Hv channels (voltage-gated proton channels) are expressed in blood cells, microglia and some types of epithelial cells. In neutrophils Hv channels regulate the production of reactive oxygen species through regulation of membrane potential and intracellular pH. Hv channels have also been suggested to play a role in sperm physiology in the human. However, the functions of the Hv channel at the whole-body level are not fully understood. In the present paper we show that Hvcn1 (voltage-gated hydrogen channel 1)-knockout mice show splenomegaly, autoantibodies and nephritis, that are reminiscent of human autoimmune diseases phenotypes. The number of activated T-cells was larger in Hvcn1-deficient mice than in the wild-type mice. Upon viral infection this was remarkably enhanced in Hvcn1-deficient mice. The production of superoxide anion in T-cells upon stimulation with PMA was significantly attenuated in the Hvcn1-deficient mice. These results suggest that Hv channels regulate T-cell homoeostasis in vivo.
- anti-DNA antibody
- pH regulation
- reactive oxygen species
- voltage-gated proton channel (Hv channel)
Hv channel (voltage-gated proton channel) currents were first described in snail neurons , then in mammalian immune cells, such as granulocytes and lymphocytes , and then in the lung epithelium . The molecular basis for Hv has been elusive for over 20 years. VSOP (voltage sensor domain-only protein) or Hv1 [4,5] (designated as VSOP/Hv1) was identified as the molecular correlate for the Hv channel. VSOP/Hv1 protein has been reported to be expressed in human sperm , immunocytes such as granulocytes, macrophage and B-cells , and microglia . HVCN1 (voltage-gated hydrogen channel 1), encoding the VSOP/Hv1 protein, is widely conserved among species from dinoflagellates  to humans.
In neutrophils and eosinophils, the Hv channel has been suggested to help the activities of the NADPH oxidase which generates superoxide anions against pathogens [10–12]. As a consequence of this production of superoxide anions, the NADPH oxidase complex containing NOX2 [NADPH oxidase 2; also known as CYBB (cytochrome b-245 β polypeptide)] transfers electrons and leaves protons in the cytoplasm, leading to excessive depolarization of the membrane if the mechanism of charge compensation is absent. The primary role of the Hv channel was thought to be compensating for this membrane charge imbalance during the activation of NADPH oxidase, thereby preventing cytoplasmic acidic pH and excessive depolarization of membrane potential [2,10]. In neutrophils it is known that the robust production of superoxide anion is induced by PMA stimulation . Previous reports showed that superoxide anion production upon PMA stimulation was less robust in neutrophils isolated from Hvcn1-deficient mice than those from the WT (wild-type) mice [8,13,14]. VSOP/Hv1 is also expressed in the microglia of the brain  and Hvcn1-deficient mice exhibit milder neuronal damage during brain ischaemia .
It has been shown that ROS (reactive oxygen species) production was reduced in Hvcn1-deficient mice, leading to attenuated BCR (B-cell receptor) signalling owing to insufficient oxidation of the protein tyrosine phosphatase SHP-1 (protein tyrosine phosphatase, non-receptor type 6) . However, the detailed mechanisms by which Hv channels regulate ROS production in B-cells remain elusive. T-cells are another immune cell type which express the Hv channel as shown by electrophysiological data from Jurkat and primary human T-cells . In T-cells ROS production is initiated upon TCR (T-cell receptor) stimulation  and plays a role in T-cell reactivity and disease susceptibility . Other studies have shown that the Hv channel is also expressed in eosinophils , basophils  and dendritic cells . Nonetheless, the macroscopic phenotypes of Hvcn1-knockout mice have not been fully described.
In the present study, we show that Hvcn1-deficient mice exhibit autoimmune disorder phenotypes and an increased number of activated T-cells in the resting state as well as in response to viral infection. We further provide evidence that VSOP/Hv1 protein contributes to ROS production in T-cells.
The following chemicals were used: Diogenes (National Diagnostics), which is the chemiluminescent reagent; SOD (superoxide dismutase; Wako); PMA (Sigma); DPI (diphenyleneiodonium; Sigma); and anti-CD4 antibody (BioLegend). The anti-VSOP/Hv1 antibody used was the polyclonal antibody against the C-terminal cytoplasmic region of mouse VSOP/Hv1 that was described previously . Alexa Fluor® 488-conjugated anti-(mouse IgG) (Invitrogen) was used to detect IgG in the mouse kidney. For differentiation into Th1 cells, IL (interleukin)-12 (PeproTech) was used. For differentiation into Th17 cells, IL-6 (Toray) and TGFβ (transforming growth factor β; PeproTech) were used. Antibodies against IFNγ (interferon γ) and IL-17 (eBioscience) were used for confirming cell differentiation.
The animal experiments were approved by animal research committees of Graduate School of Medicine, Osaka University, Osaka, Japan. HVCN1-GT mice were the same as reported previously . Mice were back-crossed against C57BL/6J mice for ten generations.
Preparation of cells
Fresh spleen, lymph nodes and thymus were dissected on a strainer and then dissociated to single cells in MACS buffer [PBS containing 5% FBS (fetal bovine serum) and 5 mM EDTA]. CD4-, CD8- and B220-positive cells were purified from spleens and lymph nodes by positive selection with antibody conjugated with magnetic beads (Miltenyi Biotech) with a purity of ~95%.
In vitro induction of non-polarized CD4+ T-cells, Th1 and Th17 cells were performed as described previously . In brief, lymph nodes and spleens from WT or HVCN1-GT mice between 8–9 weeks-of-age were harvested and CD25−CD44− naive CD4+ T-cells were sorted using a MoFlo cell sorter (Beckman Coulter). The cells were cocultured for 6 days with bone marrow-derived dendritic cells plus 1 μg/ml anti-CD3 antibodies (eBioscience) for activated non-polarized CD4+ T-cells, in the presence of 20 ng/ml rm (recombinant mouse) IL-12 (PeproTech) for Th1 cells and in the presence of 1 μg/ml of rh (recombinant human) IL-6 (Toray), 5 ng/ml rhTGFβ (PeproTech), 10 ng/ml rmIL-23 (R&D Systems) and 10 μg/ml anti-IFN-γ antibodies for Th17 cells.
Tissues were homogenized on a cell strainer with a 100-μm mesh to a single cell level then lysed for 20 min on ice in a solution of cell lysis buffer [20 mM Hepes, 150 mM NaCl, 1 mM EDTA and 1% Triton X-100 (pH 7.4)] including protease inhibitor cocktail (Sigma). After centrifugation for 15 min at 4°C (12000 g) the supernatant was used as a sample. Each sample contained 106 cells.
Measurement of superoxide anion
Superoxide anion production was measured by the chemiluminescent reagent Diogenes. Cells (approximately 1×106 cells/well for T-cells and 1×105 cells/well for B-cells) were suspended in HBSS (Hanks balanced salt solution; pH 7.4). The cell suspension was pre-incubated for 10 min at 37°C. After the addition of 1 μM PMA the change in luminescence was measured using a SH-9000Lab micro plate reader (Corona Electric). As a negative control NADPH oxidase activity was inhibited by DPI (10 μM) or SOD (50 units/ml).
For cell-surface labelling, approximately 106 cells were incubated with fluorescence-conjugated antibodies for 30 min on ice. FITC-conjugated anti-CD44, PE/Cy7 (phycoerythrin/indotricarbocyanine)-conjugated anti-CD4 and eFluor® 450-conjugated anti-CD8 (eBioscience) antibodies were used. The cells were then analysed with CyAn™ flow cytometer (Beckman Coulter). The collected data were analysed using FlowJo software (Tree Star).
In vivo LCMV (lymphocytic choriomeningitis virus) infection
Mice (8-week-old; n=4–7) were infected intravenously with LCMV clone-13 at 2×106 pfu (plaque-forming units) on day zero. On day 8 the splenocytes were counted and the activated T-cell populations with a high expression of CD44 were measured by flow cytometry.
ELISA assay of anti-dsDNA (double-stranded DNA) antibody
Anti-dsDNA IgG antibodies in serum were measured by using the ELISA kit AKRDD-61 (Shibayagi).
Histology and electron microscopy observation of the kidney
Kidney slices were embedded in paraffin and 2-μm sections were cut for light microscopic examination. After being dewaxed, the sections were stained with PAS (periodate–Schiff). For immunofluorescence the sections were incubated with Alexa Fluor® 488-conjugated anti-(mouse IgG) at a dilution of 1:2000 for 1 h, and the Alexa Fluor® 488 signals were detected by confocal microscope (Leica).
For electron microscopy the kidney slices were fixed with 2.5% glutaraldehyde and 1% paraformaldehyde in 0.1 M PBS for 4 h, and then the kidney slices were cut into 1 mm3 sections and postfixed in 2% osmium tetroxide in 0.1 M sodium cacodylate buffer for 1 h at 4°C, dehydrated in graded ethanol and embedded in epoxy resin. Ultrathin sections were cut and examined by the transmission electron microscope (H-7000, Hitachi).
Hvcn1-deficient mice exhibit autoimmune disease phenotypes
In the course of the present study with Hvcn1-deficient mice (designated as HVCN1-GT [8,14]), we noticed that many of the animals older than 6 months showed splenomegaly, although this was less often observed as mice were back-crossed to the C57BL/6J line (if splenomegaly is defined as a spleen mass over 150 mg, about half of the mice showed splenomegaly at 6-months-old before back-crossing to the C57BL/6J line compared with 10% of mice with the same age after ten back-crosses). Splenomegaly in HVCN1-GT mice was less often observed at less than 6-months of age. In the present study all of the data were obtained from mice with ten back-crosses with the C57BL/6J line. The spleen mass from HVCN1-GT mice was significantly heavier than that of the WT mice at approximately 1 year-of-age (Figure 1a). Associated with splenomegaly, the serum level of the anti-dsDNA antibody was significantly higher in HVCN1-GT mice than that of the WT mice (Figure 1b). No significant histological change in the thymus was observed (Supplementary Figure S1 at http://www.biochemj.org/bj/450/bj4500295add.htm).
It is known that when present in increased levels anti-dsDNA antibodies form immune complexes which leads to nephritis with deposition in the glomeruli in kidney [23a]. In PAS staining mesangial cells were diffusely proliferated with wire loop lesions associated with immune complex deposition (Figure 2, green arrows). Focal lymphocyte infiltration was observed around the interlobular artery indicating inflammation of the small arteries (Figure 2, yellow arrows). Immunofluorescence analysis showed IgG staining in the glomerular immune complex in the mesangium and along the capillary walls forming wire loop lesions (Figure 2, white arrow). Electron microscopy demonstrated subendothelial and paramesangial electron-dense deposits (Figure 2, yellow asterisk). These glomerular lesions are similar to lupus nephritis. In contrast, these changes were not observed in the WT mice. Lupus erythematosus is known to have an increased prevalence in females [23b]. The HVCN1-GT mice also showed a mild bias in phenotype between the sexes: 71% (5/7) of the females and 0% (0/4) of the males showed obvious splenomegaly (as defined as a spleen mass greater than 150 mg) and 77% (7/9) females and 25% (3/12) males had anti-dsDNA antibodies levels in their serum no less than three times the average level of the WT mice.
Increase of activated T-cells in HVCN1-GT mice
Defects of macrophages, B-cells and T-cells are known to lead to the development of autoimmune disease [24,25]. VSOP/Hv1 is known to be expressed in macrophage and B-cells. Several reports have shown that a reduced ability of macrophages to eliminate apoptotic cells results in autoimmune disease, such as lupus . The macrophage's ability to engulf apoptotic cells was not significantly reduced in HVCN1-GT mice as compared with that of the WT cells (results not shown). Lymphocytes are also important cells that underlie the development of autoimmune disease. The ratio of the cell populations of B220-positive B cells and the CD4+ and CD8+ T-cell profile of the spleen and lymph nodes were not significantly different between WT and HVCN1-GT mice (results not shown). The proportion of CD4+ and CD8+ T-cells was also not altered in the thymus of HVCN1-GT mice (Supplementary Figure S2 at http://www.biochemj.org/bj/450/bj4500295add.htm). The ability of CD4+ cells to differentiate into Th1 and Th17 cells was not significantly different between the WT and HVCN1-GT cells (Figure 3), indicating that the lymphocyte differentiation was normal. However, the proportion of activated T-cells (CD4+CD44high and CD8+CD44high) was increased in non-stimulated 6-month-old mice (Figure 4a).
In order to understand the roles for Hv channels of T-cells in vivo, CD4+ and CD8+ T-cell responses were analysed in HVCN1-GT mice after LCMV infection. The total number of splenocytes (Figure 4b) as well as activated CD4+ and CD8+ T-cells increased in 8-week-old HVCN1-GT mice as compared with the WT (Figure 4c). These results indicate that the survival of activated CD4+ and CD8+ T-cells was augmented in HVCN1-GT mice as compared with the WT mice, consistent with the idea that increased activated T-cells might enhance the immune response even at the basal level, which in turn developed the auto-reactive immune responses.
Expression and VSOP/Hv1 protein in T-cells
The increased number of activated T-cells could be due to the intrinsic properties of T-cells of HVCN1-GT mice. So far there has been no direct evidence at the molecular level that mouse T-cells express the VSOP/Hv1 protein. In a Western blot analysis of T-cells using an VSOP/Hv1-specific antibody (Figure 5), a positive band was detected corresponding to the expected size of the VSOP/Hv1 protein. This is in contrast with previous studies that have reported no expression of VSOP/Hv1 protein in T-cells [7,8], probably owing to a low expression level. Indeed the expression level of the VSOP/Hv1 protein in T-cells was much lower than in B-cells, consistent with previous electrophysiological measurements . In addition, quantification of the mRNA level by quantitative PCR showed that the expression level in T-cells was about 10% of that in macrophages (Supplementary Figure S3 at http://www.biochemj.org/bj/450/bj4500295add.htm). The smaller mass protein band, which was the less prevalent form in B-cells, but prevalent form in T-cells (Figure 5), might represent a protein that is translated from the second initiation site downstream of the first ATG as previously suggested . We also verified the expression of the Hv channel in T-cells by electrophysiological measurements using whole-cell patch recording. T-cells prepared from the WT mouse showed Hv channel currents, whereas T-cells prepared from the HVCN1-GT mouse did not (Supplementary Figure S4 at http://www.biochemj.org/bj/450/bj4500295add.htm).
Roles for the Hv channel in the late phase of PMA-stimulated superoxide anion production in T-cells
The production of the superoxide anion and H2O2 is reduced in neutrophils [8,13,14] and B-cells  from HVCN1-GT mice. A previous report also suggested that T-cells express the phagocyte-type NADPH oxidase that generates superoxide anion . We explored whether the Hv channel plays a role in the production of superoxide anion in T-cells. The production of superoxide anion in T-cells was induced by PMA stimulation and detected by an enhancer-containing luminol-based detection system (see the Experimental section). CD4+ and CD8+ T-cells produced significant levels of superoxide anion after PMA stimulation (Figure 6a). This activity was completely inhibited by DPI or SOD, but was not affected by catalase, confirming that the ROS detected was not H2O2, but superoxide anion (Supplementary Figure S5 at http://www.biochemj.org/bj/450/bj4500295add.htm). We also measured superoxide anion production from B-cells for comparison with T-cells (Supplementary Figure S6 at http://www.biochemj.org/bj/450/bj4500295add.htm). Superoxide anion production in T-cells from the WT was at most 10% of the amplitude in B-cells (superoxide anion production in B-cells at the basal level was 15.6±4.5- (n=9) and 12.0±8.0- (n=9) fold higher than T-cells for the WT and HVCN1-GT mice respectively).
We noted that the kinetics of superoxide anion production showed two phases (Figure 6a). The first phase attained its peak about 5 min after PMA stimulation followed by the later phase which was sustained for approximately 30 min. When comparing the responses between the WT and HVCN1-GT mice we found that the first peak remained intact, but the second phase was markedly attenuated in T-cells from the HVCN1-GT mice (Figures 6b and 6c).
Although reduced ROS production by neutrophils and an increased susceptibility to pathogen infection in vitro have been reported in HVCN1-GT mice [8,14], no macroscopic phenotype of Hvcn1-deficient mice has been reported. In the present study we report phenotypes similar to autoimmune disease, such as splenomegaly, the presence of anti-dsDNA antibody in the serum and histological changes of nephritis with deposit of immunoglobulins in Hvcn1-deficient mice. It is known that abnormal T-cell function is a common feature of autoimmune diseases, including systemic lupus erythematosus, arthritis, and, sometimes, NADPH oxidase-deficient CGD (chronic granulomatous disease) patients [26,27]. In HVCN1-GT mice, the numbers of both activated CD44high CD4+ and CD44high CD8+ T-cells were increased after LCMV infection, indicating abnormal response to T-cell receptor stimulation. Consistent with this, we found the activated CD44high T-cells were increased in non-stimulated mice. It is known that CD44high T-cells are increased in autoimmune disease patients , although the precise mechanism is not known completely. Increased CD44high T-cells might enhance the immune response at the basal level, which, in turn, develops into an autoreactive immune response.
Consistent with a previous electrophysiological study of human T-cells , the results of the present study showed that the Hv channel is expressed in mouse T-cells, raising a possibility that the autoimmune disorder phenotypes observed in HVCN1-GT mice are caused by altered intrinsic functional properties of T-cells. The functional cross-linkage between VSOP/Hv1 and the phagocyte-type NADPH oxidase NOX2 has been reported in neutrophils and B-cells [7,8,13,14]. T-cells also express NOX2 . To assess the roles of the Hv channel in the NADPH oxidase activity in T-cells, the superoxide anion production in T-cells was compared between the WT and HVCN1-GT mice using an enhancer-containing luminal-based detection system (see the Experimental section), which has high sensitivity. Superoxide anion production upon PMA stimulation occurred in two phases. In a previous study where B-cells were analysed in HVCN1-GT mice , the same measurement was employed for up to 25 min after stimulation, but a later phase corresponding to 30 min after stimulation was not examined. PMA-induced superoxide anion production during the delayed phase was almost absent in T-cells from HVCN1-GT mice, indicating that the delayed phase depends on VSOP/Hv1 activity, whereas the first phase does not. Nocodazol, Brefeldin A and monensin, inhibitors of membrane transport, did not affect the delayed phase, suggesting that targeting the NADPH oxidase complex to the plasma membrane or recycling of the NADPH oxidase complex does not occur in the delayed phase (results not shown). Apocynin, which is known to block NADPH oxidase activity, inhibited both the first and the delayed phase (Supplementary Figure S7 at http://www.biochemj.org/bj/450/bj4500295add.htm). In neutrophils the Hv channel is known to positively regulate the level of NADPH oxidase activity through its prevention of cytoplasmic acidic pH and excessive membrane depolarization [2,13]. At present we do not know whether VSOP/Hv1 regulates the activity of NADPH oxidase through regulation of membrane potential or cytoplasmic pH. More rigorous studies are needed to clarify the mechanism by which VSOP/Hv1 contributes to ROS production in T-cells.
Development of autoimmune disease in HVCN1-GT mice may involve alterations of other cellular activities due to the lack of VSOP/Hv1. It is unlikely that loss of VSOP/Hv1 in B-cell function is the major cause of the autoimmune disease phenotype in HVCN1-GT mice. Capasso et al.  showed that the response of B-cells to the T-cell-dependent antigen NP-KLH (4-hydroxy-3-nitrophenylacetyl-keyhole limpet haemocyanin) is attenuated in HVCN1-GT mice, which is the opposite to what is expected from the phenotypes of HVCN1-GT mice in the present study. To confirm that a defect in VSOP/Hv1 function in T-cells is the major cause of autoimmune disorder-related phenotypes in HVCN1-GT mice, future studies with a cell-specific knockout of Hvcn1 will be necessary.
Mari Sasaki, Akihiro Tojo, Daisuke Kamimura and Nana Miyawaki conducted the experiments. Yoshifumi Okochi, Mari Sasaki and Nana Miyawaki raised the HVCN1-GT mice. Mari Sasaki and Daisuke Kamimura analysed data. Mari Sasaki, Akihito Yamaguchi, Akihiro Tojo, Masaaki Murakami and Yasushi Okamura planned the experiments and wrote the paper. All authors read and approved the paper.
This work was supported by the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) [grant numbers 21229003 and 10J04301 (to Y.Oka. and M.S.)], the Human Frontier Science Program [grant number RGP0039/2007-C (to Y.Oka.)] and the Japan Society for the Promotion of Science (JSPS) (Research Fellowship for Young Scientists and Restart Postdoctoral Fellowship to M.S.). This work was also supported, in part, by the Special Coordination Funds for Promoting Science and Technology from MEXT for the Osaka University Program for the Support of Networking among Present and Future Women Researchers.
We thank Dr Toshio Hirano, Dr Yukihisa Sawa, Dr Hideki Ogura, Dr Yasunobu Arima and Dr Yuko Okuyama for helpful advice and technical instruction. We also thank Dr Takeshi Imamura and Mr Tomohiro Saneto for supporting analysis and Dr Betty Tsao for helpful comments on the paper.
Abbreviations: DPI, diphenyleneiodonium; dsDNA, double-stranded DNA; Hv channel, voltage-gated proton channel; HVCN1, voltage-gated hydrogen channel 1; IFNγ, interferon γ; IL, interleukin; LCMV, lymphocytic choriomeningitis virus; NOX2, NADPH oxidase 2; PAS, periodate–Schiff; rh, recombinant human; rm, recombinant mouse; ROS, reactive oxygen species; SOD, superoxide dismutase; TGFβ, transforming growth factor β; VSOP, voltage sensor domain-only protein; WT, wild-type
- © The Authors Journal compilation © 2013 Biochemical Society