https://orcid.org/0000-0001-8889-3314
Abstract
Inflammatory bowel disease (IBD) is associated with F. nucleatum, and chronic stress can increase the risk of aggravation. However, whether norepinephrine (NE) can enhance the pathogenicity of F. nucleatum to aggravate dextran sulfate sodium salt (DSS)-induced colitis is unclear.
Transcriptome sequencing was used to identify differentially expressed genes in bacteria treated with NE. Affinity testing and molecular docking were applied to calculate and predict the binding of NE and Quorum sensing regulators C (QseC). The pathogenicity of Fusobacterium nucleatum treated with NE and QseC inhibitors was examined in vitro and further verified using the IBD mouse model induced by DSS.
Norepinephrine could bind to QseC directly to upregulate the quorum sensing pathway of F. nucleatum and enhance its virulence gene expression (FadA, FomA, Fap2) and invasiveness in vitro. Meanwhile, it promoted the invasion of F. nucleatum into the intestine and increased the expression of host inflammatory cytokines (IL-6, IL-1β) to aggravate colonic inflammation in IBD mice. The QseC inhibitor LED209 inhibited the effect of NE on F. nucleatum and partially restored short-chain fatty acid (SCFA)–producing bacteria (Prevotellaceae, Lactobacillaceae) to attenuate colonic inflammation in IBD mice.
Generally, the NE-QseC axis enhanced the pathogenicity of F. nucleatum through interkingdom signaling to aggravate colonic inflammation in IBD mice. We see that QseC may be a potential target for microbiota management of IBD under chronic pressure.
Lay Summary
Norepinephrine could bind to QseC directly to enhance the pathogenicity of F. nucleatum to aggravate colonic inflammation. The QseC inhibitor inhibited the effect of NE on F. nucleatum and partially restored short-chain fatty acid–producing bacteria to attenuate colonic inflammation.
Inflammatory bowel disease (IBD) is associated with F. nucleatum and chronic stress. In this study, stress-related norepinephrine could bind to QseC directly to enhanced the pathogenicity of F. nucleatum to aggravate colonic inflammation. It provides a new basis for the management of IBD patients.
Introduction
Inflammatory bowel disease (IBD), composed of ulcerative colitis (UC) and Crohn’s disease (CD), is a chronic, recurrent intestinal nonspecific inflammatory pathology in the bowel.1 Globally, IBD is prevalent and continues to grow, especially in developing countries.2 Although the etiology is still elusive, bacterial dysbiosis in the gut and related immune disorders due to the complex interactions between genetic risk factors3 and environmental exposure4,5 are considered to be the main reasons.
Fusobacterium nucleatum (Fn), an oral resident opportunistic pathogen, was previously reported to accumulate in the intestine of IBD patients compared with that in healthy controls, which was related to the course of disease, clinical activity, and refractory behavior of UC.6–9Fusobacterium nucleatum can attack the epithelial barrier and induce abnormal inflammation, such as promoting the increased secretion of interleukin (IL)-6, IL-1β, and IL-17, to exacerbate dextran sulfate sodium salt (DSS)-induced colitis.6
Long-term high perceived stress was prompted to induce the acute exacerbation of IBD.10 The neurotransmitter norepinephrine (NE), whose secretion is increased by chronic psychological stress,11 can be partly produced in the enteric nervous system and plays a central role in the stress response in mammals.12 Norepinephrine has been shown to affect the growth, movement, adhesion, invasion, biofilm formation, and virulence of a variety of bacteria.13 This hormone also disrupted the homeostasis of the gut microbiome, which triggered an immune response and promoted DSS-induced colitis.14
Prokaryotic microorganisms exposed to host hormones in the intestinal tract and host cells exposed to bacterial quorum sensing (QS) autoinducers (AIs), such as acyl-homoserine lactone, AI-2, AI-3, can communicate with each other; this is known as the “interkingdom signaling.”15 Quorum sensing regulators C (QseC), previously recognized as a bacterial AI-3 receptor, could recognize host-derived epinephrine (Epi)/NE to change bacterial behaviors with strengthened virulence.16 The inhibitor LED209, with high selectivity for QseC, reduced the virulence of S. Typhimurium and F. tularensis and displayed a protective role in mice infected with them.17
Therefore, it is worth investigating the role of NE in the biological behavior of F. nucleatum and its possibility in IBD exacerbation. In the current study, we demonstrated that NE directly bound to the periplasmatic domain of QseC and enhanced the pathogenicity of F. nucleatum, which further aggravated DSS-induced colitis.
Methods
Ethics Approval
All animal experimental protocols were performed in accordance with the relevant provisions of national experimental animal welfare ethics, which conforms with the animal protection, animal welfare, and ethical principles and was approved by the Experimental Animal Ethics Committee of the Chengdu Medical College (2023, No.005).
Bacterial Culture and Drug Treatment
The strain was isolated and identified from fresh tumor tissues of colorectal cancer (CRC) patients by our research team. Fusobacterium nucleatum was inoculated on fastidious anaerobe broth (FAB) medium (Topbio) and cultured in an anaerobic environment at 37°C.18
The bacteria in vitro were divided into 3 groups, including the untreated Fn, Fn + NE, and Fn + NE + LED209. A final concentration of 1 ng/mL NE (Solarbio) and 10 μM LED209 (Beyotime) were added, while the control group was given the same volume of solvent. After 24 hours of culture, bacterial precipitation was collected, and we waited for further detection.
Transcriptome Sequencing
The total RNA of Fn treated with/without NE was extracted to construct a library and used for sequencing. The data were compared with the reference genome by HISAT software,19 and the new transcripts were predicted by Coding Potential Calculator software.20 The level of gene expression was calculated by the RSEM package,21 and the differentially expressed genes were selected for functional enrichment analysis.22,23
Cell Culture and Invasion Experiments of F. nucleatum
The HCT116, HT-29, and Caco-2 cells were inoculated separately in 6-well plates at 2 × 105/mL per well. Bacterial precipitates collected in vitro and labeled with 5(6)-carboxyfluorescein diacetate N-succinimidyl ester (CFSE) dye were added to a 6-well plate according to an multiplicity of infection (MOI) equal to 100.24–26 After coculture for 4 hours, the rate of bacterial invasion was detected by flow cytometry.
Affinity Test
The structure of the sensor protein QseC with ID number Q8ZLZ9 was derived from the AlphaFold DB database,27 and the structure of norepinephrine was prepared. The possible conformations and directions of the ligand at the binding site were obtained using the Lamarckian genetic algorithm (LGA) by the AutoDock Vina program.28 The protein-ligand conformation and interaction patterns were analyzed using PyMol software.
Affinity Determination
A plasmid expressing the periplasmatic domain of QseC (34-161AA) was cloned and transfected to Rosetta BL21 for expression. The expressed periplasmatic peptide was purified from inclusion bodies and stained with coomassie brilliant blue after subjection on SDS-PAGE. The periplasmatic peptide of QseC was identified according to its molecular weight at 15 KD (Figure S1), followed by the direct sequencing (MS/MS) of the band cut out from the gel (Beijing Youji Technology Co., LTD, Beijing, China). The measurement of affinity between norepinephrine and periplasmatic domain of QseC was based on biolayer interferometry (BLI), as mentioned previously.29 The data were further processed by the software fortebio data analysis 12 to calculate the affinity. The affinity constant represents the binding capacity of small molecules to proteins. It can be categorized to 4 levels: high affinity (1 × 10-7 - 1 × 10-5), middle affinity (1 × 10-5 - 1 × 10-3), low affinity (0 - 1 × 10-3), no affinity (KD >0).
Animal Model and Clinical Score
Thirty male C57BL/6 mice (6 weeks, 20 ± 2g, Jicui Yaokang Biotechnology Co., Ltd.) were randomly and equally divided into 5 groups: control, DSS, Fn + DSS, NE + Fn + DSS, and NE + Fn + LED 209 + DSS. We used 3% (w/v) DSS (MP Biomedicals) solution in free drinking water to establish the IBD model.6 Based on the usage and dosage of the drug in previous studies, we pretreated the mice 2 weeks before modeling. In detail, F. nucleatum, dissolved in FAB medium to 109 CFU/mL, was administered by gavage at 10 mL/kg.d.9 The LED 209 was intragastrically administered at 20 mg/kg.d.30 Norepinephrine was injected intraperitoneally at 2 mg/kg.d 30 minutes later.31 At the same time, the control group was given the same dose of solvent intragastrically or intraperitoneally. During the period of modeling, the disease activity index (DAI) score was calculated, including the evaluation of weight, diarrhea, and hemafecia.32
Histopathology
The colon tissue was cut into 6-µm paraffin sections, and the changes in histopathology were evaluated as previously described through hematoxylin-eosin (HE) staining.33
Quantitative real-time PCR
We used TRIzol (Yeasen) to extract total RNA from the bacteria and tissues and reverse transcribed it into cDNA using a cDNA synthesis kit (Yeasen). The template was used for quantitative real-time polymerase chain reaction (qPCR) by applying the kit from Yeasen to measure the relative mRNA expression.
The primer sequence is shown as following:
18S: F: 5’-AGTCCCTGCCCTTTGTACACA-3’;
R: 5’-CGATCCGAGGGCCTCACT-3’.
IL-6: F: 5’-CTGCAAGAGACTTCCATCCAG-3’;
R: 5’-AGTGGTATAGACAGGTCTGTTGG-3’.
IL-1β: F: 5’-GAAATGCCACCTTTTGACAGTG-3’;
R: 5’-TGGATGCTCTCATCAGGACAG-3’.
IL-17: F: 5’-CTGATCAGGACGCGCAAAC-3’;
R: 5’-TCGCTGCTGCCTTCACTGTA-3’.
Microbial Fluorescence In Situ Hybridization
The colon tissue was prepared into 5 µm paraffin sections, which hybridized with a Cy3-labeled probe (FUS664: 5ʹ- CTTGTAGTTCCGC(C/T) TACCTC-3ʹ) according to the instructions of the bacterial fluorescence in situ hybridization (FISH) kit (Focobio). Five random fields per sample at ×200 magnification were used to count the number of bacteria, and the average was blindly evaluated by the observer.34
16S DNA Sequencing
We used qualified DNA samples extracted from the mixture of stool and tissue35,36 to construct libraries and sequences to analyze the differences in gut microbiota between groups.
Statistical Analysis
Continuous variables were described with the mean ± standard deviation (SD). Data were analyzed using 1-way ANOVA for multiple comparisons by Graphpad (GraphPad Prism 8.0). P < .05 was considered to be statistically significant. The analysis of gut microbiota was carried out on the QIIME (V1.80) and R software (V3.1.1).
Results
NE Upregulates the Quorum Sensing Pathway and Enhances the Pathogenicity of F. nucleatum
To systematically study the influence of NE on the biological process of F. nucleatum, a transcriptome analysis was conducted after NE incubation. In detail, a total of 273 upregulated genes were obtained, while no genes appeared to be downregulated (Figure 1A). Kyoto encyclopedia of genes and genomes (KEGG) pathway analysis showed an obviously enhanced quorum sensing (QS) process (Figure 1B). Quorum sensing is the method for intercellular communication that regulates bacterial behavior related to pathogenicity, including proliferation, virulence, motility, and biofilm formation.37 Here, qRT-PCR was used to show the increased expression of virulence genes FadA (P < .001), FomA (P < .01), and Fap2 (P < .0001) after NE treatment (Figure 1C). We next labeled F. nucleatum with CFSE and used flow cytometry to show the enhanced invasion of colorectal cancer epithelial cells after NE intervention (P < .0001; Figure 1D).
Norepinephrine upregulates the quorum sensing pathway and enhances the virulence of F. nucleatum. A, Heatmap for DEGs of F. nucleatum treated with/without NE. The level of DEGs is represented in different colors. B, KEGG pathway enrichment of DEGs. Quorum sensing pathways of F. nucleatum were significantly associated with NE intervention. C, The expression of bacterial virulence genes (FadA, FomA, Fap2) was increased by NE treatment. D, Flow cytometry showed that the number of F. nucleatum invading colon cancer cells (Caco-2, HCT 116, HT-29) increased after NE intervention. Cells cocultured without F. nucleatum were used as a negative control. DEGs: differentially expressed genes.*P < .05; **P < .01; ***P < .001; ****P < .0001.
NE Directly Binds to QseC of F. nucleatum to Enhance Its Pathogenicity
To verify whether NE could be directly recognized by QseC, we assessed its affinity by molecular docking, showing the possible interaction site at the periplasmatic domain of QseC (Figure 2A). Experimentally, the plasmid of the predicted periplasmatic domain of QseC was constructed and expressed, followed by subjection to BLI technology. The affinity constant of the 3 repeated experiments was 6.31 × 10-04 ± 5.46 × 10-04, suggesting the confirmation of interaction (Figure 2B, Table S1). To identify whether NE-induced virulence was dependent on QseC, LED 209, a small molecule inhibitor that competitively binds to QseC, was shown to attenuate the NE-induced expression of virulence genes, including FadA (P < .001), FomA (P < .0001), and Fap2 (P < .0001; Figure 2C). Additionally, the NE-enhanced invasion ability of epithelial cells was also attenuated by LED 209 incubation (Figure 2D). Taken together, these results demonstrated that NE directly binds to the periplasmatic domain of QseC and enhances the pathogenicity of F. nucleatum.
Norepinephrine directly binds to QseC of F. nucleatum to enhance its pathogenicity. A, Binding mode of the QseC protein and NE ligand. Among them, the key amino acid residues (I63, A65, R68, M69, and H100) in the active pocket of the QseC protein could form hydrogen bonding forces with small molecules of the NE ligand. B, Representative images of affinity between NE molecules and QseC protein. KD = 6.31 × 10-04 ± 5.46 × 10-04. C, LED 209 inhibited the expression of bacterial virulence genes (FadA, FomA, Fap2) in F. nucleatum treated with NE. D, The number of F. nucleatum invading colon cancer cells (Caco-2, HCT 116, HT-29) raised by NE intervention was depressed by LED 209. Abbreviations: BLI, Biolayer interferometry; KD, dissociation constant.*P < .05; **P < .01; ***P < .001; ****P < .0001.
NE Aggravates F. nucleatum-related Colonic Inflammation Through the NE-QseC Axis
Fusobacterium nucleatum was reported to aggravate colonic inflammation in IBD mice. In this study, we successfully established an IBD mouse model induced by DSS (Figure 3A). The results showed an increased tendency of DAI score on the fourth day with NE precondition compared with the Fn + DSS group, although the difference was not significant (P = .22). In addition, these scores in the NE + Fn + LED 209 + DSS group (2.67 ± 0.82) were significantly lower than those in the NE + Fn + DSS group (6.83 ± 0.75; P < .0001; Figure 3B). The changes in body weight are shown in the figures (Figure 3B).
Norepinephrine aggravates F. nucleatum-related colitis through the NE-QseC axis. A, Schematic diagram of the animal experiment process. Diverse strategies were given according to different groups. B, The DAI scores and weight loss between groups on the fourth day. Norepinephrine increased the DAI scores of the IBD model infected with F. nucleatum, which was reduced by the QseC inhibitor LED 209. C,D, The gross colon length was measured (C). Colon shortening in IBD mice administered NE and F. nucleatum was significantly alleviated with LED 209 intervention (D). E,F, In IBD mice infected with F. nucleatum, NE promoted the expression of pro-inflammatory cytokines in the host colon, including IL-1β, IL-6, and IL-17. In addition, it could be inhibited by LED 209. G, HE staining showed that IBD mice infected by F. nucleatum had greater infiltration of inflammatory cells and more crypt damage with NE intervention, which was reduced by LED 209. *P < .05; **P < .01; ***P < .001; ****P < .0001.
The length of the colon was shorter in the DSS-treated mice than in the healthy mice at the end of the experiment. However, colon shortening in the NE + Fn + DSS group was significantly alleviated with LED 209 intervention (P = .0002; Figure 3C, 3D), suggesting partial dependence on QseC. The expression of pro-inflammatory cytokines (IL-6, IL-1), promoted by F. nucleatum in the intestinal tissue, was further enhanced by pretreatment with NE (P < .05) but decreased in the NE + Fn + LED 209 + DSS group (IL-6, P < .01; IL-1β, P < .05; Figure 3E). Although there was no statistical significance, another inflammatory cytokine, IL-17, showed a similar trend (Figure 3F). Moreover, HE staining showed that NE aggravated the degree of colonic inflammation in IBD mice treated with F. nucleatum, which could be alleviated by LED 209 (Figure 3G).
The QseC Inhibitor Alleviates the Invasion of F. nucleatum Induced by NE and Partially Recovers SCFA-producing Bacteria in IBD Mice
Specific probes were used to label F. nucleatum in colonic tissues (Figure 4A). The signal was detected in all DSS-treated mice except the healthy mice. In detail, NE promoted the invasion of F. nucleatum into the intestine (P < .05), which was alleviated by LED 209 (P < .01; Figure 4B).
The QseC inhibitor alleviates the invasion of F. nucleatum induced by NE and partially recovers SCFA-producing bacteria in IBD mice. A, Representative images of FISH in each group. Among them, red is F. nucleatum, and blue is the nucleus. B, Norepinephrine promoted F. nucleatum invasion into the tissue, which was restrained by LED 209. C, Venn diagrams based on OTU clustering. D, Analysis of the microbial α diversity between groups through the Chao index (P = .09), Shannon index (P = .43), and Simpson index (P = .28). E, Principal component analysis showed that the β diversity of gut microbiota was different between groups. F, G, The gut microbiota composition was compared at the phylum (F) and genus levels (G). H, Evolutionary branching diagram of LEfSe analysis. *P < .05; **P < .01; ***P < .001; ****P < .0001.
We next used 16S rDNA to analyze the composition of the gut microbiota in each group. A total of 557 of 2955 operational taxonomic units (OTUs) were shared in all groups (Figure 4C). Although the α diversity among groups showed no significant difference (Figure 4D), the community structures were substantially altered (Figure 4E). At the phylum level, the dominant phyla were Actinobacteria (15.06%) and Verrucomicrobia (7.50%) in the control group, while Proteobacteria (14.50%) and Deferribacteres (5.81%) in the model groups (Figure 4F). In detail, the DSS-treated mice mainly showed reductions in Allobaculum and Akkermansia, and increases in Oscillospira, Odoribacter, Mucispirillum, Helicobacter, and Escherichia at the genus level (Figure 4G). In particular, the relative abundance of Helicobacter in IBD mice pretreated with F. nucleatum (1.36%) was increased by administration of Fn + NE (2.49%), which was decreased by treatment with LED 209 (1.35%). More importantly, linear discriminant analysis effect size (LEfSe) showed that the Bacteroidaceae, Alcaligenaceae, and Helicobacteraceae families played important roles in the NE + Fn + DSS group, while SCFA-producing bacteria (Prevotellaceae, Lactobacillaceae), Turicibacteraceae, Clostridiaceae and Desulfovibrionaceae were considered biomarkers in the NE + Fn + LED 209 + DSS group (Figure 4H).
Discussion
There is no consensus or clear evidence that IBD is caused by classic pathogens in the traditional sense, but it is more likely that some symbiotic microbiota show pathogenicity in a specific context.4 A representative example is that F. nucleatum expresses high levels of FadA under the conditions of stress and disease, rather than in healthy sites or tissues.38 In this study, we found that NE could reinforce the pathogenicity of F. nucleatum by promoting the invasion ability of F. nucleatum into intestinal tissue to aggravate colitis. The NE concentration could be increased by 20 to 40 times in the case of stress compared with the physiological state.13 Chronic stress increases the pathogenicity of opportunistic pathogenic bacteria by increasing the concentration of NE in the gut, which may partially explain why long-term high perceived stress leads to the exacerbation of IBD.
Fortunately, people have paid increasing attention to the management of psychological stress in IBD treatment.39 Here, the QseC inhibitor LED209 significantly inhibited the effect of NE on F. nucleatum, which further alleviated colonic inflammation in IBD mice. Thus, QseC, which is widely distributed in gram-negative pathogens, may be a promising target for IBD therapy. Other possible bacterial adrenergic receptors have been proposed, including BasS, QseE, CpxA, and KdpD.40,41 Whether and how they can produce the same effect as QseC is worthy of further study.
Microbial interactions are complex and can be modified depending on the environment. The complexity may cause different outcomes of microbiota, including adaptability or growth and function or phenotypes.42 Quorum sensing signaling is one way microorganisms sense their environment and relay information to their own community or other species.43 In this study, we observed variation in microbial community structure between groups in mice that received different interventions. The QseC inhibitor intervention partly recovered the relative abundance of SCFA-producing bacteria (Prevotellaceae, Lactobacillaceae) and reduced colonic inflammation in the mice, although there was still a relatively high abundance of opportunistic pathogenic bacteria. From an ecological point of view, microbial communities can carry out chemical transformations that cannot be achieved by individual species. It means that we should consider the collective rather than the individual to evaluate the role of microorganisms.44 Therefore, reducing the impact of environmental factors (not just stress) on opportunistic pathogens, but not killing specific bacteria, may be effective in the treatment of IBD. In particular, the infection of antibiotic-resistant bacteria has become the top priority of global health challenges.45 Targeting quorum sensing receptors of bacteria may also be applicable in other infectious diseases.
Because of the severe bloody stool, it was difficult to collect adequate intestinal contents of the mice at the end of the experiment. We could not further detect the metabolites of the gut microbiota, especially the concentration of AIs, which makes it impossible to clarify the effect of NE on quorum sensing and what changes have taken place after LED209 intervention. In addition, quorum sensing systems exist in many microbes, and NE or LED 209 administration cannot guarantee that they only act on the target bacteria. Furthermore, we did not analyze the fungal and viral groups that may also take part in the development of IBD,46,47 which may lead to some neglected interspecific interactions of microorganisms.
Conclusions
Our study demonstrated that NE could bind to the quorum sensing receptor QseC of F. nucleatum directly, and this interkingdom signaling enhanced its pathogenicity (including virulence and invasiveness) to aggravate the colonic inflammation of IBD mice. Moreover, QseC may be a potential target for microbiota management of IBD under chronic pressure.
Supplementary Data
Supplementary data is available at Inflammatory Bowel Diseases online.
Author Contributions
L.Z. and G.M.C. conducted the experiments and analyzed the data. L.Z. drafted the preliminary manuscript. X.S.F. conceived and designed the study and approved the final manuscript. Other co-authors did helpful work. All authors contributed to the article and approved the submitted version.
Funding
This work was supported by the National Natural Science Foundation of China (No.81972315) granted to Xiangsheng Fu.
Conflicts of Interest
All authors declare that there is no conflict of interest.
Data Availability
The data that support the findings of this study are available in “Mendeley Data.” DOI: 10.17632/sf7sf9v3fy.1.
References
Author notes
Contributed equally
