Molecular hydrogen as a preventive and therapeutic medical gas: initiation, development and potential of hydrogen medicine


Molecular hydrogen (H2) has been accepted to be an inert and nonfunctional molecule in our body. We have turned this concept by demonstrating that H2 reacts with strong oxidants such as hydroxyl radical in cells, and proposed its potential for preventive and therapeutic applications. H2 has a number of advantages exhibiting extensive effects: H2 rapidly diffuses into tissues and cells, and it is mild enough neither to disturb metabolic redox reactions nor to affect signaling reactive oxygen species; therefore, there should be no or little adverse effects of H2. There are several methods to ingest or consume H2; inhaling H2 gas, drinking H2-dissolved water (H2-water), injecting H2-dissolved saline (H2-saline), taking an H2 bath, or dropping H2-saline into the eyes. The numerous publications on its biological and medical benefits revealed that H2 reduces oxidative stress not only by direct reactions with strong oxidants, but also indirectly by regulating various gene expressions. Moreover, by regulating the gene expressions, H2 functions as an anti-inflammatory and anti-apoptotic, and stimulates energy metabolism. In addition to growing evidence obtained by model animal experiments, extensive clinical examinations were performed or are under investigation. Since most drugs specifically act to their targets, H2 seems to differ from conventional pharmaceutical drugs. Owing to its great efficacy and lack of adverse effects, H2 has promising potential for clinical use against many diseases.

1. Introduction

Molecular hydrogen (H2, dihydrogen, or hydrogen gas) has been accepted to behave as an inert gas at body temperature in mammalian cells. In fact, H2 seems to react with no biological compound, including oxygen gas in the absence of catalysts at body temperature. On the other hand, in some bacteria, H2 is enzymatically catabolyzed as an energy source for providing electrons, or is a product of some types of anaerobic metabolism. These reactions are usually catalyzed by iron- or nickel-containing enzymes called hydrogenases. In contrast, mammals have no functional hydrogenase genes (Fritsch et al., 2013). Thus, it has been believed that H2 is nonfunctional in our cells.

We turned this concept in a publication in 2007 that H2 acts as a therapeutic and preventive antioxidant by selectively reducing highly strong oxidants such as hydroxyl radical (OH) and peroxynitrite (ONOO) in cells, and that H2 exhibits cytoprotective effects against oxidative stress (Ohsawa et al., 2007). Since then, a large number of studies have explored therapeutic and preventive effects of H2. These published papers cover many biological effects against oxidative stress in almost all organs (Ohta, 2011Ohta, 2012). Moreover, it has been revealed that H2 has more functions, including anti-inflammatory, anti-apoptotic, and anti-allergic effects, and that H2 stimulates energy metabolism, in most tissues of model animals. Until 2013, the number of publications on its biologically or medically beneficial effects had been increasing and had surpassed ~ 300 as shown in Fig. 1. Previous review articles mainly introduced various cellular and animal experiments (Ohta, 2011Ohta, 2012).

In addition to publications on model animal experiments, more than 10 papers on clinical examinations have been published. Thus, this article will review the results of recent clinical examinations toward actual applications. Moreover, this review article will look back the discovery process of the biological effects of H2, and propose the possible mechanisms to explain the effects of H2.

2. Oxidative stress as pathogenic sources and physiological role

2.1. Oxidative stress as pathogenic sources

Reactive oxygen species (ROS) are generated inside the body throughout our daily lives as a side-product of energy metabolism by oxidative phosphorylation in every aerobic organism. Occasionally, excess ROS are produced, such as smoking or air pollution, exposure to ultraviolet or irradiation rays, hard exercise, physical or psychological stress, and so on (Liu et al., 1996Agarwal, 2005Harma et al., 2006Tanriverdi et al., 2006Grassi et al., 2010). When ROS are produced excessively or endogenous antioxidant capacity is diminished, indiscriminate oxidation elicits harmful effects, resulting in “oxidative stress”. Acute oxidative stress arises from various different situations: inflammation, ischemia reperfusion in cardiac or cerebral infarction, organ transplantation, and cessation of operative bleeding, or others (Ferrari et al., 1991Vaziri and Rodriguez-Iturbe, 2006Reuter et al., 2010). Mounting evidence has established strong links between chronic oxidative stress and a wide variety of pathologies, including malignant diseases, diabetes mellitus, atherosclerosis, and chronic inflammatory processes as well as many neurodegenerative diseases and the aging process (Andersen, 2004Bagul and Banerjee, 2013El Assar et al., 2013Kim and Byzova, 2014). Under normal conditions, ROS induced by strenuous exercise results in muscle fatigue (Westerblad & Allen, 2011).

2.2. The process of generation of oxygen species

As a first step in generating ROS, superoxide anion radicals (O2) are the primary ROS mostly generated by electron leakage from the mitochondrial electron transport chain (Finkel and Holbrook, 2000Turrens, 2003Andersen, 2004Lin and Beal, 2006). Superoxide dismutase (SOD) enzymatically converts O2 to hydrogen peroxide (H2O2), which is metabolized to generate water (H2O). Very reactive OH is generated from H2O2 via the Fenton or Weiss reaction in the presence of catalytically active metals, such as Fe2 + and Cu+ (Halliwell & Gutteridge, 1992). Reaction of O2 with nitric oxide (NO) generates ONOO, which is a very active nitrogen species (RNS) (Radi, 2013). OH is the major cause of the oxidation and destruction of biomolecules by the direct reaction or by triggering the chain reaction of free radicals (Lipinski, 2011). Ionizing radiation, including cosmic rays also generates OH as a damaging intermediate through interaction with water, a process termed radiolysis (Schoenfeld et al., 2012Schoenfeld et al., 2011).

Other enzymes, including NADPH oxidases, cytochrome p450s, lipoxygenase, cyclooxygenase, and xanthine oxidase, also participate in ROS generation in the immuno- or detoxifying-system (Droge, 2002).

2.3. Physiological roles of H2O2

As mentioned above, ROS had historically been believed to cause cellular damage and to lack physiological functions. Indeed, accumulation of oxidative damage by ROS has been linked to multiple pathologies, as mentioned above. Cellular redox homeostasis is, however, a delicate balance between ROS production and the antioxidant system (Bashan et al., 2009Brewer et al., 2013). Oxidative stress is now appreciated to function as signaling molecules to regulate a wide variety of physiology (Liu et al., 2005Bell, Klimova, Eisenbart, Moraes, et al., 2007Bell, Klimova, Eisenbart, Schumacker and Chandel, 2007). H2O2 was shown to be required for cytokine, insulin, growth factor, AP-1, c-Jun N-terminal kinase 1 (JNK1), p53, and nuclear factor kappa B (NF-κB) signaling and to promote phosphatase-inactivation by cysteine oxidation (Finkel, 1998Chandel, Trzyna, McClintock and Schumacker, 2000Chandel, Vander Heiden, Thompson and Schumacker, 2000). These reactions provide a plausible biochemical mechanism by which ROS can impinge on signaling pathways. There have been numerous reports highlighting the importance of ROS-dependent signaling in a variety of systems (Salganik, 2001Sauer et al., 2001Collins et al., 2012).

Additionally, oxidative stress caused by H2O2 and NO induces enzymes involved in anti-oxidation and tolerance to protect cells against oxidative stress (Endo et al., 2009Ristow and Zarse, 2010). For example, translocation of NF-E2-related factor 2 (Nrf2) into the nucleus leads to the regulation of gene expression involved in defense systems against oxidative stress (Jazwa & Cuadrado, 2010) and other toxic sources including heavy metals (Gan & Johnson, 2013). Moreover, H2O2 is a key factor to regulate cellular differentiation (Tsukagoshi et al., 2010Tormos et al., 2011), the immune system (West et al., 2011Zhou et al., 2011), autophagy (Li et al., 2012Garg et al., 2013) and apoptosis (Mates et al., 2012). Thus, it is crucial not to completely eliminate H2O2 to maintain homeostasis.

2.4. Exploration of an ideal antioxidant

Although antioxidant therapy or prevention of various diseases is expected owing to the clinical importance of oxidative damage, antioxidants have been of limited therapeutic success (Steinhubl, 2008). Antioxidant supplements have exhibited little effect on preventing cancer, myocardial infarction and atherosclerosis, but rather conversely have increased mortality (Bjelakovic et al., 2007Hackam, 2007Brambilla et al., 2008Steinhubl, 2008Hercberg et al., 2010); thus, it is very important to be aware of side effects in developing an effective antioxidant for the prevention of oxidative stress-related diseases.

Under these situations, an ideal antioxidant molecule is expected to mitigate excess oxidative stress, but not disturb the redox homeostasis. In other words, the ideal molecule should not reduce signaling molecules such as H2O2 but should effectively reduce strong oxidants such as OH. After experiments, we have reached the current conclusion that the ideal antioxidant could be H2.

3. Discovery of the biological effects of molecular hydrogen

During the process of searching for the ideal antioxidant, we infused H2 into culture medium without changing pH, O2 and CO2 concentrations or other conditions. When cultured PC12 cells were exposed to oxidative stress by treatment with antimycin A, an inhibitor of the mitochondrial electron translocation chain, the cells apparently shrank and extended short fibers in response to oxidative stress (Fig. 2A). In contrast, when the cells were treated with the inhibitor in the presence of H2, the cells did not change their shape (Fig. 2B). In H2-degassed medium from H2-medium, the cells again responded to oxidative stress. This finding indicates that H2 affected no components in the original medium, but directly acted to the cells. By this finding of the first experiment, we foresaw that H2 has great potential for actual clinical use.

After this experiment, we tried to identify the target of H2 in cultured cells. H2 dissolved in culture medium did not change the cellular levels of O2 and H2O2, as judged by the fluorescent signals of MitoSOX and dichlorofluorescein-diacetate (DCF-DA), respectively. Additionally, H2 did not decrease the cellular level of NO. In contrast, H2 treatment significantly decreased levels of OH, as judged by the decrease in the fluorescent signal of hydroxyphenyl fluorescein (HPF) (Setsukinai et al., 2003). Moreover, the decrease in the cellular OH level by H2 was confirmed by spin trapping technology (Halliwell & Gutteridge, 1992).

The selective reduction of ROS can be explained by the marked oxidative strength of OH, as shown in Fig. 3, which was profiled according to published data (Setsukinai et al., 2003). This means that OH is strong enough to react with even inert H2, but that O2, H2O2, and NO are insufficient to react with H2 according to their activities. In other words, H2 is mild enough neither to disturb metabolic redox reactions nor to affect ROS that function in cellular signaling.

4. Methods of how to ingest molecular hydrogen

4.1. Safety of molecular hydrogen

Previous review articles have summarized the nature of H2 as a chemical element (Ohta, 2011Ohta, 2012). Here, the following point should be emphasized regarding the safety of H2 for medical use: H2 gas is flammable only at temperatures higher than 527 °C, and explodes by a rapid chain reaction with O2 only in the explosive range of H2 concentration (4-75%, vol/vol). Thus, H2 can be used for medical applications without surplus worries by several ingestion methods because inhalation of 1-4% H2 gas exhibits great efficacy (Ohsawa et al., 2007Hayashida et al., 2008).

4.2. Inhalation of hydrogen gas

Inhalation of H2 gas is a straightforward therapeutic method. H2 gas can be inhaled through a ventilator circuit, facemask or nasal cannula. Since inhaled H2 gas acts rapidly, it may be suitable for defense against acute oxidative stress. In particular, inhalation of gas does not affect blood pressure (Ohsawa et al., 2007); on the other hand, drip infusion of drugs increases blood pressure and causes serious obstacles during the treatment of myocardial infarction.

By a clinical examination, Ono et al. showed that inhalation of 3–4% H2 gas reached a plateau at approximately 10–20 μM in the arterial and venous blood, respectively, in about 20 min and affected no physiological parameters, suggesting no adverse effects (Ono et al., 2012a).

4.3. Oral ingestion by drinking hydrogen water

Inhalation of H2 gas may be unsuitable or impractical for continuous H2 consumption in daily life for preventive use. In contrast, solubilized H2 (H2-dissolved water; namely, H2-water) may be beneficial since it is a portable, easily administered and safe way to ingest H2 (Nagata et al., 2009Nakashima-Kamimura et al., 2009). H2 can be dissolved in water up to 0.8 mM (1.6 mg/L) under atmospheric pressure at room temperature without changing pH. Although ad libitum drinking saturated H2-water was more effective than diluted one, 80 μM of H2-water was still effective for improving obesity (Kamimura et al., 2011).

H2-water can be made by several methods: infusing H2 gas into water under high pressure, electrolyzing water to producing H2, and reacting of magnesium metal or its hydride with water. These methods may be applicable not only to water but also to other solvents. H2 penetrates the glass and plastic walls of any vessels in a short time, while aluminum containers are able to retain hydrogen gas for a long time.

4.4. Injection of hydrogen-saline

H2 is intravenously or intraperitoneally injectable as H2-saline (H2-dissolved saline), which allows the delivery of H2 with great efficacy in model animals (Cai et al., 2009Sun et al., 2011aSun et al., 2011bLi, Ji, et al., 2013).

Nagatani et al. performed an open-label, prospective, non-randomized study of intravenous H2 administration in 38 patients hospitalized for acute ischemic stroke. All patients received an H2 intravenous solution (200 mL twice a day) immediately after the diagnosis of acute ischemic stroke. Data from this study indicated that an H2 intravenous solution is safe for patients with acute cerebral infarction, including patients treated with tissue-plasminogen activator (t-PA) (Nagatani et al., 2013).

4.5. Direct incorporation of molecular hydrogen by diffusion: Eye-drop, bath and cosmetics

Alternatively, H2-loaded eye drops were prepared by dissolving H2 in saline and directly administering to the ocular surface (Oharazawa et al., 2010Kubota et al., 2011).

H2 delivery to cardiac grafts during cold preservation using a hydrogen-supplemented water bath efficiently ameliorated myocardial injury due to cold ischemia and reperfusion. This device to saturate organs with H2 during cold storage merits further investigation for possible therapeutic and preventative use during transplantation (Noda et al., 2013).

H2 should easily penetrate the skin and is distributed throughout the body via blood flow. Thus, taking a warm water bath with dissolved H2 is a method of incorporating H2 into the body in daily life. Indeed, powders that produce H2 baths are commercially available in Japan.

5. Monitoring the movement of molecular hydrogen

H2 has a number of advantages as a potential antioxidant. It has favorable distributing characteristics with its own physical ability to penetrate biomembranes and diffuse into the cytosol. H2 rapidly reaches the nucleus and mitochondria to protect nuclear DNA and mitochondria (Ohsawa et al., 2007). Moreover, H2 passes through the blood-brain barrier although most antioxidant compounds cannot.

H2 in blood can be monitored by the following method; venous or arterial blood is collected in a closed aluminum bag with no dead space, followed by the addition of a definite volume of air into the bag. H2 in the air phase transferred from the blood can be measured by gas chromatography. The inhalation of H2 actually increased H2 dissolved in arterial blood in a H2 gas concentration-dependent manner, and the H2 levels in venous blood were lower than in arterial blood; the different level between arterial and venous blood indicates the amount of H2 incorporated and consumed into tissues (Ohsawa et al., 2007).

In a clinical examination, Ono et al. also showed a difference in H2 concentrations between arterial and venous blood (Ono et al., 2012a).

The gaseous diffusion of H2 can be monitored inside various tissues by detecting with a specific electrode. For example, H2 concentration has been monitored within the rat myocardium. The electrode was inserted into the ‘at risk’ area for infarction to estimate the diffusion of H2 into the ischemic myocardium area after coronary artery occlusion. H2 concentration was increased by its diffusion even with coronary artery occlusion (Hayashida et al., 2008).

Moreover, we devised eye drops with dissolved H2 to directly administer H2 to the retina, and monitored the time course of changes in H2 levels using a needle-shaped hydrogen sensor electrode inserted through the sclera to the vitreous body in rats. H2 could reach the vitreous body by administering H2 saturated in normal saline. When H2 eye drops were administered continuously, approximately 70% H2 was detected on the ocular surface (Oharazawa et al., 2010).

When water saturated with H2 was placed into the stomach of a rat, H2 was detected at several μM in blood (Nagata et al., 2009Nakashima-Kamimura et al., 2009). Moreover, hepatic H2 was monitored with a needle-type hydrogen electrode, and hepatic glycogen maintained H2 after oral administration of H2-water, partly explaining why consumption of even a small amount of H2 over a short dwell time could efficiently improve various disease models (Kamimura et al., 2011).

When seven adult volunteers drank H2-water, the H2 content of their expired breath was measured by gas chromatography with a semiconductor (Shimouchi et al., 2012). The ingestion of H2-water rapidly increased breath H2 content to its maximal level 10 min after ingestion and thereafter decreased to the baseline level within 60 min. The loss of H2 from the water during the experimental procedures accounted for 3% or less of the H2. H2 release from the skin surface was estimated as approximately 0.1%. Based on the remaining H2 mass balance, approximately 40% of H2 that had been drunk was consumed inside the body. This report suggests that exogenous H2 is at least partially trapped by oxygen radicals, such as OH (Shimouchi et al., 2012).

6. Effects of molecular hydrogen and clinical examinations

6.1. Safety of molecular hydrogen for humans

H2 has advantages from the aspect of toxicity: H2 has no cytotoxicity even at high concentration (Abraini et al., 1994Lillo et al., 1997Fontanari et al., 2000Lillo and Parker, 2000). Safety standards have been established for high concentrations of H2 gas for inhalation since high-pressure H2 gas is used in deep-diving gas mixes to prevent decompression sickness and arterial gas thrombi (Fontanari et al., 2000). Importantly for clinical use, the safety of H2 for humans is demonstrated by its application of an extremely high concentration of H2 gas in Hydreliox, an exotic, breathing gas mixture of 49% H2, 50% helium and 1% O2, which is used to prevent decompression sickness and nitrogen narcosis during very deep technical diving (Abraini et al., 1994Lillo et al., 1997Fontanari et al., 2000Lillo and Parker, 2000). Considering that H2 is an inert gas and nonfunctional in our body, its lack of toxic effects is easily understandable. As mentioned above, inhalation of 1–4% H2 gas exhibits great efficacy (Ohsawa et al., 2007Hayashida et al., 2008).

Many reports have supported that H2 exhibited great efficacy in extensive disease models regardless of the ingestion methods of H2 as summarized in Fig. 4. This review article focuses mainly on clinical examinations.

6.2. Protective effects against reperfusion injury

Ischemia/reperfusion induces serious oxidative stress, and its injuries should be considered in many clinical treatments. Inhalation of H2 gas improved ischemia/reperfusion injuries in cerebral (Ohsawa et al., 2007) and myocardial infarction (Hayashida et al., 2008Yoshida et al., 2012). Hydrogen-saline protected renal ischemia-reperfusion injury (Wang et al., 2011). All clinical manifestations related to post-cardiac arrest (CA) syndrome are attributed to ischemia/reperfusion injury in various organs, including the brain and heart. H2 gas inhalation yielded great improvement in survival and the neurological deficit score in post-CA syndrome in a rat model (Hayashida et al., 2012).

H2 also mitigated damage during the transplantation of various organs by H2 gas (Buchholz et al., 2008), or H2-water (Cardinal et al., 2010) and H2-preservation solution (Noda et al., 2013).

Ono et al. intravenously administered H2-saline with Edaravone, a clinically approved radical scavenger, in 8 patients with acute brain stem infarction and compared magnetic resonance imaging (MRI) indices of 26 patients who received Edaravone alone. The relative diffusion-weighted images (rDWIs), regional apparent diffusion coefficients (rADCs), and pseudo-normalization time of rDWI and rADC were all improved with the combined infusion of H2 with Edaravone (Ono et al., 2011).

6.3. Protective effects against neurodegeneration

Chronic oxidative stress is widely accepted as one of the causes of neurodegeneration, including dementia and Parkinson disease (PD) (Andersen, 2004Federico et al., 2012). Experimental oxidative stress in the brain can be induced by chronic physical restraint stress and can impair learning and memory (Liu et al., 1996Abrous et al., 2005). Additionally, neural proliferation in the dentate gyrus of the hippocampus is suppressed by the restraint stress (Abrous et al., 2005). Drinking H2-water suppressed the increase in this oxidative stress and prevented this cognitive impairment (Nagata et al., 2009). Moreover, H2-water restored neural proliferation in the dentate gyrus of the hippocampus in this model animal (Nagata et al., 2009). Since antidepressants increase adult neurogenesis (Becker and Wojtowicz, 2007Sahay and Hen, 2007), H2-water may be applicable for improving depression and some mental disorders.

In PD, mitochondrial dysfunction and the associated oxidative stress are major causes of dopaminergic cell loss in the substantia nigra (Yoritaka et al., 1996Schapira, 2008). H2 in drinking water was given before or after stereotactic surgery for 6-hydroxydopamine-induced nigrostrital degeneration in a rat model of PD. H2-water prevented both the development and progression of nigrostriatal degeneration in rats (Fu et al., 2009). Moreover, drinking H2-water also suppressed dopaminergic neuronal loss in another PD mice model induced by MPTP (1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine) (Fujita et al., 2009).

In a placebo-controlled, randomized, double-blind, parallel-group clinical pilot study, the efficacy of H2-water in patients with levodopa-medicated PD was assessed (Yoritaka et al., 2013). Participants drank 1 L per day of H2-water or pseudo water for 48 weeks. Total Unified Parkinson's Disease Rating Scale (UPDRS) scores in the H2-water group (n = 9) improved, whereas UPDRS scores in the placebo group (n = 8) worsened. Despite the low number of patients and the short duration of the trial, the difference was significant (P < .05).

6.4. Preventive effects against metabolic syndrome

Drinking H2-water stimulates energy metabolism (Kamimura et al., 2011). H2-water significantly alleviated fatty liver in db/db mice, which is a type 2 diabetes model mice with obesity, as well as high fat-diet-induced fatty liver in wild-type mice. Long-term H2-water drinking significantly decreased fat and body weights, despite no increase in the consumption of diet and water, in db/db mice. Moreover, drinking H2-water decreased levels of plasma glucose, insulin, and triglyceride by stimulating energy metabolism (Kamimura et al., 2011).

Beneficial roles of H2-water in the prevention of potential metabolic syndrome were reported by 3 independent clinical studies as follows.

Kajiyama et al. performed a randomized, double-blinded, placebo-controlled, crossover study in 30 patients with diabetes mellitus type II and 6 patients with impaired glucose tolerance. The patients consumed either 900 mL H2-water or placebo water for 8 weeks, with a 12-week washout period. Statistical significance was observed in the improvement of electronegative charge-modified low-density lipoprotein (LDL)-cholesterol, small dense LDL, and urinary 8-isoprostanes. In four of six patients with impaired glucose tolerance, H2 improved their indexes of the oral glucose tolerance test to a normal level (Kajiyama et al., 2008).

Nakao et al. performed an open-label trial in 20 subjects with potential metabolic syndrome (Nakao et al., 2010b). The participants drank H2-water (1.5–2.0 L/day) for 8 weeks and showed an increase in urinary SOD; a decrease in urinary thiobarbituric acid reactive substances (TBARS), a marker of lipid peroxidation; an increase in high-density-lipoprotein (HDL)-cholesterol; and a decrease in total cholesterol.

Song et al. characterized the effects of H2-water (0.9–1.0 L/day) on the content, composition, and biological activities of serum lipoproteins in 20 patients with potential metabolic syndrome. Serum analysis showed that drinking H2-water for 10 weeks resulted in decreased serum total-cholesterol (TC) and LDL-cholesterol levels. In addition, H2-water significantly improved i) protection against LDL oxidation, ii) inhibition of tumor necrosis factor-α (TNF-α)-induced monocyte adhesion to endothelial cells, iii) stimulation of cholesterol efflux from macrophage foam cells, and iv) protection of endothelial cells from TNF-α-induced apoptosis (Song et al., 2013).

Taken together, these results strongly suggest the benefits of drinking H2-water in patients with metabolic syndrome.

6.5. Suppressive effects on inflammation

H2 reduced inflammation in experimental model animals induced by concanavalin A (Kajiya et al., 2009), dextran sodium sulfate (Kajiya et al., 2009), lipopolysaccharide (LPS) (Xu et al., 2012Chen et al., 2013), Zymosan, an inducer of generalized inflammation (Xie et al., 2010b), and polymicrobial sepsis (G.M. Li et al., 2013). H2 gas, H2-saline and H2-water decreased the levels of pro-inflammatory cytokines to suppress inflammation.

Rheumatoid arthritis (RA) is a chronic inflammatory disease characterized by the destruction of bone and cartilage. Ishibashi et al. administered H2-water containing 4–5  mg/L H2 (high H2-water) (0.5 L/day) for 4 weeks to 20 patients with RA. Urinary 8-hydroxy-deoxyguanine (8-OHdG) was significantly reduced by 14.3% (P < 0.01) on average. Disease activity (DAS28, using C-reactive protein levels also decreased (P < 0.01) during the same period. After the washout period, both the urinary 8-OHdG and the mean DAS28 had decreased compared to the end of the drinking period. During the second drinking period, the mean DAS28 was reduced (P < 0.01). All the 5 patients with early RA (duration < 12 months) who did not show antibodies against cyclic citrullinated peptides (ACPAs) achieved remission, and 4 of them became symptom-free at the end of the study. Thus, the symptoms of RA were significantly improved with H2-water (Ishibashi et al., 2012Ishibashi, 2013).

6.6. Mitigation of side effects in treatments of cancers

Inhalation of H2 gas and drinking H2-water improved mortality and body-weight loss caused by treatment with an anti-cancer drug, cisplatin, and alleviated nephrotoxicity in mice. Despite its protective effects against cisplatin-induced toxicity, H2 did not affect the anti-tumor activity of cisplatin against cancer cell lines in vitro and tumor-bearing mice in vivo (Nakashima-Kamimura et al., 2009).

Kang et al. performed a randomized placebo-controlled clinical study of H2-water (0.55–0.65 mM, 1.5–2.0 L/day) for 6 weeks in 49 patients receiving radiation therapy for malignant liver tumors. H2 suppressed the elevation of total hydroperoxide levels, maintained serum antioxidant capacity, and improved the quality of life (QOL) scores. In particular, H2-water efficiently prevented loss of appetite. There was no difference in tumor response to radiotherapy between the two groups (Kang et al., 2011).

6.7. Effects on dermatomyositis and mitochondrial diseases

Ito et al. performed an open-label trial of H2-water (1.0 L/day) for 12 weeks in 14 patients with muscular diseases, including muscular dystrophies, polymyositis/dermatomyositis, and mitochondrial myopathies. In the open-label trial, significant improvements were observed in the lactate:pyruvate ratio, fasting blood glucose, serum matrix metalloproteinase-3 (MMP3), and triglycerides. In particular, the lactate:pyruvate ratio, which is a sensitive biomarker of a compromised mitochondrial electron transport system, was decreased by 28% in mitochondrial myopathies. In addition, MMP3, which represents the activity of inflammation, was decreased by 27% in dermatomyositis (Ito et al., 2011).

Then, they performed a randomized, double-blind, placebo-controlled, crossover trial of H2-water or placebo-dehydrogenized water (0.5 L/day) for 8 weeks in 22 patients with dermatomyositis and mitochondrial myopathies. In the double-blind trial, a statistically significant improvement was observed only in serum lactate in mitochondrial myopathies by H2-water consumption, but the lactate:pyruvate ratio in mitochondrial myopathies and MMP3 in dermatomyositis also tended to be decreased by H2-water consumption (Ito et al., 2011).

6.8. Mitigating effects of hemodialysis

Nakayama et al. performed an open-label placebo-controlled crossover trial of 12 sessions of hemodialysis in eight patients (Nakayama et al., 2009Nakayama et al., 2010) and an open-label trial of 78 sessions of hemodialysis in 21 patients (Nakayama et al., 2010). In both studies, continuous sessions of hemodialysis with H2-dialysis solution decreased systolic blood pressure before and after dialysis. In the short-term study, plasma methylguanidine was significantly decreased. In the long-term study, plasma monocyte chemoattractant protein 1 and myeloperoxidase were significantly decreased.

6.9. Effects on acute erythematous skin diseases

Ono et al. treated 4 patients with acute erythematous skin diseases with fever and/or pain by intravenous administration of 0.5 L of H2-fluid for 30 min for more than 3  days. Erythema of these 4 patients and associated symptoms improved significantly after H2 treatment. In conclusion, an improvement in acute erythemtous skin diseases followed the administration of H2-fluid without compromising safety (Ono et al., 2012b).

6.10. Effects on exercise

Aoki et al. recruited ten young male soccer players and subjected them to exercise tests and blood sampling. Each subject was examined twice in a crossover double-blind manner; they were given H2-water or placebo water for one-week intervals. Subjects were requested to use a cycle ergometer at 75 % maximal oxygen uptake (VO2) for 30 min, followed by measurement of peak torque and muscle activity throughout 100 repetitions of maximal isokinetic knee extension. Acute exercise resulted in an increase in blood lactate levels in the subjects given placebo water, whereas oral intake of H2-water prevented the elevation of blood lactate during heavy exercise. Peak torque in the placebo water-group significantly decreased during maximal isokinetic knee extension, indicating muscle fatigue (Aoki et al., 2012).

7. Possible molecular mechanisms underlying various effects of molecular hydrogen

As mentioned, H2 has various beneficial effects on animal models and patients; anti-oxidation, anti-inflammation, anti-apoptosis, anti-allergy and stimulation of energy metabolism (Ohta, 2011Ohta, 2012). Their mutual relationships are not clear, but the reduction of oxidative stress may primarily lead to various subsequent effects. There remain many unresolved questions regarding the molecular mechanism to fully explain the effects of H2. In particular, although H2 apparently regulates gene expressions and the protein phosphorylation involved in signal transduction, the primary target(s) of H2 in these regulations has not been identified. Here, possible mechanisms are proposed as summarized in Fig. 5.

7.1. Direct reduction of hydroxyl radicals with molecular hydrogen

Considering the reaction rate of OH with H2 in dilute aqueous solutions, the reaction rate may be too slow to fully enable a decrease in OH for exhibiting its beneficial roles (Buxton et al., 1988). Mammalian cells are, however, highly structured with complicated biomembranes and viscous solutions with concentrated multiple components. Since collision frequency is rate-limiting in a viscous environment, the marked diffusion rate of H2 could be advantageous to overcome the slow reaction rate constant.

H2 was shown to reduce OH in an experiment using cultured cells (Ohsawa et al., 2007); however, the reduction of OH to exhibit therapeutic and preventive effects had not been directly demonstrated at tissue level at this time. Later, it was shown that H2 eye-drops directly decreased OH induced by ischemia reperfusion in retinas (Oharazawa et al., 2010). Moreover, it has been demonstrated that, at a tissue level, H2 neutralized OH that had been induced by ionizing irradiation in testes, as judged by the decreased HFP signal, and exhibited a radioprotective role of H2 (Chuai et al., 2012).

OH is known as a major trigger of the chain reaction of free radicals (Niki, 2009). Once the chain reaction occurs on biomembranes, it continues and expands causing serious damage to cells. H2 may accumulate in the lipid phase more than in the aqueous phase, especially in unsaturated lipid regions, which are the major target of the initial chain reaction. Thus, H2 may have an advantage to suppress the chain reaction, which produces lipid peroxide, and leads to the generation of oxidative stress markers, such as 4-hydroxyl-2-nonenal (4-HNE) and malondialdehyde (MDA) (Niki, 2014). Indeed, H2 decreased these oxidative markers in many studies (Ohsawa et al., 2008Ning et al., 2013Zhou et al., 2013). Additionally, OH can modify deoxy-guanine (dG) to 8-OHdG (Delaney et al., 2012Kawai et al., 2012). H2 decreased the level of 8-OHdG in most of the examined patients and animals (Ishibashi et al., 2012Kawai et al., 2012).

Thus, these observations support that sufficient H2 can efficiently mitigate tissue oxidation induced by OH. However, when animals or humans drink H2-water, the question remains whether H2-water provides a sufficient amount of H2 to efficiently scavenge OH, which are continuously generated in normal and disease states.

7.2. Direct reduction of peroxynitrite with molecular hydrogen to regulate gene expressions

As another molecular mechanism, scavenging ONOO by H2 should be considered. ONOO is known to modify tyrosine of proteins to generate nitro-tyrosine (Radi, 2013). Several studies have shown that drinking H2-water efficiently decreases nitro-tyrosine in animal models regardless of H2-water (Cardinal et al., 2010), H2 gas (Shinbo et al., 2013) or H2-saline (Chen, Manaenko, et al., 2010Yu et al., 2011Zhang et al., 2011Zhu et al., 2011). Moreover, drinking H2-water decreased nitro-tyrosine in patients with RA (Ishibashi et al., 2012). Thus, at least a part of the effect of H2 can be attributed to the decreased production of nitro-tyrosine of proteins.

Many protein factors involved in transcriptional control are nitrolated (-O-NO2) or nitrosolated (-S-NO2). It is possible that the decrease in -O-NO2 or -S-NO2 may regulate various gene expressions (Radi, 2013). However, major targets have not been identified and are under investigation.

7.3. Indirect reduction of oxidative stress by regulating gene expression

H2 reduces oxidative stress not only directly, but indirectly by inducing anti-oxidation systems, including hemeoxygenase-1 (HO-1) (Huang et al., 2010Park et al., 2010), SOD (Zhai et al., 2013), catalase (Cai et al., 2013) and myeloperoxidase (Zhang et al., 2011). Nrf2 is known to function as a defense system against oxidative stress and various poisons by inducing various genes including HO-1. HO-1, a microsomal enzyme degrading heme to carbon monoxide, free iron, and biliverdin, participates in the cell defense against oxidative stress (Jazwa & Cuadrado, 2010).

In Nrf2-deficient mice, mitigating effects by inhalation of H2 gas declined in hyperoxic lung injury accompanying with a decrease in HO-1, indicating that H2 gas can ameliorate hyperoxic lung injury in an Nrf2-dependent manner (Kawamura et al., 2013). Activation of Nrf2 is also required for amelioration of cerebral ischemia-reperfusion injury in rats by H2 (Zhai et al., 2013).

7.4. Regulation of gene expression of pro-inflammatory cytokines and hormones

H2 seems to display various effects against various pathogenic situations indirectly by up- or down-regulation of gene expression.

In most inflammatory models, H2 functions as anti-inflammation by decreasing the expressions of pro-inflammatory factors (Ohta, 2011). These pro-inflammatory factors include NF-κB (H. Chen et al., 2010), TNF-α, interleukin (IL)-1β, IL-6, IL-10, IL-12, CCL2 and interferon (INF)-γ, ICAM-1 (Buchholz et al., 2008), PGE2 and PGE2, (Kawasaki et al., 2010), and high mobility group box 1 (HMGB-1) (Xie et al., 2010a),

H2-water improves obesity and metabolic parameters as mentioned above. Analysis of gene expression revealed that a hepatic hormone, fibroblast growth factor 21 (FGF21), showed increased expression by drinking H2-water. FGF21 functions to stimulate fatty acid and glucose expenditure (Kamimura et al., 2011).

H2-water supplementation increased the gastric expression of mRNA encoding ghrelin, a growth hormone secretagogue, and ghrelin secretion. Strikingly, the neuroprotective effect of H2-water was abolished by administration of an antagonist to the ghrelin receptor. Thus, a neuroprotective effect of H2 may be mediated by enhanced production of ghrelin (Matsumoto et al., 2013).

It is unlikely that H2 could directly react with transcriptional factors to regulate the gene expressions as shown in Fig. 5.

7.5. Regulation of other genes

H2 functions toward the suppression of apoptosis by up-regulation of anti-apoptotic factors, and/or by down-regulation of pro-apoptotic factors against various pathogenic statuses. Indeed, H2 stimulated expressions of the anti-apoptoptic factors of Bcl-2, and Bcl-xL (Kawamura et al., 2010) and suppressed the expressions of pro-apoptotic factors including casapase 3 (Cai et al., 2008Sun et al., 2009), caspase 8 (Kawamura et al., 2010) and caspase 12 (J. Cai et al., 2008). In pro-apoptotic Bax, H2 did not only down-regulate its gene expression (Kawamura et al., 2010Huang et al., 2011), but also inhibited the translocation of Bax to mitochondria (Terasaki et al., 2011) by an unknown mechanism.

Expression profiling of rat liver demonstrated that H2-water has a minimal effect on the expression levels of individual genes in normal rats (Nakai et al., 2011); however, more apparent up- or down-regulation of individual genes were observed in disease models and patients against the pathogenic stimuli.

H2 seems to exhibits a variety of phenotypes toward improving many pathogenic states by regulating various gene expressions. Genes up- or down-regulated by H2 are as follows; MMP2 and MMP9 (C.H. Chen et al., 2010); MMP3 and MMP13 (Hanaoka et al., 2011); brain natriuretic peptide; intercellular-adhesion-molecule-1 (ICAM-1) and myeloperoxidase; cyclooxygenase 2 (COX-2), neuronal nitric oxide synthase (nNOS), and connexins 30 and 43 (Hugyecz et al., 2011); collagen III (Terasaki et al., 2011); and ionized calcium binding adaptor molecule 1 (Iba1) (Sun et al., 2011b).

Additionally, recent reports indicated that ventilation with H2 significantly increased expression of surfactant-related molecules, ATP synthases and stress-response molecules in lung grafts (Tanaka et al., 2012), and that H2 reduced mRNA levels of osteoclast-specific markers, including tartrate resistant acid phosphatase, calcitonin receptor, cathepsin K, metalloproteinase-9, carbonic anhydrase type-II, and vacuolar-type H+-ATPase (D.Z. Li et al., 2013).

These molecules are, probably, not primary responders to H2, but indirectly act to enable the various effects of H2. The primary target of H2 remains unknown.

7.6. Modulation of signaling

H2 attenuates the phosphorylation of FcεRI-associated Lyn and its downstream signaling molecules (Itoh et al., 2009). H2 inhibits the phosphorylation of ASK1 and its downstream signaling molecules, p38 MAP kinase, JNK, and IκB without affecting ROS production derived from NADPH oxidase. The other study also indicated that H2 inhibited the phosphorylations of some signal proteins, including MEK, p38, ERK, JNK (Cardinal et al., 2010). Pretreatment with H2 reduced fatty acid uptake and lipid accumulation after palmitate overload in HepG2 cells, which was associated with inhibition of JNK activation (Iio et al., 2013). H2 reduced the phosphorylation of p38, extracellular signal-regulated kinase, JNK, and protein kinases B stimulated with the receptor activator of NFκB ligand-induced osteoclast differentiation (D.M. Li et al., 2013). These studies suggest that H2 influences some signal transductions as an indirect modulator; however, it is unlikely that H2 could directly bind to some receptors involved in the signal transductions. The primary target molecule of H2 has not been identified in these signal transduction pathways.

8. Comparison with the other medical gasses

Several medical gasses are expected to provide more effective therapeutic interventions and preventive medicine. Hydrogen sulfide (H2S), carbon monoxide (CO), and NO are extremely toxic molecules; however, they play important roles as signaling molecules (Kimura, 2010Motterlini and Otterbein, 2010). CO is poisonous even in low concentrations due to its ability to interfere with oxygen delivery, whereas endogenous CO is important in the physiological functioning of organs and possesses anti-inflammatory properties (Motterlini & Otterbein, 2010). In contrast, H2 is advantageous to have no cytotoxicity.

In the past decade, there has been marked, rapid growth in the knowledge of gaseous molecules. Gas inhalation as disease therapy has received recent interest (Szabó, 2007Kajimura et al., 2010). It has recently been revealed that heme-based proteins play central roles in their generation- and reception-mechanisms as the primary target of these gas molecules. An integrated approach to the interaction of these gasses revealed the physiological significance of H2S, CO, and NO on mitochondrial cytochrome c oxidase, a key target and central mediator of mitochondrial respiration (Kajimura et al., 2010). As far as briefly examined (Ohsawa et al., 2007), H2 did not reduce the oxidized heme of cytochrome c. Thus, the primary target of H2 seems to differ from that of the other medical gasses.

Regarding the interaction between H2 and the other toxic medical gasses, combined therapy with H2 and CO demonstrated enhanced therapeutic efficacy via both anti-oxidant and anti-inflammatory mechanisms, and may be a clinically feasible approach for preventing ischemia/reperfusion injury in the myocardium (Nakao et al., 2010a).

Breathing NO plus H2 during ischemia/reperfusion reduced the infarct size and maintained cardiac function, and reduced the generation of myocardial nitrotyrosine associated with NO inhalation. Administration of NO plus H2 gasses for inhalation may be useful for planned coronary interventions or for the treatment of ischemia/reperfusion injury (Shinbo et al., 2013).

The production of NO, H2S or CO is catalyzed by NO synthases, cystathionine γ-lyase and cystathionine β-synthase or HO-1, respectively (Kashfi & Olson, 2013). In contrast, mammalian cells have no enzyme for producing intracellular H2.

9. Concluding remarks

This article reviewed the progress of hydrogen medicine from its initiation toward clinical applications. H2 is easily applicable because it has no adverse effects and great efficacy on nearly all pathogenic states involved in oxidative stress and inflammation. Indeed, the clinical effects of H2 were positive in patients with more than 10 various diseases. Since most pharmacological drugs specifically act on their targets, H2 seems to differ from conventional drugs because of its extensive and various effects. H2 has great potential for preventive and therapeutic applications in many diseases owing to its great efficacy and its novel concept.

Conflict of Interest Statement

The author is a Chief Scientic Officer of MitoGene, LLC (Little Rock, AR, USA).