Article Text

Download PDFPDF
Original research
Synbindin restrains proinflammatory macrophage activation against microbiota and mucosal inflammation during colitis
  1. Luoyan Ai1,
  2. Yimeng Ren1,
  3. Mingming Zhu1,
  4. Shiyuan Lu1,
  5. Yun Qian1,
  6. Zhaofei Chen1,
  7. Antao Xu2
  1. 1 State Key Laboratory for Oncogenes and Related Genes, Key Laboratory of Gastroenterology and Hepatology, Ministry of Health, Division of Gastroenterology and Hepatology, Shanghai Cancer Institute, Shanghai Institute of Digestive Disease, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China
  2. 2 Department of Rheumatology, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China
  1. Correspondence to Dr Luoyan Ai, Division of Gastroenterology and Hepatology, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China; storysparrow{at}126.com; Dr Antao Xu, Department of Rheumatology, Renji Hospital, School of Medicine, Shanghai Jiao Tong University, Shanghai, China; antaoxu{at}yeah.net

Abstract

Objective As a canonical membrane tethering factor, the function of synbindin has been expanding and indicated in immune response. Here, we investigated the role of synbindin in the regulation of toll-like receptor 4 (TLR4) signalling and macrophage response to microbiota during colitis.

Design Three distinct mouse models allowing global, myeloid-specific or intestinal epithelial cell-specific synbindin heterozygous deletion were constructed and applied to reveal the function of synbindin during dextran sodium sulfate (DSS) colitis. Effects of synbindin on TLR4 signalling and macrophage activation in response to bacterial lipopolysaccharide (LPS) or Fusobacterium nucleatum were evaluated. The colocalisation and interaction between synbindin and Rab7b were determined by immunofluorescence and coimmunoprecipitation. Synbindin expression in circulating monocytes and intestinal mucosal macrophages of patients with active IBD was detected.

Results Global synbindin haploinsufficiency greatly exacerbated DSS-induced intestinal inflammation. The increased susceptibility to DSS was abolished by gut microbiota depletion, while phenocopied by specific synbindin heterozygous deletion in myeloid cells rather than intestinal epithelial cells. Profoundly aberrant proinflammatory gene signatures and excessive TLR4 signalling were observed in macrophages with synbindin interference in response to bacterial LPS or Fusobacterium nucleatum. Synbindin was significantly increased in intestinal mucosal macrophages and circulating monocytes from both mice with DSS colitis and patients with active IBD. Interleukin 23 and granulocyte-macrophage colony-stimulating factor were identified to induce synbindin expression. Mechanistic characterisation indicated that synbindin colocalised and directly interacted with Rab7b, which coordinated the endosomal degradation pathway of TLR4 for signalling termination.

Conclusion Synbindin was a key regulator of TLR4 signalling and restrained the proinflammatory macrophage activation against microbiota during colitis.

  • ulcerative colitis
  • macrophages
  • colonic microflora
  • gut immunology

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Significance of this study

What is already known on this subject?

  • Most previously identified genetic risk loci of IBD are specially enriched for promoters that are regulated in macrophages response to lipopolysaccharide (LPS) or CSF1-induced macrophage differentiation, supporting a critical role of dysregulated proinflammatory macrophage activation against microbiota in the pathogenesis of IBD.

  • Toll-like receptor 4 (TLR4) signalling pathway couples signal transduction cascades with the intracellular trafficking process. The endosomal trafficking of TLR4 optimises the sensation of bacterial LPS and the amplitude of signalling activation.

  • Rab7b coordinates endosomal trafficking and lysosomal targeting processes of ligand-TLR4 complex for appropriate signalling termination. Rab7b interference inhibits TLR4 degradation and promotes proinflammatory macrophage activation.

  • Synbindin is the core subunit of the multisubunit tethering complexes, which function in the tethering step of vesicular trafficking process.

Significance of this study

What are the new findings?

  • Global synbindin haploinsufficiency greatly aggravates dextran sodium sulfate-induced colitis, which is abrogated by microbiota depletion and phenocopied by myeloid-specific synbindin heterozygous deletion.

  • Synbindin negatively regulates TLR4 signalling and restrains proinflammatory macrophage activation towards bacterial LPS.

  • Direct interaction and the similar function between synbindin and Rab7b in the regulation of TLR4 signalling support that synbindin may function as a critical effector of Rab7b in endosomal degradation pathway of LPS-TLR4 complex.

  • Synbindin is significantly elevated in intestinal mucosal macrophages and circulating monocytes from both experimental colitis mouse models and patients with active IBD.

How might it impact on clinical practice in the foreseeable future?

  • Synbindin restrains proinflammatory macrophage activation towards intestinal microbiota and was elevated in monocytes before being recruited to mucosal inflammation. It may be of great clinical benefit to monitor monocyte synbindin expression for disease activity and/or treatment responses. Further therapeutic approach targeting synbindin may provide a novel advantage in the management of human IBD.

Introduction

Inflammatory bowel disease (IBD) is characterised by chronic relapsing intestinal inflammation, including ulcerative colitis (UC) and Crohn’s disease (CD).1 In the past few decades, most studies on IBD have investigated the abnormal adaptive immunity.2 The focus has shifted towards mucosal innate immunity and a central role of macrophages in the pathogenesis of IBD is highlighted.3 4 Recent studies strongly support that the susceptibility to IBD may arise from dysregulated proinflammatory macrophage activation against microbiota.5

Lipopolysaccharide (LPS) is a cell wall component of gram-negative bacteria, recognition of which requires the specific pattern recognition receptor, toll-like receptor 4 (TLR4).6 7 The ligation of TLR4 initiates a complex signalling cascade and facilitates the release of proinflammatory cytokines. To maintain homeostasis, LPS-TLR4 signalling induces negative feedback controls almost at each step of the activation process.8 9 The unique cell biology of TLR4 signalling pathway allows it to couple signal transduction cascades with the intracellular trafficking process. The endocytic trafficking of TLR4 contributes to optimise the sensation of bacterial LPS and the amplitude of signalling activation. Disturbance at each step of the trafficking processes, like replenishment of TLR4 to the cell surface, internalisation of LPS-TLR4 complex or the lysosomal targeting trafficking process, could result in aberrant activation of TLR4 signalling.10–12 However, mechanisms and components involved in these processes remain largely unknown.

Recent studies reveal that Rab GTPases could coordinate intracellular trafficking of TLR4 and regulate signal transduction. Rab7b localises into late endosomes/lysosomes and facilitates the transportation of the LPS-TLR4 complex from early endosomes to late endosomes/lysosomes for degradation.13 Rab GTPases fulfil their various functions through recruitment and binding to specific effectors.14 The cooperation between Rab GTPases and their effectors ensures the specificity and directionality in membrane trafficking. The detailed membrane trafficking process mediated by Rab7b, including its specific effector proteins, still needs further investigation.

Synbindin is a core subunit of trafficking protein particle II (TRAPP II) and TRAPP III which are multisubunit tethering complexes functioning in the tethering step of vesicular trafficking process.15–17 The deletion of synbindin impairs the assembly and/or stability of TRAPP complexes as well as vesicular trafficking process.18 In the present study, we investigated the role of synbindin in the maintenance of intestinal mucosal homeostasis. To dissect the role of synbindin in mucosal innate immunity, three distinct mouse models with global, myeloid-specific or IEC-specific heterozygous deletion of synbindin were constructed and challenged with dextran sodium sulfate (DSS). The effects of synbindin on proinflammatory macrophage activation and TLR4 signalling transduction were further analysed.

Materials and methods

Patients

Colonoscopic biopsies were obtained from inflamed areas of the colons of 30 patients with active UC and 25 patients with active CD. Control samples were harvested from normal non-inflamed bowel located more than 10 cm away from the tumour from patients undergoing bowel resection for sporadic colon cancer. EDTA anticoagulated blood samples (10 mL) were obtained from 35 patients with active UC, 31 patients with active CD and 29 healthy controls. Written informed consent was obtained from all subjects.

Mice

Synbindin +/− (global heterozygous deletion), Synbindin flox and Synbindin ΔIEC (heterozygous deletion specific in IEC) mice were generated as previously described.19 Synbindin heterozygous deletion specific in myeloid cells (Synbindin ΔMYL) mice were also constructed (online supplemental methods). Interestingly, no mice with homozygous myeloid-specific synbindin deletion was observed in our present study, which may be embryonic lethal. These mice were housed in individually ventilated cages (IVC) in our institutional specific pathogen-free animal facilities. Germ-free (GF) mice on C57BL/6 background were purchased from Cyagen Biosciences and maintained in the GF facility. All animal experiments were performed in compliance with national and institutional guidelines for the care and use of laboratory animals and were approved by the institutional ethics committee of Renji Hospital, Shanghai Jiaotong University.

Supplemental material

Bioinformatics analysis

Transcriptome analysis of synbindin siRNA or control siRNA treated RAW264.7 cells in response to LPS was conducted by RNA sequencing (GSE158301).

Additional information could be found at online supplemental materials and methods.

Results

Mice with synbindin heterozygous deletion are more susceptible to DSS-induced colitis

To elucidate the role of synbindin in the maintenance of intestinal mucosal homeostasis, mice with global synbindin haploinsufficiency (Synbindin +/−) were constructed and challenged with 3% DSS. (As previously reported, only heterozygotes and wild-type (WT) mice were born at the expected Mendelian ratio.) Synbindin insufficiency was confirmed by immunohistochemistry (IHC) staining and western blot (WB) analysis (online supplemental figure S1A,B). No spontaneous colitis was observed in Synbindin +/− mice that were routinely bred for up to 12 months. Consistent with Synbindin ΔIEC mice,19 Synbindin +/− mice showed more goblet cells in colons than WT littermates (online supplemental figure S1D). When challenged with DSS, Synbindin +/− mice exhibited markedly increased disease severity as measured by weight loss (figure 1A), disease activity indices (DAI, figure 1B) and colon shortening (figure 1C). Histological analysis revealed significantly more infiltration of inflammatory cells and vast epithelial erosions in the intestinal mucosa of Synbindin+/− mice(figure 1D). To determine the immunological basis of increased colitis in Synbindin+/−mice, cytokines production in the colon were analysed. RT-PCR analysis showed that the expression of Tnfα, Il-1β, Il-6, Cox2, Il-12p40 and Il-23p19 was profoundly increased in Synbindin+/− mice in comparison to WT littermates (figure 1E). Consistently, fluorescence activated cell sorting (FACS) analysis showed that the secretion of tumour necrosis factor (TNF), interleukin (IL)-6, IL-1β and IL-17A, but not IL-10 and interferon γ were profoundly enhanced in the colon homogenates of Synbindin+/− mice (figure 1F).

Figure 1

Synbindin haploinsufficiency exacerbates dextran sodium sulfate (DSS)-induced colitis. (A) Body weight loss, (B) disease activity indices (DAI), the combined score of percentage weight loss, stool consistency and bleeding, and (C) colon lengths, in DSS-treated Synbindin +/− mice (n=20) and their wild-type (WT) littermates (n=20). (D) H&E staining of colon cross sections and histological scoring at day 8 after DSS challenge. (E) Cytokine mRNA induction in the colon of Synbindin +/− mice and WT littermates. (F) Secretion of cytokines in colon homogenates from Synbindin +/− mice and WT littermates at day 8 after DSS challenge. Statistical significance determined by unpaired two-sided Student’s t test. Data are mean±SEM. *P<0.05, **p<0.01, ***p<0.001, ****p<0.0001. IL, interleukin; TNF, tumour necrosis factor.

Synbindin+/− mice were further administrated with three cycles of DSS, which caused a non-self-limiting chronic colitis. Consistent with the observation in acute colitis, Synbindin+/− mice showed more severe intestinal inflammation than WT littermates, as demonstrated by weight loss, colon shortening and histological assessment (online supplemental figure S2A–D).

Since TNF, IL-1β, IL-6, IL-12 and IL-23 were markers of M1 macrophage activation,20 we next examined changes of cellular components involved in the mucosal inflammation with a special focus on macrophages. Under inflammatory conditions, circulating monocytes are continuously recruited to the site of inflammation and differentiate into inflammatory macrophages, finally dominate inflammatory responses.3 21 Flow cytometric analysis of the isolated lamina propria mononuclear cells (LPMCs) from Synbindin+/−  and WT mice demonstrated that significantly increased circulating monocytes, marked by MHC-II-CD11b+Ly6C+, were recruited to mucosal inflammation site in Synbindin+/− mice(figure 2A,B). IHC staining of F4/80 in colons also confirmed more macrophage infiltration in Synbindin+/− mice after DSS challenge (figure 2C). We further evaluated the proinflammatory activation state of these monocytes/macrophages by detecting TNF, IL-1β and IL-6 expression using flow cytometric analysis. As illustrated in figure 2D,E, we found significantly higher proportion of CD11b+ macrophages collected from Synbindin+/− mice expressed TNF, IL-1β and IL-6 than that from WT littermates. Moreover, the intracellular expression of these proinflammatory markers, as measured by the median fluorescent intensity, was significantly higher for Synbindin-insufficient macrophages than that for WT cells (figure 2F). Together, we observed increased infiltration and activation of macrophages in the mucosa of Synbindin+/− mice during colitis, supporting the hypothesis that synbindin haploinsufficiency-conferred susceptibility to colitis may be due to aberrant proinflammatory macrophage activation.

Figure 2

Onset of colitis in Synbindin+/− is characterised by enhanced influx and activation of monocytes/macrophages. (A, B) Representative staining and mean percentage of MHC-II CD11b+ Ly6C+ monocytes and MHC-II+ CD11b+ CD11C macrophages within the live-gated CD45+ fraction of isolated lamina propria mononuclear cells (LPMCs) in DSS-treated Synbindin +/− mice and wild-type (WT) littermates. (C) Immunohistochemistry (IHC) staining of F4/80+ macrophages in colons from Synbindin+/− and WT littermates at day 8 after dextran sodium sulfate (DSS) challenge. (D) Representative flow cytometry plots of intracellular cytokines tumour necrosis factor (TNF)-α, interleukin (IL)-6 and IL-1β protein levels gated on CD45+ CD11b+ live LPMCs isolated from DSS-treated WT and Synbindin +/− mice. (E, F) Quantification of percentages and median fluorescence intensity (MFI) of TNF-α, IL-6 and IL-1β positive cells gated on CD45+CD11b+ live LPMCs isolated from DSS-treated WT and Synbindin +/− mice. Data are mean±SEM from at least three independent experiments. Statistical significance determined by unpaired Student’s t test. *P<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Specific synbindin heterozygous deletion in myeloid/macrophages rather than IECs confers the increased susceptibility to colitis

Previously, our study showed that synbindin haploinsufficiency influences IECs survival and mucin production,19 both of which were essential to maintain normal epithelial barrier function. The results obtained from Synbindin +/− mice raised the possibility that the susceptibility to DSS-colitis could result from the vulnerability of epithelial barrier. Alternatively, synbindin haploinsufficiency may render dysregulated macrophage activation against microbiota during colitis. To discriminate between these two possibilities, we challenged Synbindin ΔMYL and Synbindin ΔIEC mice with DSS. Synbindin ΔMYL mice phenocopied the excessive intestinal inflammation observed in Synbindin +/− mice, which was indicated by greater weight loss, increased DAI, colon shortening and higher histological indices (figure 3A–D). Additionally, more F4/80+ cells infiltration (figure 3E) and much higher secreted levels of TNF, IL-1β and IL-6 (figure 3F) were also observed in the colons of Synbindin ΔMYL mice. However, colonic inflammation developed in Synbindin ΔIEC mice was comparable to that in Synbindin flox mice (figure 3G–J). Thus, the causative role of synbindin haploinsufficiency in the exacerbation of colitis was probably attributed to the dysregulated immune response of macrophages, rather than increased vulnerability of epithelial barrier function.

Figure 3

Synbindin ΔMYL mice, but not Synbindin ΔIEC mice, are more susceptible to dextran sodium sulfate (DSS)-induced colitis. (A) Body weight loss, (B) disease activity indices (DAI), and (C) colon lengths in DSS-treated Synbindin ΔMYL (n=10) and Synbindin flox littermates (n=10). (D) Representative H&E staining of colon cross sections and histological scoring of indicated mice at day 8 after DSS challenge. (E) Immunohistochemistry (IHC) staining of F4/80 on colons of Synbindin flox and Synbindin ΔMYL mice at day 8 after DSS challenge. (F) Tumour necrosis factor (TNF), interleukin (IL)-6 and IL-1β secretion in colon homogenates from Synbindin flox and Synbindin ΔMYL mice at day 8 after DSS challenge. (G) Body weight loss, (H) DAI and (I) colon lengths in DSS-treated Synbindin ΔIEC mice (n=10) and Synbindin flox mice (n=10). (J) Representative H&E staining of colon cross sections and histological scoring of Synbindin flox and Synbindin ΔIEC mice at day 8 after DSS challenge. Data are expressed as mean±SEM from at least three independent experiments. Statistical significance determined by unpaired Student’s t test. *P<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Synbindin knockdown profoundly enhances LPS-induced proinflammatory macrophage activation

Baillie et al recently reported that genetic risk loci of IBD were strongly and specifically enriched for promoters involved in macrophage activation in response to LPS.5 We next evaluated the influence of synbindin silence on the responses of macrophage to LPS stimulation in vitro. In order to get a global view of the processes and pathways that were modulated by synbindin, RNA sequencing (RNA-seq)-mediated transcriptome analysis was performed for LPS-stimulated murine macrophage cell line RAW264.7 cells with or without interference of synbindin. Remarkably, gene set enrichment map analysis (figure 4A) showed that most of the gene signatures enriched in macrophages with synbindin knockdown were related to activation of inflammatory responses, cytokines production, chemokines signalling and membrane trafficking. Macrophage activation pathway and Toll-like receptor signalling pathway were particularly affected by synbindin silence (figure 4B). The most upregulated and statistically significant genes all encoded proinflammatory cytokines or antimicrobial peptides, including Tnf, Tnfsf15, Csf2, Cxcl2, Spp1, Lyz1 and Lyz2 (figure 4C).22–24

Figure 4

Silence of synbindin profoundly enhances lipopolysaccharide (LPS)-induced proinflammatory macrophage activation. (A) Gene set enrichment analysis (GSEA) of transcriptional profiling comparison between LPS-stimulated RAW264.7 cells with (red) or without (blue) synbindin interference. Cytoscape and enrichment map were used for visualisation of the GSEA results (1% false discovery rate (FDR); p=0.005). Nodes represent enriched gene sets, which were grouped and annotated by their similarity according to the related gene sets. Enrichment results were mapped as a network of gene sets (nodes). Node size was proportional to the total number of genes within each gene set. Proportion of shared genes between gene sets was presented as the thickness of the green line between nodes. This simplified network map was manually curated by removing general and uninformative subnetworks. (B) GESA analysis of macrophage activation and Toll-like receptor pathway between LPS-stimulated RAW264.7 cells with or without synbindin interference. (C) Transcriptional profiling was displayed by volcano plot highlighting top significant genes. (D) Tumour necrosis factor (TNF), interleukin (IL)-6 and IL-1β secretion in the culture supernatants of RAW264.7 cells with or without synbindin interference and bone marrow-derived macrophages (BMDMs) isolated from Synbindin+/−  and wild-type (WT) mice after LPS or Fusobaterium nucleatum stimulation for 24 hours. (E) RAW264.7 cells with or without synbindin interference and (F) BMDMs isolated from Synbindin+/−  and WT mice were stimulated with LPS (100 ng/mL or 1000 ng/mL, respectively, as indicated) for 30 min, cell lysates were collected and analysed for p-P65, P65, phosphorylation of extracellular-signal regulated kinase (p-ERK), ERK, synbindin by western blotting. glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as loading control. (G) The human leukemic cell line (THP-1) cells were differentiated by phorbol 12-myristate 13-acetate (PMA) treatment and then transfected with synbindin siRNAs or control siRNA. After LPS (100 ng/mL) stimulation for 30 min, cell lysates were collected and analysed for p-P65, P65, p-ERK, ERK and synbindin by western blotting. (H) RAW264.7 cells were transfected with synbindin overexpression plasmid or pcDNA3.1 control plasmid. After LPS (100 ng/mL) stimulation for 30 min, cell lysates were collected and analysed for p-P65, P65, p-ERK, ERK and synbindin by western blotting. (I) TNF, IL-6 and IL-1β secretion in the culture supernatants of RAW264.7 cells with or without synbindin overexpression after LPS stimulation for 24 hours. (E–H) Data show representative results from more than three independent experiments. Right panel represent quantification of target protein bands relative to GAPDH. All data are expressed as mean±SEM from three independent experiments. Statistical significance was assessed by unpaired two-tailed Student’s t test. *P<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

In addition to the transcriptome changes, secretion of proinflammatory cytokines including TNF, IL-6 and IL-1β was profoundly enhanced from both macrophage cell line with synbindin silence and BMDMs from Synbindin +/− mice in response to bacterial LPS (figure 4D). These data strongly supported our hypothesis that synbindin haploinsufficiency resulted in markedly aberrant proinflammatory macrophage activation at the functional level.

We next evaluated the effect of synbindin on LPS-TLR4 downstream signalling, the activation of transcription factor nuclear factor-κB (NF-κB) and MAPK pathway. WB analysis confirmed that loss of synbindin significantly enhanced LPS-induced activation of NF-κB and ERK in both RAW264.7 and BMDM cells (figure 4E,F). Similar results were also obtained from THP-1 cells (figure 4G), a human monocytic cell line, which indicated that the role of synbindin was conserved in human macrophages. We further evaluated the effects of synbindin overexpression on LPS-induced macrophage activation. WB analysis demonstrated that ectopic expression of synbindin reduced LPS-induced activation of NF-κB and ERK (figure 4H). The production of secreted cytokines including TNF, IL-6 and IL-1β was also inhibited by overexpressed synbindin (figure 4I). Moreover, the excessive production of proinflammatory cytokines and inflammatory signalling by BMDMs derived from Synbindin +/− mice under LPS stimulation were markedly restrained by lentiviral transduction with WT synbindin plasmid (online supplemental figure S3). Hence, synbindin negatively regulated LPS-induced macrophage activation.

Fusobacterium nucleatum, an anaerobic Gram-negative bacterium, is a common resident of intestinal mucosa and reported to be associated with development and progression of IBD. Liu et al recently reported that F. nucleatum could promote the proinflammatory activation of macrophage and aggravate the progression of colitis.25 The exposure of macrophages to live F. nucleatum recapitulated the findings obtained from LPS stimulation. F. nucleatum caused significant higher production of TNF, IL-6 and IL-1β from RAW264.7 cells with synbindin silence (figure 4D), indicating that intestinal microbiota may be the major driver of exaggerated proinflammatory macrophage activation during colitis.

The intestinal microbiota drives exacerbated DSS colitis in mice with synbindin haploinsufficiency

Most genetic susceptibility genes of IBD are involved in immune responses against intestinal microbiota.26 27 The intestinal microbiota is probably the most important environmental factor that drives excessive inflammatory responses in genetically predisposed host. We next investigated whether synbindin insufficiency would still aggravate DSS colitis in the absence of intestinal microbiota. Both Synbindin+/− and WT mice were treated with broad-spectrum antibiotics to deplete the intestinal microbiota before and during the onset of DSS treatment. Whereas Synbindin+/− mice harbouring normal intestinal microbiota developed apparent more severe colitis than their littermates, the severity of colonic colitis was indeed comparable between antibiotics-treated Synbindin+/−  and WT littermates, as indicated by loss of body weight alike, indistinguishable DAI, similar colon length and histological evaluation of colitis (figure 5A–D). Therefore, synbindin haploinsufficiency does not confer increased susceptibility to DSS colitis in the absence of microbiota. In other words, the intestinal microbiota drives exacerbated DSS-induced colitis in mice with synbindin haploinsufficiency.

Figure 5

Excessive dextran sodium sulfate (DSS)-induced colitis in Synbindin +/− mice is driven by the gut microbiota. (A–D) Wild-type (WT) and Synbindin +/− mice were treated with antibiotics to deplete gut microbiota, and then exposed to DSS (n=10 per group). (A) Body weight loss, (B) disease activity indices (DAI) and (C) colon lengths of indicated mice. (D) Representative H&E stained colon cross sections and histological score of indicated mice at day 12 of the DSS colitis protocol. All data shown were representative of three independent experiments. All statistical analysis was performed with unpaired two-sided Student’s t tests. *P<0.05, **p<0.01, ***p<0.001, ****p<0.0001. (E–G) Faecal microbiota composition from Synbindin +/− mice and WT littermates was assayed by bacterial 16s rRNA V3–V4 region sequencing. (E) The Ace index and Chao index were calculated. (F) Microbial clustering was shown based on Bray-Curtis dissimilarity principal coordinate analysis (PCoA) metrics. Each symbol represents an individual mouse. Ellipsoids represent a 95% CI surrounding each disease group. Results are presented as operational taxonomic units (OTUs). (G) Similarities of bacterial communities between samples were compared by analysis of similarities based on Bray-Curtis.

Next, we investigated whether the gut microbiome colonised in mice with synbindin haploinsufficiency are themselves colitogenic. Synbindin +/− mice and WT littermates were housed separately by genotypes for 8 weeks after weaning (sequencing analysis at 11–12 weeks of age).28 Gut microbiota were collected and the influence of synbindin haploinsufficiency on the gut microbiota composition was investigated by conducting 16S rRNA sequencing. As shown in figure 5E, no difference in microbial species diversity was observed between Synbindin +/− mice and WT littermates. Principal coordinate analysis showed no Synbindin +/− genotype-dependent clustering of faecal microbiota communities (figure 5F). In addition, analysis of similarities at OTU level also supported that the microbiome difference between Synbindin +/− and WT littermates was not significantly greater than the difference among mice within each genotype (R=0.1009, p=0.159, figure 5G, calculated from figure 5F). Nevertheless, to exclude the possibility that non-significant microbiota alterations in Synbindin +/− mice may alter the colitis responses, faecal microbiota was transplanted from Synbindin +/− mice and WT littermates to GF hosts, respectively, followed by DSS challenge. As shown in online supplemental figure S4A–D, no significant difference about colitis severity in GF mice colonised with microbiota from Synbindin +/− mice and WT littermates was observed, indicating that synbindin haploinsufficiency did not shape gut microbiota composition or cause colitogenic microbiota changes.

Synbindin potentially serves as a critical effector of Rab7b in the endosomal degradation pathway of TLR4

We have proved that aberrant proinflammatory macrophage activation induced by excessive LPS-TLR4 signalling pathway was the most important pathogenic mechanism underlying synbindin haploinsufficiency-contributed susceptibility to colitis. Subsequently, we tried to elucidate how synbindin exert its essential role in the control of TLR4 signalling transduction. As a canonical tethering factor, synbindin facilitates the initial interaction between the vesicle and its target membrane and ensures the specificity at the same time. Recently, the intracellular endosomal trafficking process was shown to be essential in the regulation of TLR4 signalling pathway. Hence, we proposed that synbindin may be involved in endosomal trafficking process of TLR4. Immunofluorescence staining identified obvious colocalisation of TLR4 with synbindin (figure 6A). The protein levels of TLR4 and its adaptor protein MyD88 were increased after synbindin knockdown, especially following LPS stimulation (figure 6B).

Figure 6

Synbindin potentially serves as a critical effector of Rab7b in the endosomal trafficking and lysosomal targeting of toll-like receptor 4 (TLR4). (A) Confocal analysis of endogenous TLR4 and synbindin in THP-1 cells. Fluorescence intensity of synbindin (green line) and TLR4 (red line) traced along the white line in THP-1 cells using the line profiling function of the ImageJ software. (B) TLR4 and MyD88 protein levels in RAW264.7 cells with or without synbindin interference after Lipopolysaccharide (LPS; 100 ng/mL) stimulation for indicated time points. Right panel represent quantification of target protein bands relative to GAPDH. (C) Confocal analysis of endogenous synbindin (green) and Rab7b (red) in THP-1 cells. Fluorescence intensity of synbindin (green line) and Rab7b (red line) traced along the white line in THP-1 cells using the line profiling function of the ImageJ software. (D) Co-immunoprecipitation analysis of direct interaction between synbindin and Rab7b in RAW264.7 cells. (E) Immunofluorescence showing the colocalisation between synbindin and giantin, lysosomal associated membrane protein 1(LAMP-1) in THP-1 cells. (F) Immunofluorescence analysis of Rab7b distribution after synbindin interference in THP-1 cells. (G) RAW264.7 cells with Rab7b- and/or synbindin- interference were stimulated with LPS (100 ng/mL) for 30 min, cell lysates were collected and analysed for p-P65, P65, p-ERK, ERK, synbindin by western blotting. GAPDH was used as loading control. Right panel represent quantification of target protein bands relative to GAPDH. (H) Tumour necrosis factor (TNF), interleukin (IL)-6 and IL-1β secretion in the culture supernatants of RAW264.7 cells with Rab7b- and/or synbindin- interference after LPS stimulation for 24 hours.

Previously, Wang et al reported that Rab7b coordinated the endosomal trafficking and lysosomal targeting transport of LPS-TLR4 complex.13 In their study, Rab7b deletion significantly inhibited TLR4 degradation and augmented TLR4 signalling, which promoted LPS-induced proinflammatory macrophage activation. The overlap of function in the regulation of TLR4 signalling between Rab7b and synbindin led us to speculate that they interact with each other and participate in the same process. To validate our speculation, we performed double immunofluorescence and co-immunoprecipitation analysis between Rab7b and synbindin. Indeed, a direct interaction between Rab7b and synbindin was demonstrated (figure 6C,D). Synbindin also colocalised with late endosomes and/or lysosomes marker LAMP-1 as well as Golgi marker Giantin (figure 6E), which was similar to Rab7b as previously shown.29 Surprisingly, the intracellular distribution of Rab7b was greatly disturbed after synbindin silence (figure 6F), which further indicated the closely interaction and functional interdependence between them.

Further analysis showed that the exaggeration of LPS-TLR4 signalling by synbindin- or Rab7b- interference was comparable, as demonstrated by levels of NF-κB and ERK signalling activation as well as cytokine production, including TNFα, IL-1β and IL-6. (figure 6G,H) Moreover, Rab7b and synbindin double interference did not further amplify TLR4 signalling as compared with Rab7b interference alone (figure 6G,H), indicating that synbindin and Rab7b was involved in the same mechanism regulating LPS-TLR4 signalling pathway.

Taken together, Synbindin potentially served as a critical effector in the Rab7b-coordinated endosomal trafficking and lysosomal targeting transport of TLR4 for appropriate termination of signalling transduction.

Synbindin is elevated in intestinal mucosal macrophages and circulating monocytes during colitis

Our previous transcriptional profiling of LPS-stimulated macrophages (RAW264.7) with or without synbindin interference identified the upregulated genes in synbindin-knockdown RAW264.7 cells and most of which belonged to proinflammatory gene signatures, indicating an excessive pro-inflammatory macrophage activation (figure 4A). When we reanalysed the expression of these upregulated genes in a clinical cohort which included 62 inflamed colon biopsies from patients with UC and 63 colon biopsies from controls (GSE11223), we remarkably observed that 80% (36/45) of these upregulated genes could be found being significantly overexpressed in the mucosa from patients with active UC (online supplemental figure S5). Our results not only emphasised the essential role of aberrant proinflammatory macrophage activation in the pathogenesis of IBD, but also supported that the lack or insufficient elevation of synbindin to restrain macrophage activation could promote the development of IBD.

Next, we evaluated the expression of synbindin in DSS-treated WT mice and in patients with active IBD. WB analysis showed increased synbindin expression in the inflamed colonic tissue of mice with DSS colitis (figure 7A). Flow cytometry confirmed the elevated expression of synbindin in isolated LPMCs and mucosal macrophages (F4/80+ LPMCs) as well as circulating monocytes (figure 7B,C). Consistent with results in experimental colitis, significantly more synbindin-positive immune cells were observed in mucosal biopsies from patients with active UC and CD when compared with that from non-inflamed controls (figure 7D,E). The elevated synbindin expression was also validated in another independent UC datasets (GSE11223, figure 7F). Further coimmunofluorescence analysis found that synbindin mostly colocalised with macrophage markers, CD11b and CD68, but not with T cell marker, CD4, in the mucosa from patients with active UC and CD (figure 7G–H). Thus, synbindin was induced in intestinal macrophages during colitis, potentially serving as a negative feedback control of TLR4 signalling. Consistent with results from intestinal mucosa, monocytes were shown to be the major source of synbindin in PBMCs from patients with active IBDs (figure 7I). And synbindin expression was significantly elevated in monocytes from patients with active IBDs than that from healthy controls (figure 7J).

Figure 7

Synbindin is increased in monocytes/macrophages from mice with dextran sodium sulfate (DSS) colitis and patients with active IBD. (A) Western blot (WB) analysis of synbindin protein in colon homogenates of mice with DSS colitis. Flow cytometry analysis of synbindin in lamina propria mononuclear cells (LPMCs), CD45+F4/80+ LPMCs (B) and mCD14+ peripheral blood mononuclear cells (PBMCs) (C) from mice. (D) Immunohistochemistry (IHC) analysis of synbindin in colonic biopsies from patients with active UC (n=30) and controls (n=25). (E) IHC analysis of synbindin in colonic biopsies from patients with active Crohn’s disease (CD) (n=25) and controls (n=25). The IHC experiments for UC and CD samples were conducted in different batches. (F) Synbindin mRNA levels in colons from inflamed UC and uninflamed controls (GSE11223). (G, H) Immunofluorescence double staining on active UC and CD samples showed colocalisation of synbindin with both CD11b+ and CD68+, but not with CD4+ cells. Scale bars 50 µm. (I, J) PBMCs from 29 healthy donors, 35 active UC and 31 active CD were obtained. Synbindin expression was stained and determined in peripheral blood B cells (CD19+), T cells (CD3+) and monocytes (CD14+) by flow cytometry. (K) THP-1 cells were stimulated with granulocyte-macrophage colony-stimulating factor (GM-CSF) or interleukin (IL)-23, synbindin expression was analysed by WB analysis. All statistical analysis was performed with unpaired two-sided Student’s t tests. *P<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Since circulating monocytes were the major source of intestinal macrophage,3 our results indicated that the increased expression of synbindin in intestinal mucosal macrophage was probably induced before being recruited to intestine. To support the speculation, THP-1 cells were stimulated with cytokines that were reported to be increased in the serum of patients with active IBD,30 including TNFα, IL-6, IL-1β, IL-17A, IL-23, granulocyte-macrophage colony-stimulating factor (GM-CSF), IL-10 and TGF-β (figure 7K & online supplemental figure S6). Among these cytokines, IL-23 and GM-CSF significantly induced synbindin expression in THP-1 cells, which indicated that monocyte synbindin expression could be induced by IBD-related systemic inflammatory responses. It may be of great clinical benefit to monitor monocyte synbindin expression for disease activity and/or treatment responses.

Discussion

Most previously reported susceptibility loci of IBD are involved in macrophage response to bacterial LPS, implying that uncontrolled macrophage activation to intestinal microbiota is critical in the pathogenesis.5 In the present study, we report a pivotal role of synbindin in the immune regulation that it negatively regulates TLR4 signalling and restrains proinflammatory macrophage activation against microbiota during colitis. We demonstrate that synbindin is significantly elevated in intestinal mucosal macrophages and circulating monocytes from patients with active IBD, potentially acting as a negative feedback control. As a canonical tethering factor, defects in intracellular vesicular trafficking have been observed in fibroblasts derived from patients with synbindin insufficiency.18 We detect a direct interaction and colocalisation of synbindin with Rab7b, which supports synbindin being a critical effector in the Rab7b-coordinated endosomal trafficking process of TLR4 for appropriate signalling termination.

Although no signs of spontaneous colitis were observed, mice with global synbindin haploinsufficiency showed a highly increased susceptibility to DSS-induced colitis, indicating synbindin serving as a key regulator of mucosal innate immunity. To delineate the direct involvement of synbindin in epithelial and macrophage cells to colitis, mice with IEC-specific and myeloid-specific heterozygous deletion of synbindin were generated and challenged with DSS. Our previous studies showed that synbindin interference significantly inhibited epithelial cell line proliferation in vitro.31 If only to consider what had been observed in vitro, it may be hard to interpret why IEC-specific synbindin haploinsufficiency did not increase colitis susceptibility. Actually, in addition to increased apoptosis and decreased proliferation of IECs, goblet cell maturation was significantly enhanced simultaneously in Synbindin ΔIEC mice as indicated by PAS and MUC2 staining (online supplemental figure S1F). Mucin-producing goblet cells and the mucus layer made enormous contribution to the normal epithelial barrier function, deficiency of which could lead to increased microbiota-induced colonic inflammation and was found to be the typical pathologic feature of IBD, especially UC.32 Thus, the grossly normal epithelial barrier function in Synbindin ΔIEC mice, as indicated by similar sensitivity to DSS-colitis as Synbindin flox littermates, could probably be determined by the balance between decreased epithelial proliferation and enhanced goblet cell maturation.

In the present study, we focused on the role of synbindin in macrophage during colitis. As we all know, intestinal macrophages were constantly replenished from circulating monocytes and the recruitment of monocytes was dramatically increased during colitis.4 33 Consistently, we observed significantly more MHC-II-CD11b+Ly6C+ monocytes in lamina propria (LP) during colitis, especially in mice with synbindin haploinsufficiency. Recently, transcriptome analysis was performed in colonic LP macrophages and circulating monocytes from mice with or without DSS colitis, respectively.34 Surprisingly, 75% (151/200) of these most differentially expressed genes following colitis were the same in LP macrophages and circulating monocytes.34 These results supported the speculation that colonic inflammation could modulate circulating monocytes to the state similar to mucosal macrophages. In our study, we observed that synbindin was significantly elevated in both mucosal macrophages and circulating monocytes not only in mice with experimental colitis but also in patients with active UC and CD. This indicated that colitis model employed in this study could partly recapitulate human IBDs at least in this aspect. We speculated that synbindin expression in circulating monocytes was probably induced by systemic inflammatory responses in IBD with cytokines being possible mediators. Among the eight cytokines that have been reported to be elevated in peripheral blood of patients with IBD, IL-23 and GM-CSF were demonstrated to significantly induce synbindin expression. Taken together, our study provided direct evidence that the response of monocyte/macrophage to microbiota has already been modulated prior to being recruited to intestinal mucosa in inflammatory conditions, such as IBD. The transition of macrophage from a proinflammatory to a proresolving state was critical in the resolution of intestinal inflammation.4 Downregulation of the response of circulating monocytes to LPS stimulation before participating in mucosal inflammation could be the first step in this transition, especially when systemic inflammatory response was initiated. Obviously, the upregulation of synbindin failed to terminate excessive TLR4 signalling activation in IBD, implying the coexistence of other defects in immune regulation.

Rab GTPases proteins are now recognised as universal coordinators of every aspects of intracellular membrane trafficking, which can be divided into four major steps: vesicle formation, intracellular transport, vesicle tethering and membrane fusion.14 More than 60 Rab GTPases have been identified in humans, reflecting the highly complexity of trafficking pathways. Although it is still difficult to understand Rab-mediated regulation of vesicular traffic at a systemic level, it is feasible to illustrate isolated trafficking steps. Wang et al found that Rab7b localised to late endosomes/lysosomes and may regulate the maturation of early endosomes to late endosomes, especially involved in the lysosomal targeting trafficking of LPS-TLR4 complex.13 In their study, Rab7b silence amplified LPS-induced proinflammatory macrophage activation through the inhibition of TLR4 degradation. Here, we further proved that synbindin colocalised and directly interacted with Rab7b, and synbindin silence phenocopied effects of Rab7b loss in response to LPS stimulation in macrophage. Hence, synbindin was probably an essential effector of Rab7b in the endosomal trafficking of LPS-TLR4 complex. However, different from previously reported effects of Rab7b interference, only a mild increase of TLR4 level was obtained in our study after synbindin knockdown in macrophages. Similar results has been reported, Husebye et al speculated that this may be due to low TLR4 signals or its high turnover rate in monocytes.35 The ligation of TLR4 recruited MyD88, which subsequently initiated downstream signalling to elicit inflammatory immune response.36 LPS stimulation induced TLR4 and Myd88 expression.37 Our study demonstrated that synbindin knockdown also enhanced LPS-induced MyD88 expression in macrophages, providing indirect evidence for the increased TLR4 signalling level.

Homozygous myeloid-specific synbindin ablation was supposed to be embryonic lethal for the lack of such phenotype in our present study. Lyz2-mediated Lysozyme M (LysM) expression has been used as a cell-specific marker for myeloid cells and introduced into the construction of Cre-loxP-based mouse model in myeloid cell research.38 However, except for the strong expression in macrophages of peripheral organs, a recent research surprisingly observed a robust LysM-promoter activity in neurons or neuronal progenitor cells.39 Since synbindin splicing defect which resulted in aberrant splicing and significantly reduced protein levels caused severe syndromic intellectual disability in human patients, a pivotal role of synbindin in neurological development has been proposed.18 The LysM-Cre-mediated myeloid-targeted modification of synbindin expression in mouse model may also affect neuronal synbindin expression during embryonic development and lead to embryonic lethal. In the analysis of synbindin-associated neurological disorder, all the affected subjects were homozygous for the synbindin variant at splice site, while their parents were heterozygous for the variant. A significant reduction of the regular splice product of synbindin could also be found in the parents as high as 54%, but no related apparent phenotype is reported.18 Consistently, no obvious neurological symptom was found in mouse models with global or myeloid-specific synbindin haploinsufficiency in our study. However, further detailed studies focused on the neuronal expression and possible influences of synbindin haploinsufficiency on the development of neurological system are needed.

In summary, synbindin restrained proinflammatory macrophage activation against microbiota during colitis. Direct interaction and the similar function of synbindin and Rab7b in macrophage activation supported that synbindin potentially functioned as a critical effector of Rab7b in endosomal trafficking and lysosomal targeting transport of LPS-TLR4 complex. Synbindin was upregulated in circulating monocytes, rather than induced at local site of mucosal inflammation by LPS response. Monocyte synbindin level may have the potential to serve as a marker to monitor disease activity and treatment response. Further studies related to the role of synbindin in macrophage during colitis may contribute to better understanding the pathogenesis, prevention and treatment of the disease.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Ethics approval

The study protocol was approved by the ethics committee of Renji Hospital, School of Medicine, Shanghai Jiao Tong University. All the research was carried out in accordance with the provisions of the Declaration of Helsinki of 1975.

Acknowledgments

The authors thank Professor Jing-Yuan Fang, Professor Jie Xu and Doctor Yun Cui for the guidance and help in 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.

Footnotes

  • LA, YR and MZ contributed equally.

  • Contributors LY Ai, YM Ren, MM Zhu, SY Lu, Y Qian, ZF Chen and AT Xu performed the experiments and analyzed data. LY Ai and AT Xu conceived and wrote the study. AT Xu designed or/and supervised this project and revised the study.

  • Funding This project was supported by grants from the National Natural Science Foundation of China: 81800485, 81600435 and 81772517.

  • 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.