Article Text
PDF
Abstract
Objective The pathogenesis of sepsis is complex, and the sepsis-induced systemic proinflammatory phase is one of the key drivers of organ failure and consequent mortality. Akkermansia muciniphila (AKK) is recognised as a functional probiotic strain that exerts beneficial effects on the progression of many diseases; however, whether AKK participates in sepsis pathogenesis is still unclear. Here, we evaluated the potential contribution of AKK to lethal sepsis development.
Design Relative abundance of gut microbial AKK in septic patients was evaluated. Cecal ligation and puncture (CLP) surgery and lipopolysaccharide (LPS) injection were employed to establish sepsis in mice. Non-targeted and targeted metabolomics analysis were used for metabolites analysis.
Results We first found that the relative abundance of gut microbial AKK in septic patients was significantly reduced compared with that in non-septic controls. Live AKK supplementation, as well as supplementation with its culture supernatant, remarkably reduced sepsis-induced mortality in sepsis models. Metabolomics analysis and germ-free mouse validation experiments revealed that live AKK was able to generate a novel tripeptide Arg-Lys-His (RKH). RKH exerted protective effects against sepsis-induced death and organ damage. Furthermore, RKH markedly reduced sepsis-induced inflammatory cell activation and proinflammatory factor overproduction. A mechanistic study revealed that RKH could directly bind to Toll-like receptor 4 (TLR4) and block TLR4 signal transduction in immune cells. Finally, we validated the preventive effects of RKH against sepsis-induced systemic inflammation and organ damage in a piglet model.
Conclusion We revealed that a novel tripeptide, RKH, derived from live AKK, may act as a novel endogenous antagonist for TLR4. RKH may serve as a novel potential therapeutic approach to combat lethal sepsis after successfully translating its efficacy into clinical practice.
- sepsis
- probiotics
- macrophages
- inflammation
- intestinal microbiology
Data availability statement
Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information. Transcriptome data and 2bRAD sequencing for the Microbiome of mice are available in the Genome Sequence Archive (GSA) database: Bioproject PRJCA017617 and PRJCA017573. Raw data not included therein can be obtained with the consent of the corresponding author.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
The pathogenesis of sepsis is closely associated with the gut microbiome, and the therapeutic approach for the treatment of sepsis in the clinic is still limited.
Akkermansia muciniphila (AKK) is recognised as a next-generation probiotic that showed beneficial effects on many disease progressions, however, its modulatory role in sepsis development is still unknown.
WHAT THIS STUDY ADDS
This study uncovered that intestinal AKK participates in the lethal sepsis progression.
Live AKK is able to generate a novel tripeptide Arg-Lys-His (RKH), RKH protects against lethal sepsis in both murine and piglet models.
RKH directly binds to the Toll-like receptor 4 receptor and inhibits systemic inflammation during sepsis.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
This study identified RKH as a novel potential therapeutic approach to combat sepsis-induced systemic inflammation and lethality. However, it still faces many challenges in translating into clinical practice.
Introduction
Sepsis is defined as a life-threatening organ dysfunction caused by a dysregulated host response to an infection.1 2 A recent Global Burden of Disease study (2017) presented that sepsis accounts for a significant medical burden worldwide, with almost 20% of global reported deaths could be attributed to sepsis.3 A dysregulated immunological response is a common pathophysiological feature of sepsis, which may lead to the syndrome of persistent critical illness, characterised by cytokine storm, acquired immunosuppression, metabolic disorders, multiple organ failure and poor long-term functional outcomes.4–8 In the past decades, many clinical trials of treatments aiming to modulate dysregulated systemic inflammation have failed to improve outcomes, partly due to the tremendous complexity and heterogeneity of the individual immune response.9 Therefore, there is a strong need for new therapies that can improve outcomes for a subset of sepsis patients that benefit from immunomodulatory treatment.
Pattern recognition receptors (PRRs) located in immune cells are responsible for inflammatory signal transduction in response to pathogen-associated molecular patterns.10 Toll-like receptors (TLRs) are the major functional PRRs during sepsis, and strategies that restrict TLRs activation in progression of sepsis are recognised as effective approaches to alleviate sepsis-induced organ damage and death.11–14
Akkermansia muciniphila (AKK) is an anaerobic, Gram-negative bacterium with promising probiotic activities against multiple diseases involved with immune dysregulation and metabolic disorders.15 16 For example, AKK could induce intestinal adaptive immune responses during homoeostasis17 and generate bioactive molecules that can modulate remote functional cell activity and participate in the intervention of pathological development.18 19 Meanwhile, AKK was demonstrated to improve metabolic disorders such as obesity and diabetes in both humans and mice.20 21 However, its impact on sepsis development and the underlying mechanisms are still unknown.
This study demonstrates a novel mechanism that accounts for the impact of AKK on anti-inflammation in preclinical sepsis models. Our current work found that sepsis development is associated with intestinal AKK abundance alterations in patients, providing a possible link between AKK and sepsis. Subsequent in vivo and in vitro analyses demonstrated that AKK could generate a novel bioactive tripeptide that showed effective inhibitory effects on sepsis-induced systemic inflammation and ultimately protected against lethal sepsis. We expanded the pathophysiological function of AKK and provided a novel potential therapeutic approach to treat lethal sepsis in the clinic.
Materials and methods
Murine model
Specific pathogen-free male C57BL/6 mice aged 6–8 weeks were purchased from Southern Medical University (Guangzhou, China). And TLR 4 knockout (TLR4-/-) mice were purchased from GemPharmatech (Nanjing, China). In this study, the sepsis models were induced by CLP operation (a classic sepsis model that mimics polymicrobial infection and exacerbates the systemic inflammatory response) and LPS administration (another sepsis model that mimics gram-negative bacterial infection and can stably induce hyper-inflammatory responses). In brief, mice used for CLP were first anaesthetised. Then, skin preparation and a midline abdominal incision 1–1.5 cm in length were performed. The cecum was exposed and ligated below the ileocecal valve. Then two sides of the bowel wall were punctured with an 18G needle, and a small amount of faecal material was extruded. After the cecum was returned to the abdominal cavity, the abdominal wall was closed. Sham-operated mice underwent laparotomy and bowel manipulation without ligation and puncture. Mice were resuscitated by subcutaneous injection of saline (1 mL/20 g, BW). Another sepsis model was induced by intraperitoneal injection of LPS at a dose of 15 mg/kg. All survival experiments with mice were monitored for 72 hours after either CLP operation or LPS injection. Mice were sacrificed 12 hours after CLP operation or LPS injection for detection of systemic inflammation and organ damage.
Safety assessment of RKH in mice was performed. In brief, 12 mice were randomly assigned into two groups, including the RKH group and the vehicle group. Two groups of mice were gavaged with RKH (50 mg/kg) or vehicle for 7 days. Then mice were sacrificed for the detection of organ damage. All serum biochemical indices of mice were obtained from the Mindray 330e full-automatic biochemical analyzer.
All the mice had free access to water and food. Then mice were housed in a temperature-controlled colony room on a 12/12-hour light-dark cycle.22
Bone marrow-derived macrophages isolation and differentiation
Bone marrow-derived macrophages (BMDMs) were isolated and induced from mice femur and tibia under sterile conditions, culturing in DMEM (#C11995500BT, Gibco, USA) with 1% penicillin-streptomycin (#15140122, Gibco, USA), 10% fetal bovine serum (#10099141C, ThermoFisher, USA) and 20 ng/mL macrophage colony-stimulating factor (#130-101-705, Miltenyi, Germany).23 The femur and tibia were obtained after the mice were sacrificed. After removing the attached tissue, the bone marrow was washed out gently with phosphate buffered saline (PBS) (#C10010500BT, Gibco, USA). The cell suspension was centrifugated at room temperature then removed the supernatant, resuspending with the prepared medium thoroughly. BMDMs were cultured at 37°C with 5% CO2, observed continuously. BMDMs would be matured after 7 days of culture and can be used for the next step.
Tripeptide synthesis
The tripeptide RKH was commercially synthesised by Genscript ProBio Company (Nanjing, China).
Germ-free mice experiment
Germ-free (GF) mice were purchased from Cyagen Biosciences (Suzhou, China). In brief, 10 male C57BL/6J GF mice aged 6–8 weeks old were randomly assigned to the AKK-treated group and the vehicle-treated group. Experiments began after 1 week of adaptive feeding. Mice were gavaged with 300 µL live AKK (2×109 CFU/mL) or vehicle once. 48 hours after the treatment, mice plasma and cecal contents were gathered for quantitative detection of RKH.
2bRAD sequencing for microbiome (2bRAD-M)
The faeces DNA from sham and CLP mice were extracted and collected by Mag-Bind Stool DNA 96 Kit (#M4016-00, OMEGA). 2bRAD-M was carried out at the Qingdao OE Biotech (Qingdao, China). The experimental procedure includes library construction, sequencing and data analysis. The 2bRAD-M library preparation followed the original protocol developed by Wang et al with minor modifications.24 The DNA was processed through enzyme digestion and adaptors ligation reaction. The ligation products were then amplified and purified. Next, the sample-specific barcodes were imported by PCR using platform-specific barcode primers. The products were purified using a QIAquick PCR purification kit (Qiagen) and sequenced using the Illumina Nova PE150 platform. Data analysis included the identification of species-specific 2bRAD-M markers from the most comprehensive genomic databases and the calculation of relative abundance. First, a total of 173 165 microbial genomes were downloaded from the NCBI RefSeq database. Then, built-in Perl scripts were used to sample restriction fragments from microbial genomes by enzymes, which formed a 2bRAD microbial genome database. The set of 2bRAD tags sampled from each genome was assigned under the GCF number. Finally, after quality control, all sequenced 2bRAD tags were mapped against the 2bRAD marker database and the threshold of G score for a false positive discovery of microbial species was set as 5. The mapping process was based on the script created by Qingdao OE Biotech (https://github.com/shihuang047/2bRAD-M). Analysis of composition of microbiomes (ANCOM) method was performed to compare statistical differences in microbial abundance between the sham and CLP groups. The false discovery rate method was used for multiple testing in ANCOM analysis. The significant difference in microbial abundance between the two groups was measured by the adjusted p value.
16S rRNA gene sequencing
The faeces of septic patients were collected. Microbial DNA was extracted from the faeces, and then the hypervariable region V3–V4 of the bacterial 16S rRNA gene was amplified with primer by an ABI GeneAmp 9700 PCR thermocycler. An Illumina MiSeq (PE300) was used for sequencing, and the data were analysed by Majorbio Bio. Then the optimised sequences were clustered into operational taxonomic units (OTUs) using UPARSE 11 with a 97% sequence similarity level. The most abundant sequence for each OTU was selected as a representative sequence. The taxonomy of each OTU representative sequence was analysed by RDP Classifier against the 16S rRNA gene database (eg, Silva v138) using a confidence threshold of 0.7. Mothur V.1.30.2 was used to calculate the alpha diversity including the Chao index, Shannon index and Simpson index based on the OTUs information. Then principal coordinate analysis (PCoA) based on bray-curtis dissimilarity was calculated with the Vegan V.2.4.3 package.
Patient and public involvement
Patients’ samples were collected and analysed. But patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.
Statistical analysis
Unless otherwise specified, data in this study were expressed as mean±SEM. The experimental data were calculated by the two-tailed unpaired Student’s t-test or one-way analysis of variance (ANOVA). The survival rate differences of animals were analysed by log-rank (Mantel-Cox) test. The two-way ANOVA was used for statistical analysis in cellular thermal shift assay. ANCOM method was used for statistical analysis in the relative abundance of AKK from mice faeces. Wilcoxon rank-sum test was used for faecal microbiota analysis at the family and genus levels. Specific statistical methods were noted in the figure legends. The statistics of all data in this study were analysed by GraphPad Prism V.6.02. The p<0.05 was considered statistically significant and the p>0.05 was considered not statistically significant.
Other materials and methods are shown in online supplemental files.
Supplemental material
Results
Intestinal AKK participates in sepsis progression in humans and mice
First, we performed 2bRAD-M sequencing of faeces from sham and CLP mice to explore the potential association between gut microbiota and the occurrence and development of sepsis. Specifically, at the species level, a decreased abundance of AKK was observed in the faeces of the CLP group compared with the sham group (figure 1A). We subsequently verified this finding by measuring the relative abundance of AKK in mice faeces by quantitative real-time PCR (figure 1B). Importantly, in agreement with the murine model, as presented in figures 1C, 61 faecal samples from septic patients exhibited a markedly lower abundance of AKK than 27 samples from controls (relevant clinical information was shown in online supplemental table S1). To further clarify the involvement of AKK in sepsis progression, we selected a quarter of faecal samples with either very high or very low AKK abundance from 61 septic patients and divided them equally into two groups (n=8 for each of the high and low AKK abundance groups) (figure 1D). 16S rRNA gene sequencing was performed to characterise the type and relative abundance of bacterial taxa in these two groups. Our results did not reveal any significant differences in alpha diversity (online supplemental figure S1A), and the overall composition was similar between these two groups as revealed by PCoA analysis (online supplemental figure S1B). However, a significant increase in the relative abundances of AKK taxa was one of the main significant differences at both the family (Akkermansiaceae) and genus (Akkermansia) levels in the high AKK faeces group when comparison with the low AKK faeces group (online supplemental figure S1C,D). Then, we transplanted these two groups human faeces into recipient mice and performed CLP surgery. Interestingly, we found that mice received high AKK abundance faeces with a tendency for slightly higher survival time than for mice receiving the low AKK abundance faeces (figure 1E, 8.33% high AKK mice and 0.00% low AKK mice survived). In addition, mice pretreated with high AKK faeces exhibited lowered plasma aspartate transaminase (AST), creatine kinase (CK), IL-6 and IL-1β levels, compared with low AKK group (online supplemental figure S2A,B). Similarly, septic mice pretreated with high AKK faeces exhibited decreased levels of systemic inflammation factors in their peritoneal lavage fluid (PLF) and multiple organs (such as the heart, lung, liver and kidney) and displayed improved multiple organ injury when compared with the low AKK group, as indicated by qPCR and histopathology (online supplemental figure S2C–E). Although other bacteria may also be involved in the progression of sepsis, our data showed that the AKK bacterium, at least in part, was participated in the protective effect. Thus, the intestinal AKK levels may participate in sepsis progression.
Supplemental material
Supplemental material
Supplemental material
Intestinal AKK protects against lethal sepsis in a murine model. (A) Relative abundance of gut microbiota at the species level in the sham group and CLP group (sham n=6, CLP n=6). (B) The relative abundance of AKK from mice faeces was measured by quantitative PCR (qPCR) and displayed by fold change (sham n=13, CLP n=21). (C) The relative abundance of AKK from human faeces was measured by qPCR and displayed by fold change (control patients n=27, septic patients n=61). (D) The strategy for definition of high (red) and low (grey) levels of AKK in septic patient faeces (n=61). (E) Faecal microbiota transplantation (FMT) was performed on antibiotic-treated mice for 3 days using faeces from septic patients. These recipient mice then experienced CLP-induced sepsis and survival rates were determined using Kaplan-Meier curves (sham group n=5; CLP group n=24–25). (F) Mice were gavaged with PBS, live AKK or dead AKK (200 µL each time, the final concentration of AKK was 2×109 CFU/mL) for 3 days (once a day and perform surgical operation 2 hours after the third treatment) followed by CLP surgery. The survival rates were determined using Kaplan-Meier curves (sham group n=5; CLP group n=20). (G, H, J, K) Mice were sacrificed for lung and liver collection 12 hours after sham or CLP procedures. (G) Representative H&E staining images (scale bar:100 µm, magnification: ×200) and the histological score for lungs in mice pretreated with PBS, live AKK or dead AKK (sham group n=4; CLP group n=10). (H) The mRNA levels of IL-6, IL-1β, Ccl2, Ccl3 and Ccl4 in the lung (upper) and liver (lower) of mice pretreated with PBS, live AKK or dead AKK (sham group n=4–6; CLP group n=9–12). (I) Mice were gavaged with blank or live AKK supernatants (200 µL each time) for 3 days (once a day and perform surgical operation 2 hours after the third treatment) followed by CLP surgery. The survival rates were determined using Kaplan-Meier curves (sham group n=5; CLP group n=30–31). (J) Representative H&E staining images (scale bar:100 µm, magnification: ×200) and the histological score for lungs in mice pretreated with blank or live AKK supernatants (sham group n=4; CLP group n=6–7). (K) The mRNA levels of IL-6, IL-1β, Ccl2, Ccl3 and Ccl4 in the lung (left) and liver (right) of mice pretreated with blank or live AKK supernatants (sham group n=5; CLP group n=7–8). (L) Mice were gavaged with PBS and live AKK (200 µL each time, the final concentration of AKK was 2×109 CFU/mL) for 3 days (once a day and treat LPS 2 hours after the third treatment). The survival rates were determined using Kaplan-Meier curves (PBS control n=5; LPS group n=15). (M) Mice were gavaged with blank or live AKK supernatants (200 µL each time) for 3 days (once a day and treat LPS 2 hours after the third treatment). The survival rates were determined using Kaplan-Meier curves (PBS control n=5; LPS group n=15). (N, O) Mice were sacrificed for lung collection 12 hours after PBS or LPS injection. (N) Representative H&E staining images (Scale bar: 100 µm, magnification: ×200) and the histological score for lungs in mice pretreated with PBS or live AKK (PBS group n=4; LPS group n=8). (O) Representative H&E staining images (scale bar:100 µm, magnification: ×200) and the histological score for lungs in mice pretreated with blank or live AKK supernatants (PBS group n=4; LPS group n=8). *p<0.05; ns p>0.05 (ANCOM analysis in A; The unpaired two-tailed Student’s t-test in B, C, G, H, J, K, N, O; Mantel-Cox log-rank test in E, F, I, L, M; Results were expressed as the mean±SEM). AKK, Akkermansia muciniphila; ANCOM, analysis of composition of microbiomes; CLP, cecal ligation and puncture; LPS, Lipopolysaccharide; NS, not significant; PBS, phosphate buffered saline.
Live AKK protects against lethal sepsis in a murine model
Next, to determine the direct effect of AKK on sepsis, we treated mice with PBS, live AKK, or heat-killed AKK (dead AKK) followed by sham or CLP surgery. The survival time of live AKK-pretreated mice was slightly improved over that of PBS-pretreated and dead AKK-pretreated mice after the CLP challenge (figure 1F, 10.00% live AKK-pretreated mice, 0.00% PBS-pretreated mice, and 0.00% dead AKK-pretreated mice survived). We then evaluated the impact of AKK on sepsis-associated organ damage. In contrast to the sham group, the mice in the sepsis group exhibited severe pulmonary inflammation with alveolar wall thickening and inflammatory cell infiltration. Live AKK treatment significantly reduced these alterations, whereas dead AKK failed to improve these pathological changes (figure 1G). In agreement with these findings, pulmonary and hepatic inflammatory factors were only decreased on live AKK treatment (figure 1H). These findings demonstrated that live AKK, but not dead AKK, could mitigate the organ damage and lethality caused by sepsis.
The above data indicated that the protective effects of AKK may be dependent on the bioactive substrates generated from live bacteria. Then, we evaluated whether the culture supernatant from live AKK would alleviate sepsis. As expected, septic mice pretreated with supernatant from live AKK exhibited a considerable decrease in mortality, alleviated pulmonary injury, and reduced pulmonary and hepatic inflammatory factor expression compared with the septic mice pretreated with blank supernatant (figure 1I–K). These data indicated that live AKK and its culture supernatant can effectively protect against sepsis, possibly associated with reduced organ inflammatory response. The LPS-induced sepsis model is a simple model with higher reproducibility and reliability to induce hyperinflammatory responses.25 To further confirm the protective effect of live AKK and its culture supernatant against lethal sepsis, we employed an LPS-induced sepsis model. In parallel experiments, the live AKK and its supernatant effectively prolonged the survival time (figure 1L,M), ameliorated pulmonary damage (figure 1N,O and online supplemental figure S3A), and reduced hepatic proinflammatory factor expression (online supplemental figure S3B) compared with controls after LPS administration. Taken together, these findings clearly demonstrated that live AKK effectively rescued lethal sepsis in a murine model, possibly by generating protective substances.
Supplemental material
Live AKK generates a novel tripeptide Arg-Lys-His (RKH)
To further probe the protective substances secreted by live AKK, we performed non-targeted metabolomics analysis on the culture supernatant from the blank medium and live AKK culture medium. Principal component analysis and volcano plots showed that these supernatants exhibited different metabolite profiles (figure 2A,B). Surprisingly, we found that a novel tripeptide, RKH, was markedly enriched in live AKK supernatant and represented the largest fold change in the non-targeted metabolomics analysis (figure 2C,D). Further, targeted metabolomics analysis was warranted to verify significantly higher RKH levels in live AKK supernatant compared with blank medium (figure 2E). To evaluate the functional importance of live AKK in RKH generation, we colonised GF mice with live AKK. Our data indicated that GF mice had remarkably lower levels of RKH than conventional mice. Nevertheless, RKH levels were significantly elevated in both cecum contents and plasma from GF mice colonised with live AKK compared with non-colonised controls (figure 2F,G). These results demonstrated that live AKK could generate a novel tripeptide, RKH. To further explore whether RKH participates in sepsis development, we detected the faecal RKH levels in control and septic patients. Figure 2H shows that the faecal RKH level of septic patients was significantly decreased compared with the healthy control individuals (relevant clinical information was shown in online supplemental table S2). Collectively, these results demonstrated that RKH may participate in sepsis development.
Supplemental material
Arg-Lys-His (RKH) generated by live AKK protects against lethal sepsis in mice. (A) Scatter plots of PCA of non-targeted metabolomics analysis from blank and live AKK supernatants (n=6). (B) Volcano plot of non-targeted metabolomics analysis (AKK group vs blank group). (C) Molecular structure of RKH. (D) The fold change (FC) value ranking of differently expressed metabolites from live AKK versus blank supernatants. The y-axis displayed the name of each class, with the representative metabolite of each class shown in parenthesis. Compounds corresponding to serial numbers were shown in online supplemental table S6. (E) Targeted metabolomics analysis of RKH levels in blank and live AKK cultural supernatants (n=6). (F) RKH concentration in cecum contents from conventional mice and germ-free (GF) mice gavaged with live AKK or vehicle (n=4–5). (G) RKH concentration in plasma from conventional mice and GF mice gavaged with live AKK or vehicle (n=5). (H) RKH concentration in faeces from septic (n=38) and control patients (n=25). (I) Mice were gavaged with RKH (50 mg/kg) or vehicle (200 µl each time) 2 hours before CLP surgery. The survival rates were determined using Kaplan-Meier curves (sham group n=5; CLP group n=15–16). (J–L, N, O) Mice were sacrificed for sample collection 12 hours after procedures. (J) Representative H&E staining images and the histological score for lungs from sham and CLP mice (sham group n=4; CLP group n=8) (scale bar:100 µm, magnification: ×200). (K) Plasma levels of ALT, AST, CK, CREA and UREA (sham group n=3; CLP group n=6). (L) The mRNA levels of pulmonary proinflammatory factors (IL-6, IL-1β, Ccl2, Ccl3 and Ccl4) from sham and CLP mice (sham group n=4; CLP group n=7–8). (M) Mice were gavaged with RKH (50 mg/kg) or vehicle (200 µL each time) 2 hours before LPS injection. The survival rates were determined by Kaplan-Meier curves (PBS control n=5; LPS group n=15). (N) Representative H&E staining images and the histological score for lungs from PBS and LPS-treated mice (PBS group n=4; LPS group n=8) (scale bar:100 µm, magnification: ×200). (O) Plasma levels of ALT, AST, CK, CREA and UREA (PBS group n=3; LPS group n=5–6). *p<0.05; ns p>0.05 (unpaired two-tailed Student’s t-test in E, F, G, H J, K, L, N, O; Mantel-Cox log-rank test in I and M; Results were expressed as the mean±SEM). AKK, Akkermansia muciniphila; ALT, alanine aminotransferase; AST, aspartate transaminase; CK, creatine kinase; CLP, cecal ligation and puncture; CREA, creatinine; LPS, lipopolysaccharide; NS, not significant; PCA, principal component analysis; PBS, phosphate buffered saline; UREA, urea.
Supplemental material
RKH protects against lethal sepsis in a murine model
To explore the potential therapeutic effect of RKH in lethal sepsis, we administered RKH to mice before CLP surgery. Survival analysis showed that RKH pretreatment significantly extended septic mice survival compared with those received vehicle pretreatment (figure 2I). Meanwhile, histological analysis revealed that organs from mice treated with RKH, including the lungs, liver, kidneys and heart, exhibited less inflammation and injury after CLP (figure 2J and online supplemental figure S4A). Biochemistry analyses of plasma demonstrated significantly reduced levels of alanine aminotransferase, AST, CK, creatinine and urea in RKH-treated septic mice when compared with control mice. (figure 2K). Further, we observed lower mRNA levels of inflammatory cytokines in the lung, liver, kidney and heart of RKH-treated septic mice (figure 2L and online supplemental figure S4B). When we subjected mice to a lethal LPS challenge, we confirmed that RKH also significantly extended the survival time of LPS-induced septic mice (figure 2M). Consistently, pretreatment with RKH significantly alleviated acute tissue injuries (figure 2N,O and online supplemental figure S5A) and decreased the expression of inflammatory factors in the lung, liver, kidney and heart from LPS-induced septic mice (online supplemental figure S5B). Collectively, our data revealed that the live AKK product RKH could protect against lethal sepsis in a murine model.
Supplemental material
Supplemental material
RKH prevents sepsis-induced systemic inflammation
We then explored the detailed mechanism by which RKH protects against sepsis. Systemic inflammation, in terms of cytokine storm, is one of the main factors that participate in the death during the onset of sepsis.11 26 Thus, we examined whether RKH played a critical role in mitigating the severity of excessive inflammation during sepsis. We performed transcriptomic analysis of PLF from sham-pretreated, CLP-vehicle-pretreated and CLP-RKH-pretreated mice. As shown in figure 3A,B, RKH partly shifted the overall gene expression profile, and specifically, proinflammatory factor expression was diminished in CLP-RKH animals compared with CLP-vehicle animals. To verify this phenotype, we examined the systemic proinflammatory factor levels in both PLF and plasma obtained from AKK strains (live and dead), culture supernatants from live AKK, and RKH pretreated mice by RT-qPCR and ELISA. As presented in figure 3C, mRNA and protein levels of key cytokines in PLF and plasma were markedly reduced in mice treated with live AKK compared with control mice after CLP. In contrast, dead AKK could not mitigate the systemic inflammatory response. Similarly, live AKK culture supernatant and RKH also exhibited profound effects on reducing systemic inflammation compared with their corresponding control groups (figure 3D,E). The in vivo anti-inflammatory effects of live AKK, live AKK culture supernatant and RKH were also reproduced in the LPS-induced sepsis model (online supplemental figure S6A).
Supplemental material
RKH prevents sepsis-induced systemic inflammation. (A) Scatter plots depicting the PCA results of the RNAseq analysis for PLF cells obtained from sham, CLP-vehicle and CLP-RKH group (n=5). (B) Heatmap showed the expression of inflammation-related genes in PLF cells from sham, CLP-vehicle and CLP-RKH group. The colour bars represented the relative gene expression level (blue: downregulated; red: upregulated) (n=5). (C–G) Mice were sacrificed for samples collection 12 hours after sham or CLP procedures. (C) The concentrations of IL-6 and IL-1β in plasma (upper) and mRNA levels of IL-6, IL-1β and Ccl4 in PLF (lower) were analysed in mice pretreated with PBS, live AKK or dead AKK (sham group n=4; CLP group n=8–12). (D) The mRNA levels of IL-6, IL-1β and Ccl4 in PLF (upper) and plasma concentration of IL-6 and IL-1β (lower) in mice pretreated with blank or live AKK supernatants (sham group n=4–5; CLP group n=7–8). (E) The mRNA levels of IL-6, IL-1β and Ccl4 in PLF (upper) and plasma concentration of IL-6 and IL-1β (lower) in mice pretreated with vehicle or RKH (sham group n=3–4; CLP group n=7–8). (F) The frequencies of PMs (CD11bhigh F4/80high) from sham and CLP mice were detected by flow cytometry (sham group n=3–4; CLP group n=5–7). (G) The mRNA levels of IL-6, IL-1β and Ccl4 in sorting PMs after CLP (n=5–8). (H) The mRNA levels of proinflammatory factors (IL-6, IL-1β, Ccl3 and Ccl4) in BMDMs (upper) and THP-1-dMs (lower) after LPS (BMDMs: 100 ng/mL; THP-1-dMs: 500 ng/mL) stimulation for 6 hours with or without RKH (125 µM) were determined by qPCR (n=5–6). *p<0.05; ns p>0.05 (unpaired two-tailed Student’s t-test in C–H; results were expressed as the mean±SEM). AKK, Akkermansia muciniphila; BMDMs, bone marrow-derived macrophages; CLP, cecal ligation and puncture; LPS, lipopolysaccharide; NS, not significant; PBS, phosphate buffered saline; PCA, principal component analysis; PLF, peritoneal lavage fluid; PMs, peritoneal macrophages; THP-1-dMs, human monocyte-derived macrophages.
Macrophage activation plays a pivotal role in the development of cytokine storms during the early phase of sepsis.27 28 Therefore, we examined the impact of AKK and RKH on macrophage status during sepsis. The gating strategy for macrophages in PLF was displayed in online supplemental figure S6B. Compared with those in the sham groups, the proportion of macrophages was significantly increased after CLP surgery (figure 3F); however, pretreatment with live AKK, or live AKK culture supernatant, or RKH, had markedly reduced macrophage recruitment in the PLF during sepsis (figure 3F). Moreover, we sorted peritoneal macrophages using fluorescence activated cell sorting (FACS), and observed that macrophages isolated from live AKK, live AKK culture supernatant and RKH-treated mice showed markedly lower cytokine expression than control mice after CLP (figure 3G). In vitro experiments revealed that treatment with RKH did not induce cytotoxicity (online supplemental figure S7A), but rather suppressed the expression of proinflammatory cytokines in both BMDMs and human monocyte-derived macrophages (THP-1-dMs) after LPS stimulation (figure 3H). Collectively, our data verified that RKH had the potential to reduce the systemic inflammation caused by sepsis.
Supplemental material
RKH directly targets TLR4 and blocks TLR4 activation
TLRs play an essential role in the mediation of systemic responses to invading pathogens during progress in sepsis.29 To assess whether the anti-inflammatory properties of RKH are related to TLRs, we tested the effects of RKH on the expression of different types of TLR-mediated cytokines in macrophages. We found that RKH was able to significantly reduce the expression of proinflammatory factors in response to TLR4 agonist (LPS) in both BMDMs and THP-1-dMs, but such an effect was mild when inhibiting other TLRs (online supplemental figure S8A,B). In addition, RKH significantly reduced TLR4 downstream signal transduction in both macrophages (figure 4A,B and online supplemental figure S8C,D). Accordingly, the anti-inflammatory properties of RKH may be attributed to the inhibition of TLR4 signalling.
Supplemental material
RKH directly targets the TLR4 receptor and blocks TLR4 activation. (A) Western blot analysis of phosphorylated P38, P65, JNK and ERK (P-P38, P-P65, P-JNK and P-ERK) and total P38, P65, JNK and ERK in THP-1-dMs after 500 ng/mL LPS stimulation for 15 min with or without 125 µM RKH. (B) Quantitative analysis for western blot analysis and total protein served as control (n=3). (C) Flow cytometric analysis of the 100 µg/mL FITC-LPS labelled THP-1-d-Ms treated with or without 125 µM RKH for 15 min (n=3). (D) Surface plasmon resonance assays (SPR) analysis for TLR4 protein with different doses of RKH. (E) Drug affinity responsive target stability (DARTS) assay detected a marked increase in ~110 kD band on 125 µM RKH incubation in pronase-digested THP-1-d-Ms lysates. Western blot analysis of TLR4 in pronase-digested THP-1-d-Ms lysates treated with or without 125 µM RKH (n=3). (F) Western blot analysis of TLR4 degradation in THP-1-d-Ms lysates treated with or without 125 µM RKH (n=3). (G) Immunoprecipitation (IP) with anti-Flag antibody followed by Western blot to analyse the effect of RKH treatment on dimerisation of TLR4 in THP-1-dMs and BMDMs exposed to LPS. Two other independent experiments were displayed in online supplemental figure S8H–I. (H IP with anti-Flag antibody followed by Western blot to analyse the effect of RKH treatment on TLR4/MD2 complex formation in THP-1-dMs and BMDMs exposed to LPS. Two other independent experiments were displayed in online supplemental figure S8H–I. (I) Molecular docking analysis of RKH with the TLR4/MD2 complex (PDB:3FXI) (upper), and the potential interaction binding site (lower). (J) Flag-TLR4 and various mutant forms of HA-TLR4 were coexpressed in THP-1-d-Ms. IP with anti-Flag antibody followed by Western blot to analyse the effect of RKH treatment on dimerisation of TLR4. Data represented three independent experiments. (K) TLR4 knockout mice (TLR4-/-) mice were gavaged with RKH (50 mg/kg) or vehicle (200 µL each time) 2 hours before CLP surgery. The survival rates were determined (sham group n=4; CLP group n=7–8). (M–O) Mice were sacrificed for samples collection 12 hours after sham or CLP procedures. (L, M) Representative H&E staining images and the histological score for lungs from sham and CLP TLR4-/- mice (sham group n=3; CLP group n=6) (scale bar:100 µm, magnification: ×200). (N) Detection for plasma IL-6 and IL-1β in sham and CLP TLR4-/- mice (sham group n=3; CLP group n=6). (O) The mRNA levels of pulmonary proinflammatory factors (IL-6, IL-1β, Ccl2, Ccl3 and Ccl4) from sham and CLP TLR4-/- mice (sham group n=3; CLP group n=5–6). *p<0.05; ns p>0.05 (unpaired two-tailed Student’s t-test in B, C, E, M–O; Mantel-Cox log-rank test in K; Two-way ANOVA in F. Results were expressed as the mean±SEM). ANOVA, analysis of variance; BMDMs, bone marrow-derived macrophages; CLP, cecal ligation and puncture; LPS, lipopolysaccharide; NS, not significant; PBS, phosphate buffered saline; TLR4, toll-like receptor 4; THP-1-dMs, human monocyte-derived macrophages.
To explore the inhibitory mechanism of RKH on TLR4 activation, we assessed the binding ability of fluorescein isothiocyanate-labelled LPS (FITC-LPS) on macrophage membranes by measuring the mean fluorescence intensity in the presence or absence of RKH. As expected, the FITC-LPS intensity was significantly reduced in the RKH-treated group compared with untreated BMDMs and THP-1-dMs (figure 4C and online supplemental figure S8E). Thus, RKH may be capable of preventing the binding of LPS to TLR4 at the cell membrane. Next, we sought to determine how RKH interfered with LPS binding to TLR4. Various factors may affect the binding ability of LPS to TLR4, such as changes in TLR4 expression, internalisation, and occupancy of the binding site. We first examined the expression of TLR4 and its accessory protein myeloid differentiation 2 (MD2) in macrophages and found that RKH did not affect the protein expression of TLR4 and MD2 (online supplemental figure S9A). Concurrently, we examined whether RKH affects the internalisation of TLR4. Since CD14 serves as a membrane-associated protein that assists TLR4 activation and internalisation, we next examined the fluorescent signals of CD14 and TLR4 on macrophage membranes. We found that CD14 and TLR4 were highly expressed on macrophage membranes under normal conditions, and LPS stimulation induced their internalisation, which resulted in the attenuation of fluorescent signals. Compared with the LPS-stimulated group, the RKH-treated group did not have a statistically significant difference (online supplemental figure S9B,C). Thus, the negative impact of RKH on the binding between LPS and TLR4 was independent of influencing TLR4 expression or internalisation. We then hypothesised that RKH and LPS could compete for binding to TLR4. To test our hypothesis, we used surface plasmon resonance assays to determine whether RKH could directly bind to TLR4. As shown in figure 4D, the results revealed significant positive binding affinity between RKH and TLR4 in a dose-dependent manner, with an equilibrium dissociation constant (KD) of 3.34 nM. In addition, RKH enhanced the anti-protease hydrolysis of TLR4 protein (figure 4E and online supplemental figure S8F) and reduced the degradation rate of TLR4 (figure 4F and online supplemental figure S8G) in comparison with the control group in both BMDMs and THP-1-dMs. All this evidence suggests that TLR4 is a binding target of RKH.
Supplemental material
TLR4 forms heterodimers with MD2 and triggers the dimerisation of TLR4 by forming a complex (TLR4/MD2/LPS) that in turn initiates signal transduction. We wondered whether the binding of RKH to TLR4 affects TLR4 activation. We assessed dimer formation in macrophages overexpressing Flag-TLR4 and HA-MD2 or HA-TLR4. Compared with controls, LPS significantly promoted TLR4 to MD2 and TLR4 to TLR4 dimerisation; however, this phenomenon was markedly reduced in the RKH group in both BMDMs and THP-1-dMs (figure 4G,H and online supplemental figure S8H,I), suggesting that RKH inhibits TLR4 activation. In search of the key functional sites where RKH binds to TLR4, we docked RKH to the crystal structure of the human TLR4/MD2/LPS complex (PDB code: 3FXI) using molecular docking software. As shown in figure 4I, RKH was fit to the pocket of the TLR4/MD2 complex and interacted with several amino acid sites, including ARG264, LEU293, ASP294, LYS341 and LYS362 in TLR4 which occupied the space and weakened the binding of LPS with TLR4 receptor. To further assess the contribution of these sites to the inhibition of TLR4 activation by RKH, we transfected overexpression plasmids with mutations in these sites into THP-1-dMs and examined the dimerisation of TLR4. The results indicated that RKH lost its consistent and obvious suppressive effects on TLR4 activation at ARG264, LEU293, ASP294, LYS362 and all five site mutant groups but not at the LYS341 mutant site (figure 4J), suggesting that ARG264, LEU293, ASP294 and LYS362 on TLR4 may serve as the potential functional binding sites of RKH that inhibit TLR4 dimerisation.
Finally, we employed TLR4 knockout mice to evaluate the contribution of TLR4 signalling to the anti-sepsis and anti-inflammatory effects of RKH (online supplemental figure S10A). A survival rate experiment showed no significant difference between mice treated with RKH and vehicles (figure 4K), suggesting that RKH is no longer protected against septic lethality in TLR4 knockout mice. Similarly, the lung pathological scores (figure 4L,M) and the inflammatory factor levels in the plasma (figure 4N), lung (figure 4O), liver (online supplemental figure S10B) and PLF (online supplemental figure S10C) were comparable between the RKH-treated and untreated groups during sepsis. These results suggested that the protection of RKH against lethal sepsis, at least in part, occurred through the inhibition of TLR4 signalling.
Supplemental material
RKH prolongs survival time and reduces hyperinflammation and organ damage in a septic piglet model
To further expand the therapeutic potential of RKH for sepsis in the clinic, we employed the piglet sepsis model. The experimental strategy for the septic piglet model was shown in figure 5A. We observed that pretreatment with RKH significantly prolonged survival time (figure 5B) in the LPS-induced septic model, compared with the vehicle pretreatment. Moreover, levels of serum AST, γ-glutamyl transpeptidase and lactate dehydrogenase showed a decreasing trend in RKH-treated septic piglets, compared with those in the vehicle group (figure 5C). Notably, levels of serum UREA, IL-6 and IL-1β were remarkably decreased in RKH-treated piglets (figure 5C). In addition, histopathological abnormalities in various organs and the levels of proinflammatory factors were significantly reduced in septic piglets pretreated with RKH compared with controls (figure 5D–G). In conclusion, these results clearly revealed that RKH could reduce sepsis-induced mortality and ameliorate systemic inflammation and organ damage in a piglet model.
RKH prolongs survival time and reduces hyperinflammation and organ damage in a septic piglet model. (A) Schematic experiment timeline. Piglets were gavaged with 1 mg/kg RKH or vehicle 2 hours before LPS injection. LPS (300 µg/kg) was injected intravenously for 30 min. (B) The survival rates were determined using Kaplan-Meier curves (n=5). (C) The serum levels of AST, γ-GT, LDH, UREA, IL-6 and IL-1β were detected 3 hours after LPS injection in piglets pretreated with the vehicle and RKH (n=5). (D–G) Representative H&E staining sections and the histological score of (D) hearts, (E) livers, (F) lungs and (G) kidneys in piglets pretreated with the vehicle and RKH (left, scale bar: 100 µm, magnification: ×200, n=5). (D–G) The mRNA levels of IL-6, IL-1β, Ccl3 and Ccl4 in (D) hearts, (E) livers, (F) lungs and (G) kidneys of piglets pretreated with the vehicle and RKH (Right, n=5). *p<0.05; ns p>0.05 (unpaired two-tailed Student’s t-test in C–G; Mantel-Cox log-rank test in B. Results were expressed as the mean±SEM). AST, aspartate transaminase; γ-GT, γ-glutamyl transpeptidase; LDH, lactate dehydrogenase; LPS, lipopolysaccharide; NS, not significant; UREA, urea.
In vivo acute safety assessment of RKH
Next, we performed experiments to assess the acute safety of RKH in both piglet and murine models to explore the possible clinical translation of this novel tripeptide. As shown in online supplemental figure S11A, there were no significant differences in serum biochemical indices between the vehicle group and the RKH group (the dose of RKH was five times as the effective dose we used in the previous piglet sepsis model). In addition, acute treatment with RKH did not cause perceivable injury to the heart, liver, lung or kidney (online supplemental figure S11B). Similar results were also observed in the murine model (online supplemental figure S12A,B). Thus, acute administration of RKH did not cause obvious adverse effects in vivo.
Supplemental material
Supplemental material
RKH negatively regulated septic patients’ macrophage proinflammatory activity
Finally, to further evaluate the potency of RKH on human immune cells’ proinflammatory activity regulation, peripheral blood mononuclear cells were extracted from a septic patient and differentiated into macrophages in vitro. Following an LPS challenge, macrophages treated with RKH exhibited a markedly lower expression of IL-6, the most predominant cytokine that drives abnormalities during cytokine storms, when compared with the vehicle group (figure 6A). These findings further confirm the anti-inflammatory effects of RKH in human macrophages and highlight its potential therapeutic benefits.
RKH negatively regulated septic patients’ macrophage proinflammatory activity. (A) Peripheral blood monocytes from patients with sepsis were differentiated into macrophages by human recombinant GM-CSF. Macrophages were treated with 500 ng/mL LPS for 6 hours with or without 125 µM RKH. The mRNA levels of IL-6 were detected by qPCR (n=4–6). (B) CD4+ T cells and CD8+ T cells from 6 patients with sepsis were obtained by FACS. And these cells were treated with vehicle or RKH (125 µM) for 6 hours and then apoptosis was detected. (C) CD4+ T cells and CD8+ T cells from three patients (#1-#3) were treated with vehicle or RKH (125 µM) and co-cultured with CD3/CD28 antibody. Then, the apoptosis was detected. (D) In the unstimulated conditions, CD4+ T cells and CD8+ T cells from three patients (#4-#6) were treated with vehicle or RKH (125 µM), and apoptosis was detected. (E) CD4+ T cells and CD8+ T cells from 3 patients (#1–#3) were treated with vehicle or RKH (125 µM) and cocultured with CD3/CD28 antibody. Then, the cytokines levels were detected. (F) In the unstimulated conditions, CD4+ T cells and CD8+ T cells from three patients (#4-#6) were treated with vehicle or RKH (125 µM), and the cytokines levels were detected. (G) Working model: Intestinal AKK participated in sepsis progression in humans and mice. RKH, a product of live AKK, exerted protective effects against sepsis-induced death and organ damage in murine and piglet models. A mechanistic study revealed that RKH could directly bind to Toll-like receptor 4 (TLR4) and block TLR4 signal transduction in immune cells. *p<0.05; ns p>0.05 (unpaired two-tailed Student’s t-test in A; results were expressed as the mean±SEM). AKK, Akkermansia muciniphila; FACS, fluorescence activated cell sorting; GM-CSF, granulocyte-macrophage colony-stimulating factor; LPS, lipopolysaccharide; NS, not significant.
Immune suppression occurred in septic patients was another important cause of death, and was closely related to increased apoptosis of T cells and decreased T cell functions, including cytokine production diminishment.8 30 Many anti-inflammation approaches were associated with the risk of immune suppression.2 To investigate the potential immunosuppressive effects of RKH, we conducted in vitro experiments to assess its impact on T cell function in septic patients. CD4+ T and CD8+ T cells were isolated from peripheral blood samples obtained from 6 septic patients using FACS (relevant clinical information was shown in online supplemental table S3). Flow cytometry was then performed to assess T cell viability (figure 6B), and our results indicated that RKH treatment did not significantly induce T cell apoptosis in septic patients under both CD3/CD28 stimulated (figure 6C) and unstimulated conditions (figure 6D). Since T cells contributed to pathogen elimination in sepsis by releasing various cytokines,31 we further examined the impact of RKH on the cytokine secretion capacity of T cells. As shown in figure 6E,F, despite varying functional statuses of T cell among the 6 sepsis patients, RKH treatment did not interfere the secretion ability of cytokines (IFN-γ and IL-10) in either CD4+ T or CD8+ T cells under both CD3/CD28 stimulated and unstimulated conditions. Taken together, these findings provided strong evidence that RKH could negatively modulate the proinflammatory activity of macrophages from patients with sepsis, while not significantly impacting the immune function of T cells in vitro.
Supplemental material
Discussion
The gut microbiota is a well-established upstream regulator of sepsis,32–35 however, the detailed contribution of the gut microbiota, together with its bioactive products, to sepsis development is still unclear. More functional microbial molecular patterns should be studied to investigate the pathogenesis of sepsis. Here, we identified a potential association between AKK and sepsis progression, both in patients and in the murine model. Notably, live AKK could generate a novel anti-inflammatory tripeptide, RKH, which directly binds to TLR4 and suppresses its downstream signalling amplification during the inflammatory response, ultimately protecting against lethal sepsis. As graphically summarised in figure 6G, our work identified live AKK/RKH/TLR4 as a new regulatory axis that participates in sepsis progression. In addition, our current evidence clearly suggests that RKH, as a novel safe bioactive tripeptide exerted protective effects on both CLP surgery (polymicrobial infection) and LPS challenge (Gram-negative bacterial infection) models. Finally, we observed RKH was able to diminish macrophage proinflammatory activity from septic patients in vitro, while having minimal impact on the function of T cells.
Thus, our data confirm the anti-inflammatory effects of live AKK and its bioactive products and highlight its potential therapeutic benefits in sepsis.
AKK is accounting for >1% of the resident bacteria in the healthy human gut.36 37 Our findings suggest that septic patients have lower AKK levels than healthy individuals. However, the causal relationship between low levels of AKK and sepsis is complex and not fully understood. On the one hand, the decrease in AKK abundance may be due to non-septic factors, such as the use of antibiotics that disturbs the balance of gut micro-organisms. On the other hand, a reduction in AKK abundance may be related to sepsis. Sepsis triggers excessive inflammation and imbalanced microcirculation,32 leading to intestinal lining damage and an unfavourable redox environment that hinders AKK growth. Additionally, sepsis can disrupt the mucus layer and mucin, the essential energy source for AKK,38 further affecting the abundance of AKK. Our experiments on faecal transplantation have suggested that a lower level of AKK bacteria in faeces may play a subtle role in increasing susceptibility to sepsis rather than provide clear evidence of a causal relationship.
AKK is one of the most attractive probiotics that potentially translates into the clinic to treat multiple diseases. Although knowledge regarding the preventive mechanism of AKK against pathological progression is progressively being discovered, how AKK exerts anti-inflammatory effects is still poorly understood. Many studies, including our current work, have revealed that the living state of AKK is essential for its beneficial effects, indicating that the bioactive products derived and secreted from AKK play a key role in maintaining host homoeostasis. For example, the AKK metabolite harmaline could inhibit p65 activation and suppress virus-induced systemic inflammation.39 In contrast, AKK may generate functional metabolites to activate immune reactions in intestinal epithelial cells to maintain intestinal homoeostasis.40 Here, in the context of bacterial infection-induced lethal sepsis, for the first time, we disclosed a novel metabolite, RKH, generated from live AKK and demonstrated its anti-inflammatory properties. Our findings expand our knowledge about the modulatory mechanisms of live AKK on the host immune response. Although we clearly identified the molecular interaction mechanisms between RKH and host cells, how live AKK generates RKH is still unclear. We speculated that the underlying mechanism is complex and may involve several potential pathways. One possibility is that live AKK may contain a specific gene that encodes an RKH-enriched protein, and those proteins may degrade into smaller pieces of peptide, including RKH. Second, there are some functional microRNAs located in live AKK, and those microRNAs may directly translate or indirectly modulate mRNA to synthesise RKH-enriched peptides. We also cannot exclude the possibility that there are translation-independent pathways that format RKH in live AKK. Thus, the detailed mechanism by which live AKK generates RKH requires further investigation. In addition, it is noted that besides AKK, many other bacteria may also participate in the regulation of sepsis, more functional bacteria should be disclosed in future work. Similarly, besides RKH, other important compounds derived from live AKK may also modulate sepsis. More key bioactive metabolites should be disclosed in future work.
The usage of tripeptides in the clinic is still not widely implemented. However, there are several advantages of tripeptides in treating patients. Theoretically, tripeptides that are formed by natural amino acids are supposed to have low toxicity, as they are not difficult to be degraded and metabolised by the host on acute treatment. In addition, technologies for peptide linkage synthesis are mature, and we can synthesise many different types of tripeptides and further modify their structures. However, some disadvantages should also be noted. For example, the majority of tripeptides, including RKH, showed obvious hydrophilicity, specifically, it is not easy for them to enter cells to interact with many key biological molecules. They may only influence the proteins located in the cells, which restricts the usage of tripeptides. This reality may ultimately influence clinical translation. To address the flaw, we can optimise the structure of the functional tripeptide in the future to increase its hydrophobicity, which may increase the cellular bioavailability of the tripeptide.
The crystal structure of the TLR4/MD2/LPS shows that the hydrophobic pocket of MD2 could accommodate most of the lipid portion of LPS (five of six hydrophobic carbon chains), whereas the sixth chain of LPS is left outside the pocket and interact with the ectodomain of a neighbouring TLR4 molecule.41–44 Previous studies showed that the total number of acyl chains is one of the most important factors governing the inflammatory activity of LPS,45 and the correct binding of the six acyl chains is crucial to cause TLR4 signalling activation. Moreover, when part of the MD2 binding pocket becomes empty, this will destabilise the overall interaction.46 47 Our experiments revealed that RKH occupied part of the binding site of LPS on TLR4, which may lead to a localised structural change of the TLR4/MD2 dimer, weakening the binding of LPS with TLR4 and affecting the correct embedding of LPS into the hydrophobic pocket of MD2, finally facilitated RKH to compete with LPS for TLR4/MD2 binding and possibly suppressed LPS-induced TLR4/MD2 dimerisation during the onset of sepsis. The participation of TLR4 in sepsis development is clear. Many approaches that restrict TLR4 function have been demonstrated to combat sepsis and its related death.48–51 For example, we previously found that the gut microbiota-derived secondary bile acid hyodeoxycholic acid could serve as an endogenous inhibitor of the TLR4/MD2 complex and attenuate sepsis-associated systemic inflammation and organ damage.11 Meanwhile, many synthesised small compounds also exerted similar effects.52 53 These observations strengthened the hypothesis that TLR4 is one of the main targets to treat sepsis. Although many clinical trials on TLR4 inhibitors for sepsis therapy have not achieved satisfactory outcomes,54 55 the therapeutic strategies often do not take into account the heterogeneity of the individual state of the immune response. There is a growing recognition that treating sepsis patients requires a more personalised approach. The latest treatment strategy for sepsis is based on the patient’s immunological characteristics and clinical phenotype,8 which maximises effectiveness and can overcome some challenges associated with sepsis patients. A retrospective analysis of clinical trials showed that while anakinra (a recombinant IL-1 receptor antagonist) treatment do not convey benefit in the overall population, it did reduce mortality in subgroups identified by high baseline (endogenous) IL-1 receptor antagonist concentrations.56 In addition, thymosin α1 therapy may be effective in improving clinical outcomes in a targeted population of severe sepsis.57 Interestingly, an ongoing study of ImmunoSep, patients with sepsis were treated with the inflammation inhibitor IL-1RA or immunostimulant IFN-γ depending on, respectively, elevated ferritin (indicating hyperinflammation) or decreased monocyte HLA-DR amounts (indicating immune suppression)58 and the results worth expecting. Therefore, recognising the immune endotype is very important in order to choose the appropriate immunotherapeutic approach for each patient resulting in the best chance to improve the outcome. The novel treatment strategy with TLR4 inhibitors, such as the use of RKH, should target septic patients with a prominent inflammatory status to improve outcomes. Despite RKH demonstrated effectiveness in protecting against murine and piglet sepsis, there exist numerous obstacles hindering its clinical translation. The potential usage of RKH at affecting adaptive immunity needs to be carefully evaluated in the future although we proved RKH would not affect T cells function and death in vitro.
Of note, our data showed that RKH may perform better in the CLP model than the LPS model (figure 2I vs figure 2M). Beside systemic inflammation, compared with the LPS model, the pathogenesis of the CLP sepsis model is involved in polymicrobial dissemination. We speculated that it remains unclear whether RKH offers sepsis protection by influencing bacterial elimination in the development of CLP-induced sepsis. Thus, future research is needed to explore the impact of RKH on bacterial clearance to provide additional insights of the beneficial roles of RKH. In addition, the treatment of sepsis requires a combination of bacterial clearance, maintaining immune balance and protecting important organ functions. Hence, the combined approaches such as antibiotics usage combined with anti-inflammatory compounds including RKH may increase the efficiency of therapy.
In conclusion, we provided novel evidence of the host–microbiome interaction in the context of sepsis development. We expanded the pathophysiological role of live AKK by revealing a novel bioactive tripeptide, RKH, generated by live AKK. RKH may serve as a potentially effective treatment for lethal sepsis if exerting protective effects on sepsis in future clinical practice, especially for patients in a hyperinflammatory state.
Supplemental material
Supplemental material
Supplemental material
Data availability statement
Data are available on reasonable request. All data relevant to the study are included in the article or uploaded as online supplemental information. Transcriptome data and 2bRAD sequencing for the Microbiome of mice are available in the Genome Sequence Archive (GSA) database: Bioproject PRJCA017617 and PRJCA017573. Raw data not included therein can be obtained with the consent of the corresponding author.
Ethics statements
Patient consent for publication
Ethics approval
This study involves human participants and ethical approval was obtained from the Medical Ethics Committee of Nanfang Hospital, Southern Medical University (approval number NFEC-2021-139) and the Ethics Committee of the First People’s Hospital of Foshan (approval number FSYYYEC-2021-106). All procedures involving animals were approved by the Animal Care and Use Committee of Southern Medical University (Guangzhou, China, approval number SMUL2022308) and the ARRIVE reporting guidelines about animal preclinical studies were used in this study. Participants gave informed consent to participate in the study before taking part.
Acknowledgments
The authors sincerely thank all participants who have contributed to this study.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
- Data supplement 1
- Data supplement 2
- Data supplement 3
- Data supplement 4
- Data supplement 5
- Data supplement 6
- Data supplement 7
- Data supplement 8
- Data supplement 9
- Data supplement 10
- Data supplement 11
- Data supplement 12
- Data supplement 13
- Data supplement 14
- Data supplement 15
- Data supplement 16
- Data supplement 17
- Data supplement 18
- Data supplement 19
- Data supplement 20
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
- Data supplement 1
- Data supplement 2
- Data supplement 3
- Data supplement 4
- Data supplement 5
- Data supplement 6
- Data supplement 7
- Data supplement 8
- Data supplement 9
- Data supplement 10
- Data supplement 11
- Data supplement 12
- Data supplement 13
- Data supplement 14
- Data supplement 15
- Data supplement 16
- Data supplement 17
- Data supplement 18
- Data supplement 19
- Data supplement 20
Footnotes
SX, JL, FL and QX contributed equally.
Correction notice This article has been corrected since it published Online First. The third author affiliation has been updated.
Contributors SX, JL, FL, PG, YuC, MC, JB, RW, YouD, HW and YonD performed the experiments and analysed the data; XZ participated in the bioinformatics analysis; ZZ, ZC, K-XL, WG and QX collected clinical data; ZL and JZ participated in bacterial culture experiments; YeC, YiD, YJ, H-WZ and PC designed the study, interpreted the data, drafted and edited the manuscript, and supervised the study. PC is the guarantor for this paper. All authors read and approved the final manuscript.
Funding This study was supported by the National Key R&D Program of China (2022YFA0806400), and the National Natural Science Foundation of China (32071124, 32271230) to PC. National Natural Science Foundation of China (82130063), Special Support Plan for Outstanding Talents of Guangdong Province (2019JC05Y340) to YJ.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.