TRPV4 mediates low-humidity responses in epidermal keratinocytes

Significance

Low humidity is a common environmental stress that affects organismal physiology and homeostasis, yet how tissues sense and respond to atmospheric dryness remains poorly defined. Here, we identify the calcium-permeable cation channel transient receptor potential vanilloid 4 (TRPV4) as an important mediator of low-humidity responses in epidermal keratinocytes. We show that low-humidity exposure drives rapid water movement across the epidermal barrier and activates TRPV4-dependent signaling and transcriptional programs in the epidermis. These findings support previous data showing that atmospheric dryness is not merely a passive dehydration challenge, but an environmental cue sensed by keratinocytes, and provide a molecular basis by which TRPV4 contributes to epidermal responses to low humidity, offering insight into low-humidity-related skin disorders.

Abstract

Environmental humidity is a key determinant of organismal physiology and homeostasis. Although the genetic basis of humidity-driven responses is well established in invertebrates, the molecular and genetic mechanisms by which mammals respond to low humidity remain unclear. Here, using reconstructed 3D epidermal equivalents and a mouse model, we identify transient receptor potential vanilloid 4 (TRPV4) as an important mediator of low-humidity-induced signaling and transcriptional responses in epidermal keratinocytes. In 3D skin models, low-humidity exposure induced rapid water movement, characterized by enhanced apical water loss and concomitant basal-side water uptake within 90 s, followed by TRPV4-dependent Ca²⁺ influx within 3 min. Pharmacological TRPV4 inhibition suppressed MAPK activation at 30 min and both early (3 h) and late (24 h) low-humidity-induced transcriptional changes. Ca²⁺ chelation reduced low-humidity-induced ERK and p38 activation, and ERK inhibition attenuated low-humidity-induced gene expression changes at 3 h. In vivo, acute low-humidity exposure in mice elicited epidermal MAPK activation within 15 min and early transcriptional changes at 3 h, with RNA-sequencing analysis showing enrichment of gene sets associated with cell adhesion or inflammatory responses. These signaling and transcriptomic responses were attenuated in the epidermis of keratinocyte-specific Trpv4 conditional knockout mice. Collectively, our findings identify TRPV4 as an important mediator of epidermal responses to low-humidity stress, providing a molecular basis for low-humidity-driven signaling in the mammalian epidermis.

Humidity is a key environmental factor that influences the physiology and behavior of living organisms. Maintaining hydration and adapting to changes in ambient humidity are essential for survival across species (1, 2). In particular, exposure to low ambient humidity poses a desiccation risk and necessitates rapid and effective adaptation to dry air. In humans, ambient humidity affects comfort, psychological well-being, and health, and low-humidity conditions have been linked in observational studies to a higher incidence of cardiovascular events and to worsening xerosis and atopic dermatitis, although the evidence is largely correlative and confounded by season and temperature (3–5).

Previous studies have identified some genes and molecular pathways involved in humidity-dependent behavior using invertebrate models. In Drosophila, the transient receptor potential (TRP) channels Water Witch and Nanchung have been implicated in moist and dry-air–related behaviors, respectively (6). In Caenorhabditis elegans, a cGMP-gated channel has been shown to be important for sensing dry air (7). In contrast to the direct hygrosensation observed in Drosophila and C. elegans, mammalian responses to changes in ambient humidity are likely mediated primarily by altered water movement and evaporation at barrier surfaces, which may impose biophysical stress on epithelial cells, thereby engaging intracellular signaling pathways.

The skin, the primary barrier between the body and the external environment, directly and continuously encounters various environmental stimuli. The outermost layer, the epidermis, consists predominantly of keratinocytes (~90%), along with immune cells, such as T cells and dendritic cells (8, 9). Historically, epidermal innervating sensory neurons have been considered the sole detectors and transmitters of environmental cues. However, accumulating evidence has demonstrated that keratinocytes play an active and essential role in the regulation of sensory responses, including mechanotransduction and somatosensation, in the skin (10–12). In addition, previous studies have shown that reduced epidermal hydration status or low-humidity exposure induces downstream keratinocyte responses, including inflammatory mediator production and changes in epidermal gene and protein expression (3, 13–15). Consistent with this concept, we previously found that keratinocytes in a reconstructed 3D human epidermal model rapidly respond to low humidity, exhibiting prompt activation of intracellular MAPK signaling within 30 min and subsequent marked changes in gene expression profiles at 1 h (16). These observations establish that keratinocytes are direct effectors of the cutaneous response to ambient dryness. However, the molecular mechanisms by which keratinocytes detect changes in ambient humidity and convert the resulting biophysical perturbations into early intracellular signaling remain incompletely understood.

In this study, we sought to elucidate the molecular basis of the keratinocyte response to low humidity, with a particular focus on identifying the genes required for this process. Using reconstructed 3D skin models, we performed a targeted screen and identified transient receptor potential vanilloid 4 (TRPV4) as a candidate mediator of low-humidity-induced responses. We demonstrated that TRPV4 is required for transducing the physical stimulus imposed by low humidity into cellular signals, including Ca²⁺ influx and subsequent MAPK activation. Using keratinocyte-specific Trpv4 knockout mice, we further validated that keratinocyte TRPV4 is an important transducer of low humidity-driven signaling in the epidermis.

Results

Involvement of TRPV4 in Low Humidity-Induced Gene Expression and Cell Signaling.

In our previous study using a human 3D skin model, we showed that low-humidity exposure induced detectable transcriptional changes in keratinocytes within 1 h, and RNA-seq analysis revealed a set of genes with upregulated expression after 3 h (SI Appendix, Fig. S1A) (16). To identify the genes involved in keratinocyte low-humidity responses, we selected seven transcripts that were robustly upregulated at 3 h after exposure and used them as quantitative readouts (SI Appendix, Fig. S1B). Responses to humidity changes have shown to depend on transient receptor potential (TRP) thermosensory channels in Drosophila and mechanoreceptors in Caenorhabditis (6, 7). Based on these reports, we compiled a list of TRP and mechanosensory channel genes expressed in both human and mouse keratinocytes (SI Appendix, Tables S1 and S2) and selected 16 candidate genes that are potentially involved in keratinocyte responses to low-humidity exposure (SI Appendix, Table S3). The 3D skin model was preincubated with a panel of pharmacological compounds targeting 16 candidate channels, followed by exposure to low humidity (approximately 15% relative humidity [RH]) for 3 h, and the mRNA levels of the indicator genes were quantified (Fig. 1A). Although several pharmacological compounds targeting other candidate channels partially reduced the induction of some low-humidity-responsive genes (SI Appendix, Fig. S2 A–C), the TRPV4 inhibitor HC-067047 (5 μM) showed the broadest and most consistent attenuation of the representative readout genes examined (Fig. 1B) (17–19). Consistent attenuation was also observed at lower concentrations of HC-067047, including 0.5 and 1 μM (SI Appendix, Fig. S1C).

Fig. 1.

Involvement of TRPV4 in low humidity-induced gene expression and cell signaling. (A and B) 3D skin was preincubated with HC-067047 (5 μM, TRPV4 inhibitor) or vehicle for 30 min under control humidity (~90% RH), followed by exposure to low humidity (~15% RH) for 3 h before RNA extraction. (B) mRNA expression of indicated genes by RT-qPCR (mean ± SEM, n = 6, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001). P-values were calculated using two-way ANOVA test. (C) 3D skin was preincubated with HC-067047 (5 μM) or GSK2193874 (5 μM) for 30 min under control humidity (~90% RH), followed by exposure to low humidity (~15% RH) for 30 min before protein extraction. (Left) Representative immunoblots probed with antibodies against phospho-ERK1/2, ERK1/2, phospho-p38, and p38. (Right) Densitometric analysis of band intensities performed using ImageJ (mean ± SEM, n = 3, *P < 0.05, **P < 0.01 by one-way ANOVA test). (D and E) 3D skin models were preincubated with HC-067047 (5 μM) and then exposed to low humidity (~15% RH) for 30 min. Immunofluorescence staining was performed using antibodies against phospho-ERK1/2 (FITC, green) or phospho-p38 (FITC, green), with nuclei counterstained using DAPI (blue). Arrowheads indicate phospho-ERK or phospho-p38 positive cells. (Scale bar, 100 µm.) sc, stratum corneum; epi, epidermis; tm, transparent membrane. Right: Quantification of the percentage of phospho-ERK- or phospho-p38-positive cells (mean ± SEM from more than 200 cells in four different fields). *P < 0.05, ****P < 0.0001 by one-way ANOVA.

We previously showed that exposure of 3D skin to low humidity activates MAPK signaling pathways, such as ERK and p38, within 30 min (16). TRPV4 inhibition by HC-067047 or GSK2193874 suppressed ERK and p38 phosphorylation, as demonstrated by western blotting (Fig. 1C and SI Appendix, Fig. S7). Although the TRPA1 inhibitor HC-030031 partially suppressed the low humidity-induced increase in gene expression at 3 h (SI Appendix, Fig. S2C), it did not inhibit the MAPK activation triggered by exposure to low humidity (SI Appendix, Fig. S1D). Therefore, based on the combined gene expression and signaling analyses, we focused on TRPV4 as the candidate gene most consistently associated with the keratinocyte response to low humidity.

Immunostaining for phospho-ERK and phospho-p38 showed activation of these MAPKs in keratinocytes of the basal and spinous epidermal layers 30 min after exposure to low humidity (Fig. 1 D and E; H&E staining is shown in SI Appendix, Fig. S1E). Quantification of phospho-ERK- and phospho-p38-positive cells confirmed that TRPV4 inhibition attenuated this low humidity-induced MAPK activation. These results demonstrate that short-term exposure to low humidity induces MAPK signaling activation and gene expression changes in keratinocytes in 3D skin models and that both responses require TRPV4.

Low Humidity Elicited a TRPV4-Dependent Gene Expression Changes in 3D Keratinocytes.

To further investigate whether TRPV4 inhibition suppresses the downstream responses triggered by low humidity, we examined its effect on epidermal homeostasis during prolonged exposure to low humidity. While the impact of 24 h low humidity on keratinocytes in 3D skin models has not been fully characterized, we first performed RNA sequencing to compare gene expression in 3D skin exposed to low humidity (~20% RH) for 24 h with that in control skin at high humidity (~90% RH). Principal component analysis (PCA) revealed distinct differences in the gene profiles of the control and low-humidity-exposed 3D skin (Fig. 2A). Gene Ontology (GO) analysis of differentially expressed genes (DEGs) for biological processes revealed that low humidity-induced upregulation of several gene clusters, including those associated with cell adhesion, signal transduction, cell migration, cytosolic calcium ion concentration or keratinization/differentiation (Fig. 2B and SI Appendix, Tables S4 and S5). To visualize these expression patterns, we generated heatmaps using the Z-scores of normalized counts, which revealed clusters of genes upregulated by low-humidity exposure, including those related to cell adhesion and differentiation, as well as genes annotated with the GO term “inflammatory response” (SI Appendix, Fig. S2 D–F).

Fig. 2.

Low humidity elicits a TRPV4-dependent gene expression changes in 3D keratinocytes. (A–D) 3D skin was exposed to low humidity (~20% RH) for 24 h, and control skins were maintained at high humidity (~90% RH). Total RNA was extracted and subjected to RNA sequencing analysis and RT-qPCR. (A) PCA showing distinct gene expression profiles between control and low humidity exposed to 3D skin. (B) GO enrichment analysis of upregulated DEGs in skin exposed to low humidity. (C and D) 3D skin was exposed to low humidity (~20% RH) with HC-067047 (1 µM) or vehicle for 24 h. mRNA expression of indicated genes by RT-qPCR (mean ± SEM, n = 6, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by two-way ANOVA test). (E) Keratinocytes in 3D skin models were transfected with TRPV4 siRNA or nontargeting control siRNA and then exposed to low humidity for 3 h. mRNA expression of indicated genes by RT-qPCR (mean ± SEM, n = 6, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by two-way ANOVA test).

We next examined the expression of selected canonical adhesion- and differentiation-related genes. RT-qPCR analysis confirmed that the expression of cell adhesion-related genes, including integrin α6 (ITGA6), laminin α3 (LAMA3), and junction plakoglobin (JUP), and keratinocyte differentiation-related genes, including keratin 16 (KRT16), transglutaminase 1 (TGM1), and loricrin (LOR), was significantly increased in 3D skin cultured under low-humidity conditions for 24 h (Fig. 2 C and D and SI Appendix, Fig. S3A). Under the same conditions, treatment with the TRPV4 inhibitor HC-067047 suppressed the low humidity-induced upregulation of these transcripts.

To further confirm the involvement of TRPV4 in low humidity induced gene expression changes, TRPV4 expression in keratinocytes of the 3D skin was reduced by siRNA-mediated knockdown, and the effects on low humidity-induced gene expression changes after 24 h exposure were examined. TRPV4 siRNA effectively reduced TRPV4 expression at both the mRNA and protein levels (SI Appendix, Fig. S3 B and C), and attenuated the transcriptional responses induced by low humidity (Fig. 2E). These results suggest that TRPV4 is involved in low humidity-induced transcriptional changes in keratinocytes, as reflected by the upregulation of genes associated with cell adhesion and keratinocyte differentiation.

Low Humidity Induced Water Movement and Calcium Influx in Keratinocytes of 3D Skin.

TRPV4 is a thermosensitive and mechanosensitive calcium channel that responds to various physical stimuli, including moderate temperatures between 27 and 35 °C, hypotonic stress, and mechanical stimuli (20–22). To examine whether low humidity induces TRPV4-dependent Ca²⁺ influx into keratinocytes, we introduced GCaMP6f into 3D skin keratinocytes using an adenoviral expression vector (Ad-CAG-GCaMP6f), enabling live Ca²⁺ imaging (23). Upon exposure to low humidity, keratinocytes in the 3D skin exhibited a rapid increase in intracellular Ca²⁺ levels within 180 s (Fig. 3 A and B and Movies S1–S3). Ca²⁺ transients were readily detected when the imaging plane was focused on the basal-layer keratinocytes, whereas signals were minimal when the focal plane was shifted to the upper epidermal layers, indicating that the low-humidity-evoked Ca²⁺ response was predominantly observed in the basal keratinocytes in this model. Low-humidity-induced Ca²⁺ influx was suppressed with the pharmacological TRPV4 inhibitors HC-067047 or GSK2193874 (Fig. 3C), and siRNA-mediated knockdown of TRPV4 (Fig. 3D). These findings demonstrate that low humidity induces Ca²⁺ influx in epidermal keratinocytes in a TRPV4-dependent manner. The 3D skin model was maintained at a constant temperature (30 °C) during the experiments to minimize thermal fluctuations (24). Thus, the low-humidity-induced TRPV4-dependent response is unlikely to be explained by temperature fluctuations alone.

Fig. 3.

Low humidity induced water movement and calcium influx in keratinocytes of 3D skin. (A–D) Keratinocytes of 3D skin were transduced with Ad-CAG-GCaMP6f and subjected to time-lapse fluorescence microscopy during low-humidity exposure. (A–C) GCaMP6f-expressing 3D skin was treated with vehicle, HC-067047 (1 µM), or GSK2193874 (0.5 µM) for 30 min before imaging. Relative humidity was decreased from 50 to 60% RH to 15% RH within 20 s, and the time point at which RH reached 15% was defined as 0 s. Fluorescence images were acquired in FITC mode every 5 s for 180 s after reaching 15% RH. Each movie includes 15 s before 0 s and the subsequent 180 s of low-humidity exposure. (A) Representative fluorescence images at 0 s and 180 s. Arrows indicate representative cells exhibiting marked changes in GCaMP6f fluorescence. (Scale bar, 20 µm.) (B) Time course of normalized fluorescence intensity during low-humidity exposure. Fluorescence was quantified in ROIs drawn over keratinocytes after background subtraction and expressed as ΔF/F₀, where F₀ was defined as the mean fluorescence during the baseline period. (C) ΔF/F₀ measured 180 s after the onset of exposure to low-humidity or control-humidity conditions (n = 80 to 98, ****P < 0.0001 by two-way ANOVA). (D) GCaMP6f-expressing 3D skin was transfected with control siRNA or TRPV4 siRNA. GCaMP6f fluorescence intensity was quantified in FITC mode and expressed as ΔF/F₀ at 180 s after the onset of exposure to low-humidity or control-humidity conditions (n = 70 to 88, ****P < 0.0001 by two-way ANOVA). (E–H) Tritium-labeled water (³H-H₂O) was used to quantify water uptake from the culture medium, water evaporation from the skin surface, and water retained at 90 s after low humidity exposure (mean ± SEM, n = 4, **P < 0.01 by the unpaired t test). For details of the measurement procedure, see SI Appendix, Fig. S4. (I) 3D skin was preincubated with BAPTA-AM (20 μM) for 30 min under control humidity (~90% RH), followed by exposure to low humidity (~15% RH) for 30 min before protein extraction. (Left) Representative immunoblots probed with antibodies against phospho-ERK1/2, ERK1/2, phospho-p38, and p38. (Right) Densitometric analysis of band intensities performed using ImageJ (mean ± SEM, n = 4, ***P < 0.001, ****P < 0.0001 by two-way ANOVA test). (J) 3D skin was preincubated with U0126 (5 μM) for 30 min under control humidity (~90% RH), followed by exposure to low humidity (~15% RH) for 3 h. mRNA expression levels of the indicated genes were quantified by RT-qPCR (mean ± SEM, n = 5, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by two-way ANOVA test).

Next, we assessed the dynamics of water movement in response to low-humidity exposure by utilizing tritium-labeled water (³H-H₂O), as detailed in SI Appendix, Fig. S4. Specifically, three parameters were quantified: water uptake from the culture medium by the 3D skin, water loss due to evaporation from the 3D skin surface, and water retention in the 3D skin at 90 s. Upon exposure to low humidity for 90 s, water evaporation from the surface of the 3D skin almost doubled compared with that observed under high-humidity conditions (Fig. 3E). Notably, water uptake from the basal side of the tissue at 90 s also significantly increased under low-humidity conditions (Fig. 3F). The quantified water retained in the 3D skin was also higher at 90 s after exposure to low humidity (Fig. 3G), indicating that short-term low-humidity exposure transiently increased the water content of the 3D skin. TRPV4 inhibition did not affect the low-humidity-induced changes in water movement (Fig. 3H). These findings indicate that low-humidity exposure rapidly increased water movement across the 3D skin during the first 90 s.

We next examined whether the early Ca²⁺ influx observed within 3 min of low-humidity exposure contributes to subsequent MAPK activation. Pretreatment of 3D skin with the intracellular Ca²⁺ chelator BAPTA-AM reduced the low humidity-induced activation of ERK and p38 after 30 min of exposure (Fig. 3I). To further assess the contribution of MAPK signaling to low humidity-induced transcriptional changes, 3D skin was treated with the MEK inhibitor U0126 or the p38 inhibitor SB203580 before exposure to low humidity. U0126 treatment reduced the induction of low-humidity-responsive genes at 3 h (Fig. 3J), whereas SB203580 did not substantially affect their induction (SI Appendix, Fig. S3E). Western blotting confirmed that U0126 inhibited low humidity-induced ERK activation at 30 min (SI Appendix, Fig. S3D). These results suggest that Ca²⁺ signaling is involved in low humidity-induced MAPK activation, and that ERK signaling contributes to the transcriptional response to low-humidity exposure.

Trpv4 Deficiency in Keratinocytes Suppressed Epidermal Responses to Low Humidity in Mice.

To examine whether TRPV4 in keratinocytes contributes to the response to low-humidity exposure in vivo, we generated keratinocyte-specific Trpv4 conditional knockout mice using two Cre driver lines (hereafter referred to as Trpv4 cKO) (SI Appendix, Fig. S5A). Trpv4^fl/fl K5-Cre⁺ mice were used for experiments performed before 3 wk of age (Fig. 4 and SI Appendix, Fig. S5), whereas in mice 4 wk of age or older, Trpv4 deletion was induced in Trpv4^fl/fl K14-Cre ERT⁺ mice by tamoxifen administration (SI Appendix, Fig. S6). To confirm Trpv4 deletion in keratinocytes, EpCAM-positive keratinocytes were isolated from epidermal cell suspensions, which may contain Trpv4 expressing immune cells. Real-time PCR and western blotting confirmed reduced Trpv4 mRNA and TRPV4 protein levels in EpCAM-positive keratinocytes from Trpv4 cKO mice compared with control mice (Fig. 4 A and B and SI Appendix, Fig. S6A). Transepidermal water loss (TEWL), an indicator of skin barrier function, was comparable between control and Trpv4 cKO mice in both 1-wk-old K5-Cre mice and adult K14-Cre mice (Fig. 4C and SI Appendix, Fig. S6B). Histological analysis also revealed no obvious differences in epidermal structure between control and Trpv4 cKO mice (SI Appendix, Fig. S6C). In addition, immunostaining for major keratinocyte differentiation markers, including filaggrin, involucrin, and loricrin, showed similar staining patterns in control and Trpv4 cKO skin (SI Appendix, Fig. S6D).

Fig. 4.

Keratinocyte-specific Trpv4 deletion suppresses epidermal responses to low humidity in mice. (A–H) Trpv4^fl/fl mice and Trpv4^fl/fl K5-Cre⁺ mice were used as control and keratinocyte-specific Trpv4 conditional knockout mice (Trpv4 cKO), respectively. (A) Trpv4 mRNA expression in EpCAM⁺ keratinocytes isolated from the epidermis of control or Trpv4 cKO mice (mean ± SEM, n = 5, ***P < 0.001 by the unpaired t test). (B) Representative immunoblotting of epidermal lysates using antibodies against TRPV4 and β-actin. (C) Transepidermal water loss (TEWL) values measured on the dorsal skin (mean ± SEM, n = 6). (D–G) Control and Trpv4 cKO mice were maintained under control humidity (CH, 50 to 60% RH) or exposed to low humidity (LH, ~15% RH) for 3 h. (D) Total RNA was extracted from the separated epidermis and subjected to RNA-seq analysis. Heatmap showing gene expression profiles in the epidermis of control and Trpv4 cKO mice under CH or LH conditions. (E–G) mRNA expression of indicated genes by RT-qPCR (mean ± SEM, n = 5, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 by two-way ANOVA test). (H) Control or Trpv4 cKO mice were exposed to low humidity (~15% RH) for 15 to 45 min and extracted the protein from separated epidermis of the skin. (Left) Representative immunoblots probed with antibodies against phospho-ERK1/2, ERK1/2, phospho-p38, or p38. (Right) Densitometric analysis of band intensities performed using ImageJ (mean ± SEM, n = 5, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 using two-way ANOVA test).

For RNA-seq analysis of gene expression changes in response to low-humidity exposure, we used epidermal samples from 1-wk-old Trpv4 cKO mice because hair shaft contamination was frequently observed in epidermal preparations from mice older than 3 wk, making them unsuitable for reliable transcriptomic analysis. After exposing the mice to low humidity for 3 h, we extracted mRNA from the epidermis of the four groups (control or low humidity in each genotype) and performed RNA sequencing. A heatmap generated from the most variable genes revealed differences in expression patterns among the groups (Fig. 4D). In control mice, exposure to low humidity (LH; <20% RH) induced a distinct set of upregulated genes related to cell adhesion and inflammatory response ranked among the top categories (SI Appendix, Fig. S5B). At control humidity (CH; 50 to 60% RH), differential expression analysis showed minimal genotype-dependent differences between control and Trpv4 cKO mice, with only ~10 genes meeting the FDR-adjusted significance threshold (padj < 0.05). In contrast, comparison of the control and Trpv4 cKO mice under low-humidity condition identified 883 DEGs with padj < 0.01, indicating that the Trpv4 cKO genotype markedly altered the transcriptional profile under LH exposure (SI Appendix, Fig. S5C and Table S6). GO enrichment analysis of genes expressed at higher levels in the control epidermis compared with those in the Trpv4 cKO epidermis after low-humidity exposure identified biological processes related to cell adhesion, cell differentiation/keratinization, and inflammatory responses and visualized their expression patterns across the four groups using heatmaps (SI Appendix, Fig. S5 D–F).

RT-qPCR analysis verified that the selected canonical genes associated with cell adhesion (Itga6, Itgb4, Lama3, Jup, Dsc1) and keratinocyte differentiation (Keratin 16, Keratin 6, Transglutaminase1, Filaggrin, Loricrin) were comparable between control and Trpv4 cKO epidermis under control humidity. These genes were upregulated in the epidermis of control mice exposed to low humidity for 3 h, whereas their induction was attenuated in Trpv4 cKO epidermis (Fig. 4 E and F). Some of these genes were also upregulated in low-humidity-exposed 3D skin (shown in Fig. 2 C and D). Among the increased inflammatory response-related genes observed in the epidermis of control mice, the expression of IL-36 family genes, including Il36a, Il36b, and Il36rn, was significantly increased in response to low humidity, whereas this upregulation was not observed in the Trpv4 cKO epidermis (Fig. 4G). Similar low humidity-induced upregulation of these genes and its attenuation in Trpv4-deficient epidermis were also observed in 8-wk-old mice (SI Appendix, Fig. S6E).

Next, we examined whether short-term exposure to low humidity activates intracellular signaling in the mouse epidermis, as observed in keratinocytes in 3D skin (shown in Fig. 1D). Low-humidity exposure increased ERK and p38 phosphorylation in epidermal cells of control mice within 30 min, whereas this response was reduced in Trpv4 cKO epidermis (Fig. 4H). In addition, phospho-AKT quantification suggested that TRPV4 deficiency may also affect AKT signaling under low-humidity stress (SI Appendix, Fig. S5G), although the functional significance of this effect remains to be determined.

Taken together, these results suggest that TRPV4 contributes to low humidity-induced transcriptional changes in the mouse epidermis, as reflected by the upregulation of genes associated with epidermal adhesion, differentiation, and inflammatory responses.

Discussion

In this study, we demonstrated that TRPV4 expressed in epidermal keratinocytes contributes to major responses to low humidity and for mediating downstream signals. Using 3D skin equivalents, we showed that pharmacological inhibition of TRPV4 in epidermal keratinocytes suppressed low-humidity-induced MAPK activation, including ERK and p38, as early as 30 min after exposure and reduced the induction of low-humidity-responsive genes assessed at 3 h. Although HC-067047 attenuated these responses in a dose-dependent manner, 0.1 μM HC-067047 had little effect despite the reported IC50 of 48 nM. This may reflect limited diffusion or tissue architecture-dependent drug accessibility in the reconstructed 3D skin model, but it may also indicate that TRPV4 is not the sole contributor to the response and that other receptors or channels participate in low-humidity-induced signaling. Exposure to low humidity elicited rapid Ca²⁺ influx within 3 min, which was attenuated by TRPV4 inhibition or knockdown. Inhibition of Ca²⁺ entry attenuated low-humidity-induced phosphorylation of ERK and p38, indicating that Ca²⁺ influx is required, at least in part, for activation of these MAPK pathways. Furthermore, pharmacological inhibition of ERK reduced the subsequent low-humidity-induced changes in gene expression, suggesting that ERK signaling contributes to the transcriptional response downstream of early Ca²⁺-dependent signaling. Similarly, in vivo experiments using mouse skin showed that low-humidity exposure induced MAPK signaling and marked transcriptional changes in the mouse epidermis; however, these responses were significantly blunted in keratinocyte-specific Trpv4 knockout mice. Collectively, these findings support a model in which TRPV4 in keratinocytes mediates early low-humidity-induced signaling and contributes to downstream transcriptional responses elicited by low-humidity stress.

The mechanisms underlying TRPV4 activation have been well characterized in various contexts, and the channel is known to respond to changes in temperature (27 to 35 °C), hypotonic stress, and mechanical stimuli (25–29). Previous studies using Trpv4 knockout mice or TRPV4 inhibitors have demonstrated the involvement of the channel in various cutaneous physiological processes, including barrier function (30, 31), cell damage (19), UVB-induced pain (32), chronic itching (17), dry skin–associated pruritus (33), and wound healing (34). On the other hand, evidence for TRPV4 involvement in keratinocyte responses to hypotonicity is conflicting, with reports of both TRPV4-dependent activation (18, 35) and TRPV4-independent Ca²⁺ influx and volume regulation (36). In this study, low humidity elicited rapid Ca²⁺ influx within 3 min, suggesting that upstream events occurring on a subminute timescale may contribute to TRPV4-dependent signaling.

Consistent with this idea, when the 3D skin was exposed to low humidity for 90 s, we detected a marked increase in water movement, with both evaporation from the apical surface and water uptake from the basal side being elevated. Although prolonged exposure to low humidity (>10 min) ultimately reduced the total tissue water content, at 90 s, the 3D skin retained more water than the control skin maintained at 90% humidity. These observations suggest that, during the earliest phase of low-humidity exposure, rapid trans-tissue water movement, rather than immediate net dehydration, may predominate. A plausible consequence of this dynamic water movement is the transient alteration of the local osmotic and mechanical environment experienced by keratinocytes. For example, enhanced water movement and redistribution may alter local osmotic gradients near the epidermal basement membrane and could promote transient cell swelling, membrane stretching, or changes in membrane organization. Although we did not directly quantify osmolarity, cell volume, or membrane tension in the present study, this biophysical scenario may provide a possible route by which low humidity contributes to TRPV4 activation.

In this context, aquaporins may be relevant because previous studies in other epithelial systems have shown functional coupling between aquaporin 5 and TRPV4 during osmotic or cell-volume responses (37). Although aquaporin 3 (AQP3) is prominently expressed in epidermal keratinocytes (38, 39), treatment with an AQP3 inhibitor in our screening assay did not clearly attenuate the low-humidity-induced gene-expression responses (SI Appendix, Fig. S2B). Because this screen was based on selected transcriptional readouts, our data do not exclude the possibility that AQP3 or other aquaporins contribute to the initial water movement or local biophysical changes upstream of TRPV4 activation. Further studies directly assessing aquaporin-dependent water flux and TRPV4 activation during the earliest phase of low-humidity exposure will be required to clarify this possibility.

Previous studies have shown that reduced skin hydration or exposure to dry environmental conditions affects epidermal barrier function and induces inflammatory pathways in human skin and mouse models (3). For example, decreased hydration of human skin tissue has been reported to induce inflammatory mediators, including COX-2/PGE2 and S100A8/S100A9 (13, 14). In this study, exposure to low humidity affected multiple functional categories, including cell adhesion and inflammatory responses, in both 3D skin and mouse epidermis. Because our study focused on identifying early molecular responses to low humidity, we did not examine the effects of prolonged exposure on whole-skin physiology, immune cell recruitment, or systemic outcomes. Thus, our findings indicate that low humidity induces an epidermal transcriptional response that includes cytokine- and chemokine-related genes, rather than directly demonstrating overt skin inflammation. Specifically, low humidity significantly increased the expression of CXCL8 and CCL20 (SI Appendix, Fig. S1A), which are associated with the recruitment of neutrophils and T cells, respectively. These findings suggest that dry environmental conditions may create an epidermal milieu that favors immune cell recruitment if exposure is sustained. Because epidermal chemokine expression and immune cell infiltration are common features of inflammatory skin diseases (40–43), future studies should determine whether these early epidermal gene expression changes lead to immune cell recruitment, systemic immunological changes, or altered disease susceptibility in vivo. Such studies may inform strategies for preventing or treating skin disorders associated with dry environmental conditions.

In summary, our study demonstrates that TRPV4 expressed in epidermal keratinocytes is required for major aspects of the response to low humidity and contributes to downstream signaling. Although our results highlight an important role for keratinocyte TRPV4, we cannot exclude the possibility that humidity sensing and adaptation also involve neuronal pathways, as has been previously proposed. Future studies should clarify the distinct and potentially complementary roles of keratinocytes and neuronal cells in humidity detection and the resulting physiological responses. Given its role in mediating epidermal responses to low humidity, TRPV4 may provide a basis for developing strategies to prevent or manage skin disorders triggered or exacerbated by dry environmental conditions.

Materials and Methods

Human 3D Skin Models.

Human 3D skin models [LabCyte EPI-MODEL, Japan Tissue Engineering Co., Ltd. (J-TEC)] were maintained in EPI-MODEL assay medium at 37 °C in a 5% CO₂ incubator under high-humidity conditions [relative humidity (RH), ~95%] according to the manufacturer’s protocol.

For siRNA knockdown and GCaMP6f imaging experiments, 3D skin models were reconstructed from human keratinocytes obtained from J-TEC (LabCyte EPI-KIT). Keratinocytes were transfected with nontargeting or human TRPV4 siRNA and seeded onto Transwell inserts. For low-humidity exposure, 3D skin models were exposed either in a low-humidity incubator or in a custom humidity-control chamber, depending on the experimental design. Detailed procedures for siRNA transfection, adenoviral transduction, humidity exposure, and Ca²⁺ imaging are provided in SI Appendix, Materials and Methods.

Mice.

C57BL/6 mice were purchased from Japan SLC, Inc. Mice carrying loxP-flanked Trpv4 alleles (Trpv4^fl/fl) were created by Cyagen (USA). Keratinocyte-specific Trpv4 conditional knockout mice were generated by crossing Trpv4 ^fl/fl mice with either K14-CreERT or K5-Cre transgenic mice (44) (45). For inducible deletion of Trpv4, Trpv4^fl/fl K14-CreERT^+/− mice were administered tamoxifen intraperitoneally for five consecutive days. Detailed housing and humidity-control conditions are provided in SI Appendix, Materials and Methods.

All animal experiments were approved by the President of Keio University after review by the Institutional Animal Care and Use Committee of Keio University (approval no. 16075) and the Genetic Modification Safety Committee of Keio University School of Medicine (approval no. 28-029), and were conducted in accordance with institutional and national animal welfare guidelines and laws.

RT-qPCR.

Total RNA was extracted from 3D skin or mouse epidermis using the RNeasy Mini Kit or RNeasy Fibrous Tissue Mini Kit, respectively, according to the manufacturer’s instructions (QIAGEN). cDNA was synthesized using the PrimeScript RT Reagent Kit (Takara Bio). Quantitative RT-PCR was performed using TB Green (Takara Bio) on a CFX Opus Real-Time PCR System (Bio-Rad). The primer sequences used for quantitative RT-PCR are listed in SI Appendix, Tables S7 and S8. Detailed procedures are provided in SI Appendix, Materials and Methods.

RNA Sequencing Analysis.

RNA sequencing analysis of total RNA from 3D skin models and mouse epidermis was outsourced to Rhelixa Inc. Detailed methods are described in SI Appendix, Materials and Methods.

Immunofluorescence.

3D skin tissues or mouse skin samples were fixed in 4% formalin in PBS, embedded in OCT compound, and cryosectioned at 6 μm. Sections were permeabilized with 0.1% Triton X-100, blocked with 5% BSA, and incubated with the indicated primary antibodies for 1 h at 4 °C. After washing, sections were incubated with fluorophore-conjugated secondary antibodies and counterstained with DAPI. Images were acquired using a fluorescence microscope (BZ-800; Keyence). Details of the primary and secondary antibodies are provided in SI Appendix, Table S9.

Immunoblotting.

Mouse epidermis was separated from the dermis by incubation in PBS at 60 °C for 10 s. 3D skin or mouse epidermis were lysed and homogenized in RIPA buffer containing a protease inhibitor cocktail. The supernatants (10,000g, 10 min, 4 °C) were mixed with NuPAGE® LDS Sample Buffer and Reducing Agent, then heated at 70 °C for 10 min. The samples were then loaded onto a NuPAGE® 10% Bis-Tris Gel with MOPS SDS running buffer (ThermoFisher) and transferred to PVDF membrane. Membranes were incubated with primary antibodies, followed by HRP-conjugated secondary antibodies. Signals were detected using Immobilon Western Chemiluminescent HRP Substrate (EMD Millipore Corporation). Details of the antibodies and reagents are provided in SI Appendix, Table S9.

Water Flow Measurement.

3D skin was incubated with [³H]-labeled water (Perkin Elmer, NET001B, 1mCi/ml in assay medium, 1 to 60 mins) with either low or high humidity, as described in SI Appendix, Fig. S4. After washing the tissues with ice-cold PBS, tissues were homogenized in RIPA buffer (Cell signaling technology). 3D tissues-associated radioactivity was determined with an AccuFLEX LSC-8000 liquid scintillation counter (Hitachi).

Statistical Analysis.

Statistical analysis was performed using the two-tailed Student’s t test, one way, or two-way ANOVA as indicated using GraphPad Prism8. A P < 0.05 was considered statistically significant for all tests.

Supplementary Material

Appendix 01 (PDF)

Movie S1.

Vehicle-treated 3D skin.

Movie S2.

HC-067034-treated 3D skin.

Movie S3.

GSK2193874-treated 3D skin.

Data, Materials, and Software Availability

RNA-seq (raw reads and processed count matrix) data have been deposited in the DDBJ BioProject database under accession numbers PRJDB40265 (GEA: E-GEAD-1276) (46) and PRJDB40264 (GEA: E-GEAD-1275) (47). Study data are included in the article and/or supporting information.