HaCaT (human keratinocyte cell line) was kindly provided by Dr. Tae-Yoon Kim (College of Medicine, The Catholic University of Korea) and was cultured in Dulbeco's Modified Eagle Medium (DMEM, Gibco-BRL, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (FBS, Gibco- BRL) and 100 U/ml penicillin/streptomycin (Gibco- BRL), at 37℃ in an incubator containing 5% CO₂.

Water was collected from the Yong-gung oncheon (Incheon-si, Gangwha-gun, Korea) spring and filtered through 0.44 um filters. Waters were stored at 4℃, until used for the experiments. For the measurement of water osmolarity, we used a Micro-Osmometer 210 (FISKE Associate, Norwood, MA, USA).

For the 3-(4,5-dimethylthizol-2-yl) 2,5-diphenyl tetrazolium bromide (MTT, Sigma-Aldrich, St. Louis, MO, USA) assay, cells were plated onto 96-well microtiter plates at a density of 3×10⁴/200 µl in fresh medium, and then treated with hot spring water at a serially diluted concentration. Cells were cultured for 1, 4, 10, and 24 hours, respectively, to observe a time dependent effect. After the indicated time, 20 µl of MTT (5 mg/ml in phosphate-buffered saline) was added to each well, and the plates were returned to the incubator for an additional 4 hours. At the end of the incubation period, the supernatants were discarded by a suction and 200 µl dimethyl sulfoxide was added to all the wells, in order to dissolve the dark blue formazan crystals. The plates were subsequently covered with aluminum foil, gently shaken for 15 minutes, and read at a wavelength of 570 nm.

TLR agonists were treated with the following final concentrations for 24 hours. Tripamitoyl-S-glyceryl-cysteine (Pam3Cys, 1 µl/ml), heat-killed Listeria monocytogenes (HKLM, 10⁶ cells/ml), polyriboinosinic polyribocytidylic acid (poly (I : C), 10 µl/ml), lipopolysaccharide (LPS, 10 µl/ml), flagellin (10 µl/ml), and Pam2CGDPKHPKSF (FSL-1, 1 µl/m) were from InvivoGen (San Diego, CA, USA). Spa spring waters were added simultaneously or pre-treated 2 hours before the TLR agonist treatment. Cells were cultured for 1, 4, 10, or 24 hours in each treatment group.

Levels of IL-6, IL-8, granulocyte-macrophage colony-stimulating factor (GM-CSF), and TNF-α (BD OptEIA™, BD Biosciences Pharmingen, San Diego, CA, USA), IL-1α (Biolegend, San Diego, CA, USA) in HaCaT supernatant treated with TLR agonist in the presence or absence of hot spring water were quantified according to the manufacturer's protocol. IL-6 and TNF-α in mouse antigen presenting cells (APC) supernatant were also measured. In brief, the wells were coated with 100 µl of capture antibody in coating buffer (0.1 M sodium carbonate, pH 9.5) and the plates were incubated overnight at 4℃. After washing the wells with washing buffer (phosphate buffered saline [PBS] with 0.05% Tween-20), the wells were blocked with assay diluents (10% FBS in PBS) for 1 hour at room temperature (RT), followed by the addition of 100 µl/well of cell supernatant and cytokine standard solutions for 2 hours at RT. After washing, 100 µl of detection antibody and streptavidin-conjugated horseradish peroxidase reagent were added to the wells, which was then incubated for 1 hour at RT. After extensive washing, 100 µl of substrate solution (tetramethylbenzidine and hydrogen peroxide, BD Biosciences Pharmingen) was added to each well and the plates were incubated for 30 minutes at RT, in darkness. Stop solutions (2 N H₂SO₄) were added, and the absorbance was read at 450 nm within 30 minutes.

Naïve CD4⁺ T cells were purified from the mouse spleens, via magnetic isolation (Miltenyi Biotec GmbH, Gladbach, Germany). For the preparation of the spleen cell suspensions, spleens from 8 weeks-old female Balb/c mice were removed and minced with a Nylon mesh (70 µm pore). After the cells were pelleted, erythrocytes were lysed using hypotonic buffer (0.15 M NH₄Cl, 10 mM KHCO₃, 0.1 mM Na₂EDTA). The cells were washed in PBS and incubated with anti-CD4 antibody for 15 minutes, at 4℃. Directly following, the cells were conducted onto a magnetic separator to isolate the CD4⁺ cells and collected with positive selection. CD4^- cells were seeded onto a culture dish of 100 mm in diameter and incubated for 4 hours for the collection of the adherent cells (used as APC in these experiments) by gentle pipetting.

CD4⁺ naïve cells were seeded at a density of 2×10⁵ per well in 96-well plates, and were cultured in DMEM, containing 10% FBS. For each of the helper T cell differentiation, the skewing conditions are as followed; IL-12 (25 ng/ml, Biolegend) and anti-CD28 (1 µg/ml, Biolegend) for Th1, IL-4 (25 ng/ml, Biolegend) and anti-CD28 (1 µg/ml, Biolegend) for Th2, IL-6 (25 ng/ml, Biolegend), transforming growth factor (TGF)-β (2 ng/ml, R&D system) and anti-CD28 (1 µg/ml, Biolegend) for Th17, IL-2 (25 ng/ml, Biolegend) and anti-CD28 (1 µg/ml, Biolegend) for T reg. All cells stimulated with anti-CD3 antibodies with serial dilutions, in which the initial concentration was 3 µg/ml. Cells were incubated for 3 days in the presence or absence of spa spring water.

Before seeding for the differentiation of each of the helper T cell differentiation, the CD4⁺ cells were labeled with carboxyfluorescein diacetate, succinimidyl ester (CFSE, CellTrace™ CFSE Cell Prolifetration kit, Invitrogen, Paisley, UK). First, the cells were resuspended in prewarmed PBS/0.1% bovine serum albumin, at a concentration of 10⁶ cells/ml, and 10 µM of CFSE were added. After incubation at 37℃ for 10 minutes, 5 volumes of ice-cold culture media was added to the cells to quench the staining, incubated for 5 minutes on ice, and pelleted by centrifugation. The cells were resuspended in fresh media for a total of three washes, and seeded under each condition for the differentiation of Th1, Th2, Th17, and T_reg cells. After 3 days, CD4⁺ and CFSE⁺ cells were measured for the degree of proliferation, via a flow cytometry, and were analyzed using ModFit LT software (Verity Software House, Topsham, ME, USA), based on the reduction of CFSE positive cells.

APCs were stimulated with TLR3 agonist, poly (I : C) (10 µg/ml) for 24 hours with or without spa spring waters. APCs stimulated with poly (I : C) with or without hot spring water were collected, washed and stained with anti-mouse I-A/I-E (Biolegend, clone M5/114.15.2, Rat IgG2b, κ) to analyze the expression of MHC II on the surface of APCs for 20 minutes, at RT. Samples were acquired on FACSCalibur system (BD Bioscience, San Jose, CA, USA) and were analyzed using CellQuest software (BD Bioscience).

Data are expressed as the means±standard error of the mean. Non-parametric Mann-Whitney tests were applied to analyze the results for significant differences (at p<0.05), using GraphPad Prism software (GraphPad Software Inc., San Diego, CA, USA).

Because the osmolarity of spa spring water from Yong-gung oncheon was 700 mOsm/kg, the HaCaT cells were not able to maintain the growth in undiluted culture medium (Fig. 1). Therefore, we first determined to set an optimal dilution concentration for HaCaT culture conditions by the MTT assay. For this, HaCaT cells were cultured with spa spring water at a concentration of serially 2-fold dilution for 1, 4, 10, and 24 hours and the minimum concentration of spa spring water was determined. We found that each group of cells was up to 100% of viability under a 1 : 32 fold dilution (Fig. 2).

TLRs act as primary sensors that detect a wide variety of microbial components and elicit innate immune response, which guides the acquired immunity. Therefore, keratinocyte constitutively expressing TLRs may affect skin immune response, by regulating the inflammatory responses, via TLR signaling. We attempted to determine whether the spa spring water affected the TLR stimulated production of pro-inflammatory cytokines, including IL-1α, IL-6, IL-8, TNF-α, and GM-CSF, on the HaCaT cells. TLR1 agonist, through TLR6 agonist treatment, induced the attenuation of cytokine production in the exposure to spa spring in advance or simultaneously with TLR agonist treatment (Fig. 3, 4).

We subsequently attempted to ascertain the potential involvement of spa spring water in the differentiation of the helper T cells. Because the effector T cells, such as Th1, Th2, and Th17 cells act on various skin resident cells and potentiate the immune reaction, regulating T cell differentiation may modulate pro-inflammatory responses. To evaluate the in vitro suppressive capacity of spa spring water toward the helper T cell differentiation, magnetic sorted CD4⁺ T cells, from the spleen, were prepared from Balb/c mice. CFSE-labeled CD4⁺ T cells were cultured under various conditions, which polarized Th1, Th2, Th17, and T_reg with or without spa spring water. After 3 days, the cells were harvested and stained with CD4 and interferon (IFN)-γ for Th1, IL-4 for Th2, IL-17 for Th17 and Foxp3 for T_reg cells, respectively. Then, the flow cytometry was employed to evaluate the precursor frequency in terms of the T cell differentiation. As shown in Fig. 5, CD4⁺ naïve T cells were profoundly proliferated and differentiated under each skewing condition, along with anti-CD3 stimulation. Spa spring water suppressed the proliferation of Th1, Th2 and Th17 cells. In contrast to the suppressive effect on Th1, Th2 and Th17 cells, proliferation and differentiation to T_reg cells were promoted under spa spring water treatment. These results indicate that spa spring water may affect the distribution of the helper T cells in the immune response, by suppressing the polarization of the Th1, Th2 or Th17 cells.

Skin immunity is achieved by an interaction of different kinds of immune cells. In particular, APCs, such as Langerhans cells found in the epidermis, are the best-characterized dendritic cell population. They have the ability to process antigens in the periphery, transport it to the draining lymph nodes, where they are able to cluster with, and activate the antigen-specific naive T cells. APCs in the dermis may also provide alternative routes of antigen presentation, which can be important in the regulation of skin immune responses. Therefore, APCs are vital for the induction of immune responses to antigens encountered via the skin. To determine whether the expression of MHC II on the surface of APC is modulated by spa spring water, we isolated APC from BALB/C mouse spleen, and stimulated it with 20 µg/ml of poly (I : C) for 24 hours. TLR3 stimulation via poly (I : C) strongly induced MHC II expression on the APCs. By contrasts, APC treated spa spring water down-regulated the surface level of class II MHC expression, under TLR 3 stimulation (Fig. 6A). In addition, we observed that TLR3 agonist, poly (I : C) is a very potent inducer of inflammatory responses in the HaCaT cells9. Therefore, we measured and compared poly (I : C) mediated production of inflammatory cytokines from the APCs with or without spa spring water (Fig. 6B, C). APC induced TNF-α and IL-6 under poly (I : C) stimulation and cytokine production was reduced in the presence of spa spring water. These results showed that TLR-triggered inflammatory responses in APCs might also be modulated under spa spring water treatment.