Background

We have reported that molecular hydrogen (H2) has potential as a novel antioxidant in preventive and therapeutic applications [1]. Furthermore, H2exhibits not only anti-oxidative stress effects [2, 3], but also has various anti-inflammatory [4, 5] and anti-allergic effects [6]. Since the publication of the first article on the biological contribution of H2in 2007, more than 80 articles involved in H2have been published to establish the apparent activity of H2from various medical aspects [7–9].

H2reacted with strong reactive oxygen/nitrogen species including hydroxyl radical and peroxinitrite (ONOO-) in cell-free reactions and protected cultured cells depending upon the decrease of hydroxyl radicals (•OH) [1]. Subsequent and recent experiments including ours indicated that a small amount of hydrogen is also effective against various stimuli [8, 9]. When model animals consumed H2by drinking water with dissolved H2, a small amount of H2was extensively effective [10–12]; however, it may be difficult to explain that direct reduction of •OH by a very small amount of H2reveals all the functions of H2, because the saturated level of H2is only 0.8 mM and the dwelling time of •OH is very short in the body [11, 13]. In fact, drinking 0.04 or 0.08 mM H2was shown to be effective [14, 15]. Although we have recently shown that H2can be accumulated with hepatic glycogen, it is unlikely that the amount of H2is sufficient to exhibit all of its functions [15].

Moreover, H2regulated various gene expressions; however, there is no evidence that H2directly reacts with factors involved in transcriptional regulation including FGF21 [15], inflammatory cytokines [11], HMGB1 [16], and HO-1 [17]. It remains unclear whether such regulations are the cause or consequence of the effects against oxidative stress. Moreover, the primary molecular target of H2remains unknown.

ONOO-is produced by the rapid reaction of nitric monoxide (NO•)with superoxide anion radicals (•O2-) [18, 19]. We have shown that H2reduces ONOO-as well as •OH [1]. Different from •OH, ONOO-has a longer lifespan and the potential to regulate gene expression through nitration of target proteins [20, 21]. Thus, we hypothesized that one of the H2functions is caused by reducing cellular ONOO-.

Here, to verify this hypothesis, we examined protective and regulatory effects of H2on NO•-derived oxidative stress to chondrocytes. We found that H2protected chondrocytes from oxidative stress, and alternated gene expressions, contrary to the manner of transcriptional regulation by ONOO-. This study implies that at least one of the H2functions is responsible for the reduction of ONOO-.

Methods

Cartilage slice culture

A fresh hindlimb of a slaughtered male seven-month-old pig was purchased from Tokyo Shibaura Organ Co., Ltd. (Minato-ku, Tokyo, Japan). There were no possible contaminant diseases. Cartilage from the healthy porcine hindlimb (metatarsophalangeal joint) was cut into pieces for culture (2 mm width × 7 mm length × full thickness) as described previously [22]. Male Sprague-Dawley rats of 10 weeks of age were purchased from Nippon SLC (Hamamatsu, Shizuoka, Japan). Cartilage from the meniscus of a rat was also sliced into pieces (full width × full length × 0.5 mm thickness) for culture. Since the meniscus structure is not uniform and the peripheral part contains fewer chondrocytes, we used slices prepared from the middle part of the meniscus.

The slices were randomly divided into two experimental groups and incubated at 37°C in Dulbecco's modified Eagle's medium (DMEM)/Ham F-12 mixed medium (Gibco Invitrogen, Grand Island, NY, USA) supplemented with 10% fetal calf serum (FBS), penicillin (100 U/ml), and streptomycin (100 μg/ml).

The care and use of laboratory animals were in accordance with the NIH guidelines. This study was approved by the Animal Care and Use Committee of Nippon Medical School (Bunkyo-ku, Tokyo, Japan).

Hydrogen treatment

We prepared H2-dissolved culture medium as described previously [1]. In brief, we dissolved H2in the medium by bubbling H2gas to the saturated level. We also dissolved O2in a second medium by bubbling O2gas, and CO2in a third medium by bubbling CO2gas. We combined these media to give a medium consisting of 75% H2, 20% O2, 5% CO2(vol/vol/vol). We then cultured the cartilage slices in a closed culture flask filled with the medium. Control medium contained 75% N2instead of H2. The H2concentration was maintained for 24 hr as described [15].

Cell death assay

The cartilage slices were incubated for 12 - 80 hr in medium containing 0.3 - 3 mM S-nitroso-N-acetyl-D, L-penicillamine (SNAP) (Cayman Chemical, Ann Arbor, MI, USA) in the presence or absence of H2[22, 23]. Chondrocyte viability was determined using a LIVE/DEAD Viability/Cytotoxicity Kit (Molecular Probes, Eugene, OR, USA). Living, dying and dead cells were stained with green, yellow (combination of green and red) and red fluorescence, respectively, and visualized with a confocal scanning laser microscope (FLUOVIEW FV300; Olympus, Tokyo, Japan).

Immunohistochemical staining

Frozen sections of 6 μm-thick were fixed with 10% formalin and treated with 0.3% hydrogen peroxide in methanol to inhibit endogenous peroxidase activity. The sections were incubated with 10% Block Ace (DS Pharma Biomedical Co., Ltd., Suita, Osaka, Japan) in phosphate buffered saline (PBS) and then incubated with anti-nitrotyrosine monoclonal antibody (Calbiochem, San Diego, CA, USA; 1:100 dilution with 10% Block Ace in PBS) overnight at 4°C. Nitrotyrosine residues were visualized with DAB using horseradish peroxidase (HRP)-conjugated secondary antibody (Santa Cruz Biotechnology, Inc. Santa Cruz, CA, USA) and a HistoMark ORANGE kit (KPL, Gaithersburg, MD, USA). As a positive control for staining, we used sections from cartilage treated with 1 mM 3-morpholinosydnonimine (SIN-1) (Sigma-Aldrich, St. Louis, MO, USA), which generates both superoxide anion and nitric oxide that spontaneously produce peroxynitrite. The positive area was estimated using the Image J program (version 1.41; National Institutes of Health, Bethesda, MD, USA) from four sections for each group.

RNA isolation and RT-PCR

Total RNA was isolated from the cartilage using an RNeasy Mini kit (QIAGEN, Valencia, CA, USA). Complementary DNA synthesized by SuperScript II Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) was analyzed by quantitative PCR using the Thermal Cycler Dice Real Time System TP800 (TAKARA BIO Inc., Otsu, Shiga, Japan). All samples were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH) expression. Primer and probe sequences for each PCR are listed in Table 1.

Table 1 Primers and probes for RT-PCR.
Full size table

Immunoblot analysis

Specimens were homogenized with a micro-homogenizer in SDS (sodium dodecyl sulfate) buffer (1% SDS in PBS), and then centrifuged at 10,000 g for 10 min at 4°C to remove debris. Supernatants were subjected to SDS-PAGE (SDS-polyacrylamide gel electrophoresis) followed by electrotransfer onto a PVDF membrane. The blotted membranes were blocked with Block Ace (DS Pharma Biomedical Co., Ltd.) and incubated with anti-aggrecan polyclonal antibody (Abcam, Cambridge, UK; 1:1,000 dilution), anti-MMP13 polyclonal antibody (Santa Cruz Biotechnology, Inc. 1:1,000 dilution) or anti-actin monoclonal antibody (Sigma-Aldrich; 1:500 dilution) overnight at 4°C. Each band was visualized with horseradish peroxidase (HRP)-conjugated secondary antibody (Santa Cruz Biotechnology, Inc.) and an ECL plus Western blotting detection system (GE Healthcare, Piscataway, NJ, USA).

Statistical analysis

We performed statistical analysis using StatView software (SAS Institute) by applying an unpaired two-tailed Student's t-test and ANOVA followed by Fisher's exact test, as described previously [1]. Differences were considered significant at p < 0.05.

Results

H2protects chondrocytes of hyaline and fibrecartilage from cell death

It is reported that cultured chondrocytes are sensitive to exposure to SNAP, a donor of NO• [22] and that H2exhibits no direct reaction with NO• in cultured cells as well as in a cell-free reaction. To verify the hypothesis that H2protects cells by reducing ONOO-, we examined the effect of H2on cell death induced by SNAP by using hyaline cartilage slices from a porcine hindlimb metatarsophalangeal joint as a target. Chondrocyte viability was determined using a LIVE/DEAD Viability/Cytotoxicity Kit, which provides quantitative analyses of the proportion of live and dead cells in a mixed population (Figure 1A). In living cells, membrane-permeated calcein AM is cleaved by esterases to yield cytoplasmic green fluorescence, and in dead cells membrane-impermeable ethidium homodimer-1 labels nucleic acids with red fluorescence. Dying cells, whose membrane structure has been disrupted but still have some esterase activity, were double stained as yellow. We counted green, red and yellow cells for statistical analysis (Additional file 1: Table S1). Cell viability was calculated as the percentage of green cell numbers against total cell numbers (Figure 1B). Significant protection of chondrocytes by H2was observed in the treatment with 3 mM SNAP for 12 hr (Figure 1A and 1B). More evident effects were obtained with longer SNAP treatment (Figure 1B).

Figure 1
figure 1

Hydrogen protects chondrocytes of hyaline cartilage from cell death. (A) Porcine cartilage slices were incubated with 0, 1 or 3 mM SNAP in the presence or absence of hydrogen for 12 hr at 37°C. Cells were stained with a mixture of calcein AM (Live cell: green) and ethidium homodimer (Dead cell: red) as described in Materials and methods. Scale bar: 100 μm. (B) Chondrocyte viability was determined by counting green and red cells from three areas of each slice. Six slices were used for each experimental group. The slices were incubated with 0, 1 or 3 mM SNAP in the presence or absence of hydrogen for 12, 24 or 36 hr at 37°C. Data are the means ± SD (n= 6). *p< 0.05; **p< 0.01.

Full size image

Next, we examined another type of cartilage, meniscus fibrecartilage, isolated from rats instead of swine specimens. Because it is easier to isolate swine than rat cartilage, we used swine cartilage for preliminary experiments; however, for further analysis, cartilage from rats is more suitable for RNA and protein analysis because genomic databases and antibodies are available. Treatment with 1 mM SNAP induced cell death in a time-dependent manner and H2suppressed chondrocyte death at each time point (Figure 2A and 2B, Additional file 2: Table S2). H2significantly protected chondrocytes from death with various concentrations of SNAP treatment for 48 hr (Figure 2C, Additional file 3: Table S3). These results indicate that H2protects chondrocytes by stimuli derived from NO•, although H2has no potential to react with NO•.

Figure 2
figure 2

Hydrogen protects chondrocytes of fibrocartilages from cell death. (A) Meniscus fibrocartilages from SD rats was incubated with 1 mM SNAP in the presence or absence of hydrogen for 0, 6, 20, 48, or 80 hr at 37°C. Cells were stained with calcein AM (Live cell: green) and ethidium homodimer (Dead cell: red) as described in Materials and methods. Scale bar: 40 μm. (B) Chondrocyte viability was determined by counting green and red cells from three regions of each slice. Six slices were used for each experimental group. The slices were incubated with 1 mM SNAP in the presence or absence of hydrogen for the indicated periods at 37°C. Data are the means ± SD (n= 6). *p< 0.05; ***p< 0.001. (C) The slices were incubated with 0, 0.3, 1, or 3 mM SNAP in the presence or absence of hydrogen for 48 hr at 37°C. Data are the means ± SD (n= 6). *p< 0.05; ***p< 0.001.

Full size image

H2decreases nitrotyrosine in chondrocytes and matrix of cartilage induced by SNAP

ONOO-is a strong modifier of nitration in proteins. To confirm that H2decreased ONOO-derived from NO•, we examined levels of nitrotyrosine residues in cartilage immunohistologically. In fact, NO• increased the levels of nitrotyrosine, and H2restored its increase (Figure 3). Thus, H2should decrease ONOO-derived from NO•.

Figure 3
figure 3

Hydrogen decreases nitrotyrosine in chondrocytes and cartilage matrix. (A) Meniscus fibrocartilage from SD rats was incubated with 1 mM SNAP in the presence or absence of hydrogen for 3 hr at 37°C. Frozen sections were stained with anti-nitorotyrosine antibody and visualized with DAB as described in Methods. The inset shows chondrocytes. Cartilage incubated with 1 mM SIN-1 and without SNAP was used for positive and negative staining controls, respectively. Scale bar: 40 μm. (B) Levels of nitrotyrosine in cartilage were estimated from anti-nitrotyrosine immunostaining using an image analysis program, Image J program. Data are the mean ± SD (n = 4). *p < 0.05; control versus hydrogen in 1 mM SNAP-treated groups.

Full size image

H2restores down-regulation of matrix expression and up-regulation of matrix-metallo protease expression induced by SNAP

It was also reported that ONOO-down-regulates gene expressions of the cartilage matrix proteins including aggrecan and type II collagen [24]. Conversely, levels of matrix-metallo protease are known to be up-regulated by ONOO-[24]. We then investigated the effect of H2on the expression of chondrocyte-specific matrix genes. Isolated meniscus fibrecartilage was incubated in DMEM/F-12 supplemented with 10% FBS with or without 1 mM SNAP in the presence or absence of H2. The levels of mRNA for the matrix proteins of type II collagen and aggrecan core protein were quantified with real-time PCR coupled with reverse transcription (Figure 4A and 4B). Indeed, SNAP down-regulated aggrecan and collagen II gene expressions as expected. The decreased gene expressions of the matrix proteins were significantly restored by H2-dissolved culture medium, suggesting that the decreased ONOO-restored the gene expression.

Figure 4
figure 4

Hydrogen alters mRNA and protein expressions of matrix proteins and matrix-metalloproteases (MMPs). Meniscus fibrocartilage from rats was incubated with 1 mM SNAP in the presence or absence of hydrogen for 4 hr or 20 hr at 37°C. Total RNA was extracted from 4 hr-incubated cartilage and the expression levels of aggrecan (A), type II collagen (B), GAPDH (C), MMP3 (D) and MMP13 (E) were analyzed by real-time PCR coupled with reverse transcription. Data are the mean ± SD (n = 4). *p< 0.05; **p< 0.01. (F) Total protein was extracted from 20 hr-incubated cartilage and the expression levels of aggrecan, MMP13 and actin were analyzed by immunoblotting.

Full size image

The possibility cannot be ruled out that oxidative damage derived from SNAP may reduce any gene expressions in a non-specific manner. We examined therefore the effect of H2on catabolic enzyme genes induced by SNAP, because levels of matrix-metallo protease are known to be contradictorily up-regulated by ONOO-[24]. The levels of MMP-3 and MMP-13 mRNA were measured with quantitative real-time PCR after treatment with SNAP with or without H2(Figure 4D and 4E). Moreover, the alterations of the aggrecan and MMP-13 proteins corresponded to their mRNA levels (Figure 4F). Thus, SNAP up-regulated MMP-3 and MMP-13 gene expressions as expected, whereas H2significantly suppressed MMP gene expressions, suggesting that H2restored the increased expressions by decreasing ONOO-.

Discussion

Joint diseases including osteoarthritis (OA) and rheumatoid arthritis (RA) are the most common disabling diseases, especially among elderly people. Arthritis is a degenerative disease involving abnormalities in chondrocytes, articular cartilage and other joint tissue, and is mediated by a number of underlying biochemical and physical stimuli [25, 26]. Recent studies revealed that oxidative stress plays a leading role in the initiation and progression of the disease process [27, 28]. As a joint disease model of aged patients, we stimulated chondrocytes with oxidative stress derived from NO•. The cartilage consists mostly of the extracellular matrix, which is synthesized by chondrocytes [28, 29]. The extracellular matrix is composed of collagens and proteoglycans that are responsible for the important compressive and tensile properties of cartilage [28].

The major oxidative stress generated by chondrocytes is one of the most powerful oxidants ONOO-, which was produced by the rapid reaction of NO• with •O2-[18, 19]. At an earlier stage, NO• has been considered as the primary inducer of chondrocyte death [30]; however, it has been revealed that the oxidative strength of NO• is not sufficient to initiate cell death [31, 32]. A series of experiments have indicated that the major cytotoxicity attributed to NO• is rather due to ONOO-[20, 33]. Increased ONOO-formation has been observed in cartilage and subchondral bone in rodent models [34–36] and in cartilage in OA and RA patients [37–39]. ONOO-induces cell death and regulates the decreased expression of collagens and proteoglycans and increased matrix metallo proteinases in chondrocytes, resulting in matrix degradation [24, 40]. Thus, chondrocyte is a suitable target for investing the effect of H2regarding ONOO-in this study.

In this study, we show that H2protected chondrocytes from death induced by SNAP. SNAP is a donor of NO•; however, NO• has no strong toxicity itself and H2has no potential to reduce NO•. Our previous study demonstrated that H2reduces ONOO-in a cell-free system [1]. Thus, we speculate that H2would protect SNAP-treated chondrocytes by decreasing ONOO-. More importantly, it has been reported that drinking hydrogen water suppress the nitration of kidney proteins, although H2received from hydrogen water remained for only short period in the organ (less than 5 min) [11]. In this study, we have shown that H2in medium suppress the nitration of the chondrocyte proteins (Figure 3). Thus, it is possible that even a very small amount of H2exhibits anti-oxidative effects by reducing ONOO-in many situations.

Several laboratories including ours have reported that H2altered gene expressions involved in inflammation or energy metabolism when animals drank hydrogen water [15, 17]; however, it is an open question why H2alters gene expressions, because there is no evidence that H2directly influences gene expressions. On the other hand, ONOO-has the potential to regulate gene expressions through the nitration of factors involved in transcriptional regulation [20]. As mentioned above, drinking hydrogen water suppresses the nitration of proteins; thus, it is possible that the very small amount of H2consumed by drinking hydrogen water influences nitration in in vivoexperiments and results in regulatory as well as anti-oxidative effects [11]. These results agree with the present finding that H2suppressed the nitration of proteins.

Taken together, this study implies that one of the H2functions, including transcriptional alterations, is caused through reducing ONOO-derived from NO•.

Novel pharmacological strategies aimed at selective removal of ONOO-may represent a powerful method for preventive and therapeutic use of H2for joint diseases. Cartilage has no blood vessels and nutrients are supplied through fluid. Since H2has a great advantage to rapidly diffuse into tissues even without blood flow [41, 42], it may be useful to prevent joint diseases by reducing oxidative stress and by suppressing the decrease in matrix proteins and inhibiting degradation by proteinases.

Conclusions

This study implies that one of the H2functions, including transcriptional alterations, is caused through reducing ONOO-derived from NO•. Novel pharmacological strategies aimed at selective removal of ONOO-may represent a powerful method for preventive and therapeutic use of H2for joint diseases.