Gut microbiome alterations in pulmonary hypertension in highlanders and lowlanders

Wenxue Dong, Lifeng Ma, Qiumin Huang, Xu Yang, Zhendong Mei, Mengmeng Kong, Zhonghan Sun, Zhiying Zhang, Jing Li, Jiaojiao Zou, Dandan Chen, Yuxiang Dai, Daxin Zhou, Yan Zheng, Longli Kang
ERJ Open Research 2023 9: 00617-2022; DOI: 10.1183/23120541.00617-2022

Abstract

Background Alterations in the gut microbiota have been observed in patients with pulmonary hypertension (PH), though whether the roles of the gut microbiota in PH at different altitudes are the same is unknown. This study aims to evaluate the associations of the gut microbiome with PH in highlanders and lowlanders.

Methods PH patients and controls were recruited from those who permanently live on the Tibetan plateau (highlanders) or the plains (lowlanders), and underwent transthoracic echocardiography close to their altitude of residence (at 5070 m for highlanders versus6 m for lowlanders). The gut microbiome was profiled using metagenomic shotgun sequencing.

Results In total, 13 PH patients (46% highlanders) and 88 controls (70% highlanders) were included. The overall microbial composition was different in PH patients compared to controls (p=0.003). Notably, among lowlanders, a composite microbial score of pro-atherosclerotic trimethylamine-producing species was increased in PH patients compared with that in controls (p=0.028), while among highlanders no such difference was observed (p=0.087). Another composite gut microbial score including eight species of Lactobacillus, which has shown beneficial effects on cardiovascular functions, was higher in highlanders than lowlanders (p<0.01). Furthermore, this score tended to be lower in PH patients than controls among highlanders (p=0.056) but not among lowlanders (p=0.840). In addition, the gut microbiome showed a good performance in distinguishing PH patients from controls in both lowlanders and highlanders.

Conclusions Our study reported differently altered gut microbiome profiles between highland and lowland PH patients, highlighting the distinct microbial mechanism in PH in highlanders compared with lowlanders.

Abstract

Alterations in trimethylamine-producing bacteria and Lactobacillus are associated with lowland and highland pulmonary hypertension respectively https://bit.ly/3Y37inT

Introduction

Pulmonary hypertension (PH) encompasses a group of malignant vascular disorders, and its pathogenesis is not well delineated [1, 2]. Recent evidence has indicated the potential role of microbial ecology in PH development [3]. Proposed mechanisms include the gut microbiome-involved production of cardiovascular-related metabolites such as trimethylamine oxide (TMAO) and lactate [4], the modulation effects of certain bacterial species (e.g., Lactobacillus) on endothelium-dependent tissue oxygenation and the eventual remodelling of the vascular system [5]. Various studies have revealed that the altered gut microbial composition and specific bacterial taxonomies (e.g., Bifidobacterium, Streptococcus and Acidaminococcus spp.) are associated with hypertension and might drive pathological abnormalities in the cardiovascular system [6–9]. To our knowledge, there are thus far limited studies reporting gut dysbiosis and microbial metabolites in PH patients [10].

It is known that the specific pathogenic mechanisms differ to some degree between highland and lowland PH. For example, hypoxic stimuli-induced pulmonary arterial vasoconstriction and vascular remodelling has been shown to greatly contribute to the inception and progression of PH in highlanders, but this may not be the case in lowlanders [1]. Besides, different overall gut composition and relative abundances of specific bacteria groups (i.e., Firmicutes and Bacteroidetes) were observed between highlanders and lowlanders [11, 12]. However, little is known about whether the associations of the gut microbiome with PH are different between highlanders and lowlanders.

In the current study, we characterised the gut metagenome profiles of PH patients and controls enrolled from the Tibetan plateau and the plain areas. We hypothesised that distinct gut microbiome profiles are related to the difference in PH pathogenesis between highlanders and lowlanders.

Methods

Study participants

This study enrolled PH patients and controls among highlanders living on the Tibetan plateau range and lowlanders living on the plains (Shanghai) from June 2018 to June 2019. Highlanders were defined as those born, raised and living on the Tibetan plateau range. Lowlanders were defined as those born, raised and living in Shanghai. Participants were excluded if they: 1) self-reported use of antihypertension medication, pulmonary vasodilator therapies, probiotics or antibiotics in the past month; 2) suffered from general bowel diseases (such as diarrhoea, inflammatory bowel disease, irritable bowel syndrome, bowel removal) and overt systemic diseases (such as diabetes, chronic kidney disease, coronary artery disease, cancers) as determined by medical records and physical examination by clinicians; or 3) had PH secondary to left-sided cardiac disease, PH secondary to lung diseases, chronic thromboembolic pulmonary hypertension and other relevant etiologies in appropriate clinical scenarios (such as portopulmonary hypertension and PH secondary to sickle cell disease) [1]. Eventually, 101 participants (68 highlanders and 33 lowlanders) were included in the current analysis.

All participants provided their written informed consent to study procedures conforming to the latest revision of the Declaration of Helsinki standards. The ethical committee of Xizang Minzu University and Zhongshan Hospital Fudan University approved this study.

Outcomes and covariates assessment

Pulmonary artery pressure was measured by echocardiography (GE VIVID Q) at a local clinic at an altitude of 5070 m for highlanders and at Zhongshan Hospital at an altitude of 6 m for lowlanders. PH patients were defined as participants with a mean pulmonary artery pressure >30 mmHg or a systolic pulmonary artery pressure >50 mmHg [13]. Because pure tricuspid regurgitation (TR) or right atrial enlargement (RAE) represents a physiological adaptation rather than organic impaired pulmonary circulation [14], controls were further classified into: 1) TR/RAE controls if they had TR or RAE identified by echocardiography; and 2) normal controls if they did not have TR and RAE in the secondary analysis.

Data on participant characteristics, including sex, age, ever-smoking or drinking status (yes/no), medication use and medical history, were obtained from medical records and face-to-face interviews. Body mass index was defined as the weight (kilogrammes) divided by the square of the body height (metres).

Gut microbiome profiling

Stool samples were collected during the clinic visit using sterile stool collection kits containing storage solutions. Stool samples were frozen at −80°C after collection within 24 h until further processing in the central lab. Genomic DNA was extracted using the QIAamp DNA Micro Kit (QIAGEN, Hilden, Germany), and then quantified with Qubit 3.0 (Thermo Fisher Scientific, Waltham, MA, USA) and Qubit dsDNA HS Assay Kit (Thermo Fisher Scientific) according to the manufacturer's guidelines. The Illumina sequencing libraries were prepared from isolated DNA and normalised to 1-ng input per sample. Metagenomic shotgun sequencing of faecal samples was carried out on the NovaSeq 6000 platform for paired-end 150-bp reads (Illumina, San Diego, CA, USA). The data size per sample of raw sequence data was ∼12.9 Gbp with up to 85.9 million reads.

All sequenced metagenomes were quality controlled using KneadData v0.7.4 and were mapped to remove host-derived and rDNA sequences using Bowtie2 v2.4.4 against Genome Reference Consortium Human Build 37 (GRCh37) and SILVA database (version 128) [15]. Adapter sequences and low-quality bases were trimmed by Trimmomatic (version 0.39, with parameters ILLUMINACLIP: NexteraPE-PE.fa:2:30:10:8:TRUE SLIDINGWINDOW:4:20 MINLEN:75). The quality of processed metagenomes was examined using the FastQC toolkit v0.11.9. Taxonomic profiles of the relative abundances of all bacteria were quantified by MetaPhlan 3.0.13 [16], and a total of 287 microbial species (represented by 112 genera and by nine phyla) present in >10% of total samples with a relative abundance of >0.01% were included in our subsequent analysis. The profiling and quantification of trimethylamine (TMA)-synthesis enzymes and their functional genes were determined by HUMAnN v3.0.0 [16] using the MetaCyc database, UniRef orthologous gene family catalogue and Enzyme Commission gene families; only those present in >10% of samples were used in the downstream analysis.

Statistical analysis

Differences in the basic characteristics between PH patients and controls were analysed using Wilcoxon rank-sum tests (for continuous traits) or χ2 tests (for categorical traits). Based on the relative abundance of microbial species, the microbial species richness of a participant (α-diversity) was computed using the Shannon index, and microbiome dissimilarity (β-diversity) was assessed by principal coordinate analysis (PCoA) using the Bray–Curtis dissimilarity matrix. The differences in the α-diversity and the relative abundance of taxonomic and functional features were assessed by Wilcoxon rank-sum test for comparison between two groups and by the Kruskal–Wallis test for comparisons among >two groups. The significance of difference in β-diversity was examined between groups by permutational multivariate analysis of variance (PERMANOVA) with 999 permutations.

In consideration of the previously reported role of TMA-producing bacteria in PH development [10], a gut microbial (GMB)-TMA score was calculated based on the presence of 23 species with functional genes encoding TMA-synthesis enzymes detected in the current study (range 0–23; not present in participants=0, and present in participants=1). In addition, considering the prior knowledge on the beneficial effects of the probiotic Lactobacillus on cardiovascular health and in preventing PH [17, 18], a GMB-Lactobacillus score was calculated based on the detectable presence of eight Lactobacillus species (i.e., Lactobacillus mucosae, Lactobacillus rogosae, Lactobacillus ruminis, Lactobacillus salivarius, Lactobacillus kefiranofaciens, Lactobacillus kefiri, Lactobacillus oris and Lactobacillus sanfranciscensis) in the current study (range 0–8; absent in participants=0, present in participants=1).

Random forest-based models of 287 microbial species were constructed to classify PH patients and controls in highlanders or lowlanders separately and validated by 10-fold cross-validation testing (999 trees, balanced class weight, max features=square root of all features) using the R package randomForest v4.6–14 (www.stat.berkeley.edu/∼breiman/RandomForests/). The set of predictable microbial species with the minimum cross-validation error and the smallest number of species was selected as the optimal set. For each test, the receiver operating characteristic (ROC) curve was constructed to evaluate the performance of the constructed models, and the area under the parametric curve was computed using the R package pROC v1.18.0. All statistical tests were two-sided and an α value of 0.05 was considered to indicate statistical significance. All the data analyses were conducted in R v4.1.1 (https://www.R-project.org/).

Results

Basic characteristics of the study population

Among the 101 participants (mean age 43.4±14.6 years; 50.5% women) included in the final analysis, 13 (12.9%) were PH patients, including six highlanders and seven lowlanders, and the other 88 participants were controls, including 45 TR/RAE controls (97.8% were highlanders) and 43 normal controls (41.9% were highlanders). Among lowlanders, PH patients were more likely to have systemic hypertension compared with controls (p=0.028), while other characteristics showed no significant difference between PH patients and controls in both highlanders and lowlanders (p>0.05) (table 1).

View this table:
TABLE 1

Basic characteristics of the participants

The gut microbial composition was associated with PH status in both lowlanders and highlanders

The α-diversity measured by the Shannon index was similar between PH patients and controls (p=0.730, figure 1a), or among PH patients, TR/RAE controls and normal controls (p=0.580, supplementary figure S1A). However, a distinct microbial composition was observed in PH patients compared with controls (p=0.003, PERMANOVA, figure 1b), and among PH patients, TR/RAE controls and normal controls (p=0.001, supplementary figure S1B).

FIGURE 1
FIGURE 1

Altered gut microbiota composition in PH patients. a) Boxplots show the Shannon diversity of the intestinal microflora was comparable between PH patients and controls. b) Principal coordinate analysis (PCoA) plots show a distinct Bray–Curtis distance of dissimilarity in the intestinal microbiota between PH patients and controls. c) The Shannon diversity of PH patients decreased compared with controls among lowlanders but not among highlanders. d) The PCoA plots show a distinct Bray–Curtis distance of dissimilarity in the intestinal microbiota between PH patients and controls in lowlanders and highlanders. e and f) The relative abundance of the intestinal microbiome was plotted in taxa bars based on the phylum level (e) and genus level (f). L: lowlanders; H: highlanders; PH: pulmonary hypertension.

Lowland PH patients tended to have a lower α-diversity than either lowland controls (p=0.009, figure 1c) or highland PH patients (p=0.001, figure 1c). No significant difference was observed between PH patients and controls in highlanders (p=0.082, figure 1c), or among PH patients, TR/RAE controls and normal controls in highlanders (p=0.920 (H.Normal versus H.TR/RAE), p=0.090 (H.Normal versus H.PH), p=0.100 (H.TR/RAE versus H.PH), supplementary figure S1C). The gut microbial composition tended to differ between PH patients and controls among the lowlanders (p=0.050, PERMANOVA, figure 1d) but not among the highlanders (p=0.252, PERMANOVA, figure 1d). At the phylum level, Bacteroidetes and Firmicutes were the most abundant phyla in our population (figure 1e and supplementary figure S1D), and at the genus level, Prevotella was dominantly enriched in highlanders whereas Bacteroides was the most prevalent genus in lowlanders (figure 1f and supplementary figure S1E).

TMA-related species were more enriched in lowlanders with PH

Among the analysed microbial species, the relative abundances of 20 bacterial species (e.g., Clostridium symbiosum, Bacteroides massiliensis, Clostridium saccharolyticum, Gordonibacter pamelaeae and Eisenbergiella massiliensis) were increased in PH patients, and that of four species (Allisonella histaminiformans, Sellimonas intestinalis, Absiella dolichum and Bacteroides coprocola) were increased in controls (p<0.01, figure 2a). Among these different species, five species (i.e., Citrobacter portucalensis, Clostridium aldenense, Clostridium bolteae, Citrobacter youngae and Cl. saccharolyticum) carried functional genes encoding TMA-synthesis enzymes (i.e., carnitine monooxygenase gene (cntA), choline TMA-lyase gene (cutC) and betaine reductase complex component B subunit β gene (grdH)). Furthermore, among all the analysed species in the current study, 23 species captured homologous genes encoding enzymes producing TMA, including six species that contained cntA and cutC (i.e., Escherichia coli, Klebsiella pneumoniae, Citrobacter freundii, Enterobacter cloacae complex, Klebsiella quasipneumoniae and Klebsiella variicola), and five species contained cutC and grdH (i.e., Hungatella hathewayi, Dorea longicatena, Cl. aldenense, Clostridium citroniae and Clostridium clostridioforme) (figure 2b).

FIGURE 2
FIGURE 2

The gut microbiota of PH patients contained a higher abundance of trimethylamine (TMA)-producing microbiota. a) Bars show species with different relative abundances between PH patients and controls (p<0.01, Wilcoxon rank-sum test). The species carrying functional genes encoding TMA-synthesis enzymes are marked with a *. b) Heat map showing the prevalence of TMA-produced enzymes in the genomes of the intestinal microbiome. PH: pulmonary hypertension.

Based on these 23 species containing genes encoding TMA-producing enzymes, we calculated a GMB-TMA score and assessed its association with PH. Although there was no difference in the distribution of GMB-TMA scores between PH patients and all controls (p=0.330, figure 3a and p=0.520 (Normal versus PH), p=0.240 (TR/RAE versus PH), supplementary figure S2A), among lowlanders, PH patients were more likely to have a higher GMB-TMA score compared with controls (11.14±3.58 in PH patients versus 8.12±2.03 in controls, p=0.022, figure 3b). Among highlanders, no significant discrepancy was observed between PH patients and all controls (p=0.057, figure 3b and p=0.110 (H.Normal versus H.PH), p=0.058 (H.TR/RAE versus H.PH), supplementary figure S2B). Consistently, the genes encoding the TMA-synthesis enzymes were more likely to cluster in PH patients than controls (p<0.001, figure 3c and p=0.210 (Normal versus TR/RAE), p<0.001 (Normal versus PH), p<0.001 (TR/RAE versus PH), supplementary figure S2C), and lowland PH patients tended to have a higher relative abundance of these genes compared with lowland controls (p<0.001, figure 3d). In highlanders, the relative abundance of these genes was similar between PH patients and all controls (p=0.790, figure 3d and p=0.520 (H.Normal versus H.PH), p=0.940 (H.TR/RAE versus H.PH), supplementary figure S2D).

FIGURE 3
FIGURE 3

Comparison of trimethylamine (TMA)-producing microbiota in the gut between PH patients and controls in lowlanders and highlanders. Boxplots showing a) the discrepancy in the gut microbial (GMB)-TMA score between PH patients and controls, and b) between them in lowlanders or highlanders separately. Boxplots showing c) the discrepancy in relative abundance of TMA-producing enzymes between PH patients and controls, and d) between them in lowlanders or highlanders separately. L: lowlanders; H: highlanders; PH: pulmonary hypertension.

Lactobacillus was a potential microbial marker of PH among highlanders

Out of the 26 different gut microbial genera in highland and lowland normal controls, three genera (i.e., Lactobacillus, Lactococcus and Leuconostoc) involved in the production of lactate, which regulates prolonged hypoxia responses in cardiovascular disorders [19, 20], were more abundant in highland normal controls than in lowland normal controls (p<0.01, figure 4a). At the species level, most detectable species (i.e., L. mucosae, L. rogosae, L. ruminis, L. salivarius, L. kefiranofaciens, L. kefiri and L. sanfranciscensis) belonging to the genus Lactobacillus were more abundant in highland normal controls than in lowland normal controls (all p<0.05 expect for L. sanfranciscensis, supplementary figure S3).

FIGURE 4
FIGURE 4

The gut microbial (GMB)-Lactobacillus scores varied between highlanders and lowlanders. a) Bars represent difference in relative abundance of genus between normal controls in highlanders (H.Normal controls) and normal controls in lowlanders (L.Normal controls) (p<0.01, Wilcoxon rank-sum test). Boxplots show b) the discrepancy in GMB-Lactobacillus score between PH patients and controls, c) between them in lowlanders and highlanders, separately, and d) between all lowlanders and all highlanders. L: lowlanders; H: highlanders; PH: pulmonary hypertension.

Lactobacillus has been demonstrated to reduce blood pressure and be beneficial in cardiovascular functions via the modulation of immune-inflammatory and oxidative stress response [17]. Thus, we calculated a GMB-Lactobacillus score at the species level based on eight detectable species belonging to the Lactobacillus genus, including the revealed seven species and L. oris. The GMB-Lactobacillus score was comparable between PH patients and controls in either the combined population (p=0.066, figure 4b), the highlanders only (p=0.056, figure 4c) or the lowlanders only (p=0.800, figure 4c). However, notable discrepancies in the GMB-Lactobacillus score between lowlanders and highlanders were observed (0.36±0.65 in lowlanders versus 2.54±1.32 in highlanders, p<0.001, figure 4d). In subgroup analysis, highland PH patients had a higher GMB-Lactobacillus score compared with lowland patients (1.83±0.75 in highland PH patients versus 0.43±0.79 in lowland PH patients, p=0.007, figure 4c), and consistent findings were observed in controls (2.61±1.35 in highland controls versus 0.35±0.63 in lowland controls, p<0.001, figure 4c). In addition, among highlanders, TR/RAE controls presented a comparable distribution of GMB-Lactobacillus score to normal controls (p=0.660), and both control groups presented a suggestive higher GMB-Lactobacillus score than PH patients (p-value for difference in normal controls and PH patients=0.050, and p-value for difference in TR/RAE controls and PH patients=0.076, supplementary figure S4).

Gut microbial signatures could distinguish PH patients from normal controls

With the above-observed differences in gut microbial features between lowlanders and highlanders, we explored the potential of gut microbiota markers for the classification of PH patients in lowlanders and highlanders separately. Because there was only one lowland TR/RAE control in our study (table 1), we only included normal controls and PH patients to make random forest modelling in highlanders and lowlanders comparable. A model among lowlanders including 25 microbial species achieved an accuracy of 93.8% (figure 5a), and a model among highlanders of 15 species presented a performance of the accuracy of 85.7% (figure 5b). Of note, among the 25 microbial features in the lowland model, five species, i.e., Flavonifractor plautii, H. hathewayi, Cl. bolteae, K. quasipneumoniae and D. longicatena, carried functional genes encoding TMA-synthesis enzymes, and the first three species exhibited higher relative abundances in PH patients compared with controls (all p<0.05, supplementary figure S5). Among these 15 species in the highland model, the top three predictable species for PH patients, i.e., Roseburia sp CAG:182, Eubacterium sp CAG:38 and Roseburia sp CAG:471, were more abundant in PH patients compared with controls (all p<0.05, supplementary figure S6).

FIGURE 5
FIGURE 5

Random forest modelling identified a subset of taxa distinguishing pulmonary hypertension (PH) patients from normal controls in a) lowlanders or b) highlanders. The random forest algorithm was used to train the 25 and 15 most important species, in lowlanders or highlanders, to distinguish PH patients from normal controls. The species in the optimal set are ranked from top to bottom by decreasing mean decrease accuracy. Since mean decrease accuracy quantifies the strength of each respective predictor, the best predictors of PH are at the top of the plot.

Discussion

In this study of the gut metagenome in PH involving highlanders and lowlanders, we evaluated their gut microbiota profiles associated with PH pathogenesis. We observed distinct gut microbial composition in PH patients compared to controls. In addition, compared with controls, lowland PH patients carried higher abundances of bacterial species involved in the production of TMA, while highland PH patients tended to carry a decreased abundance of Lactobacillus. These divergent patterns of gut microbiota suggested a potential difference in microbial aetiology in PH between highlanders and lowlanders.

PH is a complex disease involving abnormal alterations in lung, brain and multiple other organs [21]. Data on population studies of the gut microbiota with PH are limited. Consistent with the results from the only two studies of shotgun metagenomics in PH [10, 22], we observed that the altered gut microbiota composition was associated with PH. In the current study, the further disparity of the gut microbial profile among PH patients, TR/RAE controls and normal controls suggested that the gut microbial communities possibly changed in physiological adaptations of a chronically hypoxic stimulus. Moreover, we observed a lower richness of gut microbiota in lowland PH patients than in highland PH patients, which might be partially attributed to the distinct gut microbiota composition in individuals living at different altitudes [23]. In line with the previous observations [11], one of the dominant phyla of microbiota was Prevotella in the plateau-living Tibetan individuals, and the plain-living residents exhibited high enrichment of microbes from the phylum Bacteroidetes in this study.

TMA is synthesised by gut microbes and then in liver is converted to TMAO, which promotes vascular inflammation and oxidative stress [24]. Previous studies have revealed that a higher circulating level of TMAO is a risk factor for various cardiovascular events such as hypertension, heart failure and atherosclerosis [25–27]. Alterations in the bacterial communities associated with TMAO metabolism have been linked to the progression of PH among lowlanders [18, 19]. For example, TMA/TMAO-producing bacteria (e.g., Clostridium, Citrobacter and Desulfovibrio) were enriched in PH patients [10], and circulating TMAO was related to severe PH progression [28]. In the current study, bacteria species with homologous genes encoding enzymes involved in the production of TMA were enriched in PH patients compared with controls among lowlanders, but not among highlanders. This is consistent with a previous observation that there is no significant difference in the abundance of TMA-producing bacteria and TMAO circulating levels between highland patients with coronary heart disease and their healthy counterparts [29]. In view of this, we suggest that the microbiota-dependent TMAO mechanism might be more involved in the development of PH in lowlanders, but not among highlanders.

Chronic hypoxic exposure gradually induces pulmonary artery pressure elevation, the structural remodelling of pulmonary vessels and persistent vasoconstriction for permanent high-altitude populations [30]. The hypoxic environment can accelerate the growth of strict anaerobes such as Lactobacillus due to the limited oxygen concentration in the intestinal micro-environment [31, 32]. This might explain why highlanders intrinsically carried the more abundant gut Lactobacillus than lowlanders in the current study. In hypoxia rat models, the relative abundance of Lactobacillus was lower in the PH group than in the control group [33]. The abundance of this bacteria was not obviously different between the combined PH patients and controls in our study but showed a greater disparity between PH patients and controls among highlanders. Recent studies have reported the role of Lactobacillus supplementation in maintaining healthy cardiovascular functions, partially through the hypoxia-induced Lactobacillus-derived metabolites (e.g., lactate) to reduce inflammation and oxidant damage [4, 17]. In addition, in the Yorkshire swine fed with L. plantarum, a decreased expression was observed in the ischaemic myocardial tissue of hypoxia-inducible factor 1α, a crucial regulator of the proliferative and inflammatory responses to hypoxia [34], and an increased expression of endothelial nitric oxide synthase was found [5], which promotes the production of nitric oxide, a potent vasodilator of the pulmonary circulation [35]. Collectively, these observations support Lactobacillus as a potential microbial marker of PH in highlanders, while its mechanism remains to be further clarified.

This study design is unique in that the highlanders were recruited from those who permanently lived on the Tibetan plateau at an altitude of 5070 m, the highest altitude reported in relevant studies, allowing us to reveal the distinct potential microbial etiologies in PH in different altitude environments. Our study has several limitations. First, PH patients in this study were defined by transthoracic echocardiography, not the gold standard, i.e., right heart catheterisation, which is invasive and not available in local highland clinics. As such, PH patients due to heart failure with preserved ejection fraction might be included in the PH patients, especially among an old population with a high prevalence of hypertension. Second, we cannot refer to the causal inference in the alterations in gut microbiota and PH in this observational study. Third, to the best of our knowledge, it is thus far the largest study of gut metagenomes in PH for residence at different altitudes; however, our sample size might be too small to reveal the influences of the demographic factors (e.g., sex, age and body mass index) on the differences in PH gut microbiome pathogenesis and could not support performance of multiple comparisons. Meanwhile, the ROC analysis with only a few cases tended to be overtrained, and therefore affected its real performance in predicting PH. Future population studies with large sample sizes as well as mechanism experiments are warranted to validate our findings.

In summary, this study has reported the altered gut microbiota in PH patients using whole-genome shotgun sequencing data from participants living at different altitudes. Our findings imply that the gut microbial species related to TMA may participate in PH development among lowlanders and gut microbial Lactobacillus is involved in the aetiology of PH among highlanders. These potentially microbiota-dependent mechanisms remain to be further elucidated. Our findings may provide gut microbial targets for the early prevention of PH in lowlanders and highlanders.

Supplementary material

Supplementary Material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material 00617-2022.SUPPLEMENT

Footnotes

  • Provenance: Submitted article, peer reviewed.

  • Author contributions: L.L. Kang and Y. Zheng designed and coordinated the study. L.F. Ma, X. Yang, Z.Y. Zhang, J. Li, D.D Chen and Y.X Dai enrolled participants, provided clinical information and collected samples. W.X. Dong and M.M. Kong performed analyses. W.X. Dong and Q.M. Huang interpreted the data and wrote the first draft of the manuscript. Q.M. Huang, Z.D. Mei, Z.H. Sun, J.J. Zou, D.X. Zhou and Y. Zheng revised the manuscript. All authors reviewed and approved the final manuscript. Data underlying this article will be shared on reasonable request to the corresponding author.

  • Support statement: This work was supported by National Natural Science Foundation of China (U20A20395, 31660307 and 81860329), Science and Technology Department Project of Tibet Autonomous Region (XZ201801-GB-03 and XZ2021JR0003G), and the “Peak Scholars Program” of Xizang Mizu University. Funding information for this article has been deposited with the Crossref Funder Registry.

  • Conflict of interest: None declared.

  • Received November 16, 2022.
  • Accepted February 15, 2023.
http://creativecommons.org/licenses/by-nc/4.0/

This version is distributed under the terms of the Creative Commons Attribution Non-Commercial Licence 4.0. For commercial reproduction rights and permissions contact permissions{at}ersnet.org

References

  1. ↵
    1. Mirrakhimov AE,
    2. Strohl KP
    . High-altitude pulmonary hypertension: an update on disease pathogenesis and management. Open Cardiovasc Med J 2016; 10: 19–27. doi:10.2174/1874192401610010019
    OpenUrl
  2. ↵
    1. Runo JR,
    2. Loyd JE
    . Primary pulmonary hypertension. The Lancet 2003; 361: 1533–1544. doi:10.1016/S0140-6736(03)13167-4
    OpenUrl
  3. ↵
    1. Witkowski M,
    2. Weeks TL,
    3. Hazen SL
    . Gut microbiota and cardiovascular disease. Circ Res 2020; 127: 553–570. doi:10.1161/CIRCRESAHA.120.316242
    OpenUrlCrossRef
  4. ↵
    1. Lee DC,
    2. Sohn HA,
    3. Park Z-Y, et al.
    A lactate-induced response to hypoxia. Cell 2015; 161: 595–609. doi:10.1016/j.cell.2015.03.011
    OpenUrlCrossRefPubMed
  5. ↵
    1. Aboulgheit A,
    2. Karbasiafshar C,
    3. Zhang Z, et al.
    Lactobacillus plantarum probiotic induces Nrf2-mediated antioxidant signaling and eNOS expression resulting in improvement of myocardial diastolic function. Am J Physiol Heart Circ Physiol 2021; 321: H839–H849. doi:10.1152/ajpheart.00278.2021
    OpenUrl
  6. ↵
    1. Li J,
    2. Zhao F,
    3. Wang Y, et al.
    Gut microbiota dysbiosis contributes to the development of hypertension. Microbiome 2017; 5: 14. doi:10.1186/s40168-016-0222-x
    OpenUrlCrossRefPubMed
    1. Li J,
    2. Yang X,
    3. Zhou X, et al.
    The role and mechanism of intestinal flora in blood pressure regulation and hypertension development. Antioxid Redox Signal 2021; 34: 811–830. doi:10.1089/ars.2020.8104
    OpenUrl
    1. Santisteban MM,
    2. Qi Y,
    3. Zubcevic J, et al.
    Hypertension-linked pathophysiological alterations in the Gut. Circ Res 2017; 120: 312–323. doi:10.1161/CIRCRESAHA.116.309006
    OpenUrlAbstract/FREE Full Text
  9. ↵
    1. Taylor WR,
    2. Takemiya K
    . Hypertension opens the flood gates to the Gut microbiota. Circ Res 2017; 120: 249–251. doi:10.1161/CIRCRESAHA.116.310339
    OpenUrlFREE Full Text
  10. ↵
    1. Kim S,
    2. Rigatto K,
    3. Gazzana MB, et al.
    Altered gut microbiome profile in patients with pulmonary arterial hypertension. Hypertension 2020; 75: 1063–1071. doi:10.1161/HYPERTENSIONAHA.119.14294
    OpenUrl
  11. ↵
    1. Li L,
    2. Zhao X
    . Comparative analyses of fecal microbiota in Tibetan and Chinese Han living at low or high altitude by barcoded 454 pyrosequencing. Sci Rep 2015; 5: 14682. doi:10.1038/srep14682
    OpenUrlCrossRefPubMed
  12. ↵
    1. Jia Z,
    2. Zhao X,
    3. Liu X, et al.
    Impacts of the plateau environment on the gut microbiota and blood clinical indexes in Han and Tibetan individuals. mSystems 2020; 5: e00660-19.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. León-Velarde F,
    2. Maggiorini M,
    3. Reeves JT, et al.
    Consensus statement on chronic and subacute high altitude diseases. High Alt Med Biol 2005; 6: 147–157. doi:10.1089/ham.2005.6.147
    OpenUrlCrossRefPubMed
  14. ↵
    1. Narvaez-Guerra O,
    2. Herrera-Enriquez K,
    3. Medina-Lezama J, et al.
    Systemic hypertension at high altitude. Hypertension 2018; 72: 567–578. doi:10.1161/HYPERTENSIONAHA.118.11140
    OpenUrl
  15. ↵
    1. Quast C,
    2. Pruesse E,
    3. Yilmaz P, et al.
    The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 2012; 41: D590–D596. doi:10.1093/nar/gks1219
    OpenUrlPubMed
  16. ↵
    1. Beghini F,
    2. McIver LJ,
    3. Blanco-Míguez A, et al.
    Integrating taxonomic, functional, and strain-level profiling of diverse microbial communities with bioBakery 3. eLife 2021; 10: e65088. doi:10.7554/eLife.65088
    OpenUrlCrossRefPubMed
  17. ↵
    1. Zhao X,
    2. Zhong X,
    3. Liu X, et al.
    Therapeutic and improving function of lactobacilli in the prevention and treatment of cardiovascular-related diseases: a novel perspective from gut microbiota. Front Nutr 2021; 8: 693412. doi:10.3389/fnut.2021.693412
    OpenUrl
  18. ↵
    1. Wedgwood S,
    2. Warford C,
    3. Agvatisiri SR, et al.
    The developing gut-lung axis: postnatal growth restriction, intestinal dysbiosis, and pulmonary hypertension in a rodent model. Pediatr Res 2020; 87: 472–479. doi:10.1038/s41390-019-0578-2
    OpenUrlCrossRef
  19. ↵
    1. Cui Y,
    2. Miao K,
    3. Niyaphorn S, et al.
    Production of gamma-aminobutyric acid from lactic acid bacteria: a systematic review. Int J Mol Sci 2020; 21: 995. doi:10.3390/ijms21030995
    OpenUrlPubMed
  20. ↵
    1. Stiles ME,
    2. Holzapfel WH
    . Lactic acid bacteria of foods and their current taxonomy. Int J Food Microbiol 1997; 36: 1–29. doi:10.1016/S0168-1605(96)01233-0
    OpenUrlCrossRefPubMed
  21. ↵
    1. Oliveira AC,
    2. Richards EM,
    3. Raizada MK
    . Pulmonary hypertension: pathophysiology beyond the lung. Pharmacol Res 2020; 151: 104518. doi:10.1016/j.phrs.2019.104518
    OpenUrl
  22. ↵
    1. Moutsoglou DM,
    2. Tatah J,
    3. Prisco SZ, et al.
    Pulmonary arterial hypertension patients have a proinflammatory gut microbiome and altered circulating microbial metabolites. Am J Respir Crit Care Med 2022; in press [https://doi.10.1164/rccm.202203-0490OC].
  23. ↵
    1. Lan D,
    2. Ji W,
    3. Lin B, et al.
    Correlations between gut microbiota community structures of Tibetans and geography. Sci Rep 2017; 7: 16982. doi:10.1038/s41598-017-17194-4
    OpenUrlCrossRefPubMed
  24. ↵
    1. Li T,
    2. Chen Y,
    3. Gua C, et al.
    Elevated circulating trimethylamine N-oxide levels contribute to endothelial dysfunction in aged rats through vascular inflammation and oxidative stress. Front Physiol 2017; 8: 350. doi:10.3389/fphys.2017.00350
    OpenUrlCrossRef
  25. ↵
    1. Jiang S,
    2. Shui Y,
    3. Cui Y, et al.
    Gut microbiota dependent trimethylamine N-oxide aggravates angiotensin II-induced hypertension. Redox Biol 2021; 46: 102115. doi:10.1016/j.redox.2021.102115
    OpenUrl
    1. Koeth RA,
    2. Wang Z,
    3. Levison BS, et al.
    Intestinal microbiota metabolism of l-carnitine, a nutrient in red meat, promotes atherosclerosis. Nat Med 2013; 19: 576–585. doi:10.1038/nm.3145
    OpenUrlCrossRefPubMed
  27. ↵
    1. Zhang Y,
    2. Wang Y,
    3. Ke B, et al.
    TMAO: how gut microbiota contributes to heart failure. Transl Res 2021; 228: 109–125. doi:10.1016/j.trsl.2020.08.007
    OpenUrlPubMed
  28. ↵
    1. Huang Y,
    2. Lin F,
    3. Tang R, et al.
    Gut microbial metabolite trimethylamine N-oxide aggravates pulmonary hypertension. Am J Respir Cell Mol Biol 2022; 66: 452–460. doi:10.1165/rcmb.2021-0414OC
    OpenUrl
  29. ↵
    1. Liu F,
    2. Fan C,
    3. Zhang L, et al.
    Alterations of gut microbiome in Tibetan patients with coronary heart disease. Front Cell Infect Microbiol 2020; 10: 373. doi:10.3389/fcimb.2020.00373
    OpenUrl
  30. ↵
    1. Sydykov A,
    2. Mamazhakypov A,
    3. Maripov A, et al.
    Pulmonary hypertension in acute and chronic high altitude maladaptation disorders. Int J Environ Res Public Health 2021; 18: 1–33.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Adak A,
    2. Maity C,
    3. Ghosh K, et al.
    Dynamics of predominant microbiota in the human gastrointestinal tract and change in luminal enzymes and immunoglobulin profile during high-altitude adaptation. Folia Microbiol (Praha) 2013; 58: 523–528. doi:10.1007/s12223-013-0241-y
    OpenUrl
  32. ↵
    1. Adak A,
    2. Maity C,
    3. Ghosh K, et al.
    Alteration of predominant gastrointestinal flora and oxidative damage of large intestine under simulated hypobaric hypoxia. Z Gastroenterol 2014; 52: 180–186. doi:10.1055/s-0033-1336007
    OpenUrlCrossRef
  33. ↵
    1. Hong W,
    2. Mo Q,
    3. Wang L, et al.
    Changes in the gut microbiome and metabolome in a rat model of pulmonary arterial hypertension. Bioengineered 2021; 12: 5173–5183. doi:10.1080/21655979.2021.1952365
    OpenUrl
  34. ↵
    1. Wilkins MR,
    2. Ghofrani HA,
    3. Weissmann N, et al.
    Pathophysiology and treatment of high-altitude pulmonary vascular disease. Circulation 2015; 131: 582–590. doi:10.1161/CIRCULATIONAHA.114.006977
    OpenUrlFREE Full Text
  35. ↵
    1. Xu W,
    2. Janocha AJ,
    3. Erzurum SC
    . Metabolism in pulmonary hypertension. Annu Rev Physiol 2021; 83: 551–576. doi:10.1146/annurev-physiol-031620-123956
    OpenUrl
Back to top
Email
Print
Citation Tools
Gut microbiome alterations in pulmonary hypertension in highlanders and lowlanders
Wenxue Dong, Lifeng Ma, Qiumin Huang, Xu Yang, Zhendong Mei, Mengmeng Kong, Zhonghan Sun, Zhiying Zhang, Jing Li, Jiaojiao Zou, Dandan Chen, Yuxiang Dai, Daxin Zhou, Yan Zheng, Longli Kang
ERJ Open Research May 2023, 9 (3) 00617-2022; DOI: 10.1183/23120541.00617-2022
del.icio.us logo Digg logo Reddit logo Technorati logo Twitter logo CiteULike logo Connotea logo Facebook logo Google logo Mendeley logo
Full Text (PDF)

Jump To

Subjects