The local Ethics Committee and the Institutional Review Board (IRB) at Chungnam University Hospital (IRB no. 2012-08-019) approved this study, and informed consent was obtained from all participants regarding the study's scientific aims as well as the use of saliva, DNA and clinical records for all stages. A total of 100 healthy volunteers were enrolled for the study (Korean females, aged 30~50 years) living in the same city. Skin capacitance, a widely accepted parameter of stratum corneum hydration, was measured using a Corneometer® CM 825 (Courage & Khazaka Electronic, Cologne, Germany) on the anterior aspect of left cheek. All measurement were done after sit and rest for 15 min in constant temperature and humidity conditions (22℃±2℃, 50%±5%). The tested volunteers were divided into groups according to the corneometer (CM) value: i) CM value ≤40 (dry skin), ii) 40< CM value <60, and iii) CM value ≥60 (hydrated skin) (Table 1).

The Oragene (DNA Genotek, Kanata, ON, Canada) saliva kit was used to collect highly purified DNA from the saliva of volunteers. In the GWAS, 100 subjects were genotyped by using the Axio™ Genome-Wide ASI 1 Array Plate (Affymetrix, Santa Clara, CA, USA) comprised of 598,100 SNPs. The PLINK (ver. 1.06) and R statistical programs (http://www.r-project.org/) (ver. 2.10.1) were used for quality control procedures. SNPs were filtered if the following criteria were satisfied: i) a call rate of <95% (n=15,314), ii) a minor allele frequency of <1% (n=76,211), and iii) a significant deviation from Hardy-Weinberg equilibrium of ≤1×10⁻³ (n=489). After quality control, 506,945 SNPs (84.7% of 598,375) were remaining for association analysis. Association analysis was performed using 506,945 common SNPs that satisfied the quality control criteria using quantitative trait analysis (Wald test).

Skin samples were obtained from 5 patients from skin lesions in trunk of patients with psoriasis, atopic dermatitis or ichthyosis. The tissues were fixed in 10% formalin for 24 h and embedded in paraffin. Sections of skin specimens were dewaxed, rehydrated, and then washed three times with phosphate-buffered saline. After treatment with proteinase K (1 mg/ml) for 5 min at 37℃, sections were treated with H₂O₂ for 10 min at room temperature, placed in a blocking-solution (Dako, Carpinteria, CA, USA) for 20 min, followed by reaction with the appropriate primary antibodies. Sections were incubated sequentially with peroxidase-conjugated secondary antibodies (Upstate, Lake-Placid, NY, USA) and visualized using a Chemmate Envision Detection Kit (Dako). Following antibodies were used in this study: single stranded DNA binding protein 3 (SSBP3) (Abcam, Cambridge, MA, USA). The intensity of SSBP3 in epidermis was analyzed using ImageJ analysis program (http://imagej.nih.gov/ij/docs/index.html).

Human skin tissues were obtained under the written informed consent of donors, in accordance with the ethical committee approval process of the Institutional Review Board of Chungnam National University Hospital (IRB no. 2016-07-009). Human skin tissues were obtained from trunk skin that was the remnant of benign skin tumor surgery. Primary keratinocytes were cultured according to the method previously described7. Human epidermal keratinocytes were cultured in keratinocyte-serum free medium supplemented with bovine pituitary extract and recombinant human epidermal growth factor (Life Technologies Corporation, Grand Island, NY, USA).

Total RNA was isolated from cultured skin epithelial cells using Easy-Blue RNA extraction kit (Intron, Daejeon, Korea). Two µg of total RNA was reverse transcribed with moloney-murine leukaemia virus (M-MLV) reverse transcriptase (RTase) (ELPIS Biotech, Daejeon, Korea). Aliquot of RT mixture was subjected to polymerase chain reaction (PCR) cycles with primer set for SSBP3 5′-GGCAGATCTA TGTTTGCCAAAGGCAAAGGC-3′ and 5′-TTTGTCGACT CACACACTCATCGTCATGC-3′. SSBP3 cDNA was subcloned into pENTR/CMV-Flag vector that has attL sites for site-specific recombination with a Gateway destination vector (Invitrogen, Carlsbad, CA, USA). The replication-incompetent adenoviruses were created using Virapower adenovirus expression system (Invitrogen) according to the method previously described8. Briefly, site-specific recombination between entry vector and adenoviral destination vector was achieved by LR clonase (Invitrogen). The resulting adenoviral expression vector was then transfected into 293A cells using Lipofectamine 2000 (Invitrogen). Cells were grown until 80% cytopathic effect was seen, then harvested for preparation of recombinant adenovirus. For microRNA (miR) specific for SSBP3, target sequences were designed using Invitrogen's RNAi Designer (http://rnaidesigner.lifetechnologies.com/rnaiexpress). The miR sequences were as follows: top strand 5′-TGCTGTTCAGAACCAGACTACAAGGTGTTTTGGCCA CTGACTGACACCTTGTACTGGTTCTGAA-3′ and bottom strand 5′-CCTGTTCAGAACCAGTACAAGGTGTCAGTCA GTGGCCAAAACACCTTGTAGTCTGGTTCTGAAC-3′. The double-stranded DNA oligonucleotides were synthesized and cloned into the parental vector pcDNA6.2-GW/EmGFP-miR (Invitrogen). The expression cassette for miR was moved into the pENTR/CMV vector and adenovirus was made as previously reported8.

To induce keratinocyte differentiation in vitro, we adopted a well-established calcium-induced differentiation model. After adenoviral transduction, cells were replenished with keratinocyte-serum free medium and then cultured for a further 3 days with or without calcium (1.8 mM).

Cells were lysed in Proprep solution (Intron). Total protein was measured using a BCA protein assay kit (Pierce Biotechnology, Rockford, IL, USA). Samples were run on SDS-polyacrylamide gels, transferred to nitrocellulose membranes and incubated with appropriate antibodies. Blots were then incubated with peroxidase-conjugated secondary antibodies, visualized by enhanced chemiluminescence (Intron). For determination of secreted proteins, cell culture medium was concentrated using a Protein concentration kit (Elpis Biotech). The following primary antibodies were used in this study: SSBP3 (Abcam); actin (Sigma-Aldrich, St. Louis, MO, USA).

Total RNAs were isolated using Easy-Blue RNA extraction kit. Two µg of total RNAs were reverse transcribed with M-MLV RTase (Elpis Biotech). Aliquots of RT mixture were amplified using SYBR Green real-time PCR master mix (Elpis Biotech). The following primers sequences were used: SSBP3, 5′-ATGCTTGGGACTGGATGGAA and 5′-C AAAGACCAGCTACAGGGGA; Involucrin, CCATCAGGA GCAAATGAAACAG and GCTCGACAGGCACCTTCTG; Loricrin, AATAGATCCCCCAGGGTACCA and CGGTGCC CCTGGAAAAC; K1, GGCTATGAC CCTGCTTTGTTCT and TCATGTGGGTGGTGGTCACT; GAPDH, 5′-TGCAC CACCAACTGCTTAGC and 5′-GGCATGGACTGTGGTCA TGAG.

Cells were grown at 70% confluency in a 24 well culture plate and co-transduced with reporter adenovirus and SSBP3 expressing adenovirus. After adenoviral transduction, cells were replenished with fresh medium. Cells were further incubated for 48 h, and then cellular extracts were prepared using cell lysis buffer. Luciferase activities were determined using Luciferase assay system (Promega, Madison, WI, USA), according to the recommended protocol.

RT-PCR and luciferase assay were evaluated statistically using one-way ANOVA using IBM SPSS Statistics software (ver. 22.0; IBM Co., Armonk, NY, USA). Statistical significance was set at p<0.01.

To identify significant loci associated with skin hydration, we genotyped 598,100 SNPs in 100 individuals using Axiom™ Genome-Wide ASI 1 Array Plate. The samples consisted of individuals with CM values. After quality control, we analyzed the association of 501,262 common markers using quantitative trait analysis (Wald test). Among several SNPs identified in GWAS, we focused on SSBP3 which is known to be associated with DNA replication, DNA damage repair and cell differentiation (Table 2).

First, we evaluated the endogenous expression of SSBP3 in epidermal cells, including normal human epidermal keratinocytes (NHEKs) and HaCaT cells, and found that SSBP3 was expressed in both skin cell lines (Fig. 1A). Next, we confirmed the expression of SSBP3 in normal human skin using immunohistochemical analysis. We noted that SSBP3 is expressed in all layers of epidermis. However, the intensity of SSBP3 was low in proliferative cells in the basal cell layer of epidermis but strongest in the spinous layers which begin to express proteins required for cornification (Fig. 1B).

Since the expression of SSBP3 was increased during the keratinocyte differentiation process, we investigated the putative role of SSBP3 in epidermal differentiation. To examine SSBP3 during keratinocyte differentiation, we adopted a well-established calcium-induced differentiation model. As expected, SSBP3 was upregulated in a time-dependent manner from day 3 after calcium treatment and decreased at day 7, which was consistent with the result of immunohistochemical analysis showing that SSBP3 was most prominent in the spinous layer of the epidermis (Fig. 2A). We then constructed a recombinant adenovirus expressing SSBP3 and transduced cultured human epidermal keratinocytes (Fig. 2B). Overexpression of SSBP3 led to a significant increase in mRNA levels of epidermal differentiation related markers such as K1, involucrin, and loricrin (Fig. 2C). Next, we transduced keratinocytes with each involucrin-luc and loricrin-luc reporter adenoviruses, in which about 3.7 kb of the involucrin promoter fragment and 2.0 kb of the loricrin promoter fragment were fused to the luciferase gene. Overexpression of SSBP3 significantly increased the involucrin and loricrin luciferase activities compared to the Ad/LacZ control (Fig. 2D). And we also confirmed that overexpression of SSBP3 increased the expression of involucrin and loricrin in protein levels (Fig. 2E).

We further confirmed the role of SSBP3 on keratinocyte differentiation by knockdown experiment. To this end, we transduced keratinocytes with recombinant adenovirus expressing miR specific for SSBP3 (Fig. 3A). As shown in Fig. 3B, mRNA levels of K1, involucrin, and loricrin were significantly decreased by SSBP3 knockdown even with the calcium treatment. Consistent with the quantitative PCR (qPCR) analysis results, SSBP3 downregulation led to decreased luciferase activities of loricrin and involucrin compared to the control groups (Fig. 3C). The results suggest that knockdown of SSBP3 decreased differentiation of keratinocytes through decreased transcriptional activity via specific DNA binding to the promoter regions of epidermal differentiation-related genes. Knockdown of SSBP3 also inhibited the expression of involucrin and loricrin in protein levels (Fig. 3D).

Since SSBP3 can regulate keratinocyte differentiation in vitro, we hypothesized that SSBP3 expression may have a correlation to skin disease relating to keratinocyte differentiation. We examined SSBP3 expression in ichthyosis, atopic dermatitis and psoriasis, which are skin disorders showing abnormal keratinocyte differentiation and decreased moisturization due to a defect in the skin barrier. As expected, decreased expression of SSBP3 was found in psoriasis, a multisystemic disease characterized by hyperproliferation and altered differentiation of the epidermis (Fig. 4A). SSBP3 was also decreased in the spinous layer in ichthyosis, a disease caused by a loss-of-function mutation that results in keratinocyte disruption (Fig. 4B). Finally, diffuse nuclear staining of SSBP3 was observed in the basal and lower spinous layers of epidermis in atopic dermatitis, an inflammatory skin disease showing defects in keratinocytes differentiation and the skin barrier; however, the intensity of staining was decreased in the upper spinous and granular layers (Fig. 4C).