Gut microbiota dysbiosis exacerbates heart failure by the LPS-TLR4/NF-κB signalling axis: mechanistic insights and therapeutic potential of TLR4 inhibition

Thirty SPF-grade male Sprague‒Dawley rats (weight: 200 ± 20 g) were purchased from the Qinglongshan Animal Breeding Center (Jiangning District, Nanjing, Jiangsu Province; animal production licence no. SCXK(Zhe)2019-0002). All the rats were housed under standard conditions (temperature: 22 ± 2 °C; humidity: 50 ± 10%; 12-hour light/dark cycle) with free access to food and water. A seven-day acclimatization period was provided prior to the experiments. The experimental protocol was approved by the Animal Ethics Committee of the Second Affiliated Hospital of Wannan Medical College (approval no. WYEFYLS 202012). All procedures conformed to the guidelines from Directive 2010/63/EU of the European Parliament on the protection of animals used for scientific purposes.

The drugs and reagents used were as follows: isoprenaline hydrochloride (ISO, MedChemExpress, cat. no. HY-B0468); TAK-242 (TLR4 inhibitor, MedChemExpress, cat. no. HY-11109); ELISA kits for serum brain natriuretic peptide (BNP), LPS, IL-1β, IL-6, IL-17, and TNF-α (Shanghai Weiao Biotechnology Co., Ltd., cat. nos. VA-EL-R0012, VA-EL-R0045, etc.); An HE staining kit and Masson’s trichrome staining kit (Servicebio, Wuhan, China; cat. nos. G1005, G1345).

The instruments used were as follows: a small-animal ultrasound imaging system (Vevo 3100, VisualSonics, Canada); an optical microscope (Nikon Eclipse E100, Japan); electrophoresis and transfer systems (Bio-Rad, USA); and a chemiluminescence imaging system (Tanon 5200, China).

A total of 30 rats were selected, of which 24 were used to model heart failure and 6 were used as controls. The experimental rats were acclimatized to the environment in which they were housed for 7 days, and CHF model rats were generated by the intraperitoneal injection of ISO. Prior to the start of the study, we referred to the literature [22, 23] on the ISO-induced heart failure model in rats and chose a high dose (10 mg/kg/d) of ISO [24] administered intraperitoneally; however, the rats appeared to be curled up in the corner of the cage, walked unsteadily, and often died minutes after ISO was injected. Thus, we chose to use a low dose of ISO (3 mg/kg/d), which was administered for 14 days. During the administration of ISO, the rats were closely observed for changes in mental status, exercise, food intake and fur glossiness. The mortality rate of the rats was 50% (12/24) after the first injection of ISO, mostly within 15 min. Surviving rats gradually exhibited depression, decreased activity, decreased feeding, and yellowing of fur, and some rats developed diarrhoea. Moreover, 6 rats in the control group were intraperitoneally injected with equal volumes of normal saline for 14 days. After the last injection was completed, the following cardiac function indices were measured using small-animal ultrasound: left ventricular ejection fraction (LVEF), left ventricular short-axis shortening rate (LVFS), left ventricular end-systolic diameter (LVESD), and left ventricular end-diastolic diameter (LVEDD), followed by tail vein blood sampling to measure the serum BNP concentration. Compared with that in the control group, the LVEF was significantly lower, the serum BNP level was significantly greater, indicating successful modelling.

The 12 surviving rats with heart failure were randomly divided into a heart failure group (n = 6) and a TAK-242 group (n = 6). The TAK-242 group was given intraperitoneal injections of TAK-242 (2 mg/kg/day) [24, 25] for 10 days. Rats in the heart failure group (n = 6) and the control group (n = 6) were injected with an equal volume of saline for 10 days. Compared with the predose period, after 10 days of injections, rats in the TAK-242 group showed significant improvement in mental status, activity, and feeding; the fur became shiny, and the diarrhoea ceased. On the other hand, rats in the heart failure group exhibited further deterioration in their mental state, reduced food intake, and worse diarrhoea and huddled in the corner of the feeding cage, rarely moving. No deaths occurred in the three groups of rats during this period. We re-measured the following cardiac function indicators using ultrasound in three groups of rats: LVEF, LVFS, LVESD, LVEDD, and serum BNP levels. Figure 1 showed the schedule of the rat heart failure model (ISO 3 mg/kg/d) and TAK-242 (2 mg/kg/d) intervention.

The numerical data for each group are expressed as the mean ± standard deviation (x̄±s). Statistical analysis of the experimental data was performed using SPSS 26.0 software. Significance testing between two groups was conducted using the t test, and comparisons between multiple groups were performed using one-way analysis of variance. P < 0.05 was considered statistically significant.

During the preparation of the rat HF model, deaths occurred primarily within 15 min after isoproterenol (ISO) administration, likely due to malignant arrhythmias caused by the ISO concentration and injection speed. The survival rate of the rats during model preparation was approximately 50%, whereas no deaths occurred in the control group. The day after the end of the 14-day period of ISO dosing, 6 out of the 12 surviving rats were randomly selected for echocardiography, and plasma BNP levels were measured in all surviving rats. Compared with control rats, ISO-induced HF rats presented significantly decreased LVEF and LVFS and increased LVESD, LVEDD, and BNP levels (Fig. 2). The differences were statistically significant (P < 0.05), indicating successful establishment of the HF model.

The low survival rate (50%) during model preparation may be attributed to the ISO concentration. Future experiments should reduce the ISO concentration to improve survival rates.

The gut microbiota of the rats was analysed using 16 S rRNA sequencing. Venn diagrams were used to visualize the shared operational taxonomic units (OTUs) and average OTU numbers among the groups (the TAK-242 group, HF group, and control group) (Fig. 3A). Principal coordinate analysis (PCoA) and nonmetric multidimensional scaling (NMDS) revealed significant differences in the gut microbial composition between the HF and control groups. After 10 days of TAK-242 treatment, the gut microbiota profile of HF rats showed marked improvement (Fig. 3B, C). At the phylum level, HF rats presented increased abundance of Bacteroidota and Spirochaetota but decreased abundance of Actinobacteriota and Proteobacteria. TAK-242 treatment reversed these changes (Fig. 3D, E). At the family level, Muribaculaceae, Peptostreptococcaceae, Prevotellaceae and Ruminococcaccac were enriched in the HF group, accompanied by reduced microbial diversity. TAK-242 treatment also reversed these alterations (Fig. 3F, G). Linear discriminant analysis (LDA) effect size (LEfSe) analysis revealed increased abundance of Muribaculaceae, Bacteroidota, and Ruminococcaceae in the HF group (Fig. 3I), and these increases were significantly reversed by TAK-242. Additionally, TAK-242 increased the abundance of Clostridia and Oscillospiraceae (Fig. 3H-I). KEGG pathway analysis revealed significant differences between the TAK-242 and HF groups in terms of “cell motility,” “signal transduction,” and “metabolism of other amino acids” (Fig. 3J).

These alterations in the gut microbiota may exacerbate myocardial injury by promoting the release of inflammatory factors. Prior studies have suggested a close link between gut dysbiosis, inflammation, and myocardial damage [12, 13]. Thus, TAK-242 likely improves cardiac function by restoring the gut microbiota balance and attenuating inflammation.

After 10 days of TAK-242 treatment, cardiac structure and function were improved in rats. Compared with the HF group, the TAK-242-treated group presented increased LVEF and LVFS (Fig. 4A, B) and decreased LVESD, LVEDD, and BNP levels (Fig. 4C, D, E), indicating improved cardiac function. These results suggest that improved cardiac function is correlated with gut microbiota restoration.

Compared with control rats, HF rats presented significantly elevated serum levels of IL-1β, IL-17, IL-6, TNF-α, and LPS. After TAK-242 treatment, the LPS and inflammatory cytokine levels decreased, and cardiac function improved (Fig. 5A, B). HE and Masson staining (Fig. 5C) revealed orderly myocardial arrangement, clear nuclei, and distinct striations in control rats. In contrast, HF rats presented disordered myocardial alignment, tissue swelling and hypertrophy, and lymphocyte infiltration, which were ameliorated by TAK-242 treatment.

Western blot analysis revealed significantly increased TLR4, P-IKBα/ IKBα and P-p65/p65 in HF rats compared with controls. TAK-242 treatment suppressed TLR4/NF-κB pathway activation, reduced TLR4 and P-p65/p65 expression. The expression of P-P65 in the nucleus was also significantly reduced after TAK-242 treatment (Fig. 6A-F).

The ISO hydrochloride-induced CHF rat model is a classical animal model for studying HF [28]. This study revealed significant gut microbiota restructuring in rats with ISO-induced HF. Specifically, the abundance of Bacteroidota (+ 38.7%) and Spirochaetota (+ 25.4%) increased, whereas that of Actinobacteriota (− 42.1%) and Proteobacteria (− 31.6%) decreased (Fig. 3D-E). Concurrently, the Shannon diversity index decreased by 1.5-fold (p < 0.01). This microbiota imbalance resulted in a marked increase in the serum LPS level to 2.33 ± 0.53 ng/mL (vs. 0.98 ± 0.29 ng/mL in controls, p < 0.001), thereby activating the TLR4/NF-κB signalling pathway (TLR4 protein expression increased by 3.2-fold and p65 phosphorylation by 4.1-fold, p < 0.001) and driving excessive release of the proinflammatory cytokines IL-1β, IL-6, IL-17, and TNF-α (Fig. 5A-B).

Histopathological analysis further confirmed that the myocardial fibrosis area in the model group increased to 28.5 ± 3.7% (vs. 8.2 ± 1.5% in the control group, p < 0.001), and the LVEF significantly decreased to 53.6 ± 2.5% (vs. 81.3 ± 5.5% in the control group, p < 0.001), validating the pathological gut‒heart axis association (Fig. 5A and C).

Mechanistic investigations demonstrated that intervention with the TLR4 inhibitor TAK-242 not only suppressed p65 phosphorylation (− 73.2%, p < 0.001) and IκBα expression (2.1-fold, p < 0.01) but also reversed microbiota dysbiosis (Bacteroidota abundance was reduced to baseline ± 5.2%, p < 0.05), suggesting bidirectional regulation between TLR4 signalling and microbiota homeostasis (Fig. 3H-I). This finding transcends the limitations of traditional exogenous LPS stimulation models, revealing a novel mechanism by which endogenous microbiota metabolic disorders amplify cardiac inflammation by the LPS-TLR4/NF-κB cascade. Notably, TAK-242 treatment significantly improved cardiac function (LVEF increased to 61.9 ± 3.1%, p < 0.001) and reduced the fibrosis area to 12.1 ± 2.4% (p < 0.001). Its dual effects (inhibiting inflammation and modulating the microbiota) provide a theoretical basis for targeting the gut microenvironment in treating HF.

Recent studies have increasingly indicated that chronic immune inflammation induces cardiomyocyte apoptosis and myocardial fibrosis, leading to impaired myocardial contraction and diastolic dysfunction, thereby triggering a cascade of HF symptoms [29, 30]. TLR4, a transmembrane protein in the TLR family, is highly expressed during cardiac injury and serves as a critical mediator of cardiac inflammation in cardiovascular disease progression [31]. Early studies revealed that TLR4 expression increases in cardiomyocytes during myocardial infarction-induced HF, promoting inflammatory responses and exacerbating HF [32]. In mice with LPS-induced sepsis, myocardial TLR4 and JNK protein expression, plasma TNF-α and cTnI levels, and cardiac dysfunction are elevated. Inhibiting TLR4 activation reduces JNK protein expression in cardiomyocytes and serum TNF-α levels, alleviating myocardial injury and improving cardiac function during sepsis [25]. Furthermore, TLR4/NF-κB pathway inhibition also ameliorates sepsis-induced cardiac dysfunction in rats [33].

Significant differences exist in the gut microbiota composition between HF patients and healthy controls. Studies have reported greater colonization of pathogenic bacteria in the intestines of CHF patients, with 78.3% exhibiting increased intestinal permeability. Additionally, patients with moderate-to-severe congestive HF have greater intestinal permeability than those with mild congestion [34]. Heart failure induces intestinal barrier dysfunction, promoting LPS translocation into the bloodstream and increasing TLR4 expression [35]. In TLR4-knockout mice, high-fat diet-induced metabolic abnormalities—including impaired myocardial contraction, intracellular Ca2 + deficits, reactive oxygen species (ROS) accumulation, mitochondrial damage, inflammation, and autophagy—are mitigated, indicating that TLR4 knockout confers partial protection against high-fat diet-induced cardiac remodelling and systolic/diastolic dysfunction [36].

Human intestinal epithelial cells are constantly exposed to vast bacterial populations and high concentrations of microbial metabolites. The intestinal mucosal barrier constitutes the first line of defence for the innate immune system against microbial invasion. Heart failure causes intestinal congestion, leading to microbiota dysmetabolism and bacterial translocation, resulting in a “leaky gut” phenomenon [37, 38]. Most relevant studies have employed exogenous high-dose LPS injections in rats to induce myocardial injury [39, 40]. The present study revealed that HF rats exhibited gut microbial dysbiosis at the phylum level, characterized by elevated abundance of Bacteroidota and Spirochaetota, which directly correlated with increased levels of proinflammatory mediators, including LPS, interleukin-1β (IL-1β), IL-6, IL-17, and TNF-α. TAK-242 intervention significantly reduced the relative abundance of Bacteroidota and Spirochaetota, concomitant with marked decreases in circulating LPS, IL-1β, IL-6, IL-17, and TNF-α levels, ultimately ameliorating cardiac dysfunction. Moreover, TAK-242 treatment induced genus-level microbial restructuring in HF rats, as evidenced by decreases in the abundance of Muribaculaceae, Peptostreptococcaceae, Prevotellaceae, and Ruminococcaceae, thereby rectifying HF-associated dysbiosis. Histopathological analysis revealed disordered myocardial arrangement, tissue thickening, swelling, and pronounced lymphocyte infiltration. However, treatment with the TLR4 antagonist TAK-242 markedly reduced LPS levels, decreased inflammatory cytokines, improved cardiac function, and alleviated myocardial disarray, tissue swelling, and lymphocyte infiltration. These findings suggest that alterations in the composition of the gut microbiota and changes in the concentration of the metabolite LPS in rats with HF cause changes in the levels of inflammatory factors in the blood, which contribute to altered cardiac function in these rats. However, a healthy intestinal flora is in dynamic balance, and Bacteroidota is a double-edged sword in the human intestinal system; Enterobacteriaceae can influence host growth by facilitating food digestion and nutrient acquisition [41]. Bacteroidota are the most abundant gram-negative bacilli in the human gut, and although they play an irreplaceable role in the intestinal flora between hosts, some anaplastic bacilli may be opportunistic pathogens [42, 43]. Muribaculaceae and Prevotellaceae have been identified as direct contributors to increased LPS levels [44]. Although Peptostreptococcaceae lacks intrinsic LPS-producing capacity, its overproliferation may indirectly promote LPS accumulation via disruption of intestinal barrier integrity or facilitation of proteobacterial proliferation [45, 46].

Notably, LPS, a major component of the outer wall of gram-negative bacterial cell walls, is a substance composed of lipids and polysaccharides and has been shown to be associated with systemic inflammation and chronic disease [47].

TAK-242, a small-molecule TLR4 inhibitor, is widely used to suppress TLR4/NF-κB pathway activation in studies of inflammation, cancer, and autoimmune diseases [48–50]. This study demonstrated that TAK-242 administration reduced serum LPS and inflammatory cytokine concentrations, reversed microbiota dysbiosis in HF rats, downregulated TLR4, P-IKBα and P-p65 protein expression in the TLR4/NF-κB pathway. Previous studies have indicated that NF-κB proteins in most resting cells remain inactive in the cytoplasm by binding to inhibitory IκB proteins. Upon pathway activation, IκB is degraded, releasing NF-κB into the nucleus to mediate transcriptional activation [51]. This study revealed that gut microbiota dysbiosis increases LPS production, enhances TLR4 binding, activates the TLR4/NF-κB pathway, and elevates IL-1β, IL-17, IL-6, and TNF-α levels, thereby inducing and aggravating HF in rats. TAK-242 improves gut microbiota dysbiosis and enhances cardiac function by blocking the TLR4/NF-κB pathway.

Despite these significant findings, this study has several limitations: (1) the animal model does not fully replicate the progression of chronic HF in humans; (2) specific metabolic pathways underlying microbiota‒host interactions require further elucidation via metabolomics; and (3) whether TAK-242-mediated microbiota reprogramming operates independently of TLR4 inhibition remains to be validated. Future studies will integrate faecal microbiota transplantation and single-cell transcriptomics to delineate the molecular mechanisms by which specific bacterial genera (e.g., Bacteroidota) regulate cardiomyocyte pyroptosis via metabolites.