Gut dysbacteriosis attenuates resistance to Mycobacterium bovis infection by decreasing cyclooxygenase 2 to inhibit endoplasmic reticulum stress

Abstract

The role of gut microbiota has been described as an important influencer of the immune system. Gut-lung axis is critical in the prevention of mycobacterium infection, but the specific mechanism, by which dysbiosis affects tuberculosis, has not been reported. In this study, we attempted to provide more information on how the gut-lung axis contributes to Mycobacterium bovis (M. bovis) infection. Mice are pre-treated with broad-spectrum antibiotics cocktail (Abx) to induce gut dysbiosis. Interestingly, dysbiosis of microbes showed a significant increase in the bacterial burden in the lungs and inhibited the level of COX-2. After faecal transplantation, cyclooxygenase 2 (COX-2) expression was restored and the inflammatory lesion in the lungs was reduced. Further research found that the deficiency of COX-2 inhibited endoplasmic reticulum stress (ER stress). This mechanism was completed by COX-2 interaction with BIP. Moreover, we found a positive feedback mechanism by which blocking ER stress could reduce COX-2 levels by the NF-κB pathway. Taken together, we reveal for the first time gut dysbacteriosis exacerbates M. bovis disease by limiting the COX-2/ER stress pathway. The finding strengthens the foundation of gut microbiota-targeted therapy for tuberculosis treatment.

Keywords: Gut dysbacteriosis; Mycobacterium bovis; apoptosis; cyclooxygenase 2; endoplasmic reticulum stress; faecal transplant.

Figure 1.

The effects of gut dysbacteriosis on the severity of M. bovis pathogenesis in mice. (A) The representative images of caecum showed the gross pathological changes in all experimental groups (n = 6). (B) Mice were pre-treated with a broad spectrum of antibiotics followed by M. bovis infection. Later, DNA was isolated from faecal samples of mice for quantitative PCR analysis and normalized to a universal bacterial primer. Bar graphs depict the total bacterial load; bacterial genera such as Lactobacillus, Bifidobacterium, Campylobacter, Bacteroides and Enterococcus. (C) The organ index of the spleen and lungs in all experimental groups. (D) The representative images of the lungs showed the gross pathological changes of all experimental groups (n = 6). The dots with different shapes show the number of mice and the horizontal lines indicate the mean value. Data are shown as mean ± SD, **P < 0.01. *P < 0.05.

Figure 2.

The effects of gut dysbacteriosis on histopathological changes in the lung tissues from M. bovis-infected mice (A) The representative images of the left lung lobe showed histopathological changes in uninfected (CT), M. bovis-infected mice (M. bovis), M. bovis with Abx-treated mice (Abx + M. bovis), M. bovis with Abx and faecal transplants treated group (Abx + FT + M. bovis). (B) The percentage of the lung’s area occupied by inflammatory lesions was quantified by Image J software. (C) Higher magnification of H&E staining sections of lungs and spleen showed M. bovis I-induced lesions. Scale bar: 20 μm. (D) The total number of M. bovis bacilli in the lung tissues of infected mice as determined using the CFU assay. (E) Ziehl–Neelsen staining results showing the number of M. bovis in the lungs of mice with gut dysbacteriosis. Data are shown as mean ± SD, ***P < 0.001, **P < 0.01. *P < 0.05

Figure 3.

Gut dysbacteriosis inhibited the expression of COX-2, ER stress and apoptosis. (A) Western blot detection and the relative intensity ratio of COX-2 in the gut of different groups of mice. GAPDH expression served as an internal control and was used for normalization (n = 3/group). (B) Western blot detection and the relative lung ratio of COX-2 in the lungs of a different group of mice. α-tublin expression served as a lung control and was used for normalization (n = 3/group). (C) ELISA detected the alteration of PGE2 in serum. (D) Western blot detection and the relative lung ratio of BIP, p-eIF2α and CHOP in the lungs of a different group of mice. α-tublin and eIF2α expression served as a lung control and was used for normalization (n = 3/group). (E) macrophage isolated from the lungs of a different group of mice were treated with Annexin-V and detected by flow cytometry. (F) Western blot detection and the relative lung ratio of cleaved-caspase3 in the lungs of a different group of mice. α-tublin expression served as a lung control and was used for normalization (n = 3/group). Data are shown as mean ± SD, **P < 0.01. *P < 0.05.

Figure 4.

The deficiency of COX-2 promoted M. bovis infection by inhibiting ER stress and apoptosis. (A) The representative images of the lungs showed the gross pathological changes in a different group of mice (n = 6/group). (B) The representative images of the left lung lobe showed histopathological changes in a different group of mice (n = 6). (C) Higher magnification of H&E staining sections of the lungs and spleen showed M. bovis-induced lesions. Scale bar: 20 μm. (D) The percentage of the lung’s area occupied by inflammatory lesions was quantified by Image J software. (E) The total number of M. bovis bacilli in the lung tissues of infected mice as determined using the CFU assay. (F) Ziehl–Neelsen staining results showing the number of M. bovis in the lungs of mice with gut dysbacteriosis. (G) Western blot detection and the relative lung ratio of COX-2, BIP, p-eIF2α, and CHOP in the lungs of M. bovis-infected mice, Cele+ M. bovis mice and Cele+ M. bovis + dm-PGE2. α-tublin and eIF2α expression served as a lung control and was used for normalization (n = 3/group). Data are shown as mean ± SD, ***P < 0.001, **P < 0.01. *P < 0.05.

Figure 5.

COX-2 interacts with BIP to activate ER stress. (A) Whole-cell lysates were immunoprecipitated with anti-BIP or anti-COX-2 antibody. The total lysates (Input) and IP were immunoblotted with the indicated antibodies. (B) Confocal microscopy analysis of COX-2 (green) colocalization with BIP (red) in RAW264.7 cells infected with M. bovis (MOI = 10) for 24 h. (C) Western blot detection and the relative lung ratio of Cleaved -caspase-3 in the lungs of M. bovis-infected mice, Cele+ M. bovis mice and Cele+ M. bovis + dm-PGE2. α-tublin expression served as a lung control and was used for normalization (n = 3/group). (D) RAW264.7 were treated with 5 mM 4-PBA for 3 h before infection, then were infected with M. bovis (MOI = 10) for 24 h. BIP and Cleaved-caspse-3 were detected by western blot. α-tublin expression served as a cell control and was used for normalization. Data are shown as mean ± SD, **P < 0.01. *P < 0.05.

Figure 6.

4-PBA inhibited COX-2 through the NF-κB pathway. (A) RAW264.7 were treated with 5 mM 4-PBA for 3 h before infection, then were infected with M. bovis (MOI = 10) for 24 h. COX-2 was detected by western blot. α-tublin expression served as a cell control and was used for normalization. (B,C) RAW264.7 was treated with 5 mM 4-PBA for 3 h before infection, then was infected with M. bovis (MOI = 10) and treated with or without 4-PBA for 24 h. (B) The expression of p-p65 in the cytoplasm was determined by western blot. GAPDH expression served as a cell control and was used for normalization (C) Expression of p-p65 in the nucleus was determined by western blot. Histone3 expression served as a cell control and was used for normalization. (D) RAW264.7 was treated with 80 μM PDTC for 3 h before infection, then were infected with M. bovis (MOI = 10) and treated with or without PDTC for 24 h. Expression of COX-2 in the whole-cell protein was determined by western blot. α-tublin expression served as a cell control and was used for normalization. Data are shown as mean ± SD, **P < 0.01. *P < 0.05.