GSTM4 protein expression and localization in endometrial stromal cells from endometriosis patients
To evaluate the protein expression of GSTM4 on endometriosis, we first performed immunohistochemistry staining to detect the protein expression level of endometrial cells betweeen 15 eutopic endometrial samples from patients with ovarian endometriosis and 15 control endometrial samples from woman without endometrisis. We found that GSTM4 is mainly expressed in cytoplasm of stromal cells and glandular epithelial cells in endometrial tissues. The expression levels of GSTM4 was significantly increased in eutopic endometrial lesions from patients with ovarian endometriosis than those from controls(0.205 ± 0.013 vs. 0.291 ± 0.014, P < 0.001; Fig. 1A). The immunocytochemical staining results were represented by optical density (Fig. 1B). WB also detected that the protein levels of GSTM4 were increased in the eutopic endometrium compared to control endometrium(0.899 ± 0.090 vs. 1.371 ± 0.121.p < 0.01; Fig. 1C). Densitometric analysis was used to calculate representative quantitative data (Fig. 1D). Taken together, the data suggested that increased expression of GSTM4 might involved in the pathological process in patients with endometriosis.
Inhibition of GSTM4 by NBDHEX suppresses the cell viability growth, migration, invasion of endometrial stromal cells from endometriosis patients
Although endometriosis is not a malignant disease, it has its own unique characteristics, but it also has similar characteristics with tumor cells, such as growth, migration, and invasion. In order to determine the mechanism of NBDHEX inhibiting the formation of endometriosis, we studied the effect of NBDHEX on the proliferation of human primary endometrial stromal cells from endometriosis patients. NBDHEX is an effective GSTs inhibitor. We treated primary endometriosis cells with different doses of NBDHEX (0.25 µM, 0.5 µM, 0.75 µM, 1 µM, 1.5 µM, 2 µM, 2.5 µM, 3 µM, 3.5 µM and 4 µM). According to CCK8 assay, the absorbance value of each inoculated cell well was measured. The higher the absorbance value, the more viable cells. We found that cell proliferation decreased in a dose-dependent manner after 48 hours of drug treated (Fig. 2A). The cell survival rate of the 0 µM drug treatment well was set as 1, and the ratio of the absorbance value of other concentrations to the 0 µM drug well was the relative survival rate. We fitted the dose-survival curve through Prism 7.0 and calculated the IC50 value of the drug. The result showed that IC50 values were about 1.44 µM for primary endometrial cells (Fig. 2B).
We also assessed the difference effect of 0 µM, 0.75 µM and 1.5 µM NBDHEX treatment on migration and invasion of endometrial cells. The cells on the lower surface of the membrane were counted. The result showed that migration and invasion were significantly reduced in 1.5 µM NBDHEX group compared to the 0 µM NBDHEX group. With the increase of NBDHEX concentration, the migration ability of cells decreased (Fig. 2C). There was no significant difference between 0 µM NBDHEX group and 0.75 µM NBDHEX group(238.3 ± 21.25 vs. 183.3 ± 22.94, p > 0.05), and there was significant difference between 0 µM NBDHEX group and 1.5 µM NBDHEX group(238.3 ± 21.25 vs. 128.1 ± 21.98,P < 0.001). In the invasion assay, the higher the concentration of NBDHEX, the weaker the invasion ability of cells (Fig. 2D). There were significant differences between 0 µM NBDHEX group and 0.75 µM NBDHEX group groups (154.2 ± 9.181 vs. 78.67 ± 7.315,P < 0.0011), 0 µM NBDHEX group and 1.5 µM NBDHEX group(154.2 ± 9.181 vs. 44.67 ± 3.765, P < 0.001).
Based on the above experiments, we chose NBDHEX concentration of 0 µM and 1.5 µM to treat the eutopic endometrial stroma cells to explore the effects on proliferation, migration and invasion. Western blot analysis was used to detect the changes of PCNA and MMP-9 protein levels (Fig. 2E). Expression of PCNA(2.487 ± 0.374 vs. 1.193 ± 0.374, p < 0.05) and MMP-9(3.153 ± 0.590 vs. 1.383 ± 0.085, p < 0.05) were significantly reduced in 1.5 µM NBDHEX group compared to the 0 µM NBDHEX group.
Inhibition of GSTM4 by NBDHEX induces apoptosis in endometrial stromal cells from endometriosis patients
To further evaluate the inhibitor of GSTM4 by NBDHEX mediated cell death of eutopic endometrial stroma cells, the effect of NBDHEX on apoptosis was detected by Annexin V-APC/7-AAD apoptosis kit. Apoptosis and the total apoptotic rate of cells of NBFHEX exposure are shown in Fig. 3A. In 0 µM NBDHEX group, the cell survival rate was 78.16%. However, in 1.5 µM NBDHEX group apoptosis rate was increased by about 13.01% (P < 0.05). Then the apoptosis related protein levels of Survivin, Bcl-xl and Bax were measured by western blot analysis (Fig. 3B). Compared with 0 µM NBDHEX group, Survivin (3.162 ± 0.690 vs.1.175 ± 0.112, p < 0.05) and Bcl-XL (2.529 ± 0.588 vs. 1.363 ± 0.613, p < 0.01)protein levels were significantly decreased in the 1.5 µM NBDHEX group, while Bax( 2.257 ± 0.504 vs. 3.019 ± 0.598, p < 0.05) protein levels were increased.
Inhibition of GSTM4 by NBDHEX interacting with Nrf2 induces apoptosis, but it did not affect the expression of Keap1 in endometrial stromal cells from endometriosis patients
Nrf2 is an important regulator of cellular apoptosis via inducing oxidative stress in different cell types. In order to further elucidate the role of GSTM4 in the progression of endometriosis, we detected the protein expression levels of Keap1 and Nrf2 and its downstream related gene in endometriosis. We first compared the protein expression of Keap1, Nrf2 and GSTM4 in endometriosis patients and controls (Fig. 4A). WB results showed that compared with the control group, Keap1 decreased (2.310 ± 0.286 vs. 1.517 ± 0.114, p < 0.05), Nrf2 increased (1.500 ± 0.397 vs. 2.313 ± 0.236, p < 0.05) and GSTM4 increased(1.463 ± 0.203 vs. 3.283 ± 0.090, p < 0.01) in the experimental group.
We hypothesized that NBDHEX treatment did not change the expression of Keap1 and Nrf2, but changeed the expression of GSTM4, indicating that GSTM4 was downstream of Keap1/Nrf2 signaling pathway. As show in Fig. 4B, WB results demonstrated that 1.5 µM NBDHEX group could reduce the expression of GSTM4 compared to the 0 µM NBDHEX group(2.429 ± 0.233 vs.1.449 ± 0.092, p < 0.01). However, WB results also showed that NBEHEX had no effect on the expression of Keap1 (1.363 ± 0.613 vs.1.371 ± 0.531, p > 0.05), but the expression of Nrf2 was decreased (3.49 ± 0.422 vs.1.772 ± 0.659, p < 0.05) compared to the 0 µM NBDHEX group. We consider that Nrf2 and GSTM4 may inhibit each other, that is, Nrf2 expression decreases and GSTM4 expression is inhibited, which is consistent with the common theory. However, NBDHEX inhibited GSTM4 expression, further inhibited Nrf2 expression and promoted apoptosis.
Knockdown of GSTM4 depress the cell proliferation, migration and invasion both in eutopic endometrial stroma cells with endometriosis and control endometrial stroma cells
Because in addition to inhibiting the expression of GSTM4, NBDHEX may also inhibit the expression of other GSTs family proteins. In order to further elucidate the role of GSTM4 in the progression of endometriosis, we transfected an siRNA containing an GSTM4 targeting sequence (si-GSTM4) and negative control siRNA (si-NC) in eutopic endometrial stroma cells with endometriosis and control endometrial stroma cells.
When the cells were not transfected with si-GSTM4, the expression of GSTM4 in the eutopic group was about 60% higher than that in the control group (1.337 ± 0.152 vs. 3.306 ± 0.634, P < 0.001). After cells transfected with si-GSTM4, GSTM4 protein level was down-regulated by nearly 70% in eutopic group (3.306 ± 0.634 vs. 0.987 ± 0.206, p < 0.001) and down-regulated by 68% in control group ( 1.337 ± 0.152 vs. 0.426 ± 0.309, p < 0.05).(Fig. 5A)
Silencing of GSTM4 led to proliferation inhibition of both group as assessed by CCK8 assay. The absorbance of the cells in each inoculated cell well at 450 nm was detected by enzyme-labeled analyzer. The results showed that the absorbance of the cells increased with the prolonging of culture time. The proliferation of endometriosis cells was stronger than that in the control group. After knockdown of GSTM4, the proliferation ability of cells in both groups was decreased compared with that in si-NC group(Fig. 5B).
Endometriosis cells and normal endometrial cells were cultured for 36 hours with and without si-GSTM4 and then assessed by transwell migration and invasion assay. When without si-GSTM4 transfection, the number of migration(170.3 ± 8.352 vs. 271.7 ± 9.854, p < 0.001) and invasion cells(100.4 ± 3.473 vs. 139.0 ± 7.297, p < 0.001) in the endometriosis group was significantly higher than that of the control group. After transfected with si-GSTM4 the number of migration cells(170.3 ± 8.352 vs. 125.3 ± 8.162, p < 0.05; 271.7 ± 15.460 vs. 145.2 ± 9.854, p < 0.001) and the number of invasion cells (100.4 ± 3.473 vs. 43.11 ± 3.048, p < 0.001; 139.00 ± 7.297 vs. 67.89 ± 5.184 p < 0.001) were significantly reduced in control group and eutopic group(Fig. 5C and D).
Additionally, we used western blot analysis to detect the changes of PCNA and MMP-9 protein levels. When without si-GSTM4 transfection, the protein levels of PNCA and MMP-9 also compared between endometrial cells and normal endometrial cells. The results showed that the expression levels of PCNA(1.457 ± 0.282 vs. 2.21 ± 0.310, p < 0.05) and MMP-9 (1.190 ± 0.148 vs. 1.753 ± 0.217, p < 0.01) were higher in eutopic endometrial stroma cells compared to the control group. After transfected with si-GSTM4, The protein expression level of PCNA (2.21 ± 0.310 vs.1.437 ± 0.185, p < 0.05 ) in endometriosis cells was significantly decreased. While the expression of PCNA (1.457 ± 0.282 vs.1.187 ± 0.150, p > 0.05) in normal endometrial cells was also decreased, but there was no statistical difference. Expression of MMP-9 (1.190 ± 0.148 vs. 0.833 ± 0.064, p < 0.05; 1.753 ± 0.217 vs. 1.420 ± 0.066,p < 0.05) were significantly reduced in two group cells transfected with si-GSTM4. (Fig. 5E).
Knockdown of GSTM4 induces apoptosis both in eutopic endometrial stroma cells with endometriosis and control endometrial stroma cells
The effect of knockdown of GSTM4 on apoptosis was analysised by flow cytometry. Compared with the control group, apoptosis rate of endometriosis group was statistically significant reduced (33.26 ± 5.878% vs. 19.67 ± 2.913%, p < 0.05). When the cells were transfected with si-GSTM4 compared to si-NC, the apoptotic rates were increased both in control group(33.26 ± 5.878 vs. 51.740 ± 4.754%,p < 0.05) and eutopic group(19.67 ± 2.913% vs. 36.92 ± 9.087%, p < 0.05).(Fig. 6A)
Then the apoptosis related protein levels of Survivin, Bcl-xl and Bax were detected by western blot analysis. When the cells were not transfected with si-GSTM4, the expression levels of the anti-apoptotic proteins Survivin (2.426 ± 0.278 vs. 4.186 ± 0.728, p < 0.01) and Bcl-XL (2.105 ± 0.147 vs. 2.779 ± 0.441, p < 0.05) were higher in the eutopic group than those in control group, and the expression levels of pro-apoptotic protein Bax were lower (2.257 ± 0.504 vs. 1.146 ± 0.131,p < 0.05). After cells transfected with si-GSTM4, the protein levels of Survivin (2.426 ± 0.278 vs. 0.854 ± 0.241, p < 0.05; 4.186 ± 0.728 vs.1.794 ± 0.752, p < 0.01) and Bcl-xl ( 2.105 ± 0.147 vs. 1.244 ± 0.222 p < 0.05, 2.779 ± 0.441 vs. 1.917 ± 0.493 p < 0.05) were significantly decreased, and the protein level of Bax was increased (2.257 ± 0.504 vs. 3.685 ± 0.708, p < 0.05; 1.146 ± 0.131 vs.2.398 ± 0.560, p < 0.05) in control group and eutopic group(Fig. 6B). Therefore, Knockdown of GSTM4 induces apoptosis was involved in the regulation of apoptosis in endometrial stroma cells.
Knockdown of GSTM4 interacting with Nrf2 induces apoptosis, but it did not affect the expression of Keap1 in endometriosis
We detected the protein expression levels of Keap1, Nrf2 and GSTM4 both in endometriosis and control cells by western blot analysis. The results were similar to those of inhibiting GSTM4 expression by NBDHEX. When the cells were not transfected with si-GSTM4, the protein expression levels of Nrf2 (1.802 ± 0.267 vs. 2.967 ± 0.209 P < 0.01) and GSTM4(1.337 ± 0.152 vs. 3.306 ± 0.634 P < 0.001) were higher in the eutopic group than those in control group, and the expression levels of Keap1 were lower (3.623 ± 0.265 vs. 2.073 ± 0.839 p < 0.05). After cells transfected with si-GSTM4, the protein levels of Nrf2 (1.802 ± 0.267 vs. 0.935 ± 0.366, p < 0.05; 2.967 ± 0.209 vs. 2.349 ± 0.342, p < 0.05) and GSTM4(1.337 ± 0.152 vs. 0.426 ± 0.309 p < 0.05, 3.306 ± 0.634 vs. 0.987 ± 0.206 p < 0.001) were significantly decreased, and the protein level of Keap1(3.623 ± 0.265 vs. 2.817 ± 0.410 p > 0.05, 2.073 ± 0.839 vs. 1.947 ± 0.475 p > 0.05 ) have no significant change in control group and endometriosis group (Fig. 7A).
To further confirm the activation of Nrf2 transport into the nucleus, we used nuclear protein to detect the expression of Nrf2 and GSTM4 in cell nucleus of eutopic endometrial stroma cells. Compared with si-NC group, knockdown of GSTM4 decreased Nrf2 level in cell nucleus (1.341 ± 0.550 vs. 0.896 ± 0.517* p = 0.0235) and total cell (1.919 ± 0.199 vs. 1.405 ± 0.102 *p = 0.0394), whereas GSTM4 was not expressed in nucleus (Fig. 7B).