Bacteroides vulgatus alleviates heart failure via butyric acid-TGF-β1/MAPK pathway

Graphical abstract

Highlights

• •Bacteroides vulgatus improves heart function by regulating SCFA metabolism.
• •Bacteroides vulgatus improves heart failure via the TGF-β1/MAPK pathways.
• •Patients with lower butyric acid have a higher incidence of endpoint events.

Abstract

Introduction

The gut microbiota plays a vital role in the progression of heart failure (HF), making it a potential strategy for treating HF. The Bacteroides genus has shown promising potential for treating HF. However, further research is needed to identify specific beneficial Bacteroides strains for treating HF and elucidate their potential mechanisms.

Objectives

This study aimed to elucidate the therapeutic effect and mechanism of Bacteroides strain on HF.

Methods

This study comprehensively used pharmacological evaluation, 16S rRNA genetic sequencing, short-chain fatty acid (SCFA) targeted metabolomics, transcriptome, and molecular biology methods to investigate the efficacy and mechanism of Bacteroides vulgatus (B. vulgatus) in the treatment of HF.

Results

We observed a significant decrease in the abundance of B. vulgatus in both HF patients and mice, and dysbiosis of the gut microbiota could cause cardiac dysfunction in mice. The administration of B. vulgatus to mice with HF significantly improved their cardiac function and significantly increased the levels of SCFAs, especially butyric acid. Meanwhile, we also observed a significant reduction in the level of butyric acid in the serum of HF patients, and the cardiac function of HF mice improved after the administration of butyric acid. Compared to patients with higher butyric acid level, patients with lower butyric acid level had a significantly increased incidence of endpoint events. Moreover, the transcriptome results revealed that B. vulgatus significantly regulated the mitogen-activated protein kinase (MAPK) pathway. After further molecular biology verification, B. vulgatus was confirmed to regulate the transforming growth factor-β1 (TGF-β1)/MAPK pathway in the heart through butyric acid, thereby exerting anti-HF effects.

Conclusion

This study provides evidence that B. vulgatus improves heart function by regulating SCFA metabolism and subsequently modulating the TGF-β1/MAPK pathway in myocardial tissue, confirming the potential therapeutic role of B. vulgatus in HF.

Introduction

Heart failure (HF) represents the final stage of nearly all cardiovascular conditions, endangering the health of millions globally. In the current framework of chemical drugs and medical technologies, the mortality rate of HF surpasses that of other cardiovascular diseases in the same timeframe [[1], [2], [3]]. The measures taken in clinical practice mainly focus on controlling the development stage of HF and improving the prognosis of HF, preventing the deterioration of HF in a short period of time [4,5]. However, intervention methods that can cure HF are lacking. The occurrence and development process of HF are very complex, and currently the treatment of HF remains challenging and the prognosis is not ideal [[6], [7], [8]]. Therefore, exploring new pathophysiological mechanisms of HF and providing new strategies and therapeutic targets for HF treatment have become the top priorityies in current research.

The gut microbiome is intricately linked to the body's metabolism and contributes to multiple physiological processes, including digestion and absorption. It is crucial for preserving the normal structure and physiological functions of the intestine and for regulating host physiology [[9], [10], [11]]. Moreover, metabolites produced by the gut microbiota, such as trimethylamine oxide (TMAO), short-chain fatty acids (SCFAs), amino acids, and bile acids, can impact the body positively or negatively [[12], [13], [14], [15]]. Accumulating evidence indicates that gut microbiota and its metabolites play important roles in cardiovascular disease [16,17].

SCFAs are metabolic byproducts of bacterial fermentation of carbohydrates in the gut. They mainly refer to organic compounds with 1–6 carbon atoms that can be directly absorbed by the gut and perform physiological functions [18,19]. SCFAs that partially inhibit histone deacetylase (HDAC) activity, such as butyrate, can directly regulate the pathophysiological stages of HF progression, such as cardiac inflammation, fibrosis and energy metabolism disorders, and can increase the function of T regulatory cells to achieve anti-inflammatory effects [20,21].

In this study, we used microbiome, metabolome, transcriptome, and molecular biology methods to explore the mechanisms of HF from the perspective of the gut microbiota. We found that the abundance of B. vulgatus was significantly reduced in HF patients and clarified its mechanism of improving HF by acting on the transforming growth factor-β1 (TGF-β1)/mitogen-activated protein kinase (MAPK) pathway. These results provide new insights for elucidating the physiological mechanisms of HF, as well as new targets for drug development.

Materials and methods

Patient enrollment and sample collection

Upon receiving ethical approval from the Ethics Committee of Anzhen Hospital, affiliated with Capital Medical University (ethical batch number: 2023142X), we conducted the enrollment of the subjects and sample collection in accordance with the principles of subject selection and pertinent human ethics regulations. The study's participants included 90 people, 60 of whom were HF patients, while the remaining 30 were healthy subjects matched by age and sex.

Based on the following criteria, patients and healthy controls were recruited: (1) age > 45 years; (2) for HF patients, a diagnosis of myocardial infarction and/or angina pectoris, New York Heart Association (NYHA) functional class ≥ III, and/or left ventricular ejection fraction (LVEF) < 40 %; (3) for healthy subjects, a normal electrocardiogram, negative coronary angiography, and absence of organic heart disease.

The exclusion criteria were as follows: (1) major intestinal surgery (including cholecystectomy and appendectomy) within the past 5 years; (2) conditions like ulcerative colitis or Crohn’s disease, which are types of inflammatory bowel disease (IBD); (3) acute gastroenteritis; (4) difficult Clostridium difficile infection (recurrence) or Helicobacter pylori infection; (5) persistent or chronic diarrhea; (6) chronic constipation; (7) peptic ulcers; (8) gastric or intestinal polyps; (9) digestive tract tumors; (10) irritable bowel syndrome; (11) acute and chronic cholecystitis, hepatitis; (12) recent use (within the past 3 months) of antibiotics or probiotic supplements; (13) end-stage illness (life expectancy less than 3 months); (14) inability to provide informed consent due to severe cognitive impairment; and (15) participation in other concurrent clinical trials. All study participants signed written informed consent before taking part in the research.

The fasting blood and stool samples were gathered from the study participants. Centrifugation (3000 rpm/min, 10 min) was used to extract the upper serum after the blood had been allowed to stand at room temperature for 1 h. The samples were all kept frozen at −80 °C until they were analyzed. The all-cause mortality rate of patients was tracked for 2 years through prospective follow-up. All patients were followed up regularly by trained professionals at 6, 12, 18, and 24 months after enrollment. Information about adverse prognostic events was obtained from patients or their families through telephone surveys. The endpoint of observation was defined as all-cause mortality.

Animal experiments

Experimental animals and feeding conditions

All animals were procured from the Department of Laboratory Animal Science, Peking University Health Science Center (PUHSC). Animal experiments were carried out in strict compliance with the protocol approved by the Institutional Animal Care and Use Committee of PUHSC under the ethics certificate number LA2022511. Male C57BL/6 wild-type mice, each weighing around 20 g, were kept in SPF animal facilities with controlled temperature and humidity, where they had unrestricted access to food and water.

Mice left anterior descending coronary artery (LADCA) model

A total of ten mice were randomly split into two groups, designated as the control group and the model group. The model group underwent LADCA ligation surgery as described previously and was left for a period of 4 weeks to establish a HF model [22]. In contrast, the control group did not undergo LADCA ligation. All other surgical procedures were standardized and maintained consistently across both groups. Briefly, the mice were anesthetized with isoflurane, and following surgical preparation, their chests were cut open to expose the heart. Surgical sutures were used to precisely ligate the LADCA. Postoperatively, the chest cavity was meticulously closed, and the mice were kept on a heated electric blanket to keep their body temperature stable and ensure they stayed warm until they completely recovered from anesthesia.

Fecal microbiota transplantation (FMT) experiments

In the initial phase of the investigation, the parameters for the FMT method were examined. The antibiotic (ABX) cocktail method is used to deplete the gut microbiota. The ABX cocktail consisted of vancomycin (Psaitong, 1404–90-6, 100 mg/kg), neomycin sulfate (Macklin, 1404–04-2, 400 mg/kg), metronidazole (J&K Scientific, 443–48-1, 400 mg/kg), and ampicillin (Macklin, 69–52-3, 400 mg/kg). A cohort of five mice was assigned to the FHF-FMT group, where they were treated with the ABX cocktail via gavage for a period of 14 days. Following this regimen, these mice received a 200 μL oral dose of a FMT mixture, which had been prepared from the feces of HF mice, over the course of four consecutive weeks. The process for preparing the FMT solution entailed diluting 1 g of a mouse fecal sample with 7.5 mL of sterile phosphate buffered saline (PBS), ensuring thorough mixing. Following dilution, the sample underwent centrifugation (800 × g, 4 °C) for 5 min, after which the precipitate was eliminated. To make a microbial suspension appropriate for FMT, the supernatant was mixed 1:1 with a 40 % (v/v) glycerol PBS solution. The efficacy of both the antibacterial treatment and the FMT procedure was rigorously evaluated, with the outcomes presented in Fig. S1. These results affirm the efficacy of the FMT method employed.

Utilizing an identical protocol, a cohort of five mice was allocated to the FC-FMT group and was administered the ABX cocktail for a duration of 14 days via gavage. Following the initial treatment phase, these mice received a 200 μL oral dose of a FMT mixture, which had been prepared from the feces of control mice, once every four weeks.

Mice supplemented with B. vulgatus

B. vulgatus [ATCC8482] was supplied from the BeNa Culture Collection and cultured in GAM (Hopebio, HB8518-1) at 37 °C under anaerobic conditions. The cultivation protocol for B. vulgatus was as follows: 100 mL of double-distilled water was used to dissolve 4.9 g of GAM. The solution was subsequently sterilized via high temperature and high pressure before being transferred into sterile tubes in an anaerobic incubator. For inoculation, 200 μL of the bacterial mixture was added to each 10 mL portion of GAM. The inoculated tubes were incubated for a period of 72 h until a large amount of bacterial sediment was produced at the bottom of the sterile tube.

The B. vulgatus colonization solution was prepared as follows: the culture was centrifuged at 4 °C and 4000 g for 30 min, after which the supernatant was discarded. To reach a bacterial concentration of 1 × 10⁹ colony-forming units (CFU)/mL for single-bacterial colonization, the pellet was resuspended in sterile PBS after being washed twice with sterile PBS. To obtain inactivated B. vulgatus solution, the colonization solution was sterilized via high-temperature and high-pressure methods.

Twenty mice were allocated into four experimental groups at random: five mice each for the control group, model group, B. vulgatus treatment (BV) group, and inactivated B. vulgatus treatment (ina-BV) group. HF was induced in the model, BV, and ina-BV groups via LADCA ligation. These groups were then administered physiological saline, B. vulgatus solution, or inactivated B. vulgatus solution via oral gavage for a period of 4 weeks. The control group did not undergo LADCA ligation but was subjected to the same surgical procedures and received physiological saline orally for 4 weeks.

Mice supplemented with butyric acid

Three groups, each consisting of five mice, were randomly selected from among the fifteen mice: the control group, model group, and butyric acid treatment group. HF models were established in both the model and butyric acid groups through ligation of the LADCA. The control group did not undergo LADCA ligation but was subjected to identical surgical procedures. For 4 weeks, the butyric acid group received butyric acid (SIGMA-ALDRICH, 203–532-3) in their drinking water at a concentration of 100 mM. In contrast, the model and control group were provided standard drinking water and served as controls.

Mice isoproterenol induced HF model

Four groups, each consisting of five mice, were randomly selected from among the twenty mice: the control group, isoproterenol (ISO) group, B. vulgatus treatment (BV) group, and butyric acid treatment group. The mice in the ISO group, BV group, and butyric acid group were induced to develop HF models by continuous intraperitoneal injection of 15 mg/kg/d of ISO for two weeks. At the same time, physiological saline was administered into the control group. In the following four weeks, the BV group and butyric acid group were treated as described above, while equal amounts of physiological saline were administered as a control to the ISO group and the control group.

Sample collection

The mice were fasted for 12 h and put to death with too much pentobarbital at the endpoint of the experiment. For subsequent analyses, the small intestine, intestinal contents, heart, and blood were all collected and kept at −80 °C.

Echocardiography

Isoflurane was used to deeply anesthetize the mice after the last dosage. Echocardiographic parameters were derived from 2D-guided M-mode short-axis measurements via Vevo 3100 LAZR software (VisualSonics Inc., Canada).

QPCR assay

DNA extraction from fecal samples was conducted via the SteadyPure Stool DNA Extraction Kit (Accurate Biology, China) in accordance with the producer's protocol. The abundance of B. vulgatus DNA was quantified via qPCR with the SYBR Green Premix Pro Taq HS qPCR Kit III (Accurate Biology, China) on an MXPro3005P Cycler (Agilent Technology, Germany). The primers for the detection of B. vulgatus DNA were designed on the basis of the 16S rRNA gene sequences, specifically, the forward primer TCCCCATCGTCTACCGGAAT and the reverse primer ACGTATCCAACCTGCCGTCT. The qPCR amplification cycle was conducted using 30 ng of fecal DNA, with primer concentrations of 200 nM and an annealing temperature of 60 °C.

Generate a standard curve by continuously diluting bacterial DNA extracted from B. vulgatus colonies. The amount of B. vulgatus DNA in the feces was ascertained by comparing the cycle threshold (Ct) values of each sample with the standard curve.

To further confirm the presence of viable B. vulgatus in the intestines of mice following oral gavage, fecal samples were cultured on blood agar plates under anaerobic conditions. After colony growth, qPCR was performed on the colonies, revealing the presence of B. vulgatus DNA, thus confirming the survival of the bacteria in the intestinal environment.

Histological examination

For general morphological assessment, the paraffin-embedded tissues of the heart and small intestine were cut into sections that were 5 μm thick and then histologically examined using standard hematoxylin and eosin (H&E) staining. Paraffin-embedded heart sections were stained using Masson's trichrome and Sirius red to evaluate fibrotic changes. To investigate cell apoptosis, TUNEL and DAPI double staining or TUNEL-DAPI-α-actinin triple staining were performed on tissue sections. In addition, FITC-binding wheat germ agglutinin (WGA) staining was performed to determine the cross-sectional area of the myocardial cells.

Measurement of HF biomarkers

The levels of plasma A-type natriuretic peptide (ANP, E-EL-M0166, Elabscience Biotechnology Co., Ltd., China), B-type natriuretic protein (BNP, EIAR-BNP-1, RayBiotech, USA) and N-terminal-proBNP (NT-proBNP, E-EL-M0834, Elabscience Biotechnology Co., Ltd., China) were measured via enzyme-linked immunosorbent assay (ELISA) following kit instructions.

DNA extraction of microbiota samples and 16S rRNA genetic sequencing

DNA extraction and 16S rRNA genetic sequencing of fecal samples from 60 HF patients and 30 healthy controls were performed by Metware (https://www.metware.cn/) following standard protocols, and the comprehensive procedure is shown in Supplementary Part 1.

SCFAs targeted metabolomics

SCFAs targeted metabolomics of plasma and fecal samples from humans and mice were performed by Metware (https://www.metware.cn/) and Shanghai Applied Protein Technology Co., Ltd., China (https://www.aptbiotech.com/). The detailed method is listed in Supplementary Part 2.

Transcriptome

The RNA libraries were sequenced on the Illumina sequencing platform by Smartgenomics Technology Institute (Tianjin, China). Oligo (dT) magnetic beads were used to enrich mRNA with polyA tails after total RNA was extracted from tissue or cell samples using standard extraction procedures. The NEB Fragmentation Buffer was used to fragment the isolated mRNA in accordance with the manufacturer's instructions. RNA samples were pooled based on their effective concentrations and the intended data output for further Illumina sequencing in order to create libraries. Alignment of the resultant clean reads to the reference genome was performed using the HISAT2 software, facilitating the determination of read localization on the genome. Transcripts from individual samples were assembled with StringTie, followed by the merging of all sample transcripts into a consolidated set. The featureCount was implemented to quantify reads mapping to each gene. The FPKM (Fragments Per Kilobase of transcript per Million mapped reads) values for each gene were computed, considering both the length of the gene and the quantity of reads that are mapped to it. Differentially expressed genes were identified with DESeq2, with a criterion of adjusted p values less than 0.05.

Western blot assay

Protein samples were extracted from lysates of mouse heart tissues. Specifically, heart tissue (20 mg) was homogenized in RIPA buffer (Solarbio, Beijing, China) using ultrasonication. After centrifuging the homogenates for 20 min at 12,000 rpm, the supernatants were gathered. Protein concentrations were determined, and equal amounts of protein were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). The separated proteins were then transferred onto PVDF membranes, which were blocked with 5 % skimmed milk and subsequently incubated sequentially with appropriate primary and secondary antibodies. The following primary antibodies were utilized for Western blot analysis: anti-TGF-β1 (Proteintech, catalog 21898–1-AP), anti-ERK (Proteintech, catalog 11257–1-AP), anti-p-ERK (Proteintech, catalog 28733–1-AP), anti-JNK (Proteintech, catalog 66210–1-Ig), anti-p-JNK (Cell signaling, catalog 4668), anti-p38MAPK (Cell signaling, catalog 9212), anti-p-p38MAPK (Cell signaling, catalog 9216).

Statistical analysis

Principal component analysis (PCA), hierarchical clustering heatmap analysis and α-diversity analysis were performed at https://www.bioinformatics.com.cn (last accessed on 09 Dec 2024), an online platform for data analysis and visualization. An advanced stacked bar chart was generated via the MetWare Cloud (https://cloud.metware.cn), a free online platform for data analysis (last accessed on 09 Dec 2024).

Statistical analyses were conducted via GraphPad Prism 8.0 software (GraphPad, La Jolla, CA, USA). Frequencies (n) and percentages (%) were used to summarize categorical variables, and the chi-square test was used for comparison. Continuous variables and nonnormally distributed variables are presented as the means and standard deviations (means ± SDs) and medians and interquartile ranges (IQRs), respectively. For the comparisons of normally distributed data and nonnormally distributed data, two-tailed Student's t-test, Wilcoxon test, one-way ANOVA and Mann-Whitney U test were used. A false discovery rate (FDR)-calibrated p value < 0.05 was considered significant.

Results

Significant changes in the gut microbiota in HF patients and mice

To determine the relationships between changes in the gut microbiota and the physiological functions of HF, we examined the gut microbiota of fecal samples from 60 HF patients and 30 healthy controls (Table 1). The Shannon and Chao1 indices of α-diversity were significantly different between HF patients and healthy controls, as illustrated in Fig. 1A. Principal coordinate analysis (PCoA) of 16S rRNA high-throughput sequencing data displayed that there existed a notable distinction in the microbiome between HF patients and the healthy controls (Fig. 1B). The bar chart of the gut microbiota structure of HF patients and healthy controls shows an upregulation at the phylum level, with a decrease in the relative abundance of Bacteroidetes in HF patients and an upregulation of the relative abundance of Proteobacteria at the phylum level (Fig. 1C). At the genus level, the relative abundance of Bacteroides, Prevotellaceae, Roseburia and Brevundimonas were decreased, whereas the relative abundance of Pseudomonas, Enterobacteriaceae, Alistipes and Clostridia were increased in HF patients (Fig. 1D). At the genus level, the differences between groups of Bacteroides are most significant (Fig. 1E). At the Bacteroides species level, B. plebeius, B. uniformis, B. ovatus and B. vulgatus showed the highest differences between groups (Fig. 1F), and the qPCR results showed that B. vulgatus was most significantly reduced in HF patients (Fig. S2). Moreover, the abundance of B. vulgatus has the greatest impact on the effect of inter group differences (Fig. 1G). Patients with HF had much lower relative abundances of B. vulgatus in their stools (Fig. 1H). B. vulgatus may therefore be a significant factor in the onset of HF. The gut microbiota of HF model mice differed significantly from that of control mice, and the relative abundance of B. vulgatus was clearly lower in model mouse feces, which is consistent with the findings of HF patients (Fig. S3A-E).

Table 1.

Demographic and clinical characteristics of all subjects.

Parameter (unit)	Healthy controls (n = 30)	HF patients (n = 60)	p
Female sex	10 (33.33 %)	18 (30.00 %)	0.747
BMI (kg/m2)	25.460 (23.7, 27.7)	24.770 (23.2, 25.8)	0.307
ALT (U/L)	16.000 (12.0, 22.5)	16.500 (13.0, 24.8)	0.655
AST (U/L)	18.000 (15.0, 22.0)	17.000 (13.3, 22.0)	0.772
TG (mmol/L)	1.080 (0.9, 1.4)	1.515 (1.1, 2.0)	<0.001
TC (mmol/L)	3.870 (3.6, 4.5)	4.025 (3.5, 4.7)	0.817
HDL-C (mmol/L)	1.235 (1.1, 1.3)	0.960 (0.9, 1.1)	<0.001
LDL-C (mmol/L)	2.130 (1.8, 2.6)	2.260 (1.7, 2.8)	0.784
Urea (mmol/L)	5.340 (4.6, 6.1)	7.245 (5.7, 8.9)	<0.001
Cr (µmol/L)	72.000 (63.3, 81.3)	86.350 (73.9, 98.7)	0.001
UA (µmol/L)	318.500 (282.7, 347.0)	342.000 (270.0, 428.5)	0.107
TSH (mIU/L)	1.624 (0.8, 2.1)	1.983(1.2, 3.0)	0.076
FT3 (pmol/L)	4.800 (4.4, 5.2)	4.310 (4.0, 4.6)	<0.001
FT4 (pmol/L)	10.160 (9.3, 11.4)	11.810 (10.5, 13.9)	0.001
HbA1C (%)	5.500 (5.2, 5.6)	6.900 (6.1, 7.0)	<0.001
GA (g/dL)	36.600 (33.3, 38.0)	16.440 (13.9, 19.0)	<0.001
T4 (nmol/L)	110.510 (104.0, 129.7)	125.000 (109.7, 136.8)	0.030
T3 (nmol/L)	1.410 (1.3, 1.8)	1.240 (1.1, 1.4)	0.003

Mean values (P25, P75) for each parameter (except female sex) are reported. BMI: body mass index; ALT: alanine aminotransferase; AST: aspartate aminotransferase; TG: triglyceride; TC: total cholesterol; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; Cr: creatinine; UA: uric acid; TSH: thyroid stimulating hormone; FT3: Free triiodothyronine; FT4: free thyroxine; HbA1C: glycosylated hemoglobin; GA: glycated albumin; T4: tetraiodothyronine; T3: triiodothyronine.

Fig. 1.

Alterations in heart failure (HF) patients’ gut microbiome. (A) The gut microbiota’s α-diversity as measured by the Chao1 and Shannon indices [n = 30 (Healthy controls) to 60 (HF patients), Wilcoxon test]. (B) Principal coordinate analysis (PCoA) plot of the β-diversity index. (C) The barplot of the gut microbiota structure at the phylum level. (D) The barplot of the gut microbiota structure at the genus level. (E, F) The microbial differences on the genus (E) and species (F) levels (n = 30 to 60, Kruskal-Wallis H test). (G) LEfSe investigation of the structure of the gut microbiota. (H) The relative abundance of B. vulgatus [n = 30 (Healthy controls) to 60 (HF patients), Student’s t test]. Data are shown as mean ± SD. **p < 0.01 vs. healthy controls group.

Dysbiosis of the gut microbiota can lead to cardiac dysfunction in mice

After 14 days of ABX treatment, C57BL/6 mice had their feces sampled for 16S rRNA sequencing to investigate whether gut microbiota dysbiosis might cause cardiac dysfunction in mice (Fig. 2A). The fecal microbiota from control or HF model mice was transplanted into ABX-treated mice through daily oral gavage administration over a 4-week period. The mice treated with fecal microbiota from HF mice (FHF-FMT) and mice treated with fecal microbiota from control mice (FC-FMT) showed substantial changes in their Shannon/Chao1 indices of α-diversity and PCoA (Fig. S4A and B). According to the gut microbiota structure barplots, the FHF-FMT group's relative abundance of Bacteroides was downregulated (Fig. S4C and D).

Fig. 2.

Dysbiosis of the gut microbiota can lead to cardiac dysfunction in mice. (A) Experimental study overview. (B) The levels of BNP and NT-proBNP in the FC-FMT and FHF-FMT groups (n = 5, Student’s t test). (C) Echocardiographic results of the FC-FMT and FHF-FMT groups (n = 5, Student’s t test). (D, E) H&E, Masson and WGA staining of the myocardial tissue in the FC-FMT and FHF-FMT groups (D) and quantitative results (E). (F) Small intestine H&E staining in the control, FC-FMT and FHF-FMT groups. Data are shown as mean ± SD. ^**p < 0.01, vs. control group. ^##p < 0.01, vs. FC-FMT group. FC-FMT: the mice treated with fecal microbiota from control mice; FHF-FMT: the mice treated with fecal microbiota from HF mice.

Compared with those of the control or FC-FMT group, the fractional shortening (FS) and ejection fraction (EF) of the FHF-FMT group were significantly lower, whereas the end-systolic volume (ESV) and end-diastolic volume (EDV) were greater (Fig. 2B), indicating that FMT led to impaired cardiac function in these mice. BNP and NT-proBNP were significantly upregulated in the FHF-FMT group (Fig. 2C). H&E examination of the mice's heart tissue revealed that the FHF-FMT group presented more obvious muscle filament rupture and disordered arrangement of myocardial cells compared with the control or FC-FMT group (Fig. 2D). As shown in Fig. 2, Fig. 2, in the Masson-stained sections from the FHF-FMT group, significant deposition of blue-stained collagen fibers was observed and quantified. The WGA staining results revealed that the myocardial cells in the FHF-FMT group of mice were obviously enlarged and that the number of myocardial cells decreased in the same field of view. Mice in the FHF-FMT group had fragmented and disorganized villi in their small intestines, as demonstrated by H&E staining of the slices (Fig. 2F). These findings suggest that disruption of gut microbial homeostasis directly contributes to impaired cardiac function in mice.

Therapeutic effects of B. vulgatus on HF mice

To determine the effect of B. vulgatus on HF, mice with HF were treated with B. vulgatus by daily oral gavage for 4 weeks. The abundance of B. vulgatus in control mice, HF model mice, BV group, and ina-BV group was determined by qPCR (Fig. S5). These results indicate that treatment with B. vulgatus can restore the reduced levels of B. vulgatus in the fecal samples of mice with HF. The model group exhibited significantly reduced EF and FS compared to the control group, while the B. vulgatus treatment group showed significantly higher EF and FS values (Fig. 3A). The results of HF biomarker analysis revealed that B. vulgatus treatment significantly reduced the ANP and NT-proBNP levels in HF model mice (Fig. 3B). As displayed in Fig. 3C-3E, H&E and Masson staining of heart tissue showed that the administration of B. vulgatus reduced the infiltration of fibroblasts and inflammatory cells in model mice and significantly reduced collagen fiber deposition. The WGA and TUNEL-DAPI-α-actinin triple staining results revealed that the enlarged myocardial cells and decreased cell nuclei in model mice were obviously improved by B. vulgatus treatment (Fig. 3F). The small intestine stained with H&E revealed that B. vulgatus treatment improved the phenomenon of small intestine villus rupture, shedding, and disordered arrangement (Fig. 3G). These results indicate that the colonization of B. vulgatus improves the heart function and intestinal morphology of the model mice.

Fig. 3.

Effects of B. vulgatus on heart failure (HF) in mice. (A) Echocardiographic results of each group of mice (n = 5, one-way ANOVA). (B) HF biomarkers levels in each group (n = 5, one-way ANOVA). (C-F) H&E, Masson, WGA, TUNEL-DAPI and α-actinin triple staining of each group's cardiac tissue (C) and quantitative results (D-F). (G) H&E staining of the small intestine. BV group: B. vulgatus-treated mice; ina-BV group: B. vulgatus treatment-treated mice. Data are shown as mean ± SD. ^##p < 0.01 vs. control group, *p < 0.05, **p < 0.01 vs. model group.

Significant changes in short-chain fatty acid profiles in HF patients and mice

SCFAs, the metabolic byproducts of Bacteroides in the gut microbiota, is crucial in a variety of illnesses [17]. The targeted metabolomics analysis of serum SCFAs in the above healthy controls and HF patients showed significant differences in the profiles of SCFAs (Fig. 4A). The levels of most SCFAs in the serum of HF patients tended to decrease, with butyric acid levels significantly lower than those in healthy controls. The SCFA profiles of HF and control mice also showed notable changes, and the amount of butyric acid in the plasma and feces of HF animals was much lower than in those of control mice, which is consistent with the findings of the clinical samples (Fig. 4, Fig. 4). After colonization with B. vulgatus, the plasma and feces of HF mice have higher concentrations of SCFAs, such as butyric acid and acetic acid (Fig. 4, Fig. 4).

Fig. 4.

Changes in short-chain fatty acid (SCFA) profiles in heart failure (HF) patients and mice. (A) SCFAs levels in the serum of healthy controls and HF patients [n = 30 (Healthy controls) to 60 (HF patients), Wilcoxon test]. (B) SCFAs levels in the plasma of each group of mice (n = 5, Student’s t test). (C) SCFAs levels in the feces of each group of mice (n = 5, Student’s t test). (D) Echocardiographic results of each group of mice (n = 5, one-way ANOVA). (E) HF biomarker levels in each group (n = 5, one-way ANOVA). (F) Correlation investigation between the amount of SCFAs and the gut microbiota (n = 30 to 60, Wilcoxon test). (G, H) H&E, Masson, WGA and DAPI double staining of each group’s cardiac tissue (G) and quantitative results (H). Data are shown as mean ± SD. *p < 0.05, ^**p < 0.01 vs. control group, ^#p < 0.05, ^##p < 0.01 vs. model group. BV: B. vulgatus; BA: butyric acid; IVA: isovaleric acid; HA: hexanoic acid; IBA: isobutyric acid; AA: acetic acid; PA: propionic acid; VA: valeric acid.

We added butyric acid to the drinking water of HF model mice to observe whether butyric acid improved heart function. The mice in the butyric acid supplementation group had considerably higher EFs and FSs than those in the model group (Fig. 4D). ANP and NT-proBNP levels in HF mice were markedly decreased by butyric acid administration (Fig. 4E). The level of B. vulgatus in the gut of HF patients is significantly correlated with the level of butyric acid in plasma (Fig. 4F). Masson staining and H&E of cardiac tissue showed that butyric acid could lessen fibroblast and inflammatory cell infiltration in model animals (Fig. 4, Fig. 4). The WGA and DAPI double staining, and TUNEL and DAPI double staining revealed that the enlarged myocardial cells and decreased cell nuclei in model mice were obviously improved by butyric acid treatment (Fig. 4, Fig. 4). These results indicate that supplementation with exogenous butyric acid can treat HF.

Effects of B. vulgatus on the transcriptomic profile

To further investigate the impacts of B. vulgatus on cardiac function, we performed RNA sequencing (RNA-seq) analysis. The PCA score plot of the transcriptomes of heart tissue samples from the control, model and B. vulgatus treatment groups displayed that the model group was far from the control group and that the B. vulgatus treatment group (BV group) was close to the control group (Fig. 5A). As demonstrated in the visualized volcano map of differentially expressed genes (DEGs) (Fig. S6) and hierarchical cluster analysis (Fig. 5B) of the differentially expressed genes, there were 467 genes levels significantly decreased, whereas 449 genes levels significantly increased in the model group. The levels of 209 genes were significantly restored after the administration of B. vulgatus. The aforementioned DEGs between the B. vulgatus treatment group and the model group were subjected to functional enrichment analyses using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) to elucidate their functions, signaling pathways, and biological processes (Fig. 5, Fig. 5). Metabolic pathway analysis revealed that B. vulgatus could regulate metabolic pathways such as the MAPK, apoptosis, cardiac muscle contraction, cAMP signaling pathway and others. Moreover, the expression level of TGF-β1 was increased in the model mice, and B. vulgatus and butyric acid treatment resulted in a notable decrease (Fig. 5, Fig. 5). To elucidate whether the anti-HF effect of B. vulgatus is related to the TGF-β1/MAPK pathway, the phosphorylation levels of ERK, JNK, and p38 MAPK in left ventricular tissue were detected. As shown in Fig. 5, Fig. 5, the protein expression levels of p-ERK, p-JNK, and p-p38 MAPK in the myocardial tissue of HF model mice were significantly greater than those in the myocardial tissue of control mice. After treatment with B. vulgatus, the protein expression of all these factors decreased, and the same results were observed in the butyric acid treatment group. Therefore, we demonstrate that B. vulgatus may alleviate HF by inhibiting the TGF-β1/MAPK pathway.

Fig. 5.

Effects of B. vulgatus on the transcriptomic profile. (A) PCA score plot of each group (n = 3). (B) Hierarchical cluster analysis. (C, D) GO (C) and KEGG (D) functional enrichment analysis. (E, F) The protein expression levels of TGF-β1, p-ERK, p-JNK, and p-p38 MAPK (n = 4, one-way ANOVA). BV: B. vulgatus. Data are shown as mean ± SD. ^#p < 0.05, ^##p < 0.01 vs. control group, *p < 0.05, **p < 0.01 vs. model group.

Effects of B. vulgatus and butyric acid on the cardiac function of ISO-induced hypertrophic and HF model mice

Here, we further confirmed the cardioprotective effects of B. vulgatus and butyric acid in a hypertrophy and HF mouse model induced by ISO [23]. The echocardiography results are manifested in Fig. 6A. Butyric acid and B. vulgatus markedly raised the EF and FS of ISO-induced HF model mice in comparison to those in the model group (Fig. 6B). H&E, Masson, and WGA staining also revealed that the administration of B. vulgatus and butyric acid reduced the inflammatory infiltration score, collagen fiber deposition, and hypertrophic size in the model mice (Fig. 6, Fig. 6). B. vulgatus and butyric acid significantly reduced the level of NT-proBNP, indicating that B. vulgatus and butyric acid could significantly improve heart function in ISO-induced model mice (Fig. 6E).

Fig. 6.

Effects of B. vulgatus and butyric acid on the cardiac function of ISO-induced hypertrophic and heart failure (HF) model mice. (A, B) Echocardiographic results of each group (n = 5, one-way ANOVA). (C, D) H&E, Masson and WGA staining of each group's cardiac tissue (C) and quantitative results (D). (E) NT-proBNP level in each group (n = 5, one-way ANOVA). BV: B. vulgatus. Data are shown as mean ± SD. ^##p < 0.01 vs. control group, *p < 0.05, **p < 0.01 vs. ISO group.

Predictive function of butyric acid on all-cause mortality rate of patients

Glycated albumin (GA) and high-density lipoprotein cholesterol (HDL-C) were positively correlated with B. vulgatus and butyric acid in HF patients, and most other indicators were negatively correlated with these two substances (Fig. 7A). To further evaluate whether butyric acid was associated with disease outcomes, we prospectively tracked the all-cause mortality rate of patients for 2 years through follow-up contact, with a total of 21 patients dying. According to ROC curve studies, butyric acid’s AUC for endpoint event prediction was 0.701 (95 % CI 0.561–0.858, p < 0.001) (Fig. 7B). When the concentration of butyric acid was 0.015 μg/mL, it could be used as the optimal cutoff point for predicting the primary endpoint, with a sensitivity of 57.1 % and specificity of 17.9 %. The Kaplan–Meier survival analysis results are shown in Fig. 7C. Compared with patients with higher butyric acid levels, patients with lower butyric acid levels had a significantly greater incidence of endpoint events [HR: 19.60 (95 % CI: 17.78–21.42), p = 0.000023].

Fig. 7.

Correlations between butyric acid level and clinical parameters, all-cause mortality in patients. (A) Spearman correlations between B. vulgatus (BV), butyric acid levels and clinical parameters. (B) ROC curve analysis. (C) Kaplan‒Meier survival analysis. BV: B. vulgatus; ALT: alanine aminotransferase; AST: aspartate aminotransferase; TG: triglyceride; TC: total cholesterol; HDL-C: high-density lipoprotein cholesterol; LDL-C: low-density lipoprotein cholesterol; Cr: creatinine; UA: uric acid; TSH: thyroid stimulating hormone; FT3: Free triiodothyronine; FT4: free thyroxine; HbA1C: glycosylated hemoglobin; GA: glycated albumin; T4: tetraiodothyronine; T3: triiodothyronine.

Discussion

HF is the ultimate outcome of various types of cardiovascular diseases and one of the most difficult problems to solve in cardiovascular diseases [24,25]. Patients with HF experience intestinal ischemia, intestinal mucosal damage, and changes in the gut microbiota structure. Through the regulation of host energy utilization, substance metabolism, and other activities, the gut microbiota and its metabolites can contribute to and facilitate the development of HF [[24], [25], [26]]. Therefore, this work investigates the pertinent mechanisms of HF from the standpoint of the gut microbiota, which may help promote new treatment methods and improve current treatment challenges. To our knowledge, this study is the first to determine that B. vulgatus has the ability to improve HF.

HF patients generally exhibit gut microbiota dysbiosis, characterized by an increase in Proteobacteria and Firmicutes, while a decrease in Bacteroidetes [27]. B. vulgatus is a common species of Bacteroides that has potential therapeutic effects on cardiovascular diseases, intestinal inflammation, obesity, and other diseases. B. vulgatus may alleviate atherosclerosis mice’s symptoms by lowering blood endotoxin levels [28]. By producing N-acetylglucosamine, B. vulgatus promotes the growth of Akkermansia muciniphila, thus controlling glucose metabolism and lowering obesity [29]. Through the gut–brain axis, B. vulgatus and 4-HPAA were able to reduce depressive symptoms and ameliorate intestinal inflammation [30]. B. vulgatus and its metabolite 3-HPAA can reduce H3K27 acetylation levels, affecting downstream gene transcription levels and thus inhibiting the progression of MASLD [31]. Moreover, B. vulgatus alleviated lumbar bone loss in ovariectomized rats and reduced colonic microbiota dysbiosis [32]. The results of this study showed that B. vulgatus was significantly reduced in HF patients and mice, while administration of B. vulgatus significantly improved heart function in HF mice.

B. vulgatus can ferment dietary carbohydrates to produce SCFAs, which serve as crucial nutritional and energy sources for the host while playing a pivotal role in maintaining metabolic homeostasis [33]. Previous studies have demonstrated that B. vulgatus significantly increases the levels of bile acids and SCFAs in the feces of hyperlipidemic rats, thereby promoting cholesterol excretion and improving lipid metabolic homeostasis [34]. B. vulgatus enhances intestinal barrier function and effectively suppresses inflammatory responses in LPS-induced acute inflammation mouse models by increasing the concentration of SCFAs [35]. However, there have been no reports on the impact of B. vulgatus on HF through regulating SCFAs. Our study provides the first evidence that B. vulgatus can improve cardiac function by elevating butyric acid levels in HF rats.

SCFAs can be directly absorbed by the intestine and perform physiological functions by regulating energy metabolism, insulin sensitivity, and the cellular immune response [19,36]. SCFAs can serve as key metabolites of the gut microbiota to influence the occurrence and development of HF [37]. According to our research, HF patients’ serum and HF mice’s feces had considerably lower levels of acetic acid, butyric acid, and hexanoic acid. The cardiac function of HF model mice was markedly enhanced by butyric acid treatment, which was in line with the findings of other studies [38]. By encouraging macrophage polarization, enhancing cardiac function during myocardial infarction, and averting ventricular arrhythmia, butyric acid treatment may reduce inflammation and sympathetic remodeling following myocardial infarction [39]. Supplementation with butyric acid or butyric acid-producing bacteria can increase plasma ketone body levels and promote tissue repair after myocardial infarction [40]. In addition, acetate, butyric acid and propionate play a crucial role in atherosclerosis by modulating the generation of Treg cells and inhibiting histone deacetylases [41]. Butyric acid can also inhibit the expression of NOX2 and ROS in endothelial cells through the PPAR δ/miR-181b pathway, which can improve endothelial function and prevent atherosclerosis [42].

In this study, pathological research revealed significant fibrosis in the myocardium of mice with HF. Myocardial fibrosis serves as a critical pathological basis for the development and progression of heart failure, characterized by excessive deposition of extracellular matrix leading to myocardial stiffness and functional impairment [43]. TGF-β1 plays a central role in driving fibrosis by promoting fibroblast activation and collagen synthesis, and the MAPK pathway is an atypical pathway downstream [44,45]. The three subfamilies of MAPKs that potentially take part in myocardial remodeling are p38MAPK, JNK, and ERK. Myocardial fibrosis is accelerated by abnormal activation of ERK1/2 and JNK1/2, which are widely implicated in the proliferation of cardiac fibroblasts and collagen maturation; p38MAPK primarily influences myocardial cell death [[46], [47], [48], [49]]. The activation of the TGF-β1/MAPK pathway is associated with myocardial fibrosis and HF [44]. In this study, B. vulgatus and butyric acid were shown to exert antifibrotic effects through the TGF-β1/MAPK signaling pathway.

Conclusion

In summary, this study demonstrated for the first time that the abundance of B. vulgatus was significantly reduced in HF patients, which inhibited TGF-β1/MAPK signaling pathway by mediating SCFA metabolism, especially butyric acid. This suggests the potential role of B. vulgatus in the development of HF and provides an innovative theoretical basis for the search for new therapeutic drugs.

Compliance with ethics requirements

The ethics application (Nos. 2023142X and LA2022511) was approved by the Ethics Committee of Anzhen Hospital, affiliated with Capital Medical University and the Institutional Animal Care and Use Committee of PUHSC.

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2008 (5). Informed consent was obtained from all patients for being included in the study.

All Institutional and National Guidelines for the care and use of animals (fisheries) were followed.

Funding

This work was supported by the National Key Research and Development Program of China (No. 2022YFC3501600), the Beijing Natural Science Foundation (No. 7252220 and 7252215), the Noncommunicable Chronic Diseases-National Science and Technology Major Project (No. 2024ZD0533200 and 2023ZD0503400), the Tibet Autonomous Region Science & Technology Program (XZ202301YD0008C), the National Natural Science Foundation of China (No. 82030114 and 82104543), and the Open Research Project in State Key Laboratory of Vascular Homeostasis and Remodeling (Peking University, No. 2025-SKLVHR-014).

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: