Gut Microbiome-Targeted Modulations Regulate Metabolic Profiles and Alleviate Altitude-Related Cardiac Hypertrophy in Rats

Abstract

It is well known that humans physiologically or pathologically respond to high altitude, with these responses accompanied by alterations in the gut microbiome. To investigate whether gut microbiota modulation can alleviate high-altitude-related diseases, we administered probiotics, prebiotics, and synbiotics in rat model with altitude-related cardiac impairment after hypobaric hypoxia challenge and observed that all three treatments alleviated cardiac hypertrophy as measured by heart weight-to-body weight ratio and gene expression levels of biomarkers in heart tissue. The disruption of gut microbiota induced by hypobaric hypoxia was also ameliorated, especially for microbes of Ruminococcaceae and Lachnospiraceae families. Metabolome revealed that hypobaric hypoxia significantly altered the plasma short-chain fatty acids (SCFAs), bile acids (BAs), amino acids, neurotransmitters, and free fatty acids, but not the overall fecal SCFAs and BAs. The treatments were able to restore homeostasis of plasma amino acids and neurotransmitters to a certain degree, but not for the other measured metabolites. This study paves the way to further investigate the underlying mechanisms of gut microbiome in high-altitude related diseases and opens opportunity to target gut microbiome for therapeutic purpose. IMPORTANCE Evidence suggests that gut microbiome changes upon hypobaric hypoxia exposure; however, it remains elusive whether this microbiome change is a merely derivational reflection of host physiological alteration, or it synergizes to exacerbate high-altitude diseases. We intervened gut microbiome in the rat model of prolonged hypobaric hypoxia challenge and found that the intervention could alleviate the symptoms of pathological cardiac hypertrophy, gut microbial dysbiosis, and metabolic disruptions of certain metabolites in gut and plasma induced by hypobaric hypoxia. Our study suggests that gut microbiome may be a causative factor for high-altitude-related pathogenesis and a target for therapeutic intervention.

Keywords: 16S rRNA; cardiac hypertrophy; hypobaric hypoxia; metabolome; microbiota.

FIG 1

Gut microbiome-targeted treatments alleviated HH-induced pathological cardiac hypertrophy. (A) Image of rat hearts captured at the end of the experiment. (B, C) Heart breadth (B) and the ratio of heart weight/body weight (C) were measured in each group. (D–K) The mRNA gene expressions of type I collagen (D), type III collagen (E), the ratio of type I collagen/type III collagen (F), atrial natriuretic peptide (ANP) (G), B-type natriuretic peptide (BNP) (H), cardiac myosin heavy chain beta isoform (βMHC) (I), cardiac myosin heavy chain alpha isoform (αMHC) (J), the ratio of αMHC/βMHC (K) at the apex of the heart tissue were detected by quantitative real-time PCR. Data were normalized to the expression of the glyceraldehyde-3-phosphate dehydrogenase (Gapdh) gene. Data are presented as means ± S.D., n = 6 rats/group. The statistical significance was performed by GraphPad Prism 8.0 software using one-way analysis of variance followed by Student's t test between groups. * P < 0.05, ** P < 0.01.

FIG 2

The treatments ameliorated the effect of hypobaric hypoxia on gut microbiota. (A) Principal Coordinate Analysis (PCoA) of unweighted UniFrac distances of all samples. Samples are colored by day. (B) PCoA of unweigted UniFrac on day 28. Samples are colored by different treatments. Cone shape represents NN samples; sphere shape represents HH samples. Samples were clustered by shape rather than color, indicating that hypobaric hypoxia affected the overall microbiome more than treatments. (C) Euclidean distances between NN and HH samples were reduce in three treatments compared to saline group. ** P < 0.01.

FIG 3

Heatmap of the differential abundant microbes affected by hypobaric hypoxia. (A) Linear model (ASV ∼ NN/HH environment + probiotics + prebiotics + synbiotics) was used to estimate the impact of the factors to the abundance of each ASV. The linear model coefficients of NN/HH, probiotics, prebiotics, synbiotics were visualized in the four columns of heatmap. Each row represents a ASV with significant impact of HH environment. Darker color indicates stronger impact of the factor. The heatmap shows that the treatments reversed the effect of hypobaric hypoxia on most of the ASVs to a degree. Asterisk (*) indicates False discovery rate (FDR) < 0.1. (B) Interaction plots of genus Lactococcus, Parabacteroides, Alistipes, Lachnospira, Prevotella and the ratio of Bacteroides to Prevotella. If a treatment line is unparallel to the saline line, it indicates there is interaction effects between the treatment and HH environment. Treatments alleviated HH induced changes in these genera except Lachnospira.

FIG 4

Alterations of fecal SCFA and BA profiles. (A, B) PCoA of Aitchison distances showed age affected the overall composition of fecal SCFAs (A) and BAs (B). Samples are colored by day. Cone shape represents NN samples; sphere shape represents HH samples. (C, D) Euclidean distances between NN and HH samlples in each treatment. Prebiotics reduced the distances in SCFAs (C) and synbiotics reduced the distances in BAs (D). (E) Interaction plots of differential fecal SCFAs. Treatments could not reverse the expression of these SCFAs. **, P < 0.01.

FIG 5

Effect of treatments on plasma metabolome and cytokines. (A, B) Aitchison distances between NN and HH samples was reduced by probiotics treatment in AAs (A) and NTs (B). **, P < 0.01. (C–H) Interaction plot of two-way ANOVA. If a treatment line is unparallel to a treatment line, it indicates there is interaction effects between the treatment and HH environment. The relative abundance of DCA (C), 12-methyltridecanoic acid (D), 13-methylmyristic acid (E), behenic acid (F), eucic acid (G), palmitic acid (H) in HH groups was reversed especially by synbiotics. Dot represents the mean value in each group and error bar represents standard error.

FIG 6

Correlation between the multi-omic profiles. (A) Correlation between microbiome, fecal SCFAs, fecal BA, plasma SCFAs, plasma BAs, plasma AAs, plasma FFAs, plasma NTs, plasma cytokines was analyzed by Mantel test. (B, C) Visualization of Spearman correlation between fecal SCFAs and plasma SCFAs (B), and between fecal BAs and plasma BAs (C). Dark red color indicates strong positive correlation, dark blue indicates strong negative correlation. Asterisk indicates the matched pairs were statistically significantly associated with each other. One star (*) indicates FDR < 0.05, two stars (**) indicate FDR < 0.01.

FIG 7

Pairwise correlation between fecal microbes and fecal and plasma metabolites. Each row is an ASV, each column is a metabolite. Dark red color indicates strong positive correlation, dark blue indicates strong negative correlation. Asterisk indicates the matched pairs were statistically significantly associated with each other. One star (*) indicates FDR < 0.05, two stars (**) indicate FDR < 0.01.