Introduction

The incidence and prevalence of type 2 diabetes mellitus (T2DM) are on the rise globally (Khosla et al. 2021), constituting approximately 90% of all diabetic cases (Kerner and Brückel 2014). Moreover, T2DM is accompanied by chronic, low-grade inflammation, impaired immune response, and changes in the composition and function of the gut microbiome (Gutiérrez-Salmerón et al. 2021). Conversely, inflammation can disrupt gut structure, leading to insulin resistance (IR) and reduced insulin secretion (Dyer et al. 2020; Rohm et al. 2022). Additionally, intestinal microorganisms can induce metabolic dysfunction and compromise the barrier function of the intestinal epithelium through their own activities or metabolites. This may affect the mucosal immune response of the intestines (Sittipo et al. 2018). Researchers have demonstrated that retinoic acid-inducible gene-I-like receptors (RLRs) function as pattern recognition receptors (PRRs), which are responsible for the body’s first line of defense against pathogenic microorganisms (Wicherska-Pawłowska et al. 2021). Previous studies have indicated that RLRs can induce innate immunity by identifying and combining microbe-associated molecular patterns (MAMPs) (Medzhitov 2007). Furthermore, they can initiate the RLRs signaling pathway (Kelly et al. 2004), inducing microbicidal activities that can change the composition of the gut microbiome (Privitera et al. 2021). Studies have found that commensal anaerobic bacteria in the gut can attenuate the inflammatory response by regulating NF-κB subunit p65 (RELA) of the RLRs signaling pathway. Moreover, excessive activation of RLR signaling is implicated as a pivotal factor in either triggering or facilitating the onset of autoinflammation (Zhao et al. 2017). Additionally, the microbiome serves as an extensive chemical factory, producing numerous compounds essential for its own sustenance and that of its hosts. Given that these compounds can reach the liver through the portal circulation system (Saeedi et al. 2020) and subsequently circulate throughout the body via hepatic veins, capillaries, etc., they exert a remarkable influence on the entire holobiont (Fan and Pedersen 2021). Research indicated that the intestinal microbiome is linked to diverse diseases, including metabolic syndrome, diabetes, inflammation, and cardiovascular diseases (Nicholson et al. 2012).

Tryptophan stands as an indispensable amino acid that is crucial for the human body, with its sole dietary source being derived from food intake. Approximately 4–6% of tryptophan undergoes direct metabolism by gut microbiota (Yokoyama and Carlson 1979). Furthermore, the metabolism of tryptophan plays a pivotal role in shaping the diversity of the intestinal microbiome (Gao et al. 2018). Studies have highlighted the anti-inflammatory properties of tryptophan metabolites, such as indole and skatole, in mammals (Marsland 2016). Notably, these microbial derivatives of tryptophan activate signaling pathways regulating the differentiation and function of various immune cells, including T cells (Liu et al. 2018), B cells (Piper et al. 2019), innate immune cells (ILCs) (Zelante et al. 2014), and CD4+CD8αα+ lymphocytes (Cervantes-Barragan et al. 2017). This activation contributes to sustaining the dynamic balance and functionality of immune cells (Peterson and Artis 2014). Research has demonstrated that rats subjected to a low-tryptophan diet exhibited an increased susceptibility to inflammation (Hashimoto et al. 2012). Yin and Yang deficiency syndrome, a characteristic of traditional Chinese medicine (TCM) syndrome associated with IR, is notably prevalent in patients with T2DM (Xie et al. 2011). A clinical trial revealed that, in comparison to other T2DM syndromes, the proportion of patients with Yin and Yang deficiency syndrome was the lowest, and those patients exhibited an older average age. Distinctive clinical symptoms, such as thirst, frailty, nocturnal polyuria, soreness, and weakness in the knees and waist, differentiate Yin and Yang deficiency syndrome from other syndromes of T2DM (Zhang et al. 2021). This suggests that the Yin and Yang deficiency syndrome may represent a prolonged course of T2DM.

Recent research demonstrated that the microbial tryptophan derivative, indole, has the potential to regulate glucagon-like peptide-1 (GLP1) expression. This regulatory action contributes to reducing fasting plasma glucose (FPG) levels by stimulating pancreatic β cells to secrete insulin through the activation of enteroendocrine L cells. Moreover, microbial tryptophan metabolites play a crucial role in shaping the composition of the intestinal microbiome and influencing the immune response in the intestinal mucosa. This leads to notable anti-inflammatory effects and the modulation of interactions between the intestinal microbiome and inflammatory responses (Nicholson et al. 2012). Specifically, Bacteroides and Faecalibacterium have been reported to attenuate the inflammatory response by impacting the RLRs signaling pathway through RELA (Cervantes-Barragan et al. 2017; Kelly et al. 2004). Furthermore, a study indicated that both native immune cells and adaptive immune cells establish intricate connections among T2DM, gut microbiota, and immune dysfunction (Rohm et al. 2022). It is noteworthy that PRRs, located in the cell membrane or cytoplasm, are essential prerequisites for immune cells to initiate an immune response (Ori and Kawai 2023). Remarkably, despite the intricate interplay among gut microbiota, microbial tryptophan metabolites, prolonged courses of T2DM, and the innate immune response with RLRs as a focal point, no research has comprehensively examined this relationship. Therefore, the present study aimed to reveal this complex relationship, employing bacterial gut microbiome, untargeted metabolomics, and RNA sequencing (RNA-seq) through a multi-perspective and integrated analysis. This research may provide a foundation for potential therapeutic interventions for extended courses of T2DM.

Materials and methods

Human subjects

This study and experimental protocol have been approved by the Ethics Committee of Beijing University of Chinese Medicine (BUCM) under approval number #2017BZHYLL0105. Prior to enrolment, each participant signed a written informed consent form.

The research involved 14 patients with T2DM and seven healthy controls, all recruited from Hepingli Hospital, Beijing, China. After that, the T2DM patients were then categorized into two groups: the YYLXD group (IDs: YYLXD001, YYLXD002, YYLXD003, YYLXD004, YYLXD005, YYLXD006, and YYLXD008), and the NYYLXD group (IDs: QYLXD001, QYLXD002, QYLXD003, SRKPD001, SRKPD002, YXRSD001, and YXRSD002). The healthy participants formed the normal (N) group (IDs: ZC001, ZC002, ZC003, ZC004, ZC005, ZC006, and ZC007). Each T2DM patient fulfilled diagnostic criteria with an FPG level of ≥ 7.0 mmol/L and a hemoglobin A1C (HbA1C) level of ≥ 6.5%. Additionally, following the guidelines for clinical research on new Chinese medicines for diabetes treatment set forth by the Ministry of Health of China in 2002 (Zhang et al. 2021), T2DM patients underwent simultaneous evaluation by three clinical associate chief and director physicians. This comprehensive assessment considered clinical symptoms, tongue appearance, and pulse patterns and ensured the absence of complications. Subsequently, these patients were categorized into distinct TCM syndromes based on referenced literature (Xu et al. 2013; Yang et al. 2010). Conversely, healthy subjects displayed no history of metabolic diseases, maintained standard vital signs, and exhibited normal screening laboratory results (e.g., FPG, HbA1c, total cholesterol (TC), triglycerides (TG), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C)). No antibiotics were utilized in the preceding three days, and no chronic conditions, such as heart failure, pulmonary disease, renal failure, or gastrointestinal disorders, occurred.

Venous blood and human stool samples were collected from both healthy volunteers and T2DM patients. The peripheral venous blood underwent laboratory analysis, while the human stool samples were promptly frozen and transported on dry ice to the Novegene Laboratory (Beijing, China). Subsequently, they were stored at −80 °C for metabolite and DNA extraction.

Experimental animals

All animal procedures and experimental protocols received approval from the Ethics Committee of the BUCM. For the experiments, age-matched C57BL/6J SPF mice were utilized, which were attained from Beijing Vital River Laboratory Animal Technology (Beijing, China). All mice were aged between 4–6 weeks upon the commencement of the study. They were housed under specific pathogen-free conditions at BUCM, maintained at a temperature of 20–24 °C, with a 12-h light-dark cycle, 35% ± 5% humidity, and granted unrestricted access to tap water. Their diet consisted of a standard rodent feed (Cat#1022, Beijing Huafukang Biotechnology, Beijing, China).

The experiments involved a total of 16 mice, with eight mice subjected to an 8-week high-fat diet (HFD) to induce IR, forming the model (M) group. The remaining eight mice, consuming a regular rodent diet (Cat#1022, Beijing Huafukang Biotechnology, Beijing, China), comprised the normal group (N). After a 12-h fasting period, mice from the model group received an intraperitoneal injection of freshly prepared streptozotocin (STZ) (dissolved in 0.1 M citrate buffer, pH 4.5) for 24 h to induce mild insulin secretion deficiency and subsequent hyperglycemia. The control group received equivalent volumes of saline injections. After 7 days, tail vein blood sampling was performed to examine the FPG level. Mice with an FPG level exceeding 11.1 mmol/L after 6 h were identified as diabetic. Throughout the experimental period, the FPG level was routinely monitored at bi-weekly intervals.

16S rRNA gene amplicon sequencing

Genomic DNA was extracted from human stool samples using the cetyltrimethylammonium bromide (CTAB) method (Bulgarelli et al. 2015), and the concentration and purity of the extracted DNA were assessed on 1% agarose gels. Depending on its concentration, the DNA was diluted to 1 ng/μL with sterile water. Specific primers (e.g., 16S V4: 515F806R) with a barcode were employed to amplify distinct subregions of the 16S rRNA genes. All PCR reactions were performed with 15 μL of Phusion® High-Fidelity PCR Master Mix (New England Biolabs, Ipswich, MA, USA), 2 μM of forward and reverse primers (Hess et al. 2011; Youssef et al. 2009), and approximately 10 ng template DNA. The thermal cycling was commenced with an initial denaturation at 98 °C for 1 min, followed by 30 cycles of denaturation at 98 °C for 10 s, annealing at 50 °C for 30 s, and elongation at 72 °C for 30 s. Finally, the last step was at 72 °C for 5 min. The same volume of 1× sample loading buffer was mixed with PCR products and the mixture was loaded onto a 2% agarose gel for detection by electrophoresis. The agarose band containing the mixture of PCR products was purified using a Qiagen gel extraction kit (Qiagen, Hilden, Germany). Sequencing libraries were generated via the TruSeq® DNA PCR-free sample preparation kit (Illumina, San Diego, CA, USA) following the manufacturer’s instructions, and index codes were incorporated. The final library was sequenced on an Illumina NovaSeq platform, producing 250-bp paired-end reads.

Untargeted metabolomics

Fecal samples (5–10 g) from subjects were collected in the morning. These samples were individually ground with liquid nitrogen, and the homogenate was resuspended with prechilled 80% methanol by vortexing. After incubation on ice for 5 min, the samples were centrifuged at 15,000 g for 20 min at 4 °C. A portion of the supernatant was diluted to achieve a final 53% methanol concentration using chromatography-mass spectrometry (LC-MS) grade water. The samples were subsequently transferred to a fresh Eppendorf tube and centrifuged at 15,000g for 20 min at 4 °C. The resulting supernatant was analyzed using the LC-MS/MS system. Ultra high-performance liquid chromatography (UHPLC)-MS/MS analysis was conducted using a Vanquish UHPLC system (Thermo Fisher Scientific, Waltham, MA, USA) coupled with an Orbitrap Q ExactiveTMHF mass spectrometer (ThermoFisher, Braunschweig, Germany). Raw data files generated by UHPLC-MS/MS were processed using Compound Discoverer 3.1 (CD3.1, Thermo Fisher Scientific) for peak alignment, peak picking, and quantitation of each metabolite. Metabolites were annotated using the KEGG (https://www.genome.jp/kegg/pathway.html), HMDB (https://hmdb.ca/metabolites), and LIPIDMaps (http://www.lipidmaps.org/) databases.

RNA-seq of human peripheral blood

Notably, 7–8 ml of fasting venous blood was collected from each subject and dispensed into ethylenediaminetetraacetic acid (EDTA) anticoagulation tubes. The RNA degradation and contamination were monitored on 1% agarose gels. The purity of the RNA was determined using the NanoPhotometer® spectrophotometer (IMPÜLEN. Westlake Village, CA, USA). The integrity of the RNA was verified using the RNA nano 6000 assay kit of the Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA, USA). For RNA sample preparation, 1 μg of RNA from each sample was utilized. Sequencing libraries were generated using NEBNext®UltraTM RNA library prep kit for Illumina® (NEB, Ipswich, MA, USA) on the basis of the manufacturer’s protocol, and specific index codes were added to link sequences to their respective samples. The clustering of the index-coded samples was performed on a cBot Cluster Generation system using TruSeq PE Cluster kit v3-cBot-HS (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions. After clustering, the library preparations were sequenced on an Illumina NovaSeq platform, and 150-bp paired-end reads were generated.

Histological assessment

After embedding, sectioning, and dewaxing, the colon, liver, and kidney tissues were stained with hematoxylin and eosin (H&E). The pathological characteristics of the liver, kidney, and colon tissues were then scrutinized under a microscope (magnification × 20, U-LH100-3, Olympus, Tokyo, Japan). Finally, we performed an inflammation score concerning this literature (Cheng et al. 2008; Rank et al. 1995).

Western blot analysis of murine colon

Subsequently, 50 mg of colon tissue was homogenized in 500 μL of radioimmunoprecipitation assay (RIPA) lysate with protease inhibitor, and the final homogenate was collected. After centrifugation at 12,000 rpm for 5 min at 4 °C, the clear supernatant was collected for further analysis. Protein concentrations were determined using the bicinchoninic acid assay (BCA) kit (LABLEAD, Beijing, China). Thereafter, 20 μg of protein for each sample was mixed with 5× loading buffer, respectively, and boiled for 5 min. Separated proteins were transferred to polyvinylidene fluoride (PVDF) membranes, and they were subsequently blocked with 5% skimmed milk in Tween-20 (TBST) for 1.5 h at room temperature. After thrice washing with 1× Tris-buffered saline with TBST for 10 min each, the PVDF membranes were incubated with primary antibodies [rabbit anti-TANK (1:1000; ABclonal, Woburn, MA, USA) and rabbit anti-NFκB2 (p100/p52) (1:600; Proteintech. Rosemont, IL, USA)] overnight at 4 °C. Following additional thrice washing with 1× TBST, the PVDF membranes were incubated with the corresponding secondary antibodies at room temperature for 1 h. After extensive washing with 1× TBST, proteins were visualized by enhanced chemiluminescence (ECL) detection. Densitometric analysis was carried out using the Image J software.

Immunohistochemistry (IHC)

The colon, liver, and kidney tissues were fixed in 10% neutral buffered formalin (Beijing Yili Fine Chemical Co., Ltd., Beijing, China), followed by embedding into paraffin. The paraffin-embedded tissue sections were sectioned into 4-μm-thick slices utilizing a rotary sectioning machine (Leica Biosystems, Wetzlar, Germany). These slices underwent a dewaxing process with adhesive sheets and were subsequently baked at 60 °C in an oven (Beijing Yongguangming Medical Appliance Factory, Beijing, China). The immunohistochemical procedures encompassed antigen retrieval, a blocking procedure employing normal goat serum (10% in TBST), and incubation with primary antibodies [rabbit anti-GLP1 (1:100; ABclonal, Woburn, MA, USA) and rabbit anti-NFκB2 (p100/p52) (1:100; Proteintech, Rosemont, IL, USA)]. Visualization of the results was achieved using appropriate secondary antibodies and diaminobenzidine as the chromogen. The expression levels of the corresponding antibodies in the tissues were determined under a laboratory microscope (U-HL100.3, Olympus, Tokyo, Japan) at magnifications of 20×/40×. The mean optical density values (IOD) of the positive reactions in each tissue were quantified using the Image J software for statistical evaluation.

Statistical analysis

To examine the differences between groups, comparisons between two groups following normal distribution were performed using a two-tailed student’s t-test, while one-factor ANOVA was used for comparisons of three or more groups. For nonparametric distributions, the Kruskal–Wallis H was performed for comparisons between two groups or three or more groups, respectively. All data are presented as mean ± SD, as indicated. Statistical tests used to compare conditions are indicated in figure legends (*P < 0.05; **P < 0.01; ***P < 0.001), where P < 0.05 was considered significant. R version 3.4.3 (https://www.r-project.org/), Python version 2.7.6 (https://www.python.org/), SPSS 26.0 (Statistical Package for the Social Sciences Software, SPSS Inc., Chicago, IL, USA), Image J (https://sourceforge.net/projects/imagej/), and Cytoscape version 3.8.2 (https://cytoscape.org/) were used for the generation of graphs and statistics.

Results

Clinical baseline characteristics of the subjects

A total of 21 subjects were enrolled in this study, including seven healthy subjects as the N group, 7 patients with T2DM who had a longer course as the YYLXD group, and 7 patients with T2DM and other TCM syndromes in the NYYLXD group. Compared to the N group, FPG (P < 0.05), HbA1C (P < 0.05), and age (P < 0.01) were all significantly increased in the YYLXD group; FPG (P < 0.01) was significantly increased in the NYYLXD group. Compared to the NYYLXD group, age (P < 0.05) was significantly increased in the YYLXD group (Table 1).

Table 1 Clinical characteristics of the patients, n = 7; values are presented as mean ± SD
Full size table

Tryptophan derivatives and prolonged T2DM

Untargeted metabolomics of the subjects’ fecal samples showed that three tryptophan derivatives, skatole, 5-methoxyindoleacetic acid (5MIAA), and 5-methoxytryptamine were screened in the YYLXD versus the N comparison group, but no tryptophan derivatives were screened in the YYLXD versus NYYLXD comparator group. Compared to the normal group, skatole (P < 0.01), 5MIAA (P < 0.01) and 5-methoxytryptamine (P < 0.05) were all significantly decreased in the YYLXD group (Table 2). Of these, the greater downward revision of skatole was relatively (Fig. 1a and b). In addition, 5MIAA was significantly negatively correlated with both FPG (P < 0.01) and HbA1c (P < 0.01). Skatole was significantly negatively correlated with HbA1c (P < 0.001) and negatively correlated with FPG (P < 0.05) (Fig. 1c and d).

Table 2 Tryptophan derivative information in the YYLXD and N groups
Full size table
Fig. 1
figure 1

Tryptophan derivatives and prolonged T2DM a 5MIAA data distribution and b skatole data distribution between YYLXD and N groups. c and d Correlation analysis between tryptophan derivatives and clinical and biochemical indicators of T2DM. *P < 0.05;**P < 0.01 as determined by Spearman analysis. P < 0.05 was considered significant

Full size image

Gut microbiome and prolonged T2DM

The results showed that there were differences in the composition of the bacterial gut microbiome between the YYLXD, N, and NYYLXD groups (Fig. 2a). Compared to the normal group, eight significantly different enteric genera (P < 0.05) were significantly decreased in the YYLXD group (Fig. 2b). There were no significantly different intestinal microorganisms between the YYLXD versus NYYLXD groups. In addition, Paraprevotella (P < 0.01), Faecalibacterium (P < 0.01), Alistipes (P < 0.05), Barnesiella (P < 0.01), Bacteroides (P < 0.05), Lachnospira (P < 0.01), and Fusicatenibacter (P < 0.05) were all significantly negatively correlated with HbA1c; Faecalibacterium (P < 0.05) and Lachnospira (P < 0.05) were negatively correlated with FPG (Fig. 2c). All of the above gut microorganisms were correlated with N, YYLXD, and NYYLXD groups (Fig. 2d).

Fig. 2
figure 2

T2DM correlates with changes in the community structure of the intestinal bacterial microbiome. a and c Composition differences in gut microorganisms and correlation analysis of significantly different gut microorganisms and clinical and biochemical indicators of T2DM between YYLXD, N, and NYYLXD comparison groups at the genus level. *P < 0.05; **P < 0.01 as determined by Spearman analysis. P < 0.05 was considered significant. b Gut microorganisms with significant differences between the YYLXD and N groups at the genus level. d Relationship between significant gut microorganisms and the YYLXD, N, and NYYLXD groups

Full size image

Gut microbiome and tryptophan derivatives

To explore the relationship between significant gut microorganisms and tryptophan derivatives, we performed a correlation analysis between them. The results showed that Fusicatenibacter (P < 0.01), Faecalibacterium (P < 0.01), Lachnospira (P < 0.01), and Dorea (P < 0.05) were all positively correlated with 5MIAA; Fusicatenibacter (P < 0.05), Faecalibacterium (P < 0.01), Lachnospira (P < 0.01), Paraprevotella (P < 0.05), Barnesiella (P < 0.05), Bacteroides (P < 0.01) were all positively correlated with skatole. In addition, Bacteroides (P < 0.05) had a positive correlation with 5-methoxytryptamin (Fig. 3a). Besides, compared to the normal group, the expression of the gene for Dopa decarboxylase ( DDC) (P < 0.01) was significantly decreased in the YYLXD group (Fig. 3b ). Compared to the NYYLXD group, DDC (P < 0.01) was significantly decreased in the YYLXD group (Fig. 3c).

Fig. 3
figure 3

Correlation between significant gut microorganisms and tryptophan derivatives. *P < 0.05; **P < 0.01 as determined by Spearman analysis. P < 0.05 was considered significant. b The expression of DDC between YYLXD and N groups. c The expression of DDC between YYLXD and NYYLXD groups

Full size image

RNA changes of the RLRs signaling pathway in the peripheral blood of subjects

In the present study, RNA-seq was performed on the peripheral blood samples. The results indicated that 8 significantly different mRNAs and 44 long non-coding RNAs (lncRNAs) of the RLRs signaling pathway were identified between the YYLXD group and the normal group. Among these, compared to the normal group, TANK (P < 0.05), TANK-012 (P < 0.01), RELA (P < 0.05), and RELA-005 (P < 0.05) were all upregulated in the YYLXD group (Table 3 and Supplemental Table S1). Additionally, a total of 23 significantly different mRNAs and 9 LncRNAs of the RLRs signaling pathway were identified between the YYLXD and NYYLXD groups. Compared to the NYYLXD group, TANK (P < 0.01) and TANK-013 (P < 0.05) were upregulated in the YYLXD group (Table 4 and Supplemental Table S2). In both pairs of comparison groups, the relative variability was greater for TANK. TANK encodes the TRAF family member-associated NF-κB activator.

Table 3 mRNA with significant differences in the RLRs signaling pathway in the peripheral blood of subjects between the YYLXD and N groups
Full size table
Table 4 mRNA with a significant difference in the RLRs signaling pathway in the peripheral blood of subjects between the YYLXD and NYYLXD groups
Full size table

Correlation analysis

To further investigate the correlation between tryptophan derivatives, significantly different gut microbes, and RNAs of the RLRs signaling pathway, a combined multi-omics analysis was performed. As lncRNAs can regulate the expression levels of target mRNAs through co-localization and co-expression, a lncRNA-mRNA relationship network was constructed. The results revealed that tryptophan derivatives and significantly different gut microbes were correlated with mRNAs of the RLRs signaling pathway (Fig. 4a), and there was a correlation between lncRNAs and their target mRNAs (Fig. 4b). Of these, 5MIAA and skatole were significantly negatively correlated with TANK (P < 0.001) and TANK-012 (Fig. 4f and d). Faecalibacterium and Bacteroides were negatively correlated with RELA (P < 0.05) and RELA-005. Lachnospira and Paraprevotella were negatively correlated with TANK (P < 0.05) and TANK-012 (Fig. 4e and c).

Fig. 4
figure 4

Correlation analysis. a Correlation analysis of significantly different gut microorganisms, tryptophan derivatives, and differential RLRs signaling pathway mRNAs. b Correlation analysis of differential lncRNAs and their targeted mRNAs in the RLRs signaling pathway. c Correlation analysis of significantly different gut microorganisms and differential lncRNAs. d Correlation analysis of tryptophan derivatives and differential lncRNAs. e Correlation analysis of significantly different gut microorganisms and significantly different RLRs signaling pathway mRNAs. f Correlation analysis of tryptophan derivatives and significantly different RLRs signaling pathway mRNAs. *P < 0.05; **P < 0.01 as determined by Spearman analysis. P < 0.05 was considered significant

Full size image

Changes in pathology and inflammation-related genes in the colon of mice modeled with T2DM

Subsequently, an animal model of prolonged T2DM was established. The results showed that compared to the N group, FPG (P < 0.01) was significantly increased in the model (M) group (P < 0.01), and the FPG value of the model group was more than 30.00 mmol/l (Fig. 5m). In addition, the colonic intestinal glands were neatly arranged without wrinkles, and there was no obvious inflammatory cell infiltration in the N group. Compared with mice in the N group, mice in the prolonged T2DM group exhibited abnormalities in the gut wall and intestinal glandular folds. Additionally, there was a significant presence of infiltrating inflammatory cells (Fig. 5a). Compared with the normal group, the inflammation scores (P < 0.01) of colonic tissues of mice were significantly increased in the model group (Fig. 5j). In addition, compared with the normal group, the expression of TANK (P < 0.05), p100NFκB2 (P < 0.05) and p52NFκB2 (P < 0.05) were all increased in the colonic tissues of mice in the M group ( Fig. 5n–q) while the expression of GLP1 (P < 0.01) was significantly decreased (Fig. 5d).

Fig. 5
figure 5

FPG value of mice, H&E staining, and expression of inflammation-related genes and GLP1 between N and M groups. H&E staining and inflammation score of a and j colon, b and k liver, and c and l kidney tissues in N and M groups. m FPG values of mice. d and g Expression of GLP1 in the colon between N and M groups. Expression of NFκB2 in e and h liver and f and i kidney tissues of mice. n–q Expressions of p100NFκB2, p52NFκB2, and TANK in the colon between N and M groups.*P < 0.05; **P < 0.01 as determined by Kruskal–Wallis H and one-factor ANOVA. P < 0.05 was considered significant

Full size image

Changes in liver and kidney pathology and inflammation-related genes in mice modeled with a longer course of T2DM

We performed H&E staining on the liver and kidney of mice and examined the expression of NFκB2 in murine liver and kidney using IHC. The result showed that in the N group, the liver tissue exhibited an intact lobular structure, with hepatocytes arranged radially around the central vein and centrally located nuclei. However, versus the N group, mice in the M group displayed recognizable hepatic lobules, while with disorganized hepatic cords. The central vein was congested with blood, and some of the cell nuclei were displaced to one side (Fig. 5b). Compared with the normal group, the inflammation scores (P < 0.05 ) of liver tissues were significantly increased in the model group (Fig. 5k); Regarding renal structures, the renal glomeruli and tubules of the normal mice were regularly shaped, with the epithelial cells of the tubules neatly arranged and intact, and homogeneous cytoplasm. In contrast, in the M group versus the N group, the glomeruli and tubules were congested with blood. Additionally, the tubular epithelial cells were swollen and disorganized (Fig. 5c). Meanwhile, compared with the normal group, the inflammation scores (P < 0.01) of kidney tissues of mice were significantly increased in the model group (Fig. 5l). In addition, compared with the normal group, the expression of NFκB2 (p100/p52) (P < 0.01) was increased in both kidney and liver tissues in the model group (Fig. 5i and h). The transfer of p52 NFκB2 to the nucleus was observed in both liver and kidney tissues (Fig. 5e and f).

Gut microbiome of murine

To verify the relationship between inflammatory response and intestinal microbiome and its structural changes at the animal level, 16S rRNA gene amplification sequencing was conducted on animal feces. The results showed that there were six intestinal microorganisms with significant differences between the N and the M groups. Compared with the normal group, Allobaculum (P < 0.01) and Ligilactobacillus (P < 0.01) were all significantly decreased in the M group; Faecalibaculum (P < 0.01), Helicobacter (P < 0.01), Romboutsia (P < 0.01), and Alistipes (P < 0.05) were all significantly increased (P < 0.01) ( Fig. 6).

Fig. 6
figure 6

Gut microorganisms with significant differences between N and M groups (genus level). The abundance of a Allobaculum, b Faecalibaculum, c Helicobacter, d Ligilactobacillus, e Romboutsia, and f Alistipes between N and M groups.*P < 0.05; **P < 0.01 as determined by MetaStat. P < 0.05 was considered significant

Full size image

Discussion

Gut microbes, such as Bacteroides can metabolize tryptophan to produce indole-3-acetic acid (IAA) (Smith and Macfarlane 1997) and then convert IAA to skatole by coding for enzyme DDC. Studies demonstrated that Bacteroides have positive influences on glucose metabolism (Gauffin Cano et al. 2012; Yang et al. 2017), and they can regulate glycan synthesis (Yan et al. 2022). Besides, Bacteroides may serve as one of the intestinal microorganisms that catabolize tryptophan to produce indole (Keszthelyi et al. 2009) that can stimulate the production of GLP1 by intestinal endocrine L cells, which in turn stimulates insulin secretion by pancreatic β cells and exerts hypoglycemic effect (Zhang et al. 2021). The enzyme DDC is an integral component of skatole metabolism (Ouwerkerk et al. 1999; Whitehead et al. 2008). The glucose concentration in the intestine significantly affects the activity of DDC, in turn affecting skatole production (Yokoyama and Carlson 1979). Nevertheless, an ample presence of skatole is crucial for the establishment of the microbial-indole axis (Hubbard et al. 2015).

In the present study, it was found that the proportions of Bacteroides, DDC, and skatole were significantly reduced in the YYLXD group compared with those in the N group. In addition, Bacteroides and skatole were negatively correlated with HbA1c levels. Furthermore, decreased GLP1 expression levels in the colon tissue of mice simulated the longer course of T2DM. Prior research has demonstrated that T2DM could lead to dysbiosis (Gutiérrez-Salmerón et al. 2021). This suggests that suppression of DDC activity and decrease of Bacteroides abundance may be associated with the risk of T2DM, which has a longer course and results in the reduction of skatole production. Moreover, a low abundance of Bacteroides or reduced skatole production may further lead to increased blood glucose levels by downregulating the GLP1 expression level.

In addition, Kang et al. (2019) found that the abundance of Alistipes was positively correlated with TC and TG levels in the blood (Kang et al. 2019). Previous studies have shown that Alistipes could indirectly activate gluconeogenesis by producing butyrate to participate in the process of glucose metabolism (Vadder et al. 2014; Wu et al. 2022). Moreover, scholars have demonstrated that diabetes and blood lipids exhibit a correlation, where TC and TG are integral components of blood lipids (Jia et al. 2021). An elevated TG level contributes to the increased IR, leading to reduced glucose uptake and elevated blood glucose level (Blüher et al. 2001; Dresner et al. 1999). Additionally, IR can further elevate TG levels (Boden 1997). Studies have indicated that the abundance of butyrate-producing bacteria decreases in diabetic mice with lipid metabolism disorders (Qin et al. 2012; Wang et al. 2020). The findings of the present study revealed a decreased abundance of Alistipes in the YYLXD group compared with the N group. Notable, TG and TC levels in the YYLXD group did not significantly differ from those in the control group. Thus, the results of the present study, associated with findings from other researchers, underscore the potential regulatory role of gut microbes in host lipid metabolism (Chen et al. 2019; Fu et al. 2015).

Bacteroides and Faecalibacterium can attenuate the inflammatory response by regulating RELA (Kelly et al. 2004; Zhu et al. 2021), and Faecalibacterium abundance is negatively correlated with the levels of inflammatory cytokines. Previous studies have concluded that RELA is a member of the NF-κB family, increasing the subsequent production of large amounts of pro-inflammatory cytokines (Karunakaran et al. 2021). Moreover, RELA is one of the RLRs signaling pathway-associated genes (Ding et al. 2017). The results of the present study indicated that Bacteroides and Faecalibacterium were negatively correlated with RELA. Besides, RELA expression level was upregulated, although the abundances of Bacteroides and Faecalibacterium were reduced in the YYLXD group. Therefore, these findings suggest that in prolonged T2DM, low abundances of Bacteroides and Faecalibacterium may trigger the RLRs signaling pathway by upregulating RELA expression level, which subsequently induces the release of inflammatory cytokines.

To further clarify the relationship among the RLRs signaling pathway, gut microbiota, immune dysfunction, and prolonged T2DM, an animal model of T2DM was established by intraperitoneal injection of STZ. STZ-induced T2DM is a widely accepted model for the assessment of the pathobiology of T2DM (Bathina and Das 2018). To conform to the clinicopathological course of T2DM with complications (e.g., fatigue and frailty, nocturnal polyuria, soreness, weakness of the waist and knees, dizziness and discomfort, and thin stools), the extended modeling time displayed symptoms, such as emaciation, huddling, slow movement, and lackluster body hair. Research indicated that inflammation can impact the kynurenine pathway of tryptophan metabolism, inducing significant alterations in the indolamine-2,3-dioxygenase enzyme and influencing the production of neuroactive metabolites (Widner et al. 1997). This process plays a crucial role in the development of neuropsychiatric symptoms, including fatigue and dizziness (Kim et al. 2015). Furthermore, the characteristics of physical frailty, weakness, and slowness are regulated by neuromuscular action (Westbrook et al. 2020). Gonzalez-Freire et al. (2014) discovered that chronic inflammation could lead to the decline of neuromuscular function, encompassing spinal neurons, peripheral axons, and muscle tissue (Gonzalez-Freire et al. 2014). The elevated NF-κB expression level can induce muscle and bone weakness. Additionally, skatole has the potential to inhibit inflammatory reactions by suppressing lipid oxidation (Adams et al. 1987; Liu et al. 2020). It can be inferred that the physical frailty and weakness of the waist and knees may be associated with the decline of neuromuscular function induced by the low levels of skatole.

Nocturnal polyuria is one of the main clinical symptoms of T2DM patients with a longer course (Guo et al. 2014). As the kidney is an important excretory organ, the metabolites produced by intestinal microorganisms may first enter the liver via the portal circulation system that may influence the whole organism through the capillary network, and skatole can affect the liver homeostasis (Saeedi et al. 2020). The expression level of NF-κB in murine kidneys was identified in the present study to clarify whether the phenomenon of nocturnal polyuria could be associated with the activation of the NF-κB signaling pathway by low levels of probiotics, causing inflammatory damage to the kidneys. Additionally, the expression levels of inflammation-related genes in the murine colon and liver tissues were detected to determine whether the NF-κB signaling pathway, which was triggered by the low skatole production, could cause inflammatory damage to the liver. The results revealed that upregulation of TANK expression level in the animal’s colon tissue (consistent with RNA-seq results in human peripheral blood samples) activated NFκB2, resulting in inflammatory damage to the liver, kidney, and colon. TANK is a member of the TRAF family that is associated with the NF-κB activator (Huh et al. 2020), and TANK is one of the RLR signaling pathway-associated genes (Chen and Jiang 2013; Eisenächer and Krug 2012). Both NFκB2 and RELA are members of the NF-κB family. Chawla et al. (2021) found that the NFκB2/RELA complex is present in intestinal epithelial cells (Chawla et al. 2021). RELA becomes activated, subsequently triggering NFκB2 (Ghosh and Wang 2021). In addition, it was demonstrated that NFκB2 expression level was significantly elevated in aged tissues (Bernal et al. 2014 ). Moreover, previous studies have reported that exposure to STZ alone did not activate the NF-κB signaling pathway (Wright et al. 2021). It is evident that the activation of the NFκB2 signaling observed in this study is associated with the presence of T2DM itself.

Besides, the present study revealed that dysbiosis was identified in the intestinal microbiome of mice in the M group and the intestinal microbiome of T2DM patients with a longer course, representing a significant reduction in the abundance of probiotics. Studies have demonstrated that dysbiosis is closely associated with inflammation and metabolic syndrome (Sen et al. 2017). The gut microorganisms in the normal murine tissue were mainly Allobaculum and Ligilactobacillus, and the abundances of Allobaculum and Ligilactobacillus were significantly reduced in the M group compared with those in the N group. It was reported that Allobaculum and Ligilactobacillus could produce butyric acid that has anti-inflammatory effects (Andoh et al. 1999; Yao et al. 2021; Zheng et al. 2021). On the other hand, the main human intestinal microorganisms (Bacteroides and Faecalibacterium) can also exert anti-inflammatory effects by producing butyric acid (Deb et al. 2020). It suggests that a decrease in the abundance of butyric acid-producing bacteria may induce inflammation in the longer pathological T2DM process. In addition, the abundance of pathogenic bacteria Helicobacter in the intestinal microbiome of mice in the M group was significantly elevated compared with that in the N group. Prior research has indicated that Helicobacter pylori could promote inflammation and induce indoleamine 2,3-dioxygenase 1 (IDO-1) activity (Feng et al. 2020). IDO-1 is the first rate-limiting enzyme that is responsible for the degradation of tryptophan (Barat et al. 2016). Bacteroides can exert inflammatory effects by encoding DDC to metabolize tryptophan into skatole (Smith and Macfarlane 1997). It can be concluded that in the longer pathological T2DM process, although the compositions of animal intestinal microorganisms and human intestinal microorganisms are different, they may eventually lead to dysbiosis, which is attributable to inflammation and tryptophan metabolism. Thus, the present study indicated that the RLRs signaling pathway may be triggered by intestinal flora and may further activate its downstream NF-κB signaling pathway, inducing an inflammatory response in T2DM with a longer course.

In conclusion, this study presented a fresh perspective on the intricate relationship among the RLRs signaling pathway, gut microbiota, microbial tryptophan derivatives, inflammation, and the prolonged course of T2DM. On one side, hyperglycemia could potentially suppress tryptophanase activity, and the extended duration of T2DM might reduce probiotic abundance. Conversely, the decreased abundance of Bacteroides, along with reduced DDC enzymatic activity, could result in a decline in skatole production. This reduction in Bacteroides or skatole levels could contribute to the elevated blood glucose level by decreasing GLP1 expression level. Moreover, the intestinal microbiome could act as a regulator of lipid metabolism. Additionally, a diminished abundance of probiotics could trigger an inflammatory response through the RLRs signaling pathway, subsequently activating the downstream NF-κB signaling pathway in prolonged T2DM. Throughout this process, lncRNAs, such as TANK-012 and RELA-005, could play a crucial role. These findings hold promise for the targeted use of specific microbes or microbially derived small molecules for treating T2DM with an extended course. However, certain limitations of this study should be noted, such as the small sample size and the absence of mechanistic insights, which will be addressed in future research, including experiments involving genetically modified animals and fecal microbiota transplantation in animal models of T2DM.