Tigecycline attenuates tumorigenesis in AOM/DSS treated mice.
Upon the observation of the in vitro anti-tumoral effect of tigecycline, we studied it in vivo . CRC was induced by the well-described AOM/DSS model. As evidenced by colonoscopy, the administration of either tigecycline or 5-FU exerted beneficial effects ameliorating tumoral progression (Figure 4A). The tumor score, based on the size of the tumors (Becker et al. , 2006), was significantly reduced in all treated groups in comparison with untreated control mice (Figure 4A), as well as the tumor count (Figure 4B). Thus, both treatments had a positive impact toward reducing tumor number, with a trend to lower the tumor size (Figures 4B-D and S2A). These observations were supported by the histological analysis, which confirmed the presence of a higher number of adenocarcinomas in non-treated CAC mice than in treated mice (Figure 4E). Histological staining with anti-MKI67 showed that tigecycline and 5-FU induced a reduction of the proliferation index (Figure 4F), thus supporting the anti-proliferative impact of both drugs in this model.
We sought to evaluate the molecular mechanisms underlying these effects. We could confirm that the inhibitory actions of tigecycline treatment on CAC progression are mediated by modulation of the Wnt/β-catenin signaling pathway, as observed in the in vitro evaluation. Administration of tigecycline to CAC mice inhibited STAT3 activation in the colon by reducing the ratio of phospho-STAT3/STAT3 (Figure 4G), and subsequently, ameliorated Ccnd1 and Mmp9 expressions, which were upregulated in control mice (Figure 4G). Moreover, a trend to reduce the colonic expression of CTNNB1 was observed with both doses of tigecycline and 5-FU (Figure 4G). The treatments also restored the phospho-AKT/AKT ratio, altered by AKT phosphorylation with cancer progression (Figure 4G), which could reduce the proliferative effects and angiogenesis process through VEGFA and ANGPT2 up-regulation (Revathidevi et al. , 2019). Thus, in this study, a significant reduction of Angpt2 gene expression was observed in mice treated with tigecycline versus CAC group (Figure 4G).
The flow cytometry analysis of isolated colonocytes indicated that AOM/DSS stimulus significantly increased the levels of LGR5+ and LGR5+CD44+ cancer stem cells (CSC), key drivers of the initiation and progression of the tumoral process. Tigecycline treatment significantly reduced both populations (Figure 5A). This reduction was confirmed by SNAI1 protein analysis, an EMT process marker involved in the formation of these CSCs (Zhou et al. , 2014). Thus, the treatment with tigecycline reduced the protein levels of SNAI1 compared with the CAC group (Figure 5B).
Moreover, the anti-tumor effect of tigecycline was also related to a pro-apoptotic impact. By TUNEL assay, we observed an increase of the number of apoptotic cells in mice treated with both tigecycline and 5-FU (Figure 5C). In line with the in vitro described mechanisms, apoptosis induction with tigecycline treatment involved the increase of CASP7 full length and cleaved levels (Figure 5D). Moreover, tigecycline and 5-FU treated mice also showed normalized BCL2 levels compared to CAC-mice (Figure 5D), which contributes to the pro-apoptotic effect.
The inflammatory status of the disease was assessed throughout the induction process by monitoring the DAI (Figure 6A). 5-FU treatment did not show any statistical differences in comparison with the CAC group at the end of the assay, but tigecycline did reduce DAI value (Figure 6A). Accordingly, tigecycline lowered the expression of some cytokines involved in the tumor-associated inflammatory response in the colon, like Tnfa, Il6, Il17a and Il23a (Figure 6B).
When the impact of the tigecycline on the myeloid populations from colon (CD45+CD11b+) (Figure S3) and lymphoid cells MLN (CD3+CD4+/CD3+CD8+) (Figure S4) was analyzed, we observed a reduced infiltration of myeloid cells (CD45⁺CD11b⁺) in the colon (Figure S3), and similar numbers of CD3+CD4+ cells among the T cells from MLN in the different groups, but a significant reduction of CD8⁺ cytotoxic T lymphocytes (CTLs) in CAC group (Figure 6C and 6D). When CD3+CD4+IFNγ+ (Th1 phenotype) population was considered, the highest dose of tigecycline produced a significant increase (Figure 6D). Interestingly, this increment has been associated with a protective response against CAC through favoring the recruitment of CTLs (Bhat et al. , 2017). In fact, when CTL population was analyzed, tigecycline and 5-FU increased its level in comparison with non-treated CAC-mice (Figure 6D), particularly those CTLs with IFNG production (Tc1) (Figure 6D).