Advertisement

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Csep1P protein from Campylobacter concisus induces a chemokine-dominant inflammatory state in macrophages and enhances proinflammatory response to gut bacteria

  • Christopher Yau Man Luk,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Validation, Visualization, Writing – original draft, Writing – review & editing

    Affiliation School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, New South Wales, Australia

    ⨯
  • Mohammad M. Rahman,

    Roles Methodology, Visualization

    Affiliation Department of Microbiology, Monash University Monash Biomedicine Discovery Institute, Clayton, Victoria, Australia

    ⨯
  • Xiaotian Zhou,

    Roles Data curation, Writing – original draft

    Affiliation Department of Microbiology, Monash University Monash Biomedicine Discovery Institute, Clayton, Victoria, Australia

    ⨯
  • C. Mee Ling Munier,

    Roles Methodology, Resources, Writing – review & editing

    Affiliation The Kirby Institute, University of New South Wales, Sydney, New South Wales, Australia

    ⨯
  • Fang Liu,

    Roles Methodology, Writing – review & editing

    Affiliation School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, New South Wales, Australia

    ⨯
  • Stephen M. Riordan,

    Roles Conceptualization, Funding acquisition, Writing – review & editing

    Affiliation Gastrointestinal and Liver Unit, Prince of Wales Hospital, Sydney, New South Wales, Australia

    ⨯
  • Anna Roujeinikova ,

    Roles Data curation, Methodology, Visualization, Writing – original draft, Writing – review & editing

    l.zhang@unsw.edu.au (LZ); anna.roujeinikova@monash.edu (AR)

    Affiliations Department of Microbiology, Monash University Monash Biomedicine Discovery Institute, Clayton, Victoria, Australia, Department of Biochemistry and Molecular Biology, Monash University, Clayton, Victoria, Australia

    ⨯
  • Li Zhang

    Roles Conceptualization, Funding acquisition, Project administration, Supervision, Writing – review & editing

    l.zhang@unsw.edu.au (LZ); anna.roujeinikova@monash.edu (AR)

    Affiliation School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, New South Wales, Australia

    ⨯

Abstract

Translocation of Campylobacter concisus from the oral cavity to the intestinal tract is increasingly recognised as a contributor to inflammatory bowel disease (IBD). The C. concisus secreted protein Csep1 has emerged as a molecular marker of C. concisus strains associated with Crohn’s disease, a form of IBD. However, its structure and role in inflammation remain unknown. Here, we report the X-ray crystal structure of plasmid-encoded Csep1P that reveals a unique α-helical fold with structural similarity to Helicobacter pylori cysteine-rich proteins HcpB and HcpC. Because HcpA, another Hcp family member, is known to affect monocyte differentiation, this structural similarity led us to hypothesise that Csep1P may modulate monocyte differentiation and macrophage function. Transcriptomic analysis revealed that Csep1P induced a chemokine-dominant inflammatory state in macrophages, M1-chem. Protein-level validation in both THP-1-derived and primary human macrophages confirmed this selective chemokine response. While Csep1P alone did not upregulate proinflammatory cytokines, THP-1-derived macrophages pre-incubated with Csep1P produced a higher level of proinflammatory cytokines in response to commensal Escherichia coli, which was validated on primary human macrophages. Furthermore, silencing the delta like canonical notch ligand 4 (DLL4) gene decreased the proinflammatory response of Csep1P-mediated macrophages to E. coli. Collectively, our data demonstrate that the structurally unique Csep1P reprograms macrophage response, which provides a mechanistic link between C. concisus infection and Crohn’s disease pathogenesis, and identifies Csep1P as a potential target for therapeutic intervention.

Author summary

Crohn’s disease is a form of inflammatory bowel disease (IBD) that is a life-long chronic inflammation of the gastrointestinal tract. The causes of Crohn’s disease are complex and are not fully understood, but certain bacteria are thought to play a role. Our study focuses on Campylobacter concisus, a bacterium commonly found in the human oral cavity that can move to the gut and has been linked to Crohn’s disease. We investigated a C. concisus secreted protein called Csep1P, to better understand how it might affect the immune system. Using X-ray crystallography, we determined the unique 3D structure of this protein. We found that Csep1P drives macrophages, a type of immune cell, into a state where they release specific signals that attract other immune cells. Although Csep1P alone did not trigger inflammation, it “primed” macrophages to react much more strongly to commensal gut bacteria such as Escherichia coli. Our results suggest that Csep1P may help explain how virulent C. concisus strains contribute to gut inflammation and highlight it as a potential target for new treatment for Crohn’s disease.

Citation: Luk CYM, Rahman MM, Zhou X, Munier CML, Liu F, Riordan SM, et al. (2026) Csep1P protein from Campylobacter concisus induces a chemokine-dominant inflammatory state in macrophages and enhances proinflammatory response to gut bacteria. PLoS Pathog 22(2): e1013951. https://doi.org/10.1371/journal.ppat.1013951

Editor: Sébastien Bontemps-Gallo, Centre National de la Recherche Scientifique, FRANCE

Received: August 28, 2025; Accepted: January 29, 2026; Published: February 13, 2026

Copyright: © 2026 Luk et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The coordinates and structure factors for Campylobacter concisus Csep1P have been deposited in the Protein Data Bank (http://www.rcsb.org) with accession code 8V4W. The RNA sequencing data have been deposited in the NCBI’s Gene Expression Omnibus (GEO) database under the accession number GSE251732 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE251732).

Funding: This work was supported by the University of New South Wales (https://www.unsw.edu.au; PS46772 to LZ) and New South Wales Government (https://www.nsw.gov.au; RG152309 to LZ). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Crohn’s disease is a major form of inflammatory bowel disease (IBD), which encompasses a group of chronic inflammatory disorders of the gastrointestinal tract [1]. Crohn’s disease is most diagnosed in adolescents and young adults between 15 and 35 years, although it can occur at any age [1]. The aetiology of Crohn’s disease is multifactorial, involving a complex interplay between genetic predisposition, immune system dysregulation, environmental triggers and microbial factors [1]. Alterations in gut microbiota, characterised by reduced microbiota diversity, have been consistently observed in patients with Crohn’s disease as compared to healthy controls [2]. However, the specific bacterial genera or species reported to be decreased vary substantially across studies [2].

The hallmark of Crohn’s disease pathogenesis is chronic intestinal inflammation, driven by both innate and adaptive responses [1]. Among innate immune cells, macrophages play a central role in the initiation and perpetuation of inflammatory processes. Monocytes differentiate to macrophages in response to environmental stimuli, and naïve macrophages (M0) can polarise into proinflammatory M1 macrophages or anti-inflammatory M2 macrophages. Although the M1 and M2 framework represents two ends of macrophage activation, this binary classification is increasingly recognised as an oversimplification. Accumulating evidence indicates that macrophage activation exists along a spectrum of functional states shaped by distinct stimuli and tissue contexts [3]. In Crohn’s disease, particularly in the active stage of the disease, intestinal macrophages exhibit an inflammatory M1 type, which are thought to contribute to persistent tissue inflammation [4]. Recent single-cell transcriptomic profiling of colonic tissue from patients with Crohn’s disease has revealed substantial macrophage heterogeneity, with particularly big inter-patient differences in the myeloid compartment [5].

Current evidence supports the concept that Crohn’s disease arises from a dysregulated immune response to intestinal microbes including commensal bacteria [1]. Under physiological conditions, a balanced host-microbe relationship maintains intestinal homeostasis. In Crohn’s disease, this homeostasis is disrupted; however, the mechanisms underlying this breakdown remain incompletely understood.

Campylobacter concisus is a Gram-negative bacterium commonly found in the human oral cavity [6]. Translocation of C. concisus to various regions of the gastrointestinal tract has been associated with a range of inflammatory and pathological conditions [6]. For instance, detection of C. concisus in the oesophagus has been linked to Barrett’s oesophagus [7], while its presence in the stomach and intestines has been implicated in gastric inflammation [8,9], diarrhoeal disease [10], microscopic colitis [11], and IBD [12–16]. These observations have led to the suggestion that C. concisus may contribute to the development of a subset of IBD [17]. However, Nielsen et al. examined subsequent IBD development in patients with intestinal infection of C. concisus identified by bacterial culture, and did not observe an increased risk of IBD [18]. This discrepancy across studies suggests that the relationship between C. concisus and IBD may be context-dependent and that pathogenic potential may vary between strains, with only specific C. concisus strains contributing to IBD.

Liu et al. examined the bacterial genomes of C. concisus strains isolated from patients with IBD and healthy controls and identified the csep1 gene and its association with active Crohn’s disease [19]. The csep1 gene is located either on the C. concisus plasmid pICON (Csep1P) or on the chromosome (Csep1C) of a subgroup of C. concisus strains, and it encodes a secreted 24-kDa protein, Csep1 [19]. Notably, pICON has only been found in C. concisus strains isolated from patients with Crohn’s disease [19]. Despite this disease association, the Csep1 protein structure and its role in inflammation remain unknown.

In this study, we determined the protein structure of Csep1P and investigated its impact on monocyte differentiation and macrophage activation. We found that Csep1P adopts a novel protein structure and induced a previously unrecognised macrophage activation state, characterised by predominant chemokine production but not proinflammatory cytokines. Furthermore, Csep1P-primed macrophages exhibit an enhanced proinflammatory response to Escherichia coli. These findings provide insights into the mechanistic role of specific C. concisus molecules that may contribute to Crohn’s disease pathogenesis and identify Csep1P as a potential target for therapeutic intervention.

Results

Csep1P adopts a novel α-helical fold

To gain insights that could guide our investigation into the pathogenic mechanisms of Csep1P, we determined its structure by X-ray crystallography using the single-wavelength anomalous dispersion (SAD) technique at a resolution of 1.4 Å. Analysis of the crystal packing identified no stable quaternary structure, indicating that Csep1P exists as a monomer in the crystal, consistent with its monomeric behaviour in solution [20].

Csep1P folds into a single globular α-helical domain with a hemispherical shape measuring 54 × 57 × 45 Å (Fig 1a). The structure is composed of 11 α-helices, 7 of which (coloured red and orange in Fig 1a) are arranged in an α/α-solenoid configuration. The solenoid adopts a curved horseshoe-like structure, with each of the three α-α-hairpins stabilised by an internal disulfide bond (Cys-59/Cys-68, Cys-97/Cys-106 and Cys-156/ Cys-165; Fig 1a and 1b). The remaining four helices (coloured in green and blue hues in Fig 1a) form a small subdomain at one pole of the solenoid.

thumbnail
Download:
Fig 1. Structural analysis of C. concisus Csep1P.

(a) Stereo representation of the Csep1P structure. Helices that form the α-α-solenoid moiety are coloured in red and orange, while the extra-solenoidal helices are shown in green and blue hues. (b) Topology diagram of the secondary structure elements in Csep1P. The positions of the internal disulfide bonds stabilising the α-α-haipins (Cys-59/Cys-68, Cys-97/Cys-106 and Cys-156/Cys-165) are indicated. (c) Comparison of the crystal structure of C. concisus Csep1P with H. pylori HcpB (PDB ID 1klx) and HcpC (PDB ID 1ouv). Note the similarity between the solenoid moiety of Csep1P (coloured red) and the Sel1-like tetratricopeptide repeat (TPR) solenoid folds of HcpB and HcpC. Remarkably, the positions of the internal disulfide bonds stabilising the α-α-hairpins are highly conserved in these proteins, despite their very low overall sequence identity (see S1 Fig for structure superpositions). (d) Conserved residue clusters on the surface of the solenoid moiety of Csep1P. Stereoview of the protein backbone coloured according to the evolutionary conservation of each residue among 48 C. concisus strains. The colour gradient runs from cyan (not conserved) to magenta (fully conserved). Absolutely conserved internal disulfide bonds stabilising the α-α-hairpins are shown as spheres and labelled. (e) Molecular surface of C. concisus Csep1P, coloured according to the residue conservation.

https://doi.org/10.1371/journal.ppat.1013951.g001

The comparison between the structure of C. concisus Csep1P and all structures in the Protein Data Bank (PDB) using the DALI server [21] revealed that the fold of the solenoid moiety of the molecule is remotely similar to the Sel1-like tetratricopeptide repeat (TPR) fold [22,23] (S1 Table). However, the analysis of the amino acid sequence of Csep1P using TPRpred (https://toolkit.tuebingen.mpg.de/tools/tprpred) identified no Sel1-like repeats in this protein [24]. Of particular interest are similarities to the Helicobacter cysteine-rich proteins HcpC [22] and HcpB [25]. Like in Csep1P, each of their α-α-hairpins is reinforced by an internal disulfide bridge. Although HcpC and the solenoid moiety of Csep1P share only 17% amino acid sequence identity, they can be superimposed with a root mean square deviation (RMSD) of 3.1 Å for 99 aligned Cα atoms, indicating their significant overall similarity (S1 Fig). Figs 1c and S1 illustrate that not only HcpB, HcpC and the solenoid part of Csep1P adopt a very similar fold, but the positions of the stabilising disulfide bonds are also highly conserved in these proteins. However, the similarities between Csep1P and Hcps do not extend beyond the Csep1P solenoid part – the extra-solenoidal helices α1, α6, α7, and α11 (shown in light grey) are unique to Csep1P.

Conserved residues of Csep1P map to its solenoid moiety

To analyse the Csep1P residue conservation across different Csep1P strains, we found 47 homologous amino acid sequences using PSI-BLAST search, aligned the sequences using COBALT [26] and used the alignment to generate a conservation profile with UCSF Chimera [27]. The backbone and the molecular surface representations of the structure of Csep1P were then colour-coded according to the residue conservation scores (Fig 1d and 1e). This analysis showed that the cysteine residues that form disulfide bridges (Cys-59/Cys-68, Cys-97/Cys-106 and Cys-156/ Cys-165) stabilising the α-α-hairpins of the solenoid are absolutely conserved. It also clearly showed that the surface of the solenoid moiety of Csep1P is highly conserved (Fig 1e), while most of the variable residues are mapped on the extra-solenoidal subdomain. The conservation of the disulfide-forming cysteine residues in the Csep1P sequence, and in the structures of C. concisus Csep1P, H. pylori HcpB, and H. pylori HcpC (S1 Fig) suggests they are critical for the stability of the solenoid fold and control the angle between the adjacent helices of the α-α-hairpins. The TPR-like α/α-solenoid domains often mediate protein-protein interactions [28]. The conserved surface of the solenoid moiety of Csep1P is likely important for its function and may represent the binding site(s) for (as yet unidentified) interacting proteins.

Partner proteins often bind to the concave surface of α/α-solenoids [28]. In the crystal structure of Csep1P, this solenoid surface is occupied by the protein’s C-terminus (S2 Fig). However, conservation analysis revealed a notable mismatch: highly variable C-terminus residues pack against highly conserved pockets on the solenoid surface. This pattern suggests that the C-terminus may shield a potential binding site in the crystal that could be exposed in solution.

Gene expression analysis indicates that Csep1P induced a chemokine-dominant macrophage activation

The protein structure analysis of Csep1P revealed unexpected similarity to HcpB and HcpC. A previous study reported that HcpA, a member of the Hcp protein family, promoted THP-1 monocyte-to-macrophage differentiation [29]. We therefore assessed differentiation-associated transcriptional signatures and cell morphology following Csep1P exposure, which did not support induction of a macrophage-like differentiation program under our experimental conditions, as indicated by the lack of expression in CD11b and CD68 markers, macrophage-like morphology, and enriched Gene Ontology (GO) terms (S3 Fig and S2 Table).

In macrophages, Csep1P induced an interesting activation pattern and showed an immunomodulatory effect. In THP-1-derived macrophages (M0 macrophages), Csep1P significantly upregulated 77 transcripts, including 69 protein-coding genes, seven non-coding RNA and one pseudogene (Figs 2a, 2b, S4 and S5 and Table 1). The top GO term among upregulated genes was “chemokine-mediated signalling pathway” (Fig 2c). Csep1P also significantly downregulated 24 transcripts, comprising 23 protein-coding genes and one pseudogene (Figs 2a, 2d, S4 and S5 and Table 1), with the top GO term for downregulated genes being “regulation of secretion by cell” (Fig 2e).

thumbnail
Download:
Table 1. Differentially expressed genes in THP-1-derived macrophages after incubation with Csep1P protein from transcriptomic analysis.

https://doi.org/10.1371/journal.ppat.1013951.t001

thumbnail
Download:
Fig 2. Transcriptomic analysis of global gene response to Csep1P in THP-1-derived macrophages.

THP-1-derived macrophages were incubated with Csep1P for 24 hours and then subjected to RNA sequencing. (a) Volcano plot illustrating 29,622 transcripts identified in the RNA sequencing dataset, with 77 transcripts significantly upregulated (red) and 24 transcripts significantly downregulated (blue). Gene expression differences with adjusted P-value < 0.05 and log2 fold change ≤ -1 or ≥ 1 were considered statistically significant. The volcano plot was generated using the EnhancedVolcano package. (b) Of the 77 significantly upregulated genes, 69 were protein-coding, seven were non-coding RNAs, and one was a pseudogene. (c) Enriched Gene Ontology (GO) terms describing the biological processes associated with the 77 upregulated genes, sorted according to P-value. The top three enriched terms were “chemokine-mediated signalling pathway”, “cellular response to chemokine”, and “cytokine-mediated signalling pathway”. (d) Of the 24 significantly downregulated genes, 23 were protein-coding and one was a pseudogene. (e) Enriched GO terms of the 24 downregulated genes, sorted according to P-value. The top three enriched terms were “regulation of secretion by cell”, “regulation of programmed cell death”, and “regulation of exocytosis”.

https://doi.org/10.1371/journal.ppat.1013951.g002

Among inflammatory mediator genes, Csep1P primarily induced chemokine gene expression rather than a broad classical proinflammatory cytokine profile. Specifically, Csep1P upregulated ten chemokine genes and only one proinflammatory cytokine gene, IL1B (Fig 3 and Table 2). This gene expression profile differed markedly from the canonical M1 proinflammatory macrophage activation state induced by lipopolysaccharide (LPS) and interferon-gamma (IFN-γ), which is characterised by significant upregulation of proinflammatory cytokine genes IL1A, IL1B, IL6, IL23A, and TNF (Fig 3 and Table 2) [30]. We designated the Csep1P-induced macrophage state as M1-chem, where ‘chem’ denotes the predominant upregulation of chemokines.

thumbnail
Download:
Table 2. Upregulated chemokine and cytokine genes in THP-1-derived macrophages induced by Csep1P compared to LPS + IFN-γ.

https://doi.org/10.1371/journal.ppat.1013951.t002

thumbnail
Download:
Fig 3. Comparison of upregulated M1 macrophage-associated chemokine and cytokine genes induced by Csep1P and LPS.

Csep1P induced THP-1-derived macrophages (M0) into a chemokine-dominant activation state, designated here as M1-chem. M1-chem macrophages were characterised by the upregulation of a broad range of chemokine genes, including those typically induced in classical M1 macrophages stimulated with LPS plus IFN-γ, but lacked many hallmark proinflammatory M1 cytokines. Two sets of RNA-seq data from THP-1-derived macrophages were used for analysis: the Csep1P-induced gene expression data were generated in this study, while the LPS plus IFN-γ-induced gene expression data were obtained from public database [30]. Only genes significantly upregulated (adjusted P-value < 0.05, log2 fold change ≥ 1) were included for comparison. Created in BioRender. Luk, C. (2026) (https://BioRender.com/byr6efn) is licensed under CC BY 4.0.

https://doi.org/10.1371/journal.ppat.1013951.g003

Validation of chemokine and proinflammatory cytokine protein responses support a chemokine-dominant macrophage activation state induced by Csep1P

To validate the RNA-seq findings, we quantified representative chemokines (IL-8/CXCL8 and CCL4) and proinflammatory cytokines (IL-1β and TNF-α) by ELISA in THP-1-derived macrophages and primary human macrophages treated with Csep1P or LPS. Consistent with the transcriptomic data, Csep1P induced a robust increase in chemokine production, while proinflammatory cytokines were not detected (Fig 4).

thumbnail
Download:
Fig 4. Csep1P and LPS induced chemokines and proinflammatory cytokines in macrophages.

The levels of chemokines IL-8 and CCL4, and proinflammatory cytokines IL-1β and TNF-α in the supernatants of THP-1-derived macrophages stimulated with PMA for 2 days and primary human macrophages following incubation with Csep1P (10 μg/ml) or LPS (100 ng/ml) were measured using ELISA. (a) In both THP-1-derived macrophages and human primary macrophages, Csep1P significantly increased the production of IL-8 and CCL4, while IL-1β and TNF-α were below the limit of detection. (b) LPS significantly increased the production of all four cytokines in THP-1-derived macrophages. Statistical analysis was performed using one-way analysis of variance (ANOVA) with Tukey’s post hoc test or two-tailed unpaired t-test. Bars represent the mean of triplicate experiments ± SEM. *** P < 0.001, **** P < 0.0001.

https://doi.org/10.1371/journal.ppat.1013951.g004

In THP-1-derived macrophages, Csep1P induced a significant increase in IL-8 secretion at both 4 hours (190.4 ± 3.5 pg/ml versus 31.7 ± 0.2 pg/ml in controls; P < 0.0001) and 24 hours (663.6 ± 12.4 pg/ml versus 89.2 ± 0.8 pg/ml; P < 0.0001; Fig 4a). CCL4 production also showed a significant increase at both time points: 60.8 ± 1.0 pg/ml (versus 23.6 ± 0.6 pg/ml) after 4 hours and 190.5 ± 1.5 pg/ml (versus 27.8 ± 2.7 pg/ml) after 24 hours (P < 0.0001 for both; Fig 4a). Both IL-1β and TNF-α were below the limit of detection.

In primary human macrophages differentiated from peripheral blood mononuclear cells (PBMCs; S6 Fig), Csep1P treatment for 24 hours resulted in a significant upregulation of IL-8 (2,423 ± 168.7 pg/ml compared to 149.1 ± 7.5 pg/ml in controls; P < 0.001; Fig 4a). Similarly, CCL4 production was significantly higher (573.1 ± 10.9 pg/ml compared to 80.8 ± 0.4 pg/ml in controls; P < 0.0001; Fig 4a). Both IL-1β and TNF-α were below the limit of detection.

LPS significantly increased the production of both chemokines IL-8 and CCL4, and proinflammatory cytokines IL-1β and TNF-α in THP-1-derived macrophages. After 4 and 24 hours incubation, LPS-treated THP-1-derived macrophages showed elevated IL-8 (929 ± 124 and 2,328 ± 297 pg/ml, respectively), CCL4 (1,564 ± 3 pg/ml and 2,865 ± 72 pg/ml, respectively), IL-1β (83 ± 5 and 127 ± 8 pg/ml, respectively) and TNF-α, which was undetectable at 4 hours but reached 274 ± 21 pg/ml after 24 hours of incubation (Fig 4b).

Together, these data demonstrate that Csep1P selectively induces chemokine production without triggering a canonical proinflammatory cytokine response, supporting the designation of a chemokine-dominant macrophage activation state.

Csep1P priming enhances macrophages proinflammatory responses to gut commensal bacterium Escherichia coli

To assess the functional consequences of Csep1P-mediated M1-chem state activation, we evaluated the responses of both THP-1-derived and primary human macrophages to gut bacterium E. coli strain K12 by measuring IL-8, IL-1β, and TNF-α production. Our data showed that Csep1P-primed macrophages mounted a stronger proinflammatory response to E. coli (Fig 5).

thumbnail
Download:
Fig 5. Effects of Csep1P priming on production of chemokines and proinflammatory cytokines by macrophages in response to E. coli K12.

The levels of chemokine IL-8 and proinflammatory cytokines IL-1β and TNF-α were measured using ELISA. (a) In THP-1-derived macrophages (10⁶ cells/ml), Csep1P priming for 4 hours significantly increased IL-8, IL-1β, and TNF-α production following stimulation with E. coli K12 at both MOI 1 and MOI 10 for 2 hours, compared to non-primed macrophages. (b) In human primary macrophages (3.3 × 10⁵ cells/ml), Csep1P priming for 4 hours significantly increased IL-8 and TNF-α production following stimulation using E. coli K12 at MOI 1 for 2 hours. IL-1β was below the limit of detection. Statistical analysis was performed using one-way analysis of variance (ANOVA) with Tukey’s post hoc test. Bars represent the mean of triplicate experiments ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001. MOI: multiplicity of infection.

https://doi.org/10.1371/journal.ppat.1013951.g005

Csep1P-primed THP-1-derived macrophages produced significantly higher levels of IL-8, IL-1β, and TNF-α upon stimulation with E. coli strain K12 for 2 hours at both multiplicities of infection (MOIs) of 1 and 10, compared to unprimed cells: IL-8: 4,369 ± 33.2 and 6,165 ± 197.8 pg/ml versus 1,818 ± 11.2 and 2,111 ± 53.1 pg/ml (P < 0.0001; Fig 5a); IL-1β: 38.6 ± 0.4 and 114.2 ± 1.7 pg/ml versus 19 ± 0.5 and 23.3 ± 0.6 pg/ml (P < 0.0001; Fig 5a); TNF-α: 1,199 ± 52.4 and 2,057 ± 84.7 pg/ml versus 700.7 ± 54.7 and 1,002 ± 33.2 pg/ml (P < 0.001 and P < 0.0001, respectively; Fig 5a).

Similar effects were observed in human primary macrophages. Csep1P-primed macrophages produced significantly more IL-8 and TNF-α in response to E. coli (MOI 1) than unprimed controls (1,017 ± 38.1 and 444.4 ± 0.3 pg/ml versus 739.4 ± 30.8 and 419.5 ± 5.7 pg/ml in controls; P < 0.01; Fig 5b). However, IL-1β was below the limit of detection in primary macrophage supernatants following E. coli stimulation for 2 hours, which is most likely due to the lower cell numbers used.

Delta like canonical Notch ligand 4 (DLL4) silencing reduces CCL4 and TNF-α production in Csep1P-primed macrophages following E. coli challenge

Transcriptomic analysis using RNA-seq showed that Csep1P significantly upregulated DLL4 gene expression in THP-1-derived macrophages (Fig 6a and Table 1). We validated this by quantitative real-time PCR (qRT-PCR), which showed a 3.3 ± 0.5-fold increase in DLL4 expression after 24 hours of incubation with Csep1P (P < 0.01; Fig 6a).

thumbnail
Download:
Fig 6. Upregulation of DLL4 expression by Csep1P in macrophages and its role in regulating macrophage response to E. coli.

The change in DLL4 gene expression in THP-1-derived macrophages induced by Csep1P was measured using qRT-PCR. The levels of chemokines IL-8 and CCL4, and proinflammatory cytokines IL-1β and TNF-α in the supernatants of Csep1P-primed THP-1-derived macrophages with and without DLL4 silencing were measured using ELISA. (a) Csep1P induced significant upregulation of DLL4 gene expression, demonstrated by both RNA-seq and qRT-PCR. Statistical significance was determined by two-tailed unpaired t-test. (b) Silencing DLL4 gene expression significantly decreased the production of CCL4 and TNF-α in response to E. coli strain K12 following 2 hours incubation, but not IL-8. IL-1β was below the limit of detection. Statistical significance was determined by one-way analysis of variance (ANOVA) with Tukey’s post hoc test. Bars represent the mean of triplicate experiments ± SEM. *P < 0.05, **P < 0.01, ****P < 0.0001. MOI: multiplicity of infection.

https://doi.org/10.1371/journal.ppat.1013951.g006

To further determine the role of DLL4 in the enhanced response of Csep1P-primed macrophages to E. coli stimulation, we silenced DLL4 in THP-1-derived macrophages using small interfering RNA (siRNA) and confirmed knockdown by qRT-PCR (S7 Fig). Macrophages were primed with Csep1P for 4 hours, followed by a 2-hour stimulation with E. coli K12 (MOI 1). DLL4 silencing significantly reduced CCL4 and TNF-α production compared to negative controls (CCL4: 413 ± 17.3 pg/ml versus 497.3 ± 44 pg/ml; P < 0.01; TNF-α: 140.8 ± 2.3 pg/ml versus 170.5 ± 1.5 pg/ml; P < 0.0001; Fig 6b). In contrast, IL-8 production was not significantly affected (633.2 ± 19.6 pg/ml versus 659 ± 11.3 pg/ml; P > 0.05; Fig 6b), and IL-1β was below the limit of detection under these conditions.

Discussion

In this study, we determined the structure of the C. concisus Csep1P protein and investigated its immunomodulatory role.

Structural analysis revealed that Csep1P adopts a novel α-helical fold incorporating an α/α-solenoid moiety (Fig 1). The solenoid region showed remote structural similarity to H. pylori HcpB and HcpC proteins, indicating shared evolutionary pressures favouring a conserved functional scaffold. Although the extra-solenoidal domain of Csep1P is unique to this protein, the conservation of the solenoid fold across distantly related bacterial genera suggests this domain is functionally important for host-pathogen interactions [31]. While the precise biological functions of most solenoid Hcp proteins remain poorly understood, the evidence of positive selection targeting surface-exposed residues in Hcp proteins suggests they play active roles in modulating bacterial-host interactions. A previous study reported that H. pylori HcpA, but not HcpC, promoted THP-1 monocyte-to-macrophage differentiation [29]. The precedent of HcpA’s role in immune modulation prompted us to investigate whether Csep1P similarly functions in host immune manipulation, albeit potentially through distinct mechanisms. Under the experimental conditions used, we did not detect evidence of differentiation based on transcriptional signatures or cell morphology (S2 Table and S3 Fig).

However, our investigation on the effects of Csep1P on macrophages revealed interesting results. Csep1P induced a novel activation state in THP-1-derived macrophages. The chemokine and cytokine gene expression profile of this state was distinct from that of classical M1 macrophages induced by LPS plus IFN-γ (Fig 3 and Tables 1 and 2) [30]. We designate this distinct Csep1P-induced state as M1-chem, reflecting its predominant upregulation of chemokine genes.

We further validated the Csep1P-induced M1-chem state identified by transcriptomic analysis at the protein level by measuring representative chemokines and cytokines associated with M1-macrophage in both THP-1-derived and primary human macrophages. Csep1P stimulation led to increased secretion of chemokines IL-8 and CCL4, but not proinflammatory cytokines IL-1β and TNF-α (Fig 4). This protein expression profile was consistent with the transcriptomic data, except for IL-1β. Although Csep1P upregulated IL1B expression following 24 hours incubation, it did not increase IL-1β secretion (Figs 3 and 4a and Table 1). We speculate that Csep1P alone does not provide the necessary signals required for inflammasome activation to cleave pro-IL-1β into its mature IL-1β form for detectable secretion [32]. In contrast, LPS stimulation induced robust secretion of IL-8, CCL4, IL-1β, and TNF-α (Fig 4b). Since chemokines such as IL-8 and CCL4 are potent immune cell recruiters [33], these findings suggest that Csep1P may promote immune cell recruitment prior to overt inflammation, thereby reshaping the immune landscape.

Macrophage polarisation is increasingly recognised as a continuum rather than a M1 and M2 classification, with multiple intermediate or context-dependent activation states described [3]. In this context, the Csep1P-induced M1-chem state represents a non-canonical, chemokine-dominant macrophage activation state that is distinct from classical M1 polarisation. The selective induction of chemokines in the absence of strong proinflammatory cytokine secretion suggests that Csep1P does not directly trigger full inflammatory activation but instead establishes a primed macrophage state poised for enhanced responses to secondary microbial stimulation, while also promoting immune cell recruitment. Indeed, in our experiments, although Csep1P alone did not directly induce proinflammatory cytokine secretion, it altered macrophage responsiveness to commensal E. coli. Both THP-1-derived and human primary macrophages primed with Csep1P exhibited significantly increased secretion of IL-8 and TNF-α upon exposure to E. coli strain K12 (Fig 5), supporting the role of Csep1P in promoting exaggerated proinflammatory response to other gut bacteria.

Crohn’s disease predominantly affects the terminal ileum [34]. Despite the introduction of numerous anti-inflammatory therapies, relapse rates remain high, possibly because current treatments do not address microbial triggers. Each day, approximately 1 – 1.5 litres of saliva pass from the oral cavity to the intestines, providing a route for the repeated translocation of potential Crohn’s disease triggers, such as C. concisus, to the gastrointestinal tract. Virulent C. concisus strains can invade intestinal epithelial cells, induce epithelial cell death and produce proinflammatory secondary metabolites [35]. The pathogenesis of IBD is increasingly recognised to result from inappropriate immune responses to commensal gut bacteria, however the exact triggers remain unclear [36]. Our finding that Csep1P activates macrophage to a M1-chem state and enhances macrophage proinflammatory responses to E. coli strain K12 suggests that virulent C. concisus strains may act as sensitisers by secreting immunomodulatory molecules such as Csep1P. This, in turn, could reshape the landscape of the mucosal immune system and reprogram macrophages to mount heightened proinflammatory responses to gut bacteria, thereby triggering the initiation or relapse of gut inflammation in Crohn’s disease.

A previous study by Nielsen et al. did not find an association between intestinal infection with C. concisus and an increased risk of subsequent development of IBD [18]. Our findings in this study raise the possibility that strain-specific factors, such as Csep1 expression, may be important, and that future studies assessing infection with Csep1-positive C. concisus strains are needed to clarify their potential role in Crohn’s disease.

The DLL4 gene was the only signalling-related gene upregulated by Csep1P (Fig 6a and Table 1). DLL4, a ligand of the Notch signalling pathway, has been shown to promote proinflammatory macrophage activation and to influence T-cell differentiation, particularly toward Th1 and Th17 lineages [37–39]. To assess its role in Csep1P-mediated macrophage responses, DLL4 expression was silenced in THP-1-derived macrophages (S7 Fig). DLL4 silencing significantly reduced secretion of CCL4 and TNF-α in Csep1P-primed macrophages following E. coli challenge (Fig 6b), indicating that DLL4-Notch signalling contributes to the enhanced macrophage inflammatory response induced by Csep1P. In contrast, IL-8 secretion was not affected by DLL4 silencing (Fig 6b), likely reflecting the involvement of additional receptors and signalling pathways in E. coli-induced macrophage activation. Beyond its role in macrophages, DLL4 expressed by antigen-presenting cells has been reported to modulate T-cell responses through cell-cell Notch signalling interactions [40]. In the intestinal mucosa, such macrophage and T cell crosstalk may contribute to the amplification of Th1/Th17-type immune responses characteristic of Crohn’s disease [40]. Although direct effects on T cells were not examined in this study, the induction of DLL4 by Csep1P suggests a potential mechanism by which C. concisus-derived molecules may influence both innate and adaptive immune responses through macrophage-mediated signalling.

In conclusion, our study demonstrates that the structurally unique Csep1P induces a chemokine-dominant inflammatory state in macrophages and reprograms macrophage to enhance proinflammatory response to E. coli. These findings provide a potential mechanistic rationale for the role of virulent C. concisus strains in Crohn’s disease pathogenesis and identify Csep1P as a potential therapeutic target.

Materials and methods

Determination of Csep1P protein structure

Protein crystallisation and data collection.

Native crystals of recombinant Csep1P (lacking the N-terminal signal peptide) were grown using the hanging-drop method as previously described [20]. The crystals belonged to space group P64, with unit cell dimensions a = b = 85.8 Å, c = 55.2 Å, and contained one molecule per asymmetric unit. The platinum derivative was obtained by soaking crystals overnight in 1 mM potassium hexabromoplatinate. To perform data collection at cryogenic temperatures, crystals were loop-mounted in a cryoprotecting solution containing 36% w/v PEG 4000, 100 mM ammonium acetate, 80 mM sodium acetate trihydrate (pH 4.6), and 10% (v/v) glycerol, then flash-cooled by plunging into liquid nitrogen. Native X-ray diffraction data and SAD data for the platinum derivative (λ = 1.071 Å for both) were collected to 1.4 Å resolution on the MX2 beamline at the Australian Synchrotron. All data were processed and scaled using XDS [41] and AIMLESS [42] from the Collaborative Computational Project, Number 4 (CCP4) suite [43]. Data collection and processing statistics are summarised in S3 Table.

Structure determination.

Autosol [44] from the PHENIX software suite [45] was used to locate the platinum sites and build an initial partial model, comprising 173 residues (R = 0.239, Rfree = 0.241). The model was then manually completed using Coot [46] and refined to R = 0.125, Rfree = 0.156 using REFMAC [47], with waters placed by ARP/WARP [48]. This model was used for phasing the data for the native crystal by molecular replacement. The final model of Csep1P (refined to R = 0.126, Rfree = 0.167), contains all the expected amino acid residues (22–222) and 225 water molecules. Refinement statistics and stereochemistry are shown in S4 Table. All the non-glycine residues lie within permitted regions of the Ramachandran plot, with 99% in the most favoured regions. Representations of crystal structures were prepared using PYMOL (The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC) and UCSF Chimera [27].

Csep1P protein production and experimental optimisation

The expression vector for recombinant Csep1P protein was based on the csep1 gene sequence from the pICON plasmid of C. concisus strain P2CDO4 [19], minus the N-terminal signal peptide. Protein expression and purification were performed by GenScript (NJ, USA; S8 Fig). Protein identity was confirmed by mass spectrometry at the Mark Wainwright Analytical Centre, University of New South Wales, and no LPS was detected in the sample. Prior to incubation with macrophages, Csep1P was sterilised by filtering through a 0.22 µm filter (Merck Millipore). Preliminary experiments were conducted to determine the ability of Csep1P to induce responses in macrophages, and to optimise protein concentration and incubation times. Based on the results, 10 μg/ml of Csep1P and 24-hour incubation were used for RNA-seq, while 10 μg/ml of Csep1P and 4-hour incubation were used for priming experiments (S5 Table).

Cells used in this study

The human monocytic leukaemia cell line THP-1 (ATCC No. TIB-202) was differentiated into macrophages using 10 nM phorbol 12-myristate 13-acetate (PMA; Sigma-Aldrich, NSW, AU) for 2 days as previously described [49]. Both undifferentiated and differentiated THP-1 cells were used. Primary macrophages were prepared from PBMCs obtained from a healthy individual and differentiated using 50 ng/ml recombinant human macrophage colony-stimulating factor (M-CSF; Sigma-Aldrich) for 7 days [50]. All cell types were cultured in RPMI 1640 medium (Thermo Fisher Scientific, CA, USA) supplemented with 10% foetal bovine serum (FBS; Cytiva, MA, USA), 100 U/ml penicillin, and 100 µg/ml streptomycin (Thermo Fisher Scientific), which we refer to as complete RPMI medium. Cells were maintained at 37°C in a humidified incubator with 5% CO2.

Macrophage differentiation was confirmed by quantifying expression of colony-stimulating factor 1 receptor (CSF1R) using qRT-PCR (Forward primer: GCTGCCTTACAACGAGAAGTGG; Reverse primer: CATCCTCCTTGCCCAGACCAAA; OriGene; SKU: HP208389). On top of gene expression change, the cell morphology of macrophages was also confirmed using light microscopy and detection of CD11b and CD68 expression by fluorescence microscopy [51].

Examination of the effects of Csep1P on THP-1 monocyte differentiation by fluorescence staining

THP-1 monocytes were seeded on coverslips in a 12-well plate at a concentration of 5 × 105 cells/ml in complete RPMI medium with Csep1P (10 μg/ml), with fresh Csep1P added every 24 hours for 3 days. THP-1 monocytes incubated with 10 nM PMA were used as a positive control. THP-1 monocytes without treatment were used as an untreated control. The THP-1 cells were fixed with 3.6% paraformaldehyde for 15 minutes, permeabilised with 0.1% triton for 10 minutes, and blocked with 1% bovine serum albumin (BSA) for 1 hour. The filamentous actin (F-actin) and nuclei were then stained with Alexa Fluor 488 phalloidin (Cell Signaling Technology, MA, USA; Cat. no. 8878S) and Hoechst 33342 (Invitrogen, CA, USA), respectively. The cells were mounted onto glass slides and examined using a fluorescent microscope (Olympus BX61; Olympus, Tokyo, Japan) with FITC (Excitation wavelength: 480 nm; Emission wavelength: 520 nm) and DAPI (Excitation wavelength: 365 nm; Emission wavelength: 430 nm) filters to visualise the F-actin and nucleus, respectively.

RNA-seq and transcriptomic analysis

Global gene expression in response to Csep1P protein in naïve M0 macrophages was examined using RNA-seq. THP-1 monocytes and THP-1-derived macrophages were treated with Csep1P (10 μg/ml) in RPMI medium without FBS and antibiotics for 24 hours. Untreated THP-1-derived macrophages serve as the negative control. Following incubation, total RNA was extracted from THP-1-derived macrophages using the ISOLATE II RNA Mini Kit (Bioline, NSW, Australia; Cat. no. BIO-52072) and submitted to the Australian Genome Research Facility Ltd for RNA-seq. Experiments were conducted in triplicate.

RNA-seq data were analysed as previously described [52]. Briefly, the quality of raw sequencing reads was assessed using FastQC (version 0.11.8) (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/). Adapter sequences and low-quality bases were trimmed using Trimmomatic (version 0.38), with a minimum Phred score of 3 for both leading and trailing bases, and a sliding window of 4:15. Reads shorter than 30 bp following trimming were discarded [53]. Trimmed reads were aligned to the human reference genome GRCh38.p14 (GenBank ID: GCA_000001405.29) using HISAT2 (version 2.1.0) [54]. The resulting SAM files were converted to BAM format using SAMtools (version 1.11) [55], and gene-level quantification was performed using featureCounts from the Subread package (version 2.0.1) [56]. Differentially expressed genes (DEGs) between Csep1P-primed and untreated control samples were identified using the DESeq2 package (version 1.36.0) with default normalisation. Genes were considered significantly differentially expressed if they met the thresholds of adjusted P-value < 0.05 and log2 fold change ≤ -1 or ≥ 1 [57]. The list of DEGs was submitted to the Enrichr database for GO enrichment analysis to identify associated biological processes [58–60].

Comparison of gene expression of chemokines and proinflammatory cytokines induced by Csep1P and LPS

The chemokine and proinflammatory cytokine genes upregulated by Csep1P in this study were compared with previously reported M1-macrophage associated cytokines induced by LPS plus IFN-γ [30]. Genes that were significantly differentially expressed, with the thresholds of adjusted P-value < 0.05 and log2 fold change ≤ -1 or ≥ 1, were included for comparison.

Measurement of secreted cytokines by macrophages in response to Csep1P

To validate the expression of upregulated chemokines and proinflammatory cytokines induced by Csep1P in the transcriptomic analysis, we measured the protein levels of two representative chemokines including IL-8 and CCL4, and two representative proinflammatory cytokines including IL-1β and TNF-α. The selection of these cytokines was based on their association with M1 macrophage and IBD [61–63].

THP-1-derived macrophages were seeded in 6-well plates at a density of 106 cells/ml and incubated with Csep1P (10 μg/ml) in RPMI medium without FBS and antibiotics for 4 hours and 24 hours; untreated cells served as a negative control. Supernatants were collected and chemokine and cytokine levels were measured using commercially available ELISA kits (Invitrogen; Cat. no. IL-8: CHC1301, CCL4: 88–7034, IL-1β: CHC1213, TNF-α: CHC1753). Human primary macrophages (3.3x105 cells/ml) were also used to validate the 24-hour Csep1P incubation results. All measurements were conducted in triplicate. LPS (from E. coli O55:B5; 100 ng/ml; Sigma-Aldrich; Cat. no. L6529) incubated with THP-1-derived macrophages for 4 hours and 24 hours was used as a positive control for measurement of IL-8, CCL4, IL-1β, and TNF-α [64].

Investigating the impact of Csep1P on macrophage response to commensal gut bacteria

To investigate the potential role of Csep1P protein in modulating macrophage function, we compared the production levels of IL-8, IL-1β, and TNF-α in THP-1-derived macrophages following exposure to microbial stimulation with the commensal E. coli strain K12, with and without Csep1P priming. Briefly, THP-1-derived macrophages (106 cells/ml) were incubated in RPMI medium with or without Csep1P (10 μg/ml). After 4 hours, cells were washed with Dulbecco’s phosphate-buffered saline (DPBS), resuspended in RPMI medium without FBS and antibiotics, and stimulated with heat-killed E. coli K12 at MOI 1 or 10. Primary macrophages (3.3x105 cells/ml) were also used to validate the MOI 1 results. After a 2-hour incubation, supernatants were collected, and IL-8, IL-1β, and TNF-α levels were measured using ELISA kits. All measurements were conducted in triplicate.

Investigation of the potential involvement of DLL4 in Csep1P upregulated macrophage response to E. coli

Our transcriptomic analysis identified upregulation of the DLL4 gene, we confirmed it using qRT-PCR (Forward primer: AACTACTGCACCCACCACT, Reverse primer: GCCATCCTCCTGGTCCTTACA) [65]. We then examined whether silencing the DLL4 gene expression impacts the Csep1P-upregulated macrophage response to E. coli. The DLL4 gene encodes a ligand for Notch receptors [37].

For silencing DLL4 gene expression, THP-1-derived macrophages were seeded in 6-well plates at a density of 106 cells/ml and transfected with 40 nM of two siRNA sets (SASI_Hs01_00174509 and SASI_Hs02_00352665; Sigma-Aldrich) for 16 hours. Transfection was carried out using Lipofectamine RNAiMAX reagent (Thermo Fisher Scientific) diluted in Opti-MEM Reduced Serum Medium (Thermo Fisher Scientific) [66]. MISSION GAPDH SiRNA (SASI_Hs01_00140986; Sigma-Aldrich) and MISSON Universal Negative Control #1 (Sigma-Aldrich, Cat. No. SIC001) were used as positive and negative controls, respectively. The efficiency of DLL4 gene silencing was validated using qRT-PCR [67].

DLL4-silenced cells were washed with DPBS, resuspended in RPMI medium without FBS and antibiotics, and incubated with Csep1P (10 μg/ml) for 4 hours. Cells were then washed again with DPBS and incubated with heat-killed E. coli K12 at MOI 1 in RPMI medium without FBS and antibiotics for 2 hours. Supernatants were collected and levels of IL-8, IL-1β, CCL4, and TNF-α were measured using ELISA kits. All measurements were conducted in triplicate.

Statistical analysis

All measurements for RNA-seq and ELISAs were conducted in triplicate. For comparisons involving multiple treatment groups and the control group, statistical significance was assessed using one-way analysis of variance (ANOVA) followed by Tukey’s post hoc test. For comparisons between a single treatment group and the control, a two-tailed unpaired t-test was used. All statistical analyses were performed using GraphPad Prism (v9.3.1). P < 0.05 was considered statistically significant. All data were expressed as mean of triplicate experiments ± standard error of the mean (SEM), unless otherwise stated.

Supporting information

S1 Fig. Structure superpositions of C. concisus Csep1P protein with H. pylori HcpB and HcpC.

The superpositions of the solenoid moiety (red) of the structure of C. concisus Csep1P with the crystal structures of H. pylori HcpB (PCD ID 1klx) and HcpC (PDB ID 1ouv), highlighting the structural conservation of the internal disulfide bonds (shown as spheres) stabilising the α-α-hairpins in these proteins.

https://doi.org/10.1371/journal.ppat.1013951.s001

(TIF)

S2 Fig. Conserved residue clusters on the surface of the solenoid moiety of Csep1P. Stereo diagram showing the C-terminus lodged within the concave groove of the solenoid moiety.

The side chains of highly variable residues at positions 203 and 209 (cyan) are accommodated within highly conserved (magenta) pockets on the concave surface of the solenoid.

https://doi.org/10.1371/journal.ppat.1013951.s002

(TIF)

S3 Fig. Fluorescence microscope images and Gene Ontology (GO) analysis of THP-1 monocytes incubated with Csep1P or PMA for 72 hours.

(a) THP-1 monocytes incubated with media only (untreated), Csep1P, or PMA were observed after 72 hours. Cell nucleus and F-actin were stained with Hoechst 33342 and Alexa Fluor 488 phalloidin and visualised using DAPI and FITC filters, respectively. Incubation with Csep1P resulted in slight increase in cell size and F-actin aggregation when compared with the untreated cells. PMA-treated THP-1 monocytes were used as a positive control for macrophage-like differentiation. Both Csep1P-treated and untreated THP-1 monocytes showed adherence to the cover slip. Scale bars for 20× and 100 × magnifications represent 50 μm and 10 μm, respectively. (b) Enriched GO terms of the 15 significantly upregulated genes, sorted according to P-value. The top three enriched terms were “regulation of defence response”, “natural killer cell degranulation”, and “neutrophil degranulation”. (c) Enriched GO terms of the 32 significantly downregulated genes, sorted according to P-value. The top three enriched terms were “myeloid cell activation involved in immune responses”, “protein K48-linked deubiquitination”, and “protein K63-linked deubiquitination”.

https://doi.org/10.1371/journal.ppat.1013951.s003

(TIF)

S4 Fig. Principal component analysis (PCA) of RNA-seq data.

A total of 6 samples were included in the analysis, with red dots representing Csep1P-treated and blue dots representing untreated control THP-1-derived macrophages. The plot shows clear separation between control and treatment groups, indicating significant changes in global gene expression in THP-1-derived macrophages after 24-hr incubation with Csep1P. Gene counts were log2-transformed and normalised prior to PCA. The PCA plot was generated using the ggplot2 package.

https://doi.org/10.1371/journal.ppat.1013951.s004

(TIF)

S5 Fig. Heatmap of differentially expressed genes in Csep1P-treated THP-1-derived macrophages.

The heatmap shows 101 genes that were differentially expressed (P < 0.05; log2 fold change ≤ -1 or ≥ 1). The gene read counts of the experimental triplicates obtained using featureCounts from the Subread package (version 2.0.1) were log2-normalised and expressed in a colour scheme. Heatmap was generated using the pheatmap package.

https://doi.org/10.1371/journal.ppat.1013951.s005

(TIF)

S6 Fig. Confirmation of primary macrophage differentiation using microscopy and qRT-PCR.

(a) Light microscopy image (40×) showing that primary macrophages differentiated from peripheral blood mononuclear cells (PBMCs) using macrophage colony-stimulating factor (M-CSF), display a characteristic elongated phenotype. (b) M-CSF treatment significantly upregulated CSF1R gene expression in primary macrophages (1.6 ± 0.04-fold change, *** = P < 0.001). Fold change was calculated using the comparative threshold cycle CT (2-ΔΔCT) method, with target gene expression normalised to the housekeeping gene GAPDH and calculated relative to the untreated control. Statistical significance was assessed using two-tailed unpaired t-test. Bars represent the mean of triplicate experiments ± SEM. *** = P < 0.001. (c) Immunofluorescence staining of cell nuclei (Hoechst 33342, DAPI filter), CD11b (anti-CD11b antibody, FITC filter), and CD68 (anti-CD68 antibody, CY5 filter), visualised at 100 × magnification. Treatment of macrophages with M-CSF resulted in increased fluorescence intensity for both CD11b and CD68 compared to untreated cells. Scale bars represent 10 μm.

https://doi.org/10.1371/journal.ppat.1013951.s006

(TIF)

S7 Fig. Confirmation of DLL4 gene silencing in THP-1-derived macrophages with and without Csep1P and E. coli stimulation.

DLL4 gene expression in DLL4-silenced THP-1-derived macrophages was measured by qRT-PCR to validate the gene silencing efficacy of the transfected siRNA. (a) DLL4 silencing was confirmed by a significant reduction in DLL4 gene expression, showing a 0.37 ± 0.03-fold change (P < 0.0001). (b) DLL4 silencing significantly reduced DLL4 expression in Csep1P-primed THP-1-derived macrophages incubated with E. coli, showing a 62.6 ± 5.5-fold and 34.1 ± 4.7-fold change for the siRNA negative control and DLL4-silenced cells, respectively (P < 0.01). DLL4 gene fold change is shown relative to the untreated control and normalised to the housekeeping gene GAPDH. Statistical significance was assessed by two-tailed unpaired t-test for (a) and by one-way analysis of variance (ANOVA) with Tukey’s post-hoc test for (b). Bars represent the mean of triplicate experiments ± SEM. ** = P < 0.01, **** = P < 0.0001. MOI = multiplicity of infection.

https://doi.org/10.1371/journal.ppat.1013951.s007

(TIF)

S8 Fig. Coomassie Blue-stained SDS-PAGE gel of recombinant C. concisus Csep1P used for transcriptomic and cytokine production analyses in macrophages.

The N-terminally His6-tagged protein was expressed in E. coli BL21 Star (DE3) and purified by GenScript. Its identity was confirmed by mass spectrometry. LPS was not detectable in the purified protein using the Pierce Chromogenic Endotoxin Quant Kit. BSA: bovine serum albumin.

https://doi.org/10.1371/journal.ppat.1013951.s008

(TIF)

S1 Table. Results of the similarity search of Csep1P against the protein structures deposited in the PDB.

https://doi.org/10.1371/journal.ppat.1013951.s009

(XLSX)

S2 Table. Differentially expressed genes in THP-1 monocytes after incubation with Csep1P protein from transcriptomic analysis.

https://doi.org/10.1371/journal.ppat.1013951.s010

(XLSX)

S3 Table. X-ray data collection and processing statistics.

https://doi.org/10.1371/journal.ppat.1013951.s011

(XLSX)

S4 Table. Refinement statistics.

https://doi.org/10.1371/journal.ppat.1013951.s012

(XLSX)

S5 Table. Optimisation of Csep1P dosage.

https://doi.org/10.1371/journal.ppat.1013951.s013

(XLSX)

Acknowledgments

We thank the staff at the Australian Synchrotron for their assistance with X-ray diffraction data collection.

References

  1. 1. Torres J, Mehandru S, Colombel J-F, Peyrin-Biroulet L. Crohn’s disease. Lancet. 2017;389(10080):1741–55. pmid:27914655
  2. 2. Pittayanon R, Lau JT, Leontiadis GI, Tse F, Yuan Y, Surette M, et al. Differences in gut microbiota in patients with vs without inflammatory bowel diseases: a systematic review. Gastroenterology. 2020;158(4):930-946.e1. pmid:31812509
  3. 3. Lendeckel U, Venz S, Wolke C. Macrophages: shapes and functions. ChemTexts. 2022;8(2):12. pmid:35287314
  4. 4. Zhang K, Guo J, Yan W, Xu L. Macrophage polarization in inflammatory bowel disease. Cell Commun Signal. 2023;21(1):367. pmid:38129886
  5. 5. Garrido-Trigo A, Corraliza AM, Veny M, Dotti I, Melón-Ardanaz E, Rill A, et al. Macrophage and neutrophil heterogeneity at single-cell spatial resolution in human inflammatory bowel disease. Nat Commun. 2023;14(1):4506. pmid:37495570
  6. 6. Zhang L, Budiman V, Day AS, Mitchell H, Lemberg DA, Riordan SM, et al. Isolation and detection of Campylobacter concisus from saliva of healthy individuals and patients with inflammatory bowel disease. J Clin Microbiol. 2010;48(8):2965–7. pmid:20519479
  7. 7. Macfarlane S, Furrie E, Macfarlane GT, Dillon JF. Microbial colonization of the upper gastrointestinal tract in patients with Barrett’s esophagus. Clin Infect Dis. 2007;45(1):29–38. pmid:17554697
  8. 8. Ferreira EO, Lagacé-Wiens P, Klein J. Campylobacter concisus gastritis masquerading as Helicobacter pylori on gastric biopsy. Helicobacter. 2022;27(2):e12864. pmid:34820966
  9. 9. Luk CYM, Lee SA, Naidovski N, Liu F, Tay ACY, Wang L, et al. Investigation of Campylobacter concisus gastric epithelial pathogenicity using AGS cells. Front Microbiol. 2024;14:1289549. pmid:38274743
  10. 10. Lindblom GB, Sjögren E, Hansson-Westerberg J, Kaijser B. Campylobacter upsaliensis, C. sputorum sputorum and C. concisus as common causes of diarrhoea in Swedish children. Scand J Infect Dis. 1995;27(2):187–8. pmid:7660089
  11. 11. Nielsen HL, Dalager-Pedersen M, Nielsen H. High risk of microscopic colitis after Campylobacter concisus infection: population-based cohort study. Gut. 2020;69(11):1952–8. pmid:32111632
  12. 12. Zhang L, Man SM, Day AS, Leach ST, Lemberg DA, Dutt S, et al. Detection and isolation of Campylobacter species other than C. jejuni from children with Crohn’s disease. J Clin Microbiol. 2009;47(2):453–5. pmid:19052183
  13. 13. Man SM, Zhang L, Day AS, Leach ST, Lemberg DA, Mitchell H. Campylobacter concisus and other Campylobacter species in children with newly diagnosed Crohn’s disease. Inflamm Bowel Dis. 2010;16(6):1008–16. pmid:19885905
  14. 14. Mahendran V, Riordan SM, Grimm MC, Tran TAT, Major J, Kaakoush NO, et al. Prevalence of Campylobacter species in adult Crohn’s disease and the preferential colonization sites of Campylobacter species in the human intestine. PLoS One. 2011;6(9):e25417. pmid:21966525
  15. 15. Mukhopadhya I, Thomson JM, Hansen R, Berry SH, El-Omar EM, Hold GL. Detection of Campylobacter concisus and other Campylobacter species in colonic biopsies from adults with ulcerative colitis. PLoS One. 2011;6(6):e21490. pmid:21738679
  16. 16. Kirk KF, Nielsen HL, Thorlacius-Ussing O, Nielsen H. Optimized cultivation of Campylobacter concisus from gut mucosal biopsies in inflammatory bowel disease. Gut Pathog. 2016;8:27. pmid:27252786
  17. 17. Zhang L. Oral Campylobacter species: initiators of a subgroup of inflammatory bowel disease? World J Gastroenterol. 2015;21(31):9239–44. pmid:26309350
  18. 18. Nielsen HL, Dalager-Pedersen M, Nielsen H. Risk of inflammatory bowel disease after Campylobacter jejuni and Campylobacter concisus infection: a population-based cohort study. Scand J Gastroenterol. 2019;54(3):265–72. pmid:30905214
  19. 19. Liu F, Ma R, Tay CYA, Octavia S, Lan R, Chung HKL, et al. Genomic analysis of oral Campylobacter concisus strains identified a potential bacterial molecular marker associated with active Crohn’s disease. Emerg Microbes Infect. 2018;7(1):64. pmid:29636463
  20. 20. Rahman MM, Goff B, Zhang L, Roujeinikova A. Refolding, characterization, and preliminary X-ray crystallographic studies on the Campylobacter concisus plasmid-encoded secreted protein Csep1P associated with Crohn’s disease. Cryst. 2018;8(10):391.
  21. 21. Holm L, Laiho A, Törönen P, Salgado M. DALI shines a light on remote homologs: one hundred discoveries. Protein Sci. 2023;32(1):e4519. pmid:36419248
  22. 22. Lüthy L, Grütter MG, Mittl PRE. The crystal structure of Helicobacter cysteine-rich protein C at 2.0 A resolution: similar peptide-binding sites in TPR and SEL1-like repeat proteins. J Mol Biol. 2004;340(4):829–41. pmid:15223324
  23. 23. Mittl PRE, Schneider-Brachert W. Sel1-like repeat proteins in signal transduction. Cell Signal. 2007;19(1):20–31. pmid:16870393
  24. 24. Karpenahalli MR, Lupas AN, Söding J. TPRpred: a tool for prediction of TPR-, PPR- and SEL1-like repeats from protein sequences. BMC Bioinform. 2007;8:2. pmid:17199898
  25. 25. Luthy L, Grutter MG, Mittl PRE. The crystal structure of Helicobacter pylori cysteine-rich protein B reveals a novel fold for a penicillin-binding protein. J Biol Chem. 2002;277(12):10187–93. pmid:11777911
  26. 26. Papadopoulos JS, Agarwala R. COBALT: constraint-based alignment tool for multiple protein sequences. Bioinformatics. 2007;23(9):1073–9. pmid:17332019
  27. 27. Pettersen EF, Goddard TD, Huang CC, Couch GS, Greenblatt DM, Meng EC, et al. UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem. 2004;25(13):1605–12. pmid:15264254
  28. 28. Perez-Riba A, Itzhaki LS. The tetratricopeptide-repeat motif is a versatile platform that enables diverse modes of molecular recognition. Curr Opin Struct Biol. 2019;54:43–9. pmid:30708253
  29. 29. Dumrese C, Slomianka L, Ziegler U, Choi SS, Kalia A, Fulurija A, et al. The secreted Helicobacter cysteine-rich protein A causes adherence of human monocytes and differentiation into a macrophage-like phenotype. FEBS Lett. 2009;583(10):1637–43. pmid:19393649
  30. 30. Panahipour L, Micucci C, Gruber R. Inflammatory response of THP1 and U937 cells: the RNAseq approach. Cells. 2024;13(24):2062. pmid:39768153
  31. 31. Ogura Y, Ooka T, Asadulghani, Terajima J, Nougayrède J-P, Kurokawa K, et al. Extensive genomic diversity and selective conservation of virulence-determinants in enterohemorrhagic Escherichia coli strains of O157 and non-O157 serotypes. Genome Biol. 2007;8(7):R138. pmid:17711596
  32. 32. Lopez-Castejon G, Brough D. Understanding the mechanism of IL-1β secretion. Cytokine Growth Factor Rev. 2011;22(4):189–95. pmid:22019906
  33. 33. Sokol CL, Luster AD. The chemokine system in innate immunity. Cold Spring Harb Perspect Biol. 2015;7(5):a016303. pmid:25635046
  34. 34. Caprilli R. Why does Crohn’s disease usually occur in terminal ileum? J Crohns Colitis. 2008;2(4):352–6. pmid:21172238
  35. 35. Park J-D, Lee SR, Dhennezel C, Taylor N, Dame A, Kadoki M, et al. Elucidating the role of Campylobacter concisus-derived indole metabolites in gut inflammation and immune modulation. Proc Natl Acad Sci U S A. 2025;122(34):e2514071122. pmid:40825123
  36. 36. Zhang L, Liu F, Xue J, Lee SA, Liu L, Riordan SM. Bacterial species associated with human inflammatory bowel disease and their pathogenic mechanisms. Front Microbiol. 2022;13:801892. pmid:35283816
  37. 37. Pabois A, Pagie S, Gérard N, Laboisse C, Pattier S, Hulin P, et al. Notch signaling mediates crosstalk between endothelial cells and macrophages via Dll4 and IL6 in cardiac microvascular inflammation. Biochem Pharmacol. 2016;104:95–107. pmid:26826491
  38. 38. Pagie S, Gérard N, Charreau B. Notch signaling triggered via the ligand DLL4 impedes M2 macrophage differentiation and promotes their apoptosis. Cell Commun Signal. 2018;16(1):4. pmid:29321062
  39. 39. Mochizuki K, He S, Zhang Y. Notch and inflammatory T-cell response: new developments and challenges. Immunotherapy. 2011;3(11):1353–66. pmid:22053886
  40. 40. Mukherjee S, Schaller MA, Neupane R, Kunkel SL, Lukacs NW. Notch ligand Dll4 enhances T cell differentiation by promoting IL-17 production and RORγT expression. J Immunol. 2009;182(12):7381.
  41. 41. Kabsch W. XDS. Acta Crystallogr D Biol Crystallogr. 2010;66(2):125–32.
  42. 42. Evans PR, Murshudov GN. How good are my data and what is the resolution? Acta Crystallogr D Biol Crystallogr. 2013;69(Pt 7):1204–14. pmid:23793146
  43. 43. Winn MD, Ballard CC, Cowtan KD, Dodson EJ, Emsley P, Evans PR, et al. Overview of the CCP4 suite and current developments. Acta Crystallogr D Biol Crystallogr. 2011;67(Pt 4):235–42. pmid:21460441
  44. 44. Terwilliger TC, Adams PD, Read RJ, McCoy AJ, Moriarty NW, Grosse-Kunstleve RW, et al. Decision-making in structure solution using Bayesian estimates of map quality: the PHENIX AutoSol wizard. Acta Crystallogr D Biol Crystallogr. 2009;65(Pt 6):582–601. pmid:19465773
  45. 45. Liebschner D, Afonine PV, Baker ML, Bunkóczi G, Chen VB, Croll TI, et al. Macromolecular structure determination using X-rays, neutrons and electrons: recent developments in Phenix. Acta Crystallogr D Struct Biol. 2019;75(Pt 10):861–77. pmid:31588918
  46. 46. Emsley P, Lohkamp B, Scott WG, Cowtan K. Features and development of Coot. Acta Crystallogr D Biol Crystallogr. 2010;66(Pt 4):486–501. pmid:20383002
  47. 47. Murshudov GN, Skubák P, Lebedev AA, Pannu NS, Steiner RA, Nicholls RA, et al. REFMAC5 for the refinement of macromolecular crystal structures. Acta Crystallogr D Biol Crystallogr. 2011;67(Pt 4):355–67. pmid:21460454
  48. 48. Langer G, Cohen SX, Lamzin VS, Perrakis A. Automated macromolecular model building for X-ray crystallography using ARP/wARP version 7. Nat Protoc. 2008;3(7):1171–9. pmid:18600222
  49. 49. Mahendran V, Liu F, Riordan SM, Grimm MC, Tanaka MM, Zhang L. Examination of the effects of Campylobacter concisus zonula occludens toxin on intestinal epithelial cells and macrophages. Gut Pathog. 2016;8:18. pmid:27195022
  50. 50. Jin X, Kruth HS. Culture of macrophage colony-stimulating factor differentiated human monocyte-derived macrophages. J Vis Exp. 2016;2016(112):54244.
  51. 51. Xiang C, Li H, Tang W. Targeting CSF-1R represents an effective strategy in modulating inflammatory diseases. Pharmacol Res. 2023;187:106566. pmid:36423789
  52. 52. Lee SA, Liu F, Yuwono C, Phan M, Chong S, Biazik J. Emerging Aeromonas enteric infections: their association with inflammatory bowel disease and novel pathogenic mechanisms. Microbiol Spectr. 2023.
  53. 53. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–20. pmid:24695404
  54. 54. Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotype. Nat Biotechnol. 2019;37(8):907–15. pmid:31375807
  55. 55. Danecek P, Bonfield JK, Liddle J, Marshall J, Ohan V, Pollard MO, et al. Twelve years of SAMtools and BCFtools. Gigascience. 2021;10(2):giab008. pmid:33590861
  56. 56. Liao Y, Smyth GK, Shi W. featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics. 2014;30(7):923–30. pmid:24227677
  57. 57. Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. pmid:25516281
  58. 58. Kuleshov MV, Jones MR, Rouillard AD, Fernandez NF, Duan Q, Wang Z, et al. Enrichr: a comprehensive gene set enrichment analysis web server 2016 update. Nucleic Acids Res. 2016;44(W1):W90-7. pmid:27141961
  59. 59. Chen EY, Tan CM, Kou Y, Duan Q, Wang Z, Meirelles GV, et al. Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool. BMC Bioinform. 2013;14:128. pmid:23586463
  60. 60. Xie Z, Bailey A, Kuleshov MV, Clarke DJB, Evangelista JE, Jenkins SL, et al. Gene set knowledge discovery with Enrichr. Curr Protoc. 2021;1(3):e90. pmid:33780170
  61. 61. Kerneur C, Cano CE, Olive D. Major pathways involved in macrophage polarization in cancer. Front Immunol. 2022;13:1026954. pmid:36325334
  62. 62. Neurath MF. Cytokines in inflammatory bowel disease. Nat Rev Immunol. 2014;14(5):329–42. pmid:24751956
  63. 63. Danese S, Gasbarrini A. Chemokines in inflammatory bowel disease. J Clin Pathol. 2005;58(10):1025–7. pmid:16189145
  64. 64. Wu T-T, Chen T-L, Chen R-M. Lipopolysaccharide triggers macrophage activation of inflammatory cytokine expression, chemotaxis, phagocytosis, and oxidative ability via a toll-like receptor 4-dependent pathway: validated by RNA interference. Toxicol Lett. 2009;191(2–3):195–202. pmid:19735705
  65. 65. Ye J, Coulouris G, Zaretskaya I, Cutcutache I, Rozen S, Madden TL. Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinform. 2012;13:134. pmid:22708584
  66. 66. Yang S-S, Yu D-Y, Du Y-T, Wang L, Gu L, Zhang Y-Y, et al. Inhibition of Delta-like Ligand 4 enhances the radiosensitivity and inhibits migration in cervical cancer via the reversion of epithelial-mesenchymal transition. Cancer Cell Int. 2020;20:344. pmid:32742191
  67. 67. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001;25(4):402–8. pmid:11846609