centrifuged at 1000 ×g for 3 min at 4°C, and the supernatant was collected into a new tube. This step was repeated three times to remove intact cells and cell debris. The supernatant was then centrifuged at 17,000 ×g for 10 min at 4°C. The supernatant was collected as the cytoplasmic fraction and the pellet was collected as the mitochondrial fraction. DNA from the cytosolic fraction and cell culture supernatant were purified using the QIAquick Nucleotide Removal Kit (Qiagen). Pellets containing mitochondria were lysed with cell lysis buffer (100 mM Tris-HCl [pH 8.0], 5 mM EDTA [pH 8.0], 0.2% SDS, 200 mM NaCl) containing 1/100 volume of protease K at 56°C for 2 h. DNA from mitochondrial lysates was precipitated with isopropanol and washed with 70% ethanol. The amount of mitochondrial DNA in each fraction was analyzed by qPCR using PowerUP SYBR Green PCR Master Mix (Applied Biosystems) with a Step One Real-Time PCR System (Applied Biosystems).

ELISA

BMDCs were seeded onto 96-well plates at a density of 1.0 × 10⁵ cells per well. For IL-1β measurements, cells were preincubated with 20 ng/ml of LPS for 3 h and stimulated with 50 μg/ml IMQ, 50 μg/ml RSQ, 5 mM ATP, or 5 μM nigericin for 30 min or 3 h. In the case of pDCs, the cells were seeded onto 96-well plates at a density of 2.0 × 10⁵ cells per well and stimulated with ligands for 24 h. The cytokine levels of IL-6, TNF-α, and IL-1β in the culture supernatant were measured using a mouse IL-6 DuoSet ELISA (R&D Systems), mouse TNF-α ELISA kit (Invitrogen), and mouse IL-1β ELISA kit (Invitrogen), respectively, according to the manufacturer’s instructions.

Western blot analysis

BMDCs were seeded onto six-well plates at a density of 1.0 × 10⁶ cells per well. For caspase-1, the cells were pretreated with 20 ng/ml LPS for 3 h and stimulated with 50 μg/ml IMQ for 30 min. The cells were washed with PBS and lysed with 1× SDS sample buffer (25 mM Tris-HCl [pH 6.8], 5% glycerol, 1% SDS, 50 mM DTT, and BPB). For PERK and eIF2α, cells were stimulated with IMQ, RSQ, or 1 mM DTT for 0.5–6 h and lysed with lysis buffer (150 mM NaCl, 1% Nonidet P-40, 1 mM EDTA, 20 mM Tris-HCl [pH 7.5]) containing cOmplete™ Protease Inhibitor Cocktail (Roche). The cell lysates were subjected to SDS-PAGE and transferred to an Immun-Blot PVDF membrane (Bio-Rad). The membrane was then immunoblotted with anti-caspase-1 (p20) (AG-20B-0042-C100, AdipoGen), anti-phospho-PERK (Thr980) (#3179, CST), anti-PERK (#3192, CST), anti-phospho-eIF2α (Ser51) (#3398, CST), or anti-eIF2α (#9722, CST) antibodies. Horseradish-peroxidase-conjugated anti-mouse, rabbit, or goat IgG (Sigma-Aldrich) were used as secondary antibodies. Protein bands were visualized using Western Lightning Plus-ECL (PerkinElmer) and analyzed on a LAS-4000 (Fujitsu Life Sciences). Image files were processed using ImageJ software 1.8.0_172 (NIH).

Ca²⁺ flux analysis

BMDCs or keratinocytes were incubated with Quest Fluo-8 (ABD Bioquest) for 30 min in the dark at room temperature. The cells were washed three times with HBSS containing 1 mM MgCl₂, 2 mM CaCl₂, and 1% FBS, and the Ca²⁺ flux was measured after stimulation with 50 μg/ml IMQ or 50 μg/ml RSQ. Fluo-8 intensity was monitored every 5 seconds on an LSM700 or LSM710 (Zeiss). Keratinocytes were washed with CnT-PR and the Ca²⁺ flux was measured. Ionomycin was used as a positive control. Ca²⁺-insensitive fluorescence was subtracted from each wavelength before calculations to normalize fluorescence values. The values were then plotted against time and shown as F/F0.

RNA isolation, reverse transcription PCR, and quantitative real-time PCR

Total RNA was isolated using Trizol reagent (Invitrogen) and cDNA was synthesized using ReverTra Ace (Toyobo) according to the manufacturer’s instructions. To amplify spliced and unspliced Xbp1 for RT-PCR, Go Taq Master Mix (Promega) was used with the primers described in Supplementary Table 11. PCR conditions were 2 min at 95°C, 35 cycles of denaturation at 95°C for 30 sec, annealing at 60°C for 30 sec, and extension at 72°C for 2 min. A final extension was performed at 72°C for 7 min. PCR products were electrophoresed on 2.5% agarose, and spliced and unspliced Xbp1 bands were visualized using Midori Green Advance DNA Stain (Nippon Genetics). Real-time quantitative PCR (RT-qPCR) was performed using PowerUP SYBR Green PCR Master Mix with Step One. The primers used for RT-qPCR are described in Supplementary Table 12.

Luciferase reporter assay

To measure Il23a promoter activity, 7.5 × 10⁵ HEK293T cells were seeded onto 24-well plates and transfected with 100 ng of pGL3-Il-23p19 and 500 ng of expression plasmid. As an internal control, 10 ng of pRL-TK (Promega) was transfected simultaneously. The medium was changed 6 h after transfection. After 24 h of transfection, luciferase activity was measured using a TriStar2 LB 942 Multidetection Microplate Reader (Berthold) with the Dual-Glo Luciferase System (Promega) according to the manufacturer’s instructions. To measure UPR activity, transfected HeLa cells were stimulated with 50 μg/ml IMQ or 50 μg/ml RSQ for 9 h and the luciferase activity was analyzed.

Induction of IMQ-induced psoriasis-like skin inflammation

Mice aged 8–12 weeks were treated with 10 μg tunicamycin or 20 μg 2-APB applied to the ears or injected intraperitoneally with 25 μg carbachol, followed by the application of 12.6 mg IMQ cream (5%) (Mochida Pharmaceutical) to the ears, for 6–7 consecutive days. As a control, DMSO or EtOH was applied to the opposite ear. Ear samples were fixed in 4% paraformaldehyde phosphate buffer overnight at 4°C and then embedded in paraffin. Sections were cut at 5 μm and stained with hematoxylin and eosin.

Identification of IMQ-binding proteins

[Immobilization of IMQ onto magnetic beads]

Carboxyl-functionalized magnetic beads (2.5 mg; Tamagawa Seiki Co., Ltd.) were washed several times with N,N-dimethylformamide (DMF) and activated with N-hydroxysuccinimide (NHS, 1 M in DMF) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide hydrochloride (EDC·HCl). After incubation at room temperature for 2 h, the beads were washed with DMF and resuspended to obtain NHS-activated beads. For ligand coupling, NHS-activated beads (1.25 mg per reaction) were incubated with a ligand solution containing 2 mM IMQ and 4 mM triethylamine in DMF for 70 min at room temperature with gentle mixing. Control beads were treated in parallel with DMF alone. Following the coupling reaction, residual NHS groups were quenched with 1 M ethanolamine for 2 h. Finally, the beads were washed several times with 50% methanol and resuspended in 50% methanol for storage until use.

[Preparation of BMDC-derived protein lysates]

BMDCs (2.0 × 10⁶ cells) were collected, washed once with PBS, and pelleted again. The cell pellets were resuspended in KCl buffer (150 mM KCl, 0.5% Nonidet P-40, 20 mM HEPES [pH 7.9], 1 mM MgCl₂, 0.2 mM CaCl₂, 0.2 mM EDTA, 1 mM DTT, supplemented with protease inhibitor cocktail). After incubation on ice for 5 min, lysates were clarified by centrifugation at 20,000 ×g for 5 min at 4°C, and the supernatants were collected as protein extracts.

[Purification of IMQ-binding proteins]

IMQ-conjugated beads were washed with KCl buffer and separated using a magnetic rack. Protein lysates prepared as described above were added to the beads, and the mixture was incubated at 4°C for 4 h with gentle rotation. After incubation, the beads were washed five times with KCl buffer without DTT. Bound proteins were eluted by adding 3× SDS sample buffer containing 15% 2-mercaptoethanol and protease inhibitors, followed by heating at 95°C for 5 min. The supernatants containing eluted proteins were collected after cooling on ice.

[SDS-PAGE and silver staining]

Proteins eluted from the beads were separated on a 12% polyacrylamide gel prepared using the TGX FastCast Acrylamide Kit (Bio-Rad) according to the manufacturer’s instructions. Samples were loaded onto the gel and electrophoresed in running buffer (0.25 M Tris, 192 mM glycine, 1% SDS) at 20 mA for 60 min. Separated proteins were visualized by silver staining using the Silver Stain MS Kit (Wako) and subsequently subjected to mass spectrometry analysis.

Pulldown assay

HEK293T cells were seeded onto 6-well plates (Falcon) coated with Cellmatrix Type I-C (Nitta Gelatin) at 8.0 × 10⁵ cells per well. Cells were transfected with FLAG-Gelsolin expression plasmid using Lipofectamine 2000 (Invitrogen) according to the manufacturer’s protocol. Two days post-transfection, cells were washed with PBS, and lysed in 150 μL of lysis buffer (20 mM HEPES-HCl [pH 7.5], 150 mM NaCl, 1% NP-40, 1 mM EDTA) supplemented with protease inhibitors. Lysates were incubated on ice for 5 min and clarified by centrifugation at 20,000 ×g for 5 min at 4°C. The supernatants were collected as cell lysates. Cell lysates were incubated with IMQ-conjugated beads (described before) overnight at 4°C with gentle rotation. After incubation, the bead-bound proteins were eluted, separated by SDS-PAGE, subjected western blot analysis for FLAG-Gelsolin and β-Actin.

Immunofluorescence

WT and Gsn-deficient MEFs were seeded and transfected with pLZR-c-FLAG-empty or pLZR-c-FLAG-mGelsolin plasmid, and cultured for 24 h. For immunofluorescence, cells were then replated onto poly-L-lysine–coated coverslips (Matsunami) in 24-well plates and cultured overnight. Cells were stimulated with IMQ (50 μg/mL) for 0-15 min, washed with PBS, and fixed with 4% paraformaldehyde for 20 min. After permeabilization and blocking, cells were incubated with primary antibodies overnight at 4°C: anti-FLAG Ab (SigmaAldrich), anti-Calnexin Ab (Abcam), and anti-TOM20 Ab (CST). After washing, cells were incubated with Alexa Fluor–conjugated secondary antibodies (Invitrogen) for 1 h and counterstained with Hoechst 33342 (Dojindo). Images were acquired on an LSM900 confocal microscope (Zeiss) using 405, 488, and 561 nm lasers, and processed with ZEN software.

Generation of Gelsolin-deficient MEFs

MEFs were electroporated with a px330 plasmid expressing Cas9 and a single-guide RNA targeting the Gelsolin coding sequence (Table S13). Electroporation was performed using Neon Transfection System (Thermo Fisher) according to the manufacturer’s instructions. After recovery, cells were cultured under standard conditions, and single cells were sorted by flow cytometry into 96-well plates to establish clonal populations. Genomic DNA was extracted from individual clones, and the target locus was amplified by PCR and analyzed by Sanger sequencing to confirm the presence of indels and determine the genotype (Figure S9).

RNA sequencing

Keratinocytes were stimulated with 50 μg/ml IMQ in the presence or absence of 25 μM 4μ8c. After 3 h of stimulation, total RNA was isolated from keratinocytes using the FastGene™ RNA Premium Kit (Nippon Genetics) and RNA quality was analyzed using the Agilent 2100 Bioanalyzer. Illumina next-generation sequencing analysis was performed by Genewiz (Tokyo, Japan). RNA-seq raw datasets were collected from the gene expression omnibus (GEO) of the National Center for Biotechnology Information database. GSE289485 and GSE117405 were used to analyze mouse psoriasis-like dermatitis and human psoriasis samples, respectively.

Statistics

Statistical significance was determined using the Student’s t-test and ANOVA with Tukey’s test. P < 0.05 was considered statistically significant.

Study approval

All murine experimental protocols were approved by the Animal experimental Committee at the Nara Institute of Science and Technology (the approval no.2310).

Data availability

All data supporting the findings of this study and its supplementary information files are available within the article.

Acknowledgements

We thank C. Suzuki for secretarial assistance, and K. Takahashi and A. Takara for technical support. This work was supported by the Japan Society for the Promotion of Science KAKENHI (grant numbers 17H04066, 19K22533, 20H03468, 23K27392, and 24K22057 to T. Kawai; 17K15726, and 21K14817 to D. Ori; 19K07608 to T. Kawasaki; 23K14546 to N. Kano). This work was also supported by ONO PHARMACEUTICAL CO., LTD., the Takeda Scientific Foundation to T. Kawai and K. Kohno, and Japan Agency for Medical Research and Development (grant number 223fa727001h0001).

Additional information

Author contributions

DO, HO, TKawasaki, KJI, KN, MY, KKohno, and TKawai designed the experiments. DO, HO, RK, MM, SH, ST, TT, RT, and TKawasaki performed the experiments. KKobiyama, HN, MS, YK, and AT prepared the materials. DO, HO, and TKawai wrote the manuscript. All authors edited and approved the final draft. DO was listed first for having a larger role in performing experiments and analyzing data.

Funding

MEXT | Japan Society for the Promotion of Science (JSPS) (17H04066)

• Taro Kawai

MEXT | Japan Society for the Promotion of Science (JSPS) (19K22533)

• Taro Kawai

MEXT | Japan Society for the Promotion of Science (JSPS) (20H03468)

• Taro Kawai

MEXT | Japan Society for the Promotion of Science (JSPS) (23K27392)

• Taro Kawai

MEXT | Japan Society for the Promotion of Science (JSPS) (24K22057)

• Taro Kawai

MEXT | Japan Society for the Promotion of Science (JSPS) (17K15726)

• Daisuke Ori

MEXT | Japan Society for the Promotion of Science (JSPS) (21K14817)

• Daisuke Ori

MEXT | Japan Society for the Promotion of Science (JSPS) (19K07608)

• Takumi Kawasaki

MEXT | Japan Society for the Promotion of Science (JSPS) (23K14546)

• Norisuke Kano

Ono Pharmaceutical (Ono Pharmaceutical Co., Ltd.)

• Taro Kawai

Takeda Science Foundation (TSF)

• Kenji Kohno

Takeda Science Foundation (TSF)

• Taro Kawai

Japan Agency for Medical Research and Development (AMED) (223fa727001h0001)

• Taro Kawai

Additional files

Supplementary Figure 1-9.

Supplementary Table 1.

Supplementary Table 2.

Supplementary Table 3.

Supplementary Table 4.

Supplementary Table 5.

Supplementary Table 6.

Supplementary Table 7.

Supplementary Table 8.

Supplementary Table 9.

Supplementary Table 10.

Supplementary Table 11.

Supplementary Table 12.

Supplementary Table 13.