Carbon dioxide‐induced decrease in extracellular pH enhances the production of extracellular matrix components by upregulating TGF‐β1 expression via CREB activation in human dermal fibroblasts

Mild acidification caused by transcutaneous administration of carbon dioxide (CO2) has been reported to improve some epidermal skin impairments, such as desquamation and inflammation; however, its effects on dermal tissue remain unclear. Here, we examined the effect and mechanism of mild acidity on extracellular matrix (ECM) protein production in normal human dermal fibroblasts (NHDFs). To achieve this, the skin permeability of CO2 and its effect on intradermal pH were evaluated by treating reconstructed human skin equivalents (HSEs) with a CO2‐containing formulation. Additionally, NHDFs were cultured in a pH‐adjusted medium (pH 6.5). CO2 successfully permeated HSEs and reduced the intradermal pH. Decreased extracellular pH activated CREB, upregulated TGF‐β1 expression, promoted the production of elastic and collagen fibres, and increased hyaluronan concentration in NHDFs. Additionally, the low pH‐induced increase in TGF‐β1 expression was attenuated via the RNAi‐mediated suppression of the expression of CREB1 and proton‐sensing G protein‐coupled receptors (GPCRs), including GPR4 and GPR65. Moreover, low pH‐induced CREB activation was suppressed by the inhibition of the cAMP/PKA and PLC/PKC signalling pathways. Taken together, a CO2‐induced decrease in intradermal pH may promote ECM production in NHDFs via the upregulation of TGF‐β1 expression, which was mediated by the activation of the GPCR signalling pathway and CREB, indicating that CO2 could be used to treat ultraviolet radiation‐induced photoaging, intrinsic ageing and ECM deterioration.


| INTRODUC TI ON
Carbon dioxide (CO 2 ) is an end-product of aerobic cellular respiration and plays various roles in human physiology.2][3][4][5][6][7] Although these therapeutic benefits are believed to be achieved via CO 2 -induced vasodilation or improvement in microcirculation, another important mechanism underlying the effects of CO 2 is the Bohr effect, which seems to be responsible for acute microenvironmental changes after CO 2 application, such as decreased oxygen-haemoglobin concentration, increased partial pressure of oxygen and decreased pH in peripheral tissues. 8reover, recent findings suggest that transcutaneous application of CO 2 with concomitant mild acidification of the stratum corneum (SC) can prevent skin impairment (decreased skin barrier and moisturizing functions and surface roughening) 9 and improve desquamation in xerotic skin, thereby leading to improvement in scaling. 10Additionally, CO 2 has been reported to improve acne symptoms via the normalization of keratinization. 11Furthermore, CO 2 was observed to exert anti-inflammatory effects by suppressing NF-κB activation via extracellular acidification in human keratinocytes, and the topical application of CO 2 inhibited ultraviolet (UV)-B radiation-induced erythema formation, indicating that CO 2 can suppress skin inflammation in vivo. 12However, the effects of transdermal administration of CO 2 on dermal tissue remain to be understood.
The dermis comprises cellular components and an extracellular matrix (ECM) that provides tensile strength, elasticity and moisture to the skin. 13,14[20][21] Transforming growth factorβ (TGFβ) is a pleiotropic cytokine expressed in various cells, and ligand stimulation is mediated by the TGFβ/SMAD signalling pathway that consists of TGFβ receptors and a group of SMAD proteins, which regulate several cellular processes, including cell growth and apoptosis, differentiation, migration and metastasis. 22Among the three TGFβ isoforms identified in mammals (TGF-β1, TGF-β2 and TGF-β3), 23 TGF-β1 is the most prevalent and has been reported to play an important role in promoting the production and deposition of ECM proteins. 24Several studies have reported that the TGFβ/SMAD signalling pathway regulates ECM-related factors in dermal fibroblasts, such as promoting elastin gene expression, 25 type I collagen production 26 and hyaluronan synthase (HAS) expression, 27 and suppressing the expression of metalloproteinase (MMP)-1 28 and hyaluronan-binding protein involved in hyaluronan depolymerization (HYBID). 29Moreover, TGF-β1 is a well-known autocrine factor produced by fibroblasts 30,31 ; therefore, it is believed that ECM production could be mediated through the regulation of TGF-β1 expression.
Here, we examined the effect and mechanism of action of transcutaneous CO 2 administration on ECM production in normal human dermal fibroblasts (NHDFs), with an emphasis on the effect of CO 2induced decrease in intradermal pH.

| Test samples
The CO 2 formulation used in the present study, which is a waterbased emulsion dissolved at a saturated concentration of CO 2 gas (1700 mg L −1 ), was prepared according to previously described procedures. 10,11The foam, immediately after dispensing from the aerosol can, contains 8000-9000 ppm of CO 2 (measured by the alkaline trap method).The placebo formulation contained the same ingredients except for CO 2 .For the control, hydrochloric acid (HCl) was added to the placebo formulation to adjust the pH to that of the CO 2 formulation (pH 6.0).

| Cell culture
Normal human dermal fibroblasts, including NHDFs (Kurabo) and Detroit 551 cells (American Type Culture Collection), were cultured in Dulbecco's modified eagle medium (DMEM) (Sigma-Aldrich) supplemented with 10% (v/v) foetal bovine serum (FBS) (Sigma-Aldrich) at 37°C in a humidified atmosphere containing 5% or 15% CO 2 for experiments involving cells cultured under high concentrations of CO 2 .Normally, the pH of DMEM is maintained at 7.4-8.0,and the medium was used as the control for each experiment (denoted as normal pH in Figures).

| Immunocytochemistry
Normal human dermal fibroblasts were seeded on a cover glass in 24-well plates (3.0 × 10 5 cells/well) containing DMEM supplemented with 5% (v/v) FBS for 24 h.Thereafter, the medium was replaced with DMEM/Ham's F-12 (Sigma-Aldrich) with low pH

| Enzyme-linked immunosorbent assay
To measure hyaluronan concentration, NHDFs were seeded in 12well plates (4.0 × 10 5 cells/well) containing DMEM supplemented with 2% (v/v) FBS for 24 h.Thereafter, the medium was replaced with low pH (6.5) DMEM (without FBS) and was further incubated for 48 h.The hyaluronan concentration of the NHDF culture medium was determined using the QnE Hyaluronic Acid Enzyme-linked immunosorbent assay (ELISA) Assay Kit (Biotech Trading Partners), according to the manufacturer's instructions.To measure TGF-β1 concentration, NHDFs or Detroit 551 cells were seeded in 6-well plates (8.0 × 10 5 cells/well) containing DMEM supplemented with 2% (v/v) FBS for 24 h, followed by the replacement of the medium with DMEM (without FBS) and further incubation for 5 days.
Subsequently, the cells were washed with PBS and incubated in low pH (6.5) DMEM supplemented with 2% (v/v) FBS for 24 or 48 h.The TGF-β1 concentration in the culture media was measured using the Human TGF-beta 1 Quantkine ELISA kit (R&D Systems), according to the manufacturer's instructions.S1).The expression levels of target genes were normalized to that of 60S acidic ribosomal protein (RPLP0).

| Immunological detection of CREB phosphorylation
Normal human dermal fibroblasts or Detroit 551 cells were seeded in 6-well plates (5.0 × 10 5 cells/well) containing DMEM supplemented with 2% (v/v) FBS for 24 h, followed by the replacement of the medium with DMEM (without FBS).After 6 days of incubation, the cells were washed with PBS and re-incubated with low pH (6.5) DMEM (without FBS) for 10 min, 30 min, 90 min, 6 or 24 h.Subsequently, the cells were washed with ice-cold PBS two times and solubilized in 250 μL of cell lysis buffer (RayBiotech) supplemented with 1% protease/phosphatase inhibitor cocktail (100×) (Cell Signaling Technology) and incubated for 30 min at 4°C.The cell lysates were centrifuged at 14 000 g for 10 min, and the resulting supernatants were subjected to immunoblotting using PhosphoPlus® CREB (Ser 133) Antibody Duet (Cell Signalling Technology).In experiments with inhibitors, NHDFs were seeded in 6-well plates (6.0 × 10 5 cells/well) containing DMEM supplemented with 2% (v/v) FBS for 24 h, followed by the replacement of the medium with DMEM (without FBS).
After incubation for 4 days, the cells were pre-treated with H-89 and/or Go 6983 or DMSO as solvent control for 24 h.Subsequently, the cells were washed with PBS and re-incubated with low pH (6.5) DMEM (without FBS) for 20 min in the presence or absence of inhibitors and then solubilized as described above.

| RNA interference
Normal human dermal fibroblasts were seeded in 24-well plates (5.0 × 10 4 cells/well) or 12-well plates (1.0 × 10 5 cells/well) containing DMEM supplemented with 2% (v/v) FBS for 24 h, followed by transfection with 25 nM of specific siRNA against CREB1, GPR4, and GPR65 or control non-target siRNAs (Silencer® Select siRNAs, Table S2, Thermo Fisher Scientific) using Lipofectamine RNAiMAX transfection reagent (Thermo Fisher Scientific), according to the manufacturer's instructions.After 24 h, the culture medium was changed to fresh DMEM (without FBS) and further incubated for 24 h to 6 days.
To suppress both GPR4 and GPR65 expressions (double knockdown), cells were transfected using 25 nM of each siRNA (total of 50 nM).

| Cell viability assay
Normal human dermal fibroblasts were seeded in 24-well plates (2.0 × 10 5 cells/well) containing DMEM supplemented with 2% (v/v) FBS for 24 h, followed by the replacement of the medium with DMEM (without FBS) and further incubation for 5 days.Cells treated with H-89 or Go 6983 for 3 days were washed with PBS, and the culture medium was changed to fresh DMEM (without FBS) containing 10% alamarBlue Cell Viability Reagent (Thermo Fisher Scientific).After incubation for 3 h, fluorescence (Ex/Em = 560/590) in each well was measured using a microplate reader (Infinite® M200PRO; Tecan Trading AG).

| Statistical analysis
Data were compared using Student's t-test, Dunnett's test and Tukey's multiple comparison test, and mean values were considered statistically significant at p < 0.05.All statistical analyses were performed using IBM SPSS Statistics 25.0 (IBM) or Kyplot software (KyensLab Inc.).

| CO 2 penetrates from the SC side to the dermis, lowering the intradermal pH of HSEs
The permeability of CO 2 from the skin surface to the dermal layer and its effect on intradermal pH was examined, as per the protocol described in a previous study that evaluated the intercellular pH of the epidermis using a three-dimensional (3D) cultured epidermal model and a fluorescent pH indicator. 10Treatment with the placebo formulation (pH 6.9) did not significantly affect intradermal pH.In contrast, treatment with the CO 2 formulation (pH 6.0) caused a rapid decrease in intradermal pH in HSEs, with pH decreasing to approximately 6.04 within 12 min, followed by a gradual recovery to almost the initial state after 90-100 min (Figure 1).Moreover, the application of the formulation containing HCl (pH 6.0) did not cause a significant pH change (Figure 1).Overall, these results suggest that CO 2 permeated the dermal layer of the skin to reduce transient pH after the application of the CO 2 -containing formulation.

| Low extracellular pH promotes elastic and collagen fibre formation and hyaluronan synthesis in NHDFs
The effect of the CO 2 -induced decrease in intradermal pH on the production and pericellular deposition of elastic fibres, collagen fibres and hyaluronan in NHDFs was examined.There was an increase in tropoelastin mRNA (ELN) expression and a decrease in neprilysin (skin fibroblast-derived elastase) mRNA (NEP) expression in NHDFs after 24, 48 and 72 h of culture in low pH medium (Figure S1a,b).Similarly, immunofluorescence assays showed an increase in elastic fibre formation in NHDFs cultured in a low pH medium (pH 6.5) for 15 days (Figure 2A), which was confirmed by binarized images (Figure 2B).Additionally, NHDFs cultured in a low pH medium had higher mRNA expression of COL1A1, a type 1 collagen molecule, after 24 h of exposure (Figure S1c).Similarly, there was a significant increase in collagen fibre formation after 72 h under low pH (pH 6.5) culture condition (Figure 2C,D).Moreover, low pH condition (pH 6.5) significantly increased the mRNA expression of HAS2, a hyaluronan synthase, in NHDFs after 24 h of exposure (Figure S1d), and increased hyaluronan concentration after 48 h of exposure (Figure 2E).

| Low extracellular pH upregulates TGF-β1 protein synthesis in NHDFs
Furthermore, we examined the effect of low pH on the production of ECM proteins in NHDFs; particularly, we focused on TGF-β1, a TGFβ superfamily ligand.Compared to the normal culture medium, the low pH culture medium (pH 6.5) increased the production of TGF-β1 by NHDFs (Figure 2F).Moreover, there was a significant increase in TGF-β1 mRNA expression in NHDFs after culturing for 8 h in a low pH medium (Figure 2G).Based on these results, we hypothesized that TGF-β1 transcriptional activation was induced as an initial response to low environmental pH and that the upregulation and autocrine action of TGF-β1 induced the activation of the TGFβ/SMAD signalling, followed by changes in the expression of various ECM-related factors.Consistent with this hypothesis, the mRNA expression of SMAD3, one of the major factors responsible for signal transduction in the TGFβ/SMAD signalling, was upregulated in NHDFs in response to a low pH (Figure S2).In addition, NHDFs cultured under a 15% CO 2 environment showed similar expression changes in various ECMrelated genes.Under this condition, a decrease in the pH of the culture medium was also induced (Figure S3).Additionally, the above results were confirmed in HSEs cultured in low pH (6.5) medium (Figure S4).

| Low extracellular pH increases TGF-β1 production in NHDFs via CREB activation
Previous studies have demonstrated that the cAMP response element (CRE) is present in the promoter region of TGF-β1, and its transcriptional activation is induced by phosphorylated CREB. 32CREB is a transcription factor that binds to CRE sequences and is known to play a key role in regulating cell proliferation, differentiation and adaptation processes through its ability to increase or decrease gene transcription. 33,34Immunoblot analysis using a phosphorylated CREB (Ser133) monoclonal antibody showed an increase in CREB phosphorylation in NHDFs after 10 and 30 min of exposure to low pH (6.5) medium (Figure 3A).Moreover, replacement with another pH-adjusted medium (pH 7.0) also induced a considerable increase in CREB phosphorylation, although to a lower extent than that using the low pH (pH 6.5) culture condition, indicating that there was a linear relationship between the degree of acidity and CREB activation (Figure S5).CREB knockdown experiments were conducted to elucidate its role in regulating TGF-β1 expression in NHDFs.Transfection with two CREB1-specific siRNAs suppressed CREB1 mRNA expression by 80% after 48 h of transfection (Figure 3B) and suppressed CREB protein expression by approximately 80% after 72 h of transfection (Figure 3C).Additionally, CREB1 knockdown suppressed low pH-induced TGF-β1 mRNA and protein expression in NHDFs (Figure 3D,E), indicating that the low pH-induced increase in TGF-β1 production was mediated by CREB activation.

| NHDFs detect low environmental pH via GPR4 and GPR65
Furthermore, the receptors through which NHDFs sense a decrease in the pH of the extracellular environment were identified.As CREB is generally known to be activated by G protein-coupled receptors (GPCRs), 35,36 we focused on proton-sensing GPCRs (pH-GPCRs) 37,38 and performed gene knockdown experiments.NHDFs were transfected with siRNAs for four pH-GPCRs, including GPR4, TDAG8 (T cell death-associated gene 8; GPR65), OGR1 (ovarian cancer G protein-coupled receptor 1; GPR68) and G2A (G2 accumulation protein; GPR132) 38 ; however, we could not achieve sufficient suppression of GPR68 and GPR132 (data not shown).Therefore, we evaluated the role of GPR4 and GPR65 only.GPR4 and GPR65 mRNA expressions were suppressed in NHDFs by approximately 90% and 70%, respectively, after 48 h of transfection with siRNAs (Figure 4A,B).Moreover, double knockdown of GPR4 and GPR65 significantly suppressed the low pH-induced increase in TGF-β1 production (Figure 4C), indicating that GPR4 and GPR65 were activated by a decrease in extracellular pH and contributed to downstream signalling.

F I G U R E 1
Carbon dioxide (CO 2 ) induces a temporary decrease in intradermal pH in reconstructed human skin equivalents.Reconstructed human skin equivalents were pre-treated with a pH-sensitive fluorescent indicator, 2′,7′-bis(carboxyethyl)-4 or 5-carboxyfluorescein (BCECF), to measure intradermal pH.The fluorescence was measured from the bottom of the culture plate using a microplate reader during a 100-min treatment with the CO 2 , placebo or HCl formulation.The measurements were obtained from nine points per well.The intercellular pH within the dermis was calculated based on the standard curve of the pH standard solution.

| DISCUSS ION
Transdermal absorption of CO 2 is believed to induce two initial responses in tissues, including transient hypoxia or low pH caused by proton dissociation from carbonic acid.In the present study, we examined the effects of CO 2 on dermal tissue.Consistent with a previous finding, 10 the application of CO 2 formulation caused a decrease in pH in the lower fluid of HSEs.Although further studies are required to clarify this effect, which was not observed in the application of the placebo formulation adjusted to the same pH, we speculate that the small molecular weight and amphiphilic property of CO 2 may be major contributing factors.As a point of concern, the CO 2 -induced decrease in pH in HSEs was transient, which was contrary to a previous finding in the 3D cultured epidermal model, in which a CO 2 -induced decrease in pH was sustained. 10The discrepancies in results could be attributed to differences between the 3D cultured epidermal model and human skin, such as SC structure, the composition of the epidermal living cell layer and buffering function.
Therefore, it is assumed that even among 3D skin models, the presence of a dermal layer can affect CO 2 permeability (proton transmission) and the buffering capacity of HBSS.As the dermal layer was well-simulated in the present study, we concluded that the results may be more consistent with the actual situation in human skin.
However, in addition to penetrating the skin, the CO 2 contained in the formulation used in this study volatilizes into the atmosphere over time.Therefore, if the volatilization of CO 2 can be suppressed by optimizing the composition to allow for its prolonged retention in the formulation, the pH decrease may be sustained in the dermis layer as well.
As an existing technology that utilizes CO 2 , the effect of carboxytherapy, a mesotherapy-like non-surgical procedure involving the infusion of CO 2 gas below the skin, is partly attributed to the stimulation of collagen synthesis.However, the underlying mechanism of carboxytherapy is thought to be the induction of collagen reconstruction through physical exfoliation of the dermal layer by the injection pressure of CO 2 . 39,40Moreover, CO 2 has been reported to promote skin wound healing. 41Several possible mechanisms have been proposed to explain the therapeutic effect of carbonated hot spring water on skin wound healing, such as a high temperature-induced increase in blood flow and circulation, 2,42 or antimicrobial effects of the components of carbonated hot spring water on the skin. 43Liang et al. 41 found that hot spring water suppressed macrophage infiltration and promoted MMP-2 expression and activity.Additionally, pH has also been reported to be partially involved in this mechanism.5][46] In the present study, we speculated that the CO 2 -induced decrease in extracellular pH may promote ECM synthesis in NHDFs.Our results also support those of the study by Park et al. 47 reporting that acidic pH (<6.04) slightly upregulated MMP-1 and slightly downregulated COL1A1 levels via ROS generation and the p38 signalling pathway in human dermal fibroblasts.Although the data were not shown, in our preliminary study, we also visually observed that prolonged exposure of NHDFs to pH 6.0 medium caused scattered damage, which we judged to be an inappropriate condition for obtaining stable results.Therefore, we believe that pH 6.5 is the optimal pH to elicit positive effects at the cellular level.Notably, the control condition used in the present study (pH 7.4-8.0)was comparable with the pH range of the human dermal tissue reported in a previous study showing pH changes in the dermis during the tuberculin skin test (mean pH 7.58, SD 0.05, n = 6). 48Therefore, it could be concluded that the conditions in the present study were similar to those of the human skin; however, studies are yet to confirm if the transcutaneous application of CO 2 can reduce intradermal pH in human skin.
Regarding the detection of the pH of the external environment in NHDFs, clear conclusions could not be drawn mainly for two reasons.First, acid-sensing ion channels (ASICs), 49,50 which have a sensing range of pH 4-7, can also be activated at pH 6.5.Moreover, the role of ASIC activation-induced increase in intracellular Ca 2+ levels, which stimulates Ca 2+ /calmodulin-dependent protein kinase (CaMK) and enhances CREB phosphorylation, was not examined in the present study.Moreover, only the roles of two pH-GPCRs among four identified in NHDFs, including GPR4 and GPR65, were evaluated in the present study owing to differences in siRNA-induced gene suppression efficiency.Among the remaining two, GPR132 can be excluded as a candidate as it has been reported to be consistently activated under physiological pH conditions. 51,52In the case of GPR68, it would be worthwhile to obtain siRNAs capable of suppressing its expression to elucidate its role in CO 2 -induced increase in ECM synthesis.However, double knockdown of GPR4 and GPR65 significantly suppressed the low pH-induced increase in TGF-β1 expression in NHDFs.
Furthermore, decreased pH of the extracellular environment significantly upregulated SMAD3 expression.As it has been reported that SMAD3 expression is regulated by TGFβ, and SMAD3 plays an important role in ECM production, 53 it could be concluded that its upregulation contributed to the positive effects of low pH in NHDFs.
When considering the contribution of the TGFβ pathway, the phosphorylation state is typically considered more important than the expression of each factor in the pathway.In the present study, SMAD2 phosphorylation (a receptor-regulated SMAD) or SMAD4 phosphorylation (a common partner SMAD) was not significantly upregulated in NHDFs exposed to low pH medium for 10, 30 or 90 min (data not shown).Based on this result, it was concluded that the decrease in the pH of the extracellular environment did not directly activate the SMAD-dependent TGFβ pathway and that TGF-β1 acted on the autocrine system to enhance the expression of various ECM-related factors.However, further studies are necessary to comprehensively elucidate the timing of the activities of TGF-β1.For example, as the expression of ECM-related factors fluctuated after 24 h of incubation in a low pH medium, the promotion of TGF-β1 secretion might be induced prior to the promotion of synthesis.Moreover, it would be desirable to verify the effects of TGF-β1 using neutralizing antibodies for TGF-β1 or inhibitors of TGF-βR1.
This study had some limitations.First, although we investigated the effects of culturing in a low-pH medium in most cell experiments, we also attempted, based on a previous report, 12 to culture fibroblasts under a 15% CO 2 environment to evaluate the direct effect of CO 2 in a monolayer culture system.Although the gene expression changes under this condition were similar to those observed when cells were cultured in a low pH (6.5) medium, we observed that the pH of the medium decreases when cultured in a 15% CO 2 environment.Therefore, we considered that we were essentially observing the same effect and that we cannot extend our research any further.
In the future, we will use more complex experiments to further study the effects of the CO 2 formulation in detail using the 3D skin model Reconstructed human skin equivalents (HSEs) comprising both epidermal and dermal layers (T-Skin; Nikoderm Research Inc.) were preincubated in the assay medium for 24 h.After the medium was changed to Hanks' Balanced Salt Solution (HBSS) containing 10 μM of 2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein (BCECF; Dojindo), a pH-sensitive fluorescent indicator, the test samples (approximately 0.7 g) were applied above the SC of HSEs and fluorescence was measured from the bottom of the culture plate every 2 min for 100 min using a microplate reader (Infinite® M200PRO; Tecan Trading AG; Ex/Em = 440/530, 500/530).Intercellular pH was calculated from the 500/440 excitation ratio by calibration using the data from HSEs pre-treated with the buffer provided in the Intracellular pH Calibration Buffer Kit (Thermo Fisher Scientific) containing 10 μM BCECF.

( 6 . 5 )
and further incubated [15 days in DMEM/Ham's F-12 supplemented with 2% (v/v) FBS for elastic fibre formation evaluation; 3 days in DMEM/Ham's F-12 supplemented with 0.5% (v/v) FBS for collagen fibre formation evaluation].NHDFs fixed in ice-cold methanol were blocked for non-specific immunoreactivity using phosphate-buffered saline (PBS) containing 5% (v/v) bovine serum albumin and then incubated with diluted mouse anti-human elastin (Merck Millipore; 1:200 dilution) or rabbit anti-human collagen-1 antibody (Cedarlane Laboratories Ltd.; 1:100 dilution) and subsequently with diluted Alexa Fluor 555-labelled anti-mouse IgG (Thermo Fisher Scientific; 1:200 dilution) or Alexa Fluor 488-labelled anti-rabbit IgG antibody (Thermo Fisher Scientific; 1:200 dilution).Cells attached to the cover glass were mounted on a slide using ProLong Diamond antifade reagent with DAPI (Thermo Fisher Scientific), and immunofluorescence images were captured using a confocal laser scanning microscope LSM 710 (Carl Zeiss).Binarization of the obtained images and subsequent quantitative analysis of the signal regions were performed using ImageJ software.

2. 7 |
Quantitative real-time RT-PCR Total RNA was extracted from NHDFs, Detroit 551 cells and HSEs using RNeasy Mini Kit (Qiagen) and reverse-transcribed to generate singlestranded cDNA using a High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific).Quantitative real-time RT-PCR (qRT-PCR) for mRNA expression was performed using the StepOnePlus Real-Time PCR System (Thermo Fisher Scientific) and TaqMan Gene Expression Assays (Thermo Fisher Scientific) with specific primers for each target gene (Table

3 . 6 |
Low extracellular pH enhances CREB phosphorylation via the activation of cAMP/PKA and phospholipase C (PLC)/PKC signalling in NHDFsThe effects of cAMP/PKA or PLC/PKC signalling inhibition, which are known as downstream signals of GPCRs,35 on CREB activation were examined.Treatment with 30 μM of the PKA inhibitor H-89 or with 30 μM of the PKC inhibitor Go 6983 significantly reduced cell viability (Figure4D,E).Therefore, we evaluated the effect of 10 μM of H-89 and 3 or 10 μM of Go 6983 on the low pH-induced increase in CREB phosphorylation in NHDFs.Treatment with each compound (H-89 and Go 6983) suppressed the low pH-induced increase in CREB phosphorylation, and the effect was further increased by combined treatment with both compounds (Figure4F).Overall, these results indicated that CREB phosphorylation was mediated by the kinase activities of PKA and PKC (Figure4F).Additionally, these results emphasized the role of pH-GPCRs in sensing changes in extracellular pH and indicated that cAMP/PKA and PLC/PKC signals are both activated downstream of the GPCRs.

F I G U R E 2
Promotion of elastic fibre and collagen fibre formation and hyaluronan synthesis in normal human dermal fibroblasts via a low pH-induced increase in TGF-β1 production.(A) Elastic fibre formation in normal human dermal fibroblasts (NHDFs) cultured in a normal pH (7.6) or low pH (6.5) medium for 15 days.Red: Elastin, Blue: DAPI.Bars = 100 μm.(B) Photographs showing only binarized elastin signal.Each binarization area ratio (%) showing elastin signal was calculated.Values are expressed as means ± standard deviation (SD; n = 3).Student's t-test was used for statistical analysis.***p < 0.001.(C) Collagen fibre formation in NHDFs cultured in a normal pH (7.6) or low pH (6.5) medium for 72 h.Green: Collagen-1, Blue: DAPI.Bars = 100 μm.(D) Photographs showing binarized collagen-1 signal.Each binarization area ratio (%) showing collagen-1 signal was calculated, and values were expressed as means ± SD (n = 3).Student's t-test was used for statistical analysis.*p < 0.05.(E) Hyaluronan (HA) content in NHDFs cultured in normal pH (7.8) or lower pH (7.0 or 6.5) medium for 48 h.The HA concentrations of the culture media were quantified by ELISA, and values were expressed as means ± SD (n = 3).Dunnett's test was used for statistical analysis.**p < 0.01.(F) TGF-β1 content in NHDFs cultured in a normal pH (7.7) or low pH (6.5) medium for 24 or 48 h.The TGF-β1 concentrations of the culture media were quantified by ELISA.Values were expressed as means ± SD (n = 3).Student's t-test was used for statistical analysis.***p < 0.001; *p < 0.05.(G) TGF-β1 mRNA expression in NHDFs cultured in a normal pH (8.0) or low pH (6.5) medium for 8 h was determined using qRT-PCR.Values were expressed as means ± SD (n = 3) and represented a fold increase in mRNA expression relative to that of NHDFs cultured in normal pH (8.0) medium.Student's t-test was used for statistical analysis.***p < 0.001.

F I G U R E 3
or ex vivo human skin tissues.Nevertheless, although most of the analyses were conducted using NHDFs in this study, the Detroit 551 cell line was also used to evaluate the activation of CREB and the increase in TGF-β1 expression upon pH decrease, which are key points in a series of intracellular responses.The results showed that Detroit 551 cells behaved similarly to NHDFs (FiguresS6 and S7), suggesting that the mechanism revealed in the present study is widely available in dermal fibroblasts.In summary, the results of the present study showed that the CO 2 -induced decrease in intradermal pH may stimulate the production of ECM, such as elastic fibres, collagen fibres and hyaluronan, which are necessary for maintaining dermal elasticity.Additionally, the positive effects of low extracellular pH were mediated by the upregulation of TGF-β1 expression via CREB activation in NHDFs.A schematic diagram representing the intracellular response to low pH in the extracellular environment is illustrated in Figure S8.Overall, the findings of the present study suggest that the transcutaneous administration of CO 2 might be useful for anti-ageing.Role of CREB in low pH-induced increase in TGF-β1 expression in normal human dermal fibroblasts.(A) Low environmental pH induced an increase in CREB phosphorylation in normal human dermal fibroblasts (NHDFs).NHDFs were cultured in normal pH (7.6) medium (Time 0) or low pH (6.5) medium for 10 min, 30 min, 90 min, 6 or 24 h.After incubation, cells were lysed and subjected to SDS-PAGE.p-CREB and CREB expression levels were detected using western blot analysis, with β-Actin as an internal control.Band intensities were quantified (right graph).(B) CREB1 mRNA expression level in NHDFs treated with negative control siRNA or two types of CREB1 siRNA (25 nM) was detected at 24 and 48 h post-transfection using qRT-PCR.(C) CREB protein expression in NHDFs treated with negative control siRNA or two types of CREB1 siRNA (25 nM) was detected 72 h post-transfection using western blot analysis, with β-Actin as an internal control.Band intensities were quantified.(D) NHDFs were treated with negative control siRNA or two types of CREB1 siRNA (25 nM) and incubated for 48 h after transfection.TGF-β1 mRNA expression in NHDFs cultured in normal pH (7.7) or low pH (6.5) medium for 24 h was detected using qRT-PCR.(E) NHDFs were treated with negative control siRNA or two types of CREB1 siRNA (25 nM) twice (1 and 4 days after cell seeding) and incubated for 3 days after second transfection.TGF-β1 concentrations in NHDFs cultured in normal pH (8.0) or low pH (6.5) medium for 24 h were quantified by ELISA.(B, C) Values are expressed as means ± standard deviation (SD; n = 3) and represented as fold increase relative to that of NHDFs treated with control siRNA.Dunnett's test was used for statistical analysis.***p < 0.001; *p < 0.05.(D, E) Values are expressed as means ± SD (n = 3).Tukey's test was used for statistical analysis.***p < 0.001; **p < 0.01; *p < 0.05.