Fig 2: Effect of GPER activation on ERK1/2 activity in the CIM model.CIM model was induced by i.p. injection 5-FU (30 mg/kg/day) for 5 days. G-1 (0.03 mg/kg/day) and/or G15 (0.3 mg/kg/day) were administrated intraperitoneally with 5-FU together. G-1 was administrated alone or with G15 together in the CIM model to test the effect of GPER activation. On the 5th day, the ileum was collected for immunofluorescence staining and western blot. Data were expressed as mean ± SEM (*P < 0.05, **P < 0.01). a The GSVA analysis showed that MAPK signaling pathway was inactivated under the chemotherapeutic drug methotrexate treatment, with the enhancement of DNA damage and cell cycle arrest. b Representative western blots photographs for MAPK signal pathway from the control group and from the 5-FU treated group with or without G-1, G15 application. c Statistical analysis of P-ERK1/2 within four subgroups to show the effect of GPER activation on the activity of ERK1/2 in the CIM model (n = 5). d Statistical analysis of P-JNK within four subgroups to show the effect of GPER activation on the activity of JNK in the CIM model (n = 5). e Statistical analysis of P-p38 within four subgroups to show the effect of GPER activation on the activity of p38 in the CIM model (n = 5). f Immunofluorescence staining for P-ERK1/2 in ileum from the control group and from the 5-FU treated group with or without G-1, G15 application (Scale bars: 20 µm).

Fig 3: Effect of GPER activation on crypt cell proliferation in 5-FU induced CIM.CIM model was induced by i.p. injection 5-FU (30 mg/kg/day) for 5 days. G-1 (0.03 mg/kg/day) and/or G15 (0.3 mg/kg/day) were administrated intraperitoneally with 5-FU together. G-1 was administrated alone or with G15 together in the CIM model to test the effect of GPER activation. On the 5th day, the ileum was collected for immunohistochemical staining and western blot. Data were expressed as mean ± SEM (*P < 0.05, ***P < 0.001). a Representative images for immunohistochemical staining of BrdU following G-1 administration with or without G15 in the CIM model (scale bars: 20 µm). b Representative images for immunohistochemical staining of Ki67 following G-1 administration with or without G15 in the CIM model (scale bars: 20 µm). c Statistical graph for the number of BrdU+ cells per crypt within four subgroups to show the effect of GPER activation on proliferation of crypt cells in the CIM model. At least 20 crypts were counted randomly for each sample, and the mean value was calculated for single sample (n = 6). d Statistical graph for the number of Ki67+ cells per crypt within four subgroups to show the effect of GPER activation on proliferation of crypt cells in the CIM model. At least 20 crypts were counted randomly for each sample, and the mean value was calculated for single sample (n = 6). e Statistical graph of villous height following G-1 administration with or without G15 in the CIM model. At least 20 villi were counted randomly for each sample, and the mean value was calculated for single sample (n = 10). f Statistical graph of crypt depth following G-1 administration with or without G15 in the CIM model. At least 20 crypts were counted randomly for each sample, and the mean value was calculated for single sample (n = 10). g Representative western blots photograph to show the effect of G-1 administration with or without G15 on cyclin D1 and cyclin B1 expression in the CIM model. h Statistical analysis of cyclin D1 and cyclin B1expression within four subgroups to show the effect of GPER activation on cyclin D1 and cyclin B1 expression in the CIM model (n = 5).

Fig 4: The role of regulating ERK1/2 activity in GPER activation protecting 5-FU induced CIM.CIM model was induced by i.p. injection 5-FU (30 mg/kg/day) from day 0 to day 5. G-1 (0.03 mg/kg/day) and/or ERK1/2 inhibitor PD0325901 (5 mg/kg/day) were administrated intraperitoneally together with 5-FU. On the 5th day, the ileum was collected for H& E staining, immunofluorescence staining, immunohistochemical staining and western blot. Data were expressed as mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). a Representative immunohistochemical images of ?H2AX staining in the ileum of the CIM model following G-1 administration or in combination with PD0325901 (Scale bars: 20 µm). b Statistical graph of ?H2AX+ cells in ileal crypt within three subgroups to show the role of regulating ERK1/2 activity in GPER inhibiting DNA damage induced by CIM. 15 crypts were randomly calculated in each section, and the average value was obtained (n = 5). c Effect of ERK1/2 inhibitor PD0325901 on G-1 in reducing intestinal mucosal damage score and crypt damage score of CIM model (n = 9). d Effect of ERK1/2 inhibitor PD0325901 on G-1 inhibition of crypt depth reduction and villus shortening in the CIM model (n = 9). At least 20 villi were counted randomly for each sample, and the mean value was calculated for single sample. e Effect of ERK1/2 inhibitor PD0325901 on G-1 inhibition of weight loss in the CIM model (n = 9). f Effect of ERK1/2 inhibitor PD0325901on G-1 protected ileal mucosal permeability in the CIM model (n = 6). The mucosal permeability was evaluated by the concentration of Fluorescein isothiocyanate (FITC) -dextran in blood. g Representative immunohistochemical images of BrdU staining within four subgroups (scale bars: 20 µm) and statistical graph of BrdU+ cell to show the effect of ERK1/2 inhibitor PD0325901 on G-1 protecting proliferation of crypt cells in the CIM model. At least 20 crypts were counted randomly, and the mean value was calculated for single sample (n = 6). h Representative immunohistochemical images of Ki67 staining within four subgroups (scale bars: 20 µm) and statistical graph of Ki67+ cell to show the effect of ERK1/2 inhibitor PD0325901 on G-1 protecting proliferation of crypt cells in the CIM model. At least 20 crypts were counted randomly, and the mean value was calculated for single sample (n = 6). i Representative images of TUNEL staining within four subgroups (scale bars: 50 µm) and statistical graph of TUNEL + cell to show the effect of ERK1/2 inhibitor PD0325901 on G-1 inhibiting apoptosis of crypt cells in the CIM model. At least 20 crypts were counted randomly, and the mean value was calculated for single sample (n = 6). j Representative immunohistochemical images of Muc-2 staining within four subgroups (scale bars: 50 µm) and statistical graph of Muc-2+ cells to show the effect of ERK1/2 inhibitor PD0325901 on G-1 protecting the mucus barrier in the CIM model. At least 20 villi were counted randomly, and the mean value was calculated for single sample (n = 6).

Fig 5: Effect of GPER on body weight and ileum histopathological damage in 5-FU induced CIM and on anti-tumor effect of 5-FU.CIM model was induced by i.p. injection 5-FU (30 mg/kg/day) for 5 days. The control group was intraperitoneally injected with the same amount of saline. Selective GPER agonist G-1 (0.03 mg/kg/day), selective GPER blocker G15 (0.3 mg/kg/day) were administrated intraperitoneally together with 5-FU. G-1 was administrated alone or with G15 together in the CIM model to test the effect of GPER activation. G15 was used alone to test the effect of endogenous GPER blocking. Data were expressed as mean ± SEM (*P < 0.05, **P < 0.01, ***P < 0.001). a Representative images of H&E staining in the ileum following 5-FU administration to show the effect of G-1 on the histopathological damage in CIM model (scale bars: 50 µm). b The statistical graph of mucosal damage score and crypt damage score within four subgroups to show the effect of GPER activation on histological injury in the CIM model (n = 10). At least 20 villi and crypts of each slide were observed randomly to get the score of intestinal mucosa injury and crypt injury. c The effect of GPER activation with G-1 on the body weight loss from day 0 to day 5 in 5-FU induced CIM (n = 10). The data were expressed as a percentage change compared to pre-treatment. d Representative images of H&E staining in the ileum following 5-FU administration to show the effect of G15 treatment alone on the histopathological damage in the CIM model (scale bars: 50 µm). e The statistical graph of mucosal damage score and crypt damage score within three subgroups to show the effect of G15 blocking endogenous GPER on the histological damage of CIM (n = 6). f The effect of endogenous GPER blocking with G15 on the weight loss from day 0 to day 5 in the CIM model (n = 6). The data were expressed as a percentage change compared to pre-treatment. g Effects of 5-FU on tumor size in tumor-bearing mice with or without G-1. 4 × 106 LL/2 (Lewis lung carcinoma) cells were injected into the dorsal flank of the male C57BL/6 mice to establish the tumor-xenograft model (n = 5).