Gut Microbiome in Pulmonary Arterial Hypertension – An Emerging Frontier

1. Preclinical Studies of the Gut-Lung Axis in Pulmonary Hypertension

Multiple studies demonstrate that pulmonary hypertension (PH) rodent models have alterations in the gut microbiome commonly referred to as gut dysbiosis [14,15,16,17,18,19,20,21,22]. Monocrotaline (MCT) PAH rats have increased intestinal permeability as measured by fluorescein isothiocyanate (FITC)-dextran, lipopolysaccharide (LPS), and soluble CD14 in the blood [23] and altered intestinal morphology, with greater muscularis layer thickness and fibrosis and diminished villus length and goblet cell number [20,24]. Sugen hypoxia (SuHx) rats have an elevated fecal Firmicutes-to-Bacteroidetes ratio and less abundance of short chain fatty acid (SCFA)-producing bacteria, such as those that are acetate and butyrate-producing, with no change in lactate-producing bacteria [14]. The milder hypoxic rat PH model also has altered and distinct gut microbiota with more arginine and arginine-producing bacteria, Blautia and Bifidobacterium, and the trimethylamine N-oxide (TMAO) biosynthetic bacteria, Streptococcus [18]. Hypoxic mice have disrupted gut microbiome composition with increases of the genera Prevotella, Oscillospira, and Ruminococcus and decreases in Lactobacillus [25]. Alterations in the gut microbiome also occur in the large animal, bovine brisket disease PH model, with lower total volatile fatty acids and alpha diversity (richness and evenness of bacteria) in rumen fluid [26]. The specific intestinal microbiome differences observed in various PH rodent models are comprehensively reviewed elsewhere [27].

Several preclinical studies have modified the gut microbiome to assess its effects on PAH. Altering the microbiome with a broad-spectrum antibiotic cocktail (ampicillin, vancomycin, neomycin, and metronidazole) prior to SU5416 administration mitigates PH in SuHx rats [28]. Additionally, diet impacts PH pathogenesis and severity. Apolipoprotein E (ApoE) knockout mice given a Paigen (high fat, high cholesterol) diet develop PH [29] compared to ApoE knockout mice fed normal chow. Double ApoE/IL1R1 knockout mice consuming a Paigen diet have even worse PAH, implicating the role of IL-1 in PAH pathogenesis [29]. Western diet increases right ventricular systolic pressure (RVSP) and RV myocardial lipid deposition and reduces RV function in mice [30]. Metformin decreases RVSP and RV lipid and ceramide accumulation [30]. Moreover, high soluble-fiber diet attenuates hypoxia-induced pulmonary vascular remodeling by increasing abundance of SCFA-producing bacteria (Bacteroides, Anaerostripes, and Anaerocolumna), diminishing pro-inflammatory bacteria (Romboutsia, Mammalicoccus, Staphylococcus, Clostridioides, and Streptococcus), and reducing lung interstitial macrophages, dendritic cells, and nonclassical monocytes [31]. The serum metabolites, phosphatidylcholines, lysophosphatidylcholines, ceramides, and hexosylceramides, are lower while propionylcarnitine and probetaine are greater in high soluble-fiber fed mice compared to low soluble-fiber fed hypoxic mice [31]. Interestingly, treatment with a phosphodiesterase-5 inhibitor (tadalafil) or endothelin pathway inhibitor (macitentan) in SuHx rats reduces the plasma levels of many phosphatidylcholines [32]. However, not all suspected advantageous dietary interventions have been effective in PAH. Although intermittent fasting augments RV function and extends survival in MCT rats, there is minimal effect on PH severity [24], suggesting that more targeted dietary interventions may be needed to alter PH severity.

One of the primary ways that the gut microbiota interact with their host is via metabolites which are intermediate or end products of microbial metabolism [33]. The gut metabolites are generated from bacterial metabolism of dietary or other host substrates. Supplementation with the SCFA, butyrate, which is an endogenous histone deacetylase (HDAC) inhibitor, attenuates pulmonary vascular remodeling and accumulation of alveolar (CD68+) and interstitial (CD68+ and CD163+) lung macrophages in hypoxic PH [34]. Butyrate’s positive effects on pulmonary vascular remodeling have been corroborated in other unpublished studies [35,36] that suggest that butyrate also regulates endothelial cell inflammatory activation and migration. However, the improvement in PH with butyrate treatment appears to only be seen in prevention models but not when given after two- and four-weeks of hypoxic exposure [34].

Trimethylamine N-oxide (TMAO), derived from trimethylamine (TMA), is a bacterial metabolite generated by the breakdown of dietary choline, carnitine, and betaine. It is elevated in intermediate and high risk idiopathic PAH (IPAH) patients [37]. Administering TMAO to hypoxic mice worsens PH through macrophage secretion [37]. Treating hypoxic and MCT rodents with a structural analog of choline that inhibits TMAO synthesis, 3,3-dimethyl-1-butanol (DMB), reduces RV systolic pressure and pulmonary vascular thickness/muscularization and suppresses cytokine and chemokine signaling, with the strongest associations with Cxcl6 and Il6 [37,38]. While TMAO worsens PH, long-term TMAO treatment of MCT rats may be beneficial to RV function by preserving fatty acid oxidation and decreasing pyruvate metabolism, thus preserving mitochondrial energy metabolism and mitigating the development of RV dysfunction in PAH [39]. The role of other microbial metabolites [33,39,40,41], such as amino acid metabolites and retinoic and bile acids, in PAH and RV failure is not well-defined.

There are few publications investigating the outcomes of direct manipulation of the microbiome or supplementation of specific bacterial genera/species in PAH. Fecal transplant from angiotensin-converting enzyme 2 (Ace2) overexpressing mice, which have less hypoxia-induced PH, to wild-type hypoxic mice attenuates PH development [25,42]. Administration of Lactobacillus reuteri to postnatal growth restriction pups exposed to hyperoxia mitigates PH severity and RV hypertrophy [43]. Lactobacillus rhamnosus supplementation two weeks after MCT injection does not affect pulmonary vascular remodeling but enhances RV function [44].

Other approaches to modulate the microbiome involve human umbilical cord blood-derived mesenchymal stem cells (MSCs), which are nonhematopoietic cells that can self-renew and secrete anti-bacterial peptides [45]. There is recent interest in their ability to regulate the gut microbiome and alleviate inflammatory bowel diseases [46,47,48]. Treatment with MSCs rebalances the gut microbiome (reducing the disease-associated and increasing the anti-inflammatory bacteria) and attenuates hypoxia [16]- and MCT-induced PH [49,50].

Most of the animal studies modifying the gut microbiota in PH thus far use prevention models, intervening prior to or immediately after providing a stimulus to generate PH. Promising indirect evidence supports a need to further explore the role of the gut microbiome in PH. In SuHx mice, treatment with Ang1-7 four weeks after hypoxia initiation mitigates PH, partially attenuates disease-associated changes in gut microbiota, and enhances the beneficial metabolites, butyric acid and tryptophan [51]. dAdministering irbesartan, an angiotensin II receptor blocker, 30 days after starting exposure to hypoxia mitigates PH, partially normalizes the Firmicutes-to-Bacteroidetes ratio, increases intestinal abundance of Lactobacillaceae and Lachnospiraceae, and decreases Prevotellaceae and Desulfovibrionaceae in high-altitude PH hypobaric hypoxia rats [21]. Future studies need to evaluate the efficacy of altering the gut microbiome after PAH is established. Due to the many potential environmental confounders in clinical microbiome research, animal studies are needed to establish a robust foundation for biotherapeutics targeting host microbiome composition and its systemic effects on respiratory health. Lastly, the potential gut-brain-lung axis in PAH [20,25,42] should be further explored.

2. Clinical Evidence of the Gut-Lung Axis in PAH

Clinical studies have shown that PAH patients have gut dysbiosis or a leaky gut with increased bacterial translocation from the intestinal lumen to systemic circulation [23]. Kim et al. [52] completed one of the initial clinical studies describing the distinct gut microbiome composition in PAH by studying the fecal microbiome of 18 PAH patients and 12 age- and sex-matched healthy controls. Compared to controls, PAH patients have a distinct microbiome composition with lower alpha diversity, fewer bacteria associated with polysaccharide fermentation and SCFA production (Butyrivibrio crossotus, Bacteroides cellulosilyticus, Eubacterium siraeum, Bacteroides vulgatus, Akkermansia muciniphila), and more bacteria associated with the proinflammatory metabolites, TMA and TMAO [52]. PAH patients also have a disparate intestinal virome [52].

In a subsequent single-center pilot study of 20 PAH and 20 healthy controls cohabiting with PAH patients (20 matched pairs), Jose et al. [53] observed no difference in alpha (within a specific sample) or beta diversity (between samples). In the largest PAH microbiota study to date of 72 patients, Moutsoglou et al. [54] demonstrated that the gut microbiome is less diverse in PAH patients compared to healthy controls and family members residing in the same household. Gut microbiome diversity correlates with measures of pulmonary vascular disease (mean pulmonary artery pressure, pulmonary vascular resistance, and pulmonary arterial compliance), but not RV function [54], suggesting that the alterations in the gut microbiome are not due to RV failure and intestinal congestion. PAH patients have reduced abundance of gut bacteria containing genes encoding for the production of anti-inflammatory metabolites, specifically SCFAs (Eubacterium ramulus, Firmicutes sp. coabundance gene 110, Coprococcus comes, Dorea longicatena, Bifidobacterium adolescentis, Gemmiger formicilis, Fusicatenibacter saccharivorans, Eubacterium hallii, Anaerostipes hadrus, Gordonibacter pamelaeae, Ruminococcus torques, Coprococcus catus, Coprococcus eutactus, and Blautia obeum) and secondary bile acids (Collinsella aerofaciens, Coprococcus eutactus, Anaerostipes hadrus, Eubacterium ramulus, Blautia obeum, Eubacterium hallii, Ruminococcus bicirculans, Ruminococcus torques, Eubacterium eligens, Fusicatenibacter saccharivorans, Roseburia faecis, Dorea longicatena, Coprococcus catus, and Roseburia hominis) and increased relative abundance of bacteria with genes encoding for the production of TMAO (Clostridium bolteae, Escherichia coli, and Klebsiella pneumoniae) [54]. In a study of 35 patients with IPAH [37] and another of 124 PAH patients [38], TMAO levels are elevated in higher risk patients, suggesting that TMAO is associated with worse outcomes. The role of the other intestinal microorganisms (fungi, protozoans, and archaea) in clinical PAH is not well-established.

Environmental effects, such as high altitude and hypoxia, may also alter the gut microbiome. A small study of six highlander PH patients living on the Tibetan plateau range and seven lowlander PH patients (residents of Shanghai) showed that while there are overall divergent gut microbial signatures between PH patients and controls, altitude contributes to the gut microbiota differences [55]. TMA-synthesis enzymes are enriched in lowlanders with PH and there is no difference in TMA-producing microbiota between highlander controls and highlander PH patients [55].

Whether there is a causal relationship between the gut microbiome/metabolites and PAH is not yet determined in clinical studies. Mendelian randomization of data from the MiBioGen consortium, the largest genome-wide meta-analysis of intestinal microbiota [56], investigated the potential direct link between gut microbiota, metabolites, diet, and PAH [57]. The bacteria, Alistipes and Victivallis, correlate with increased PAH risk while Coprobacter, Erysipelotrichaeae, Lachnospiraceae, and Ruminococcaceae protect against PAH [57]. However, SCFAs, TMAO, and dietary patterns were not causally associated with PAH in the Mendelian randomization analysis [57]. In another recent Mendelian randomization study, Su et al. identified 11 gut microbial taxa, including Bifidobacteriaceae, Eubacterium eligens group, and Sutterella, and 24 bacterial metabolites that are linked to PAH pathogenesis by regulating the expression of ITPR2, IDE, NRIP1, and IGF1 genes in lung tissue [58]. Limitations of these studies are that the design only evaluates the impact of genetic factors on intestinal microbiota abundance and the development of PAH, small effect size with the use of single nucleotide polymorphisms (SNPs) for metabolites and bacteria, Mendelian randomization being prone to false positives, and cohorts primarily of European ethnicity, restricting the generalizability of the findings. Thus, more clinical studies are needed to determine whether there is a casual connection between the intestinal microbiota/metabolites and PAH.

There are now emerging data that not only is the fecal microbiome changed in PAH, but airway microbiome composition may also be distinct. In a study of PH patients of various etiologies (Group 1 PAH, PH due to lung disease, and chronic thromboembolic PH), Zhang et al. [59] observed higher alpha diversity (Ace and Sobs indices and Simpson index) and increased Streptococcus, Lautropia, and Ralstonia in the airways of PH patients compared to reference controls. Intratracheal instillation of Streptococcus induced PH in rats [60]. A recent study identified disparate airway mycobiomes or fungal compositions between PH and healthy controls [61]. Further research should assess the microbiome/mycobiome in other parts of the body.

3. Potential Approaches to Modulate the Gut-Lung Axis to Treat PAH

There are several different strategies to restructure the microbiome including diet/prebiotics, probiotics, postbiotics [62] (inanimate microorganisms or their components), microbiota transplant, medications, vaccines [63], exercise [64], and mesenchymal stromal cell therapy [50,65] (Figure 2). There have not yet been many clinical trials investigating how altering the gut microbiome affects PAH. There is an ongoing study evaluating the impact of microbiota transplant from healthy controls to PAH patients [66]. Unfortunately, the SARS-CoV-2 pandemic impeded the enrollment and initiation of the trial assessing the use of chlorhexidine mouthwash and oral nitrate therapy in PH patients (NCT03787082). Challenges to translating preclinical findings to PAH patients include genetic factors, age, sex, concomitant chronic diseases, and different environmental exposures [67] (medications, diet, timing of eating, toxin use, chemicals in environment/products, hygiene, air pollution, daylight exposure, etc.). Fetal/maternal or perinatal microbiota exposures may also complicate the efficacy of microbiota clinical trials. For example, supplementing omega-3 polyunsaturated fatty acids in the diet of pregnant rats improves RV systolic pressure and survival in pups exposed to hyperoxia at time of birth [68]. Additionally, the timing and duration of gut microbiome modulation needed to confer benefits in PAH are unknown. Despite these possible confounders, the vast potential health benefits of altering the microbiome in PAH should be explored to a greater extent.

4. Contribution of Infections in PAH Pathogenesis

Infectious agents, including bacteria, viruses, fungi, and parasites, can cause pulmonary arterial injury and inflammation by impacting vascular cells, leading to severe pulmonary vascular remodeling and PAH [69]. The pathophysiology and mechanisms by which infections lead to pulmonary vascular disease are more extensively reviewed elsewhere [70]. Infections in the setting of a dysregulated immune system may elevate susceptibility to PH development [71,72]. There is now growing interest in determining how host microbiota promote or resist infections [73].

A common global cause of PAH is infection by the intravascular parasite, Schistosoma mansoni. Schistosomiasis affects over 200 million people with about 1-10 million chronically infected people at risk for developing PAH [74,75]. Schistosomiasis cases occur in many regions worldwide, but the majority are observed in Africa and Asia, although longstanding epidemiological and socioeconomic challenges may underestimate its global impact. Schistosomiasis disrupts both gut and lung microbiota [76]. S. mansoni egg exposure decreases lung alpha diversity, mainly by impacting the relative abundance of the phylum, Ascomycota, while the pulmonary Firmicutes-to-Bacteroidetes ratio remains unchanged [76]. In contrast, Schistosomiasis increases gut microbiota alpha diversity and the Firmicutes-to-Bacteroidetes ratio [76], suggesting significant differences in lung and gut microbiome responses to Schistosomiasis infection.

HIV infection is also a well-recognized cause of PAH [77]. Host factors and geography contribute to the fecal microbiota disruptions that occur after HIV infection [78]. Geographic location has a greater effect on fecal microbiota composition than HIV infection status [78]. Interestingly, while HIV infection is known to disrupt gut epithelial barrier function, there are regional differences in immune activation with elevated soluble CD14 levels in HIV-infected individuals from all three regions studied (United States, Botswana, and Uganda), but there is increased intestinal fatty-acid binding protein in HIV-infected individuals from only the United States and Botswana [78]. Thus, distinct gut microbial alterations due to host region are major confounders in microbiota studies.

Unquestionably, additional work is needed to unravel the precise mechanisms contributing to the complex interactions between infections and host microbiota. Future studies should further ascertain whether microbiome alterations after infection impact susceptibility to PAH.

5. Conclusions

Animal and clinical studies reveal altered intestinal microbiota composition in PAH. Modulating the gut microbiome diminishes systemic inflammation and pulmonary immune cell infiltration. Additional research is needed to delineate whether restructuring the microbiome attenuates PAH after PAH has developed and explore the role of the gut and airway/lung microbiome in infection-associated PAH.