Mycobacterial Hsp65 antigen delivered by invasive *Lactococcus lactis* reduces intestinal inflammation and fibrosis in TNBS-induced chronic colitis model

Inflammatory bowel diseases (IBDs), comprising Crohn's disease (CD) and ulcerative colitis, are characterized by chronic and recurrent intestinal mucosal inflammation with rising global incidence. CD is incurable and marked by abdominal pain, diarrhea, fever, fatigue, weight loss, and malabsorption. A major complication is intestinal fibrosis—excessive collagen-rich extracellular matrix accumulation, especially in the submucosa and muscle layers—occurring in roughly 30–50% of patients and leading to stenosis, fistulas, abscesses, and driving many surgical resections. Despite therapies that modulate inflammation, effective treatments to prevent or reverse CD-associated fibrosis are lacking. Heat shock proteins (HSPs) are promising antifibrogenic candidates due to their roles in regulating effector T cells. HSP60 can downregulate Th1 responses and upregulate Th2 and Treg responses, increasing IL-10 and decreasing IFN-γ and TNF-α. The authors previously engineered Lactococcus lactis NCDO2118 to express Mycobacterium leprae Hsp65 via a xylose-inducible system and demonstrated prevention or amelioration of murine colitis, including via induction of regulatory T cells and IL-10/TLR2 pathways. A newer recombinant invasive L. lactis NCDO2118 strain co-expressing Staphylococcus aureus fibronectin-binding protein A (FnBPA+) and M. leprae Hsp65 [L. lactis NCDO2118 FnBPA+(pXYCYT:Hsp65)] efficiently delivers Hsp65 to affected sites and reduces acute TNBS colitis severity. Therefore, this study aimed to evaluate the anti-inflammatory and antifibrotic capacities of this invasive Hsp65-producing L. lactis strain in a TNBS-induced chronic colitis mouse model.

Prior work shows HSP60 modulates immune responses by downregulating Th1 and upregulating Th2/Treg responses, elevating IL-10 while reducing IFN-γ and TNF-α. The research group constructed L. lactis NCDO2118 expressing M. leprae Hsp65 using a xylose-inducible system and showed it can prevent or ameliorate colitis, including DSS-induced inflammation via IL-10- and TLR2-dependent pathways with increased CD4+Foxp3+ and CD4+LAP+ Tregs. Wild-type L. lactis NCDO2118 also exerts anti-inflammatory and immunomodulatory effects during remission in DSS colitis. A newer invasive strain expressing FnBPA+ and Hsp65 [L. lactis NCDO2118 FnBPA+(pXYCYT:Hsp65)] reduced acute TNBS-induced colitis, associated with decreased IL-12/IL-17 and increased IL-10 and sIgA. Together, these suggest Hsp65-delivering L. lactis as a candidate for managing inflammation and potentially fibrosis in CD.

Design: Chronic colitis was induced in female BALB/c mice (6–7 weeks) by TNBS using a standard protocol. On day 1, mice were presensitized with 150 µL of 1% TNBS in acetone/olive oil (4:1) on shaved dorsal skin. Starting day 8, mice received weekly intrarectal (i.r.) 100 µL TNBS in 50% ethanol at increasing concentrations on days 8, 15, 22, 29, 36, and 43 (0.75%, 1.5%, 2.5% w/v). Body weight was recorded weekly. All mice were euthanized on day 54. Bacterial preparation and administration: The invasive L. lactis NCDO2118 FnBPA+ (pXYCYT:Hsp65) strain was cultured at 30 °C in M17 medium with 0.5% glucose and antibiotics (chloramphenicol 10 µg/mL, erythromycin 5 µg/mL). Hsp65 expression was induced with 2% xylose and 0.5% galactose for ~8 h to OD600 ~2. Treatment consisted of intragastric gavage (i.g.) with 100 µL bacterial suspension containing 1×10^9 CFU in 0.9% saline, administered once daily for 4 consecutive days per week, for 6 consecutive weeks, starting the day after each i.r. TNBS dose. Experimental groups: Four groups (n=6 mice/group per experiment) were used: (1) C− (negative control): saline i.g. and saline i.r.; (2) C+ (positive control): saline i.g. and TNBS i.r.; (3) NFX: invasive L. lactis FnBPA+ (pXYCYT:Hsp65) not expressing Hsp65 i.g. and TNBS i.r.; (4) NFXi: invasive L. lactis FnBPA+ (pXYCYT:Hsp65) expressing Hsp65 i.g. and TNBS i.r. Data are reported from three independent experiments for some outcomes. Histopathology: Colon samples were stained with H&E for inflammation and Gomori trichrome for fibrosis. Inflammation scoring (0–4 per layer for acute and chronic components across mucosa, submucosa, muscle, serosa) summed to a maximum of 32. Fibrosis scoring ranged 0–5 based on collagen deposition across layers, multiplied by extent (1–4 reflecting 0–100% involvement). Myeloperoxidase (MPO): MPO activity was assayed in colon tissue homogenates using TMB substrate with spectrophotometric readout at 450 nm after standard buffer extraction and freeze–thaw cycles. Cytokines: Colon homogenate cytokines (IFN-γ, IL-12, IL-6, IL-13, IL-17, TGF-β, IL-10) were quantified by ELISA with absorbance at 492 nm. Secretory IgA: Intestinal lavage fluids (PBS washes) were assayed for sIgA by capture ELISA using goat anti-mouse antibodies and HRP detection (492 nm). Statistics: Data presented as mean ± SD. One-way ANOVA with Tukey’s post-test (GraphPad Prism 6.0). p<0.05 considered significant. Ethical approval: CEUA UFMG Registration Number 341/2017.

- Clinical course: TNBS-treated groups (C+, NFX, NFXi) exhibited significant weight loss versus C− at weeks 6–7; earlier weeks showed less difference, and by week 8 weights converged toward C−. - Histology: C− had no inflammation or fibrosis. TNBS groups (C+, NFX) showed marked transmural inflammation (erosion, hyperemia, hemorrhage, neutrophil/mononuclear infiltrates, necrosis) and extensive fibrosis, predominantly submucosal. NFXi mice had significantly reduced inflammation and fibrosis scores compared with C+ and NFX (p<0.05), with only mild lesions. - MPO: No significant differences among groups. - Cytokines: No group differences in IFN-γ or IL-12. NFXi showed significantly lower IL-6 versus C+ (p<0.05), significantly lower IL-13 versus NFX (p<0.05), and significantly lower IL-17 versus C+ and NFX (p<0.05). NFXi had significantly decreased TGF-β and increased IL-10 compared with NFX (p<0.05). Discussion text also notes TGF-β reductions versus both C+ and NFX. - sIgA: No differences among groups. Overall, expression of Hsp65 by the invasive L. lactis strain was necessary for the observed anti-inflammatory and antifibrotic effects; the invasive strain without Hsp65 (NFX) did not improve outcomes relative to TNBS controls.

The invasive L. lactis NCDO2118 FnBPA+ strain expressing mycobacterial Hsp65 reduced both intestinal inflammation and fibrosis in a chronic TNBS colitis model, addressing the lack of therapies targeting CD-associated fibrosis. The benefits correlated with modulation of mucosal cytokines: reductions in IL-13 and TGF-β, key profibrotic mediators, and IL-17, a proinflammatory cytokine with profibrotic functions, alongside increased IL-10, a central regulatory cytokine that restrains mucosal inflammation and can inhibit collagen deposition. Lack of differences in IFN-γ and IL-12 likely reflects measurement at a late timepoint relative to their acute-phase peaks. MPO and sIgA were not altered in this chronic setting. These findings support intracellular delivery of therapeutic proteins to the intestinal mucosa as a strategy to re-establish pro/anti-inflammatory balance and mitigate fibrosis in CD.

Oral administration of invasive L. lactis NCDO2118 FnBPA+ expressing M. leprae Hsp65 significantly attenuated histological inflammation and fibrosis in chronic TNBS-induced colitis. The therapeutic effect was associated with reduced IL-13, TGF-β, and IL-17 and increased IL-10 in colonic tissue. The data validate a protein-delivery approach using engineered probiotics to modulate mucosal immunity and fibrogenesis in experimental CD, suggesting its consideration as an alternative strategy for managing fibrotic complications.

- Cytokines IFN-γ and IL-12 showed no differences, likely due to sampling at day 54 after their peak in acute inflammation, limiting conclusions about early Th1 responses. - MPO activity and sIgA levels did not differ among groups, potentially reflecting model timing or assay sensitivity in chronic disease. - Body weight differences diminished by week 8 as TNBS colitis is harder to sustain in mice older than 8 weeks, which may temper clinical readouts. - The study was conducted in a murine TNBS model; translation to human CD requires further investigation.