Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Quantification of hydrogen production by intestinal bacteria that are specifically dysregulated in Parkinson's disease

  • Anzu Suzuki,

    Roles Data curation, Investigation, Writing – original draft

    Affiliations Division of Neurogenetics, Center for Neurological Diseases and Cancer, Nagoya University Graduate School of Medicine, Nagoya, Japan, Department of Pathophysiological Laboratory Sciences, Nagoya University Graduate School of Medicine, Nagoya, Japan

  • Mikako Ito,

    Roles Data curation, Writing – original draft

    Affiliation Division of Neurogenetics, Center for Neurological Diseases and Cancer, Nagoya University Graduate School of Medicine, Nagoya, Japan

  • Tomonori Hamaguchi,

    Roles Investigation

    Affiliation Division of Neurogenetics, Center for Neurological Diseases and Cancer, Nagoya University Graduate School of Medicine, Nagoya, Japan

  • Hiroshi Mori,

    Roles Formal analysis

    Affiliation Genome Evolution Laboratory, Center for Information Biology, National Institute of Genetics, Mishima, Japan

  • Yuka Takeda,

    Roles Investigation

    Affiliation Department of Pathophysiological Laboratory Sciences, Nagoya University Graduate School of Medicine, Nagoya, Japan

  • Ryuko Baba,

    Roles Data curation, Methodology

    Affiliation Laboratory of Soil Biology and Chemistry, Department of Biological Mechanisms and Functions, Nagoya University Graduate School of Bioagricultural Sciences, Nagoya, Japan

  • Takeshi Watanabe,

    Roles Data curation, Methodology

    Affiliation Laboratory of Soil Biology and Chemistry, Department of Biological Mechanisms and Functions, Nagoya University Graduate School of Bioagricultural Sciences, Nagoya, Japan

  • Ken Kurokawa,

    Roles Supervision

    Affiliation Genome Evolution Laboratory, Center for Information Biology, National Institute of Genetics, Mishima, Japan

  • Susumu Asakawa,

    Roles Supervision

    Affiliation Laboratory of Soil Biology and Chemistry, Department of Biological Mechanisms and Functions, Nagoya University Graduate School of Bioagricultural Sciences, Nagoya, Japan

  • Masaaki Hirayama,

    Roles Conceptualization

    Affiliation Department of Pathophysiological Laboratory Sciences, Nagoya University Graduate School of Medicine, Nagoya, Japan

  • Kinji Ohno

    Roles Conceptualization, Funding acquisition, Supervision, Writing – review & editing

    ohnok@med.nagoya-u.ac.jp

    Affiliation Division of Neurogenetics, Center for Neurological Diseases and Cancer, Nagoya University Graduate School of Medicine, Nagoya, Japan

Quantification of hydrogen production by intestinal bacteria that are specifically dysregulated in Parkinson's disease

  • Anzu Suzuki, 
  • Mikako Ito, 
  • Tomonori Hamaguchi, 
  • Hiroshi Mori, 
  • Yuka Takeda, 
  • Ryuko Baba, 
  • Takeshi Watanabe, 
  • Ken Kurokawa, 
  • Susumu Asakawa, 
  • Masaaki Hirayama
PLOS
x

Abstract

Oral administration of hydrogen water ameliorates Parkinson’s disease (PD) in rats, mice, and humans. We previously reported that the number of putative hydrogen-producing bacteria in intestinal microbiota is low in PD compared to controls. We also reported that the amount of hydrogen produced by ingestion of lactulose is low in PD patients. The decreased hydrogen production by intestinal microbiota may be associated with the development and progression of PD. We measured the amount of hydrogen production using gas chromatography by seven bacterial strains, which represented seven major intestinal bacterial groups/genera/species. Blautia coccoides and Clostridium leptum produced the largest amount of hydrogen. Escherichia coli and Bacteroides fragilis constituted the second group that produced hydrogen 34- to 93-fold lower than B. coccoides. Bifidobacterium pseudocatenulatum and Atopobium parvulum constituted the third group that produced hydrogen 559- to 2164-fold lower than B. coccoides. Lactobacillus casei produced no detectable hydrogen. Assuming that taxonomically neighboring strains have similar hydrogen production, we simulated hydrogen production using intestinal microbiota that we previously reported, and found that PD patients produce a 2.2-fold lower amount of intestinal hydrogen compared to controls. The lower amount of intestinal hydrogen production in PD was also simulated in cohorts of two other countries. The number of hydrogen-producing intestinal bacteria may be associated with the development and progression of PD. Further studies are required to prove its beneficial effect.

Introduction

Parkinson’s disease (PD) is a common neurodegenerative disorder, which is characterized by muscular rigidity, bradykinesia, resting tremor, impairment of postural reflex, and non-motor symptoms including dementia. Pathological hallmark of PD is abnormally aggregated α-synuclein protein (Lewy body) in dopaminergic neurons in substantia nigra in the midbrain [1]. The mechanisms underlying abnormal aggregation of α-synuclein have not been fully elucidated. Braak et al. observed in autopsies of idiopathic PD patients and healthy elderly people that α-synuclein pathology begins in the dorsal nucleus of vagus nerve, and ascends to raphe nuclei, locus ceruleus, substantia nigra, and cerebral cortex [1]. In 44.6% of PD patients, constipation starts on average 18.1 years before the onset of motor symptoms [2]. Similarly, rapid eye movement (REM) sleep behavior disorder and depression precede the onset of motor symptoms approximately 10 and 5 years, which represent abnormalities in Raphe nucleus and locus ceruleus, respectively [3]. Truncal vagotomy for peptic ulcer in the past markedly decreases a risk for developing PD in Denmark [4] and Sweden [5]. Accumulating evidence supports the notion that abnormal α-synuclein first aggregates in the neural plexus in the gastrointestinal tract, and ascends into the central nervous system.

In PD, intestinal permeability is increased, and its degree is positively correlated with intestinal staining for (i) Escherichia coli, (ii) nitrotyrosine, a marker for protein oxidation, and (iii) α-synuclein [6]. Similarly, in a mouse model, hyperpermeability of large intestine and abnormal aggregation of phosphorylated α-synuclein in large intestine are induced by intraperitoneal injection of lipopolysaccharide [7]. Additionally, the effect of intestinal microbiota on PD is underscored by the worsening of PD in a mouse model by transplantation of intestinal microbiota of PD patients [8]. Oxidative stress produced by macrophages in the luminal wall due to a hyperpermeabilized intestinal wall may account for the accumulation of α-synuclein in the intestinal mucosa. As the intestinal microbiota is likely to have a marked effect on the hyperpermeability-induced oxidative stress, the intestinal microbiota may be causally associated with α-synuclein pathology in the enteric nervous system in PD.

Gut microbiota in PD patients have been recently reported by eight groups including us [916]. Scheperjans et al. compared gut microbiota of 72 PD patients and 72 control subjects by analyzing 16S ribosomal RNA genes. They found that the relative abundance of Enterobacteriaceae was positively associated with the severity of postural instability and gait difficulty [9]. Keshavarzian et al. analyzed gut microbiota in 38 PD patients and 34 healthy controls. They observed that anti-inflammatory butyrate-producing bacteria from the genera Blautia, Coprococcus, and Roseburia were low in PD patients [10]. Unger et al. analyzed fecal concentrations of small chain fatty acids (SCFA) and 9 bacterial phyla/groups by qRT-PCR in 34 PD patients and 34 age-matched controls. They found that SCFA concentrations were low in PD. They also observed that Enterobacteriaceae were increased and Prevotella were decreased in PD [12], as was observed by Scheperjans et al. in Finland [9].

We analyzed 45 PD patients and 34 healthy cohabitants by quantitative RT-PCR of 16S or 23S rRNA of 19 fecal bacterial groups/genera/species [11]. A two-year follow-up study later revealed that a low count of Bifidobacterium at year 0 was associated with worsening of PD in two years [17]. We also found that fecal counts of putative hydrogen-producing bacteria were low in PD patients [11]. Among the six fecal bacterial groups/genera/species that were predicted to produce hydrogen, Bacteroides fragilis group, Clostridium perfringens, Pseudomonas possess hydrogenases to produce hydrogen [18]. Similarly, as most strains in Enterobacteriaceae, Blautia and Clostridium produce hydrogen, we assumed that Enterobacteriaceae, Blautia coccoides group, and Clostridium leptum subgroup that we analyzed by qRT-PCR produce hydrogen, but hydrogen-productivity of each bacterium remains to be experimentally determined. Similarly, hydrogen-productivity of the remaining 12 bacteria that we analyzed by qRT-PCR remains unknown.

We reported that hydrogen water prevents the development and progression of PD in a rat model [19]. Similarly, hydrogen in drinking water reduces dopaminergic neuronal loss in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced mouse model of PD [20]. Furthermore, drinking hydrogen water significantly suppresses progression of PD evaluated by Unified Parkinson's Disease Rating Scale (UPDRS) in a randomized placebo-control study of 19 PD patients [21]. A large amount of hydrogen is produced by intestinal anaerobic bacteria in human and rodents [22, 23]. The effects of hydrogen produced by intestinal bacteria are reported in a Concanavarin A (ConA)-induced mouse hepatitis model [24]. They showed that suppression of intestinal microbiota by antibiotics increased the severity of ConA-induced hepatitis, while supplementation of hydrogen-producing E. coli, but not hydrogen-deficient mutant E. coli, ameliorated the ConA-induced hepatitis.

In order to examine the roles of hydrogen-producing intestinal bacteria in PD, we individually cultured six bacterial strains representing six most prevalent intestinal bacterial groups/genera/species, and measured the amounts of hydrogen produced by these bacteria. We also analyzed hydrogen production by Lactobacillus, the count of which was higher than controls in PD [11].

Materials and methods

Bacterial strains

Blautia coccoides (JCM 1395) was provided by Japan Collection of Microorganisms, RIKEN BRC, which is participating in the National BioResource Project of the MEXT, Japan. Clostridium leptum (ATCC 29065), Bacteroides fragilis (ATCC 25285), Bifidobacterium pseudocatenulatum (ATCC 27919), Atopobium parvulum (ATCC 33793), and Lactobacillus casei (ATCC 334) were purchased from ATCC. Escherichia coli (W3110) was kindly provided by Dr. Akira Okamoto at the Aichi University of Education.

Medium

We made culture medium for each bacterium according to the protocols by DSMZ (Table 1). Blautia coccoides JCM 1395 and B. fragilis ATCC 25285 were revived in modified ATCC 1490. The constituents of each medium are shown in S1 Table. The experiments were carried out in a 31-mL screw lip test tube sealed with a butyl rubber stopper and a crimped screw cap with a hole (Sanshin Industrial Co. Ltd.). The test tube contained 5 mL of liquid medium. The headspace in the test tube was flushed with 100% N2 gas for DSMZ 58, DSMZ 104, ATCC 416, ATCC 1490 and LB, or with an 8:2 mixture of N2 and CO2 gas for DSMZ 104c, and the test tube was autoclaved.

thumbnail
Table 1. Bacterial strains used for quantification of hydrogen production and the culture medium.

https://doi.org/10.1371/journal.pone.0208313.t001

Measurement of hydrogen gas concentration

Hydrogen gas concentration was measured as previously described [25] with partial modifications. In brief, each bacterium was first cultured in 5 ml of bacterial culture medium in a 31-mL screw lip test tube, and 100 μl each was inoculated into three or four test tubes containing 5 or 7 ml culture medium (Table 1) followed by static incubation at 37°C under an anaerobic condition.

Bacterial growth was monitored by measuring the absorbance at 600 nm with Miniphoto 518R (Taitec Corp.) every 0.5 to 1.0 hours. The numbers of bacteria were counted at four or more points over a course of culture using Bacteria Counter (Sunlead Glass Corp.) to enable estimation of the bacterial counts from the 600-nm absorbance. A fraction of the headspace gas (100 μl) was intermittently sampled using a gas-tight syringe, and the concentration of hydrogen was measured by a GC-7A gas chromatography equipped with thermal conductivity detector (TCD) (Shimadzu Scientific Instruments, Inc.). For Blautia coccoides JCM 1395 and B. fragilis ATCC 25285, the concentration of hydrogen was measured by a gas chromatography (EAG analyzer GS-23, SensorTec Co. Ltd.) in a high biosafety laboratory. We confirmed that GC-7A with TCD and EAG analyzer GS-23 gave similar values by analyzing hydrogen gas produced by E. coli W3110. All gas measurements were performed at room temperature.

Manual annotation of hydrogenase genes in representative human gut strains

Six strains cultured in this study were genome sequenced previously. Genome sequence of Blautia coccoides JCM 1395 was not available, and we instead used genome sequence of Blautia coccoides YL58. We manually annotated hydrogenase genes of these seven strains using HydDB [26] (S2 Table). Four strains (Blautia coccoides YL58, Clostridium leptum ATCC 29065, Bacteroides fragilis ATCC 25285, and Escherichia coli W3110) carried hydrogenase genes, and three strains (Bifidobacterium pseudocatenulatum ATCC 27919, Atopobium parvulum ATCC 33793, and Lactobacillus casei ATCC 334) did not.

Hydrogenase genes in phylogenetic relatives of the four hydrogenase-carrying strains were similarly annotated according to a report by Wolf and colleagues [27] (S3 Table). Phylogenetic relatives of C. leptum ATCC 29065, B. fragilis ATCC 25285, E. coli W3110 were identified using (i) our previous taxonomic coverage analysis [28], which used the same primers as YIF-Scan [11], and (ii) 16S rRNA-based phylogenetic tree of type strains from The All-Species Living Tree Project database (https://www.arb-silva.de/fileadmin/silva_databases/living_tree/LTP_release_128/LTPs128_SSU/LTPs128_SSU_tree.pdf) [29]. Phylogenetic relatives of B. coccoides JCM 1395 were identified according to a previously reported collation of B. coccoides group [30].

Statistical analysis

All analyses were performed with Prism 6 (GraphPad Software Inc.). To compare hydrogen productions by seven bacterial groups/genera/species, one-way ANOVA and the Tukey-Kramer posthoc test were used. To compare hydrogen productions in PD and controls, Student unpaired t-test was used. Statistical significance was considered when p-value was less than 0.05.

Results

Among the twelve fecal bacterial groups/genera/species that we previously analyzed by qRT-PCR of 16S or 23S rRNA (YIF-Scan) (Table 2 in [11]), we quantified the amount of hydrogen production of six bacterial strains (Blautia coccoides JCM 1395, Clostridium leptum ATCC 29065, Bacteroides fragilis ATCC 25285, Escherichia coli W3110, Bifidobacterium pseudocatenulatum ATCC 27919 and Atopobium parvulum ATCC 33793) representing six most dominant bacterial groups/genera/species according to YIF-Scan (Blautia coccoides group, Clostridium leptum subgroup, Bacteroides fragilis group, Enterobacteriaceae, Bifidobacterium, and Atopobium cluster), respectively. These six prevalent bacterial groups/genera/species constitute 71.3 ± 9.4% (mean and SD) of total fecal bacteria according to the estimation by hybridization with a generic probe Eub338 [31]. The total number of the other six bacterial groups/genera/species in YIF-Scan was 1051- and 223-fold lower than the total number of six prevalent bacterial groups/genera/species in controls and PD patients, respectively [11]. However, as the count of Lactobacillus was significantly higher in PD [11], we analyzed hydrogen production by a representative strain Lactobacillus casei ATCC 334. We thus cultured seven representative strains individually, and quantified temporal profiles of hydrogen productions by each strain (Fig 1).

thumbnail
Fig 1. Temporal profiles of the number of bacteria (closed circles) and hydrogen production (open circles).

The numbers of bacteria per ml culture medium are plotted on the left axis (closed circles). The hydrogen concentration was quantified by a gas chromatography, and the number of hydrogen molecules per ml culture medium was calculated and plotted on the right axis (open circles). Arrows point to the transition between the growth and stationary phases when the hydrogen concentration accumulated in the gaseous phase was used to calculate the hydrogen production. Note that scales on the right vertical axes are identical, whereas scales on the left vertical axes and on the horizontal axes are different. Mean and SD are indicated (n = 3 or 4 culture tubes).

https://doi.org/10.1371/journal.pone.0208313.g001

We normalized the amount of hydrogen production in two ways. One was to normalize the hydrogen production by the number of bacteria at the transition between the growth and stationary phases (Method I). The other was to normalize the hydrogen production by the cumulative number of bacteria from time zero to the transition between the growth and stationary phases (Method II). Values normalized by Methods I and II had a Pearson’s correlation coefficient of 0.950 (Table 2 and Fig 2). As hydrogen accumulated during the growth phase was measured at the transition between the growth and stationary phases, Method II was likely to be more dependable than Method I. Indeed, hydrogen production by Clostridium leptum ATCC 29065 was 13.7 times lower with Method II compared to Method I (Table 2 and Fig 2). This was likely because Clostridium leptum ATCC 29065 grew slower than the other hydrogen-producing bacteria.

thumbnail
Fig 2. The amount of hydrogen production by seven intestinal bacteria.

The amount of hydrogen production was normalized in two ways. Black bars represent the hydrogen production normalized by the number of bacteria at the transition between the growth and stationary phases (arrows in Fig 1) (Method I). Blue bars represent the hydrogen production normalized by the cumulative number of bacteria from time zero to the transition between the growth and stationary phases (arrows in Fig 1) (Method II). The normalized hydrogen productions are indicated in a bar graph on a linear scale (A) and in a scattered plot on a logarithmic scale (B). (A) Mean and SD are indicated (n = 3 or 4 culture tubes). P values by one-way ANOVA are < 0.0001 for both Methods I and II. Different symbols in Methods I (a, b, and c) and II (d and e) indicate p < 0.05 by Tukey-Kramer posthoc test. (B) The Pearson’s correlation coefficient between Methods I and II is 0.950 (p = 0.001). Note that Lactobacillus is not plotted because the hydrogen production was not detected.

https://doi.org/10.1371/journal.pone.0208313.g002

thumbnail
Table 2. The amounts of hydrogen production by seven bacterial strains representing major intestinal bacterial groups/genera/species.

https://doi.org/10.1371/journal.pone.0208313.t002

Blautia coccoides JCM 1395 and Clostridium leptum ATCC 29065 produced larger amounts of hydrogen gas than the other bacterial strains (Table 2 and Fig 2). Escherichia coli W3110 and Bacteroides fragilis ATCC 25285 also produced hydrogen, but the amounts were less than those in Blautia coccoides JCM 1395 and Clostridium leptum ATCC 29065. Bifidobacterium pseudocatenulatum ATCC 27919 and Atopobium parvulum ATCC 33793 constituted the third group of hydrogen-producing bacteria, and the amounts of hydrogen production were much less compared to the aforementioned four bacteria. Lactobacillus casei ATCC 334 produced no detectable hydrogen. Using the hydrogen production by Method II in each bacterium (Table 2), we estimated the amount of hydrogen production by intestinal microbiota in 45 PD patients and 34 healthy cohabitants that we previously analyzed [11]. We multiplied the rate of hydrogen production (nmoles/bacterial cell/hour) (Table 2) by the count of fecal bacterial groups/genera/species estimated by qRT-PCR, and summed up the hydrogen production by seven bacterial strains in each patient or control. The estimated amount of hydrogen production in PD was 2.2-fold lower on average than those in controls (Fig 3). We also estimated intestinal hydrogen production using 16S rRNA amplicon data in Helsinki, Finland (72 PD patients and 72 controls) [9] and Alabama, USA (197 PD patients and 130 controls) [13]. Similar to our cohort, hydrogen production was predicted to be low in PD patients in both countries (S1 Fig).

thumbnail
Fig 3. Estimation of the amount of hydrogen production by intestinal microbiota in 45 PD patients and 34 healthy cohabitants that we previously analyzed [11].

The amount of hydrogen production per g stool per hour was calculated using the amounts of hydrogen production by seven bacterial strains measured in this study (Method II). (A) Geometric distribution plot. Bars represent mean and SD. P value was calculated with Student’s t-test. (B) Cumulative distribution function (CDF) plot.

https://doi.org/10.1371/journal.pone.0208313.g003

Discussion

We quantified the amounts of hydrogen production by seven bacterial strains representing major intestinal bacterial groups/genera/species. There are two possible drawbacks in our study. First, each culture condition was optimized for a specific strain, and was different from the intestinal environment. Simulation of intestinal environment [32, 33] would have yielded more accurate data. Second, we analyzed a single representative strain in each bacterial groups/genera/species. Multiple strains may exist in a single individual, and the dominant strain may be different from individual to individual. Similarly, as the number of genera in intestinal microbiota in human was estimated to be 89 [34], and as not all bacterial species can be cultured in vitro, quantification of hydrogen production by all observed strains would be technically unattainable. Despite these limitations, the seven bacterial strains and their taxonomic relatives constituted more than 70% of total fecal bacteria [31].

Among the seven representative bacteria that we analyzed for hydrogen production, three bacterial groups/genera/species (Blautia coccoides group, Clostridium leptum subgroup, and Bacteroides fragilis group) were decreased, whereas Lactobacillus was increased in PD [11]. The remaining three bacterial groups/genera/species (Enterobacteriaceae, Bifidobacterium, and Atopobium cluster) were not changed in PD. We found that Blautia coccoides JCM 1395 in Blautia coccoides group and Clostridium leptum ATCC 29065 in Clostridium leptum subgroup produced larger amounts of hydrogen gas than the other species. In addition, Bacteroides fragilis ATCC 25285 in Bacteroides fragilis group constituted the second group of hydrogen production. Thus, the three bacterial strains in three bacterial groups/genera/species that were decreased in PD, produced large amounts of hydrogen. In contrast, Lactobacillus casei ATCC 334 in Lactobacillus, which was increased in PD, did not produce hydrogen. Lack of hydrogen production in Lactobacillus casei has been previously reported [35]. Lactobacillus casei produces two lactate molecules from a single glucose molecule (homo lactic acid fermentation) under an anaerobic condition, which does not produce hydrogen.

Hydrogen is either a substrate or a product of hydrogenase. Production of hydrogen is mediated by reduction of protons [36]. The functions of hydrogenases are diversified, and those functions can be predicted by their amino acid sequences [37]. Based on the manual annotation, Blautia coccoides JCM 1395, Clostridium leptum ATCC 29065, and Bacteroides fragilis ATCC 25285 that we cultured have at least one [FeFe]-hydrogenase gene. One of the hydrogenase genes in B. coccoides JCM 1395 and C. leptum ATCC 29065 encodes a putative ferredoxin- and NAD-dependent [FeFe]-hydrogenase that is reversibly bifurcates protons from hydrogen to ferredoxin and NAD [38]. In addition, C. leptum ATCC 29065 and B. fragilis ATCC 25285 have at least one putative ferredoxin oxidizing [FeFe]-hydrogenase gene. Escherichia coli W3110 has putative [NiFe]-hydrogenase coding genes, at least one of which is likely to encode formate hydrogenlyase to metabolize formate and produce hydrogen [39]. Based on the genome information, these four strains have a genetic potential to produce hydrogen. On the other hand, although a very small amount of hydrogen production was observed in Bifidobacterium pseudocatenulatum ATCC 27919 and Atopobium parvulum ATCC 33793, they do not have hydrogenase genes in the genome, implying the presence of hydrogenase-independent hydrogen production.

According to the phylogenetic distribution of hydrogenase genes (S3 Table), all but one gut strains in close proximity to B. coccoides JCM 1395, C. leptum ATCC 29065, and B. fragilis ATCC 25285 have hydrogenase genes. The presence of [FeFe]-hydrogenase genes and [NiFe]-hydrogenase genes in prokaryotes genomes are relatively rare (approximately 9.1% and 26.7%, respectively [36]), suggesting that hydrogen metabolism is important for these taxa to occupy their niches in human gut. On the other hand, 13 of 21 taxonomic relatives of E. coli W3110 carried hydrogenases, whereas 8 of them did not, suggesting that the human gut inhabitants of Enterobacteriaceae have variable genetic potential of hydrogen production. Although E. coli W3110 was the third most hydrogen-producing strain, the amount of hydrogen produced by this strain was one magnitude less than those of B. coccoides JCM 1395 and C. leptum ATCC 29065. Thus, strain-to-strain variability of hydrogen production in Enterobacteriaceae is unlikely to have a major effect on our prediction of intestinal hydrogen production.

A large proportion of hydrogen produced by reduction of ferredoxin, formate and NAD(P)H in hydrogenogens is immediately reoxidized rarely within the same hydrogenogens and mostly by neighboring hydrogenotrophs (acetogens, methanogens, fumarate reducers, and sulfate reducers), which would yield no detectable hydrogen [27, 40]. Thus, the presence of hydrogenase genes does not directly indicate the production of colonic hydrogen pool. According to the healthy Japanese gut metagenome data, almost all of the genes related to methanogenesis are underrepresented compared to the metagenome data in other countries [41]. On the other hand, the genes related to the acetogenesis are overrepresented, and the genes related to the dissimilatory sulfate reduction are similar compared to the metagenome data in other countries [41]. As no shotgun metagenome data are currently available for gut microbiome in Japanese PD patients, we have to perform metagenome-based comparison of the abundance of hydrogenase genes in the future.

We previously reported breath hydrogen concentrations after taking 6 g lactulose in 28 healthy controls and 37 PD patients [42]. A synthetic disaccharide, lactulose, is metabolized by intestinal bacteria and not by human cells. Oral intake of lactulose produces hydrogen as a byproduct [43]. The hydrogen concentrations gradually increased in both controls and PD patients. However, the hydrogen concentrations at 180 min in controls was 1.5 times higher than those in PD, which coincided with our simulation that the amount of hydrogen produced by intestinal microbiota in controls was 1.69 times higher than those in PD (Fig 3). These observations support the notion that hydrogen-producing bacteria are less abundant in PD compared to controls.

Oral administration of lactulose improves chronic portal-systemic encephalopathy [44], atopic dermatitis [45], hyperammonemia [46], and chronic constipation [47]. Although the pharmacological mechanisms of lactulose on human diseases remain to be elucidated, lactulose may exert its effect by increasing hydrogen production by intestinal microbiota [48]. Although we observed no effect of lactulose on 6-OHDA-induced hemi-parkinsonism in rats [42], lactulose may be able to compensate for decreased hydrogen production in PD patients.

Molecular hydrogen reduces hydroxyl radicals in vitro [49]. However, hydrogen is unlikely to scavenge hydroxyl radicals in our cells, because the reaction rate constant between hydrogen and hydroxyl radical (4.2 x 107 L•mol-1•s-1) and the concentration of hydrogen in our bodies (0.005–0.020 mM) [42] are two to three orders of magnitude lower than those of physiological biomolecules (e.g. amino acids [50], glucose [51], and chloride [52]) that can reduce hydroxyl radicals with the reaction rate constants of 2–10 x 109 L•mol-1•s-1 and at concentrations of 5–60 mM. Another possible mechanism is that hydrogen induces hormetic responses. First, hydrogen induces mitochondrial unfolded protein response in C. elegans [53]. Second, multiple reports point to the notion that hydrogen activates the Nrf2 signaling pathway [5462]. In addition, the effect of hydrogen is not observed in Nrf2-knockout mice [55]. Third, we recently reported by meta-analysis of expression arrays that hydrogen induces the heat shock response [63]. The hormetic mechanisms are also inferred from increased reaction oxygen species (ROS) by hydrogen in human [64], rodents [65, 66], and cultured cells [62]. Further studies are required to prove the underlying mechanisms of the effects of hydrogen on PD in mouse [20], rat [19], human [21], and on other diseases [48, 67].

Supporting information

S1 Fig.

Estimation of the amount of hydrogen production by intestinal microbiota in 72 PD patients and 72 controls in Helsinki, Finland (A) [9] and 197 PD patients and 130 controls in Alabama, USA (B) [13]. Relative amount of hydrogen production was calculated using the amounts of hydrogen production by seven bacterial strains measured in this study (Method II). In contrast to our YIF-Scan (Fig 3), with which the absolute numbers of living bacteria were estimated by qRT-PCR [11], only relative abundance of each bacterium could be calculated with 16S rRNA amplicon-seq [9, 13]. Geometric distributions normalized to the mean of controls are plotted with mean and SD. P value was calculated with Student’s t-test.

https://doi.org/10.1371/journal.pone.0208313.s001

(EPS)

S2 Table. Manual annotation of hydrogenase genes in seven investigated strains.

https://doi.org/10.1371/journal.pone.0208313.s003

(DOCX)

S3 Table. Phylogenetic distribution of hydrogenase genes in representative human gut strains analyzed by YIF-Scan.

https://doi.org/10.1371/journal.pone.0208313.s004

(DOCX)

S4 Table. Raw data obtained in the current studies.

https://doi.org/10.1371/journal.pone.0208313.s005

(XLSX)

Acknowledgments

We would like to thank Dr. Akira Okamoto at the Aichi University of Education for providing E. coli strain W3110. We also would like to thank Tomomi Minato, Ayaka Minemura, Keiichi Takimoto, and Tatsunori Sakakura at Nagoya University for their technical supports and helpful critiques.

References

  1. 1. Braak H, Del Tredici K, Bratzke H, Hamm-Clement J, Sandmann-Keil D, Rub U. Staging of the intracerebral inclusion body pathology associated with idiopathic Parkinson's disease (preclinical and clinical stages). J Neurol. 2002;249 Suppl 3:III/1–5. Epub 2003/01/17. pmid:12528692.
  2. 2. Kalia LV, Lang AE. Parkinson's disease. Lancet. 2015;386(9996):896–912. pmid:25904081.
  3. 3. Tolosa E, Poewe W. Premotor Parkinson disease. Neurology. 2009;72(7 Suppl):S1. pmid:19256022.
  4. 4. Svensson E, Horvath-Puho E, Thomsen RW, Djurhuus JC, Pedersen L, Borghammer P, et al. Vagotomy and subsequent risk of Parkinson's disease. Ann Neurol. 2015;78(4):522–9. pmid:26031848.
  5. 5. Liu B, Fang F, Pedersen NL, Tillander A, Ludvigsson JF, Ekbom A, et al. Vagotomy and Parkinson disease: A Swedish register-based matched-cohort study. Neurology. 2017;88(21):1996–2002. pmid:28446653; PubMed Central PMCID: PMCPMC5440238.
  6. 6. Forsyth CB, Shannon KM, Kordower JH, Voigt RM, Shaikh M, Jaglin JA, et al. Increased intestinal permeability correlates with sigmoid mucosa alpha-synuclein staining and endotoxin exposure markers in early Parkinson's disease. PLoS One. 2011;6(12):e28032. pmid:22145021; PubMed Central PMCID: PMCPMC3228722.
  7. 7. Kelly LP, Carvey PM, Keshavarzian A, Shannon KM, Shaikh M, Bakay RA, et al. Progression of intestinal permeability changes and alpha-synuclein expression in a mouse model of Parkinson's disease. Mov Disord. 2014;29(8):999–1009. pmid:24898698; PubMed Central PMCID: PMCPMC4050039.
  8. 8. Sampson TR, Debelius JW, Thron T, Janssen S, Shastri GG, Ilhan ZE, et al. Gut Microbiota Regulate Motor Deficits and Neuroinflammation in a Model of Parkinson's Disease. Cell. 2016;167(6):1469–80 e12. pmid:27912057; PubMed Central PMCID: PMCPMC5718049.
  9. 9. Scheperjans F, Aho V, Pereira PA, Koskinen K, Paulin L, Pekkonen E, et al. Gut microbiota are related to Parkinson's disease and clinical phenotype. Mov Disord. 2015;30(3):350–8. pmid:25476529.
  10. 10. Keshavarzian A, Green SJ, Engen PA, Voigt RM, Naqib A, Forsyth CB, et al. Colonic bacterial composition in Parkinson's disease. Mov Disord. 2015;30(10):1351–60. pmid:26179554.
  11. 11. Hasegawa S, Goto S, Tsuji H, Okuno T, Asahara T, Nomoto K, et al. Intestinal Dysbiosis and Lowered Serum Lipopolysaccharide-Binding Protein in Parkinson's Disease. PLoS One. 2015;10(11):e0142164. Epub 2015/11/06. pmid:26539989; PubMed Central PMCID: PMCPMC4634857.
  12. 12. Unger MM, Spiegel J, Dillmann KU, Grundmann D, Philippeit H, Burmann J, et al. Short chain fatty acids and gut microbiota differ between patients with Parkinson's disease and age-matched controls. Parkinsonism Relat Disord. 2016;32:66–72. pmid:27591074.
  13. 13. Hill-Burns EM, Debelius JW, Morton JT, Wissemann WT, Lewis MR, Wallen ZD, et al. Parkinson's disease and Parkinson's disease medications have distinct signatures of the gut microbiome. Mov Disord. 2017;32(5):739–49. pmid:28195358.
  14. 14. Bedarf JR, Hildebrand F, Coelho LP, Sunagawa S, Bahram M, Goeser F, et al. Functional implications of microbial and viral gut metagenome changes in early stage L-DOPA-naive Parkinson's disease patients. Genome Med. 2017;9(1):39. pmid:28449715; PubMed Central PMCID: PMCPMC5408370.
  15. 15. Petrov VA, Saltykova IV, Zhukova IA, Alifirova VM, Zhukova NG, Dorofeeva YB, et al. Analysis of Gut Microbiota in Patients with Parkinson's Disease. Bull Exp Biol Med. 2017;162(6):734–7. pmid:28429209.
  16. 16. Heintz-Buschart A, Pandey U, Wicke T, Sixel-Doring F, Janzen A, Sittig-Wiegand E, et al. The nasal and gut microbiome in Parkinson's disease and idiopathic rapid eye movement sleep behavior disorder. Mov Disord. 2018;33(1):88–98. pmid:28843021; PubMed Central PMCID: PMCPMC5811909.
  17. 17. Minato T, Maeda T, Fujisawa Y, Tsuji H, Nomoto K, Ohno K, et al. Progression of Parkinson's disease is associated with gut dysbiosis: Two-year follow-up study. PLoS One. 2017;12(11):e0187307. pmid:29091972; PubMed Central PMCID: PMCPMC5665539.
  18. 18. Vignais PM, Billoud B. Occurrence, classification, and biological function of hydrogenases: an overview. Chem Rev. 2007;107(10):4206–72. pmid:17927159.
  19. 19. Fu Y, Ito M, Fujita Y, Ito M, Ichihara M, Masuda A, et al. Molecular hydrogen is protective against 6-hydroxydopamine-induced nigrostriatal degeneration in a rat model of Parkinson's disease. Neurosci Lett. 2009;453(2):81–5. pmid:19356598.
  20. 20. Fujita K, Seike T, Yutsudo N, Ohno M, Yamada H, Yamaguchi H, et al. Hydrogen in drinking water reduces dopaminergic neuronal loss in the 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine mouse model of Parkinson's disease. PLoS One. 2009;4(9):e7247. pmid:19789628; PubMed Central PMCID: PMCPMC2747267.
  21. 21. Yoritaka A, Takanashi M, Hirayama M, Nakahara T, Ohta S, Hattori N. Pilot study of H(2) therapy in Parkinson's disease: a randomized double-blind placebo-controlled trial. Mov Disord. 2013;28(6):836–9. Epub 2013/02/13. pmid:23400965.
  22. 22. Christl SU, Murgatroyd PR, Gibson GR, Cummings JH. Production, metabolism, and excretion of hydrogen in the large intestine. Gastroenterology. 1992;102(4 Pt 1):1269–77. Epub 1992/04/01. S0016508592001598 [pii]. pmid:1551534.
  23. 23. Strocchi A, Levitt MD. Maintaining intestinal H2 balance: credit the colonic bacteria. Gastroenterology. 1992;102(4 Pt 1):1424–6. Epub 1992/04/01. pmid:1551553.
  24. 24. Kajiya M, Sato K, Silva MJ, Ouhara K, Do PM, Shanmugam KT, et al. Hydrogen from intestinal bacteria is protective for Concanavalin A-induced hepatitis. Biochem Biophys Res Commun. 2009;386(2):316–21. pmid:19523450.
  25. 25. Baba R, Asakawa S, Watanabe T. H2-Producing Bacterial Community during Rice Straw Decomposition in Paddy Field Soil: Estimation by an Analysis of [FeFe]-Hydrogenase Gene Transcripts. Microbes Environ. 2016;31(3):226–33. pmid:27319579; PubMed Central PMCID: PMCPMC5017798.
  26. 26. Sondergaard D, Pedersen CN, Greening C. HydDB: A web tool for hydrogenase classification and analysis. Sci Rep. 2016;6:34212. pmid:27670643; PubMed Central PMCID: PMCPMC5037454.
  27. 27. Wolf PG, Biswas A, Morales SE, Greening C, Gaskins HR. H2 metabolism is widespread and diverse among human colonic microbes. Gut Microbes. 2016;7(3):235–45. pmid:27123663; PubMed Central PMCID: PMCPMC4939926.
  28. 28. Okai S, Usui F, Yokota S, Hori IY, Hasegawa M, Nakamura T, et al. High-affinity monoclonal IgA regulates gut microbiota and prevents colitis in mice. Nat Microbiol. 2016;1(9):16103. pmid:27562257.
  29. 29. Yarza P, Richter M, Peplies J, Euzeby J, Amann R, Schleifer KH, et al. The All-Species Living Tree project: a 16S rRNA-based phylogenetic tree of all sequenced type strains. Syst Appl Microbiol. 2008;31(4):241–50. pmid:18692976.
  30. 30. Hayashi H, Sakamoto M, Kitahara M, Benno Y. Diversity of the Clostridium coccoides group in human fecal microbiota as determined by 16S rRNA gene library. FEMS Microbiol Lett. 2006;257(2):202–7. pmid:16553854.
  31. 31. Matsuda K, Tsuji H, Asahara T, Matsumoto K, Takada T, Nomoto K. Establishment of an analytical system for the human fecal microbiota, based on reverse transcription-quantitative PCR targeting of multicopy rRNA molecules. Appl Environ Microbiol. 2009;75(7):1961–9. pmid:19201979; PubMed Central PMCID: PMCPMC2663197.
  32. 32. Van de Wiele T, Van den Abbeele P, Ossieur W, Possemiers S, Marzorati M. The Simulator of the Human Intestinal Microbial Ecosystem (SHIME((R))). In: Verhoeckx K, Cotter P, Lopez-Exposito I, Kleiveland C, Lea T, Mackie A, et al., editors. The Impact of Food Bioactives on Health: in vitro and ex vivo models. Cham (CH)2015. p. 305–17.
  33. 33. Aguirre M, Venema K. Challenges in simulating the human gut for understanding the role of the microbiota in obesity. Benef Microbes. 2017;8(1):31–53. pmid:27903093.
  34. 34. Li J, Jia H, Cai X, Zhong H, Feng Q, Sunagawa S, et al. An integrated catalog of reference genes in the human gut microbiome. Nat Biotechnol. 2014;32(8):834–41. pmid:24997786.
  35. 35. Fugelsang KC, Edwards C. G. Lactic Acid Bacteria. Boston, MA: Springer; 2007.
  36. 36. Peters JW, Schut GJ, Boyd ES, Mulder DW, Shepard EM, Broderick JB, et al. [FeFe]- and [NiFe]-hydrogenase diversity, mechanism, and maturation. Biochim Biophys Acta. 2015;1853(6):1350–69. pmid:25461840.
  37. 37. Greening C, Biswas A, Carere CR, Jackson CJ, Taylor MC, Stott MB, et al. Genomic and metagenomic surveys of hydrogenase distribution indicate H2 is a widely utilised energy source for microbial growth and survival. ISME J. 2016;10(3):761–77. pmid:26405831; PubMed Central PMCID: PMCPMC4817680.
  38. 38. Calusinska M, Happe T, Joris B, Wilmotte A. The surprising diversity of clostridial hydrogenases: a comparative genomic perspective. Microbiology. 2010;156(Pt 6):1575–88. pmid:20395274.
  39. 39. McDowall JS, Murphy BJ, Haumann M, Palmer T, Armstrong FA, Sargent F. Bacterial formate hydrogenlyase complex. Proc Natl Acad Sci U S A. 2014;111(38):E3948–56. pmid:25157147; PubMed Central PMCID: PMCPMC4183296.
  40. 40. Carbonero F, Benefiel AC, Gaskins HR. Contributions of the microbial hydrogen economy to colonic homeostasis. Nat Rev Gastroenterol Hepatol. 2012;9(9):504–18. pmid:22585131.
  41. 41. Nishijima S, Suda W, Oshima K, Kim SW, Hirose Y, Morita H, et al. The gut microbiome of healthy Japanese and its microbial and functional uniqueness. DNA Res. 2016;23(2):125–33. pmid:26951067; PubMed Central PMCID: PMCPMC4833420.
  42. 42. Ito M, Hirayama M, Yamai K, Goto S, Ito M, Ichihara M, et al. Drinking hydrogen water and intermittent hydrogen gas exposure, but not lactulose or continuous hydrogen gas exposure, prevent 6-hydorxydopamine-induced Parkinson's disease in rats. Med Gas Res. 2012;2(1):15. Epub 2012/05/23. pmid:22608009; PubMed Central PMCID: PMCPMC3407490.
  43. 43. Florent C, Flourie B, Leblond A, Rautureau M, Bernier JJ, Rambaud JC. Influence of chronic lactulose ingestion on the colonic metabolism of lactulose in man (an in vivo study). J Clin Invest. 1985;75(2):608–13. Epub 1985/02/01. pmid:3973020; PubMed Central PMCID: PMCPMC423537.
  44. 44. Bircher J, Muller J, Guggenheim P, Haemmerli UP. Treatment of chronic portal-systemic encephalopathy with lactulose. Lancet. 1966;1(7443):890–2. pmid:4159616.
  45. 45. Perlamutrov YN, Olhovskaya KB, Zakirova SA. Double-blind controlled randomised study of lactulose and lignin hydrolysed combination in complex therapy of atopic dermatitis. Microb Ecol Health Dis. 2016;27:30418. pmid:27341938; PubMed Central PMCID: PMCPMC4920936.
  46. 46. Farahmand F, Khodadad A, Fallahi G, Motaharizad D, Sabery Nejad J. Hyperammonemic Induced Coma by Bacterial Overgrowth in a Child With Hirschsprung's Disease. Acta Med Iran. 2016;54(9):614–6. pmid:27832696.
  47. 47. Voskuijl W, de Lorijn F, Verwijs W, Hogeman P, Heijmans J, Makel W, et al. PEG 3350 (Transipeg) versus lactulose in the treatment of childhood functional constipation: a double blind, randomised, controlled, multicentre trial. Gut. 2004;53(11):1590–4. Epub 2004/10/14. pmid:15479678; PubMed Central PMCID: PMCPMC1774276.
  48. 48. Ohno K, Ito M, Ichihara M, Ito M. Molecular hydrogen as an emerging therapeutic medical gas for neurodegenerative and other diseases. Oxid Med Cell Longev. 2012;2012:353152. Epub 2012/06/22. pmid:22720117; PubMed Central PMCID: PMCPMC3377272.
  49. 49. Ohsawa I, Ishikawa M, Takahashi K, Watanabe M, Nishimaki K, Yamagata K, et al. Hydrogen acts as a therapeutic antioxidant by selectively reducing cytotoxic oxygen radicals. Nat Med. 2007;13(6):688–94. pmid:17486089.
  50. 50. Masuda T, Nakano S, Kondo M. Rate constants for the reactions of OH radicals with the enzyme proteins as determined by the p-nitrosodimethylaniline method. J Radiat Res. 1973;14(4):339–45. pmid:4369113.
  51. 51. Halliwell B, Gutteridge JM, Aruoma OI. The deoxyribose method: a simple "test-tube" assay for determination of rate constants for reactions of hydroxyl radicals. Anal Biochem. 1987;165(1):215–9. pmid:3120621.
  52. 52. Liao CH, Kang SF, Wu FA. Hydroxyl radical scavenging role of chloride and bicarbonate ions in the H2O2/UV process. Chemosphere. 2001;44(5):1193–200. pmid:11513408.
  53. 53. Sobue S, Inoue C, Hori F, Qiao S, Murate T, Ichihara M. Molecular hydrogen modulates gene expression via histone modification and induces the mitochondrial unfolded protein response. Biochem Biophys Res Commun. 2017;493(1):318–24. pmid:28890349.
  54. 54. Spulber S, Edoff K, Hong L, Morisawa S, Shirahata S, Ceccatelli S. Molecular hydrogen reduces LPS-induced neuroinflammation and promotes recovery from sickness behaviour in mice. PLoS One. 2012;7(7):e42078. pmid:22860058; PubMed Central PMCID: PMCPMC3409143.
  55. 55. Kawamura T, Wakabayashi N, Shigemura N, Huang CS, Masutani K, Tanaka Y, et al. Hydrogen gas reduces hyperoxic lung injury via the Nrf2 pathway in vivo. Am J Physiol Lung Cell Mol Physiol. 2013;304(10):L646–56. Epub 2013/03/12. pmid:23475767; PubMed Central PMCID: PMCPMC3652058.
  56. 56. Zhai X, Chen X, Shi J, Shi D, Ye Z, Liu W, et al. Lactulose ameliorates cerebral ischemia-reperfusion injury in rats by inducing hydrogen by activating Nrf2 expression. Free Radic Biol Med. 2013;65:731–41. Epub 2013/08/21. pmid:23954468.
  57. 57. Li DZ, Zhang QX, Dong XX, Li HD, Ma X. Treatment with hydrogen molecules prevents RANKL-induced osteoclast differentiation associated with inhibition of ROS formation and inactivation of MAPK, AKT and NF-kappa B pathways in murine RAW264.7 cells. J Bone Miner Metab. 2014;32(5):494–504. Epub 2013/11/08. pmid:24196871.
  58. 58. Xie Q, Li XX, Zhang P, Li JC, Cheng Y, Feng YL, et al. Hydrogen gas protects against serum and glucose deprivationinduced myocardial injury in H9c2 cells through activation of the NFE2related factor 2/heme oxygenase 1 signaling pathway. Mol Med Rep. 2014;10(2):1143–9. Epub 2014/06/04. pmid:24890947.
  59. 59. Song G, Zong C, Zhang Z, Yu Y, Yao S, Jiao P, et al. Molecular hydrogen stabilizes atherosclerotic plaque in low-density lipoprotein receptor-knockout mice. Free Radic Biol Med. 2015;87:58–68. Epub 2015/06/29. pmid:26117323.
  60. 60. Li Y, Li Q, Chen H, Wang T, Liu L, Wang G, et al. Hydrogen Gas Alleviates the Intestinal Injury Caused by Severe Sepsis in Mice by Increasing the Expression of Heme Oxygenase-1. Shock. 2015;44(1):90–8. Epub 2015/04/22. pmid:25895145.
  61. 61. Li Y, Xie K, Chen H, Wang G, Yu Y. Hydrogen gas inhibits high-mobility group box 1 release in septic mice by upregulation of heme oxygenase 1. J Surg Res. 2015;196(1):136–48. Epub 2015/03/31. pmid:25818978.
  62. 62. Murakami Y, Ito M, Ohsawa I. Molecular hydrogen protects against oxidative stress-induced SH-SY5Y neuroblastoma cell death through the process of mitohormesis. PLoS One. 2017;12(5):e0176992. pmid:28467497; PubMed Central PMCID: PMCPMC5415102.
  63. 63. Nishiwaki H, Ito M, Negishi S, Sobue S, Ichihara M, Ohno K. Molecular hydrogen upregulates heat shock response and collagen biosynthesis, and downregulates cell cycles: meta-analyses of gene expression profiles. Free Radic Res. 2018;52(4):434–45. pmid:29424253.
  64. 64. Aoki K, Nakao A, Adachi T, Matsui Y, Miyakawa S. Pilot study: Effects of drinking hydrogen-rich water on muscle fatigue caused by acute exercise in elite athletes. Med Gas Res. 2012;2(1):12. Epub 2012/04/24. pmid:22520831; PubMed Central PMCID: PMCPMC3395574.
  65. 65. Eckermann JM, Chen W, Jadhav V, Hsu FP, Colohan AR, Tang J, et al. Hydrogen is neuroprotective against surgically induced brain injury. Med Gas Res. 2011;1(1):7. pmid:22146427; PubMed Central PMCID: PMCPMC3231979.
  66. 66. Matchett GA, Fathali N, Hasegawa Y, Jadhav V, Ostrowski RP, Martin RD, et al. Hydrogen gas is ineffective in moderate and severe neonatal hypoxia-ischemia rat models. Brain Res. 2009;1259:90–7. pmid:19168038.
  67. 67. Ichihara M, Sobue S, Ito M, Ito M, Hirayama M, Ohno K. Beneficial biological effects and the underlying mechanisms of molecular hydrogen—comprehensive review of 321 original articles. Med Gas Res. 2015;5:12. pmid:26483953; PubMed Central PMCID: PMCPMC4610055.