Ginkgo biloba derivative ginkgetin inhibits breast cancer growth by regulating the miRNA-122-5p/GALNT10 axis

To the Editor: Current therapeutic strategies against breast cancer are limited by poor prognosis and unwanted side effects, especially in triple-negative breast cancer (TNBC). Ginkgetin (Gk), a natural, non-toxic bioflavonoid extracted from Ginkgo biloba leaves, is a promising therapeutic candidate against breast cancer,[1] but the underlying mechanisms remain obscure. In this study, we evaluated the anti-tumor activity of Gk on multiple breast cancer cell lines, including both human (MDA-MB-231, MDA-MB-453, MDA-MB-436, MCF-7, and SUM-159) and mouse (4T1) cell lines in vitro. We next used high-throughput sequencing and bioinformatics analysis to identify the alterations in the expression of microRNA (miRNA) and messenger RNA (mRNA) and search for the potential regulating signaling pathways after Gk treatment. The findings were further validated using the dual luciferase reporter gene assay, Western blotting (WB), etc. Finally, the anti-tumor effect of Gk and the involvement of the miR-122-5p/polypeptide N-acetylgalactosaminyltransferase 10 (GALNT10) axis were assessed in an MDA-MB-231 xenograft mouse model. The anti-proliferative activity of Gk on breast cancer was evaluated with the cell counting kit-8 (CCK-8) assay. As shown in Supplementary Figure 1 and 2A,B, https://links.lww.com/CM9/C112, Gk exhibited dose-dependent cytotoxicity in all the tested cell lines, with the half maximal inhibitory concentration (IC50) ranging from 5.67 μmol/L to 11.5 μmol/L at 48 h and 3.7 μmol/L to 9.3 μmol/L at 72 h. The formula of Gk is C32H22O10 (molecular weight, 566.51) [Supplementary Figure 2C, https://links.lww.com/CM9/C112]. Compared to the control group, Gk treatment reduced the cell numbers and induced apparent morphology changes in breast cancer cells, including cell shrinkage, wrinkled membrane, blebbing, and chromatin condensation [Supplementary Figure 2D, https://links.lww.com/CM9/C112]. We next applied the Annexin V/PI apoptosis detection assay after Gk treatment to investigate the type of cell death. We found that Gk significantly increased the proportion of apoptotic cells (Annexin V+PI− and Annexin V+PI+) dose-dependently [Supplementary Figure 1B and 3A–C, https://links.lww.com/CM9/C112]. These results were supported by TUNEL staining [Supplementary Figure 3D, https://links.lww.com/CM9/C112]. Next, WB was used to examine the expression of apoptosis-related proteins. As shown in Supplementary Figure 1C, https://links.lww.com/CM9/C112, Gk treatment drastically increased the expression of cleaved caspase 3 (CASP3), BAX, and cytochrome C (Cyto C) but reduced the expression of BCL-2 in MDA-MB-231 cells. Results from the wound healing assay indicated that Gk inhibited the migration of MDA-MB-231 cells [Supplementary Figure 3E, https://links.lww.com/CM9/C112]. After exposure to Gk for 24 h, the number of colonies dramatically decreased compared to the untreated group, becoming almost invisible when treated with 20 μmol/L of Gk [Supplementary Figure 1D, https://links.lww.com/CM9/C112]. Therefore, Gk can effectively inhibit cell proliferation and migration and induce cell apoptosis in breast cancer. Since the most-effective apoptosis was observed in MDA-MB-231 cells upon Gk exposure, which can be divided into early- and end-stage apoptosis [Supplementary Figure 1B, https://links.lww.com/CM9/C112], we chose MDA-MB-231 cells treated with 10 μmol/L Gk for the transcriptome sequencing. RNA sequencing results indicated that over 1000 genes were differentially expressed in MDA-MB-231 cells after treatment with Gk compared to the control group [Supplementary Figure 4A, https://links.lww.com/CM9/C112]. Similarly, there are 351 miRNAs whose expression was significantly altered by Gk treatment, including 182 up-regulated and 169 down-regulated miRNAs [Supplementary Figure 1E, https://links.lww.com/CM9/C112]. KEGG and GO enrichment analyses were performed to clarify the functions of those differentially expressed genes (DEGs). The top seven altered terms of the KEGG pathways and GO plots are displayed in Supplementary Figure 4B–E, https://links.lww.com/CM9/C112. The results indicated that plenty of signaling pathways, biological processes, cellular components, and molecular functions were significantly changed by Gk treatment. For instance, as for the biological process, cytoplasmic translation and sensory perception of the light stimulus were overactivated, while other functions such as MAPK cascade, cell junction organization, and regulation of cell adhesion were inhibited by Gk. Since its first discovery in 1993, miRNA has been proven to be critical in almost all biological processes, including cell homeostasis, proliferation, differentiation, survival, and metabolism.[2,3] miRNAs are non-coding nucleic acids composed of 18–25 nucleotides, which can modulate gene expression by inhibiting or degrading their target mRNA.[3] We can see from the volcano plot that the miRNAs showing a log2 fold change of >5 or <−5 include hsa-miR-215-5p, hsa-miR-122-5p, hsa-miR-190b-5p, and hsa-miR-27a-5p, among which has-miR-122-5p changed the most significantly [Supplementary Figure 5A, https://links.lww.com/CM9/C112]. Real-time qPCR (qRT-PCR) results proved that Gk treatment significantly increased the expression of miR-122-5p in a dose-dependent manner [Supplementary Figure 5B, https://links.lww.com/CM9/C112]. To explore the function of miR-122-5p in Gk-mediated anti-tumor effects, MDA-MB-231 cells were transfected with a miRNA inhibitor against miR-122-5p (122i), followed by Gk treatment. Inhibitor NC (NCi) was used as the control. As a result, inhibition of miR-122-5p relieved the apoptosis-inducing [Supplementary Figure 5C, https://links.lww.com/CM9/C112] and anti-proliferative [Supplementary Figure 1F, https://links.lww.com/CM9/C112] activities of Gk. Meanwhile, in the presence of 122i, Gk displayed a reduced ability to prevent the migration of MDA-MB-231 cells [Supplementary Figure 5D, https://links.lww.com/CM9/C112]. These results proved that the anti-tumor effects of Gk on breast cancer cells may be mediated by, at least partly, the up-regulation of miR-122-5p. To ascertain the potential downstream target of miR-122-5p, we intersected the target mRNAs of miR-122-5p with the downregulated genes after Gk treatment, which was confirmed with three databases: “mirecords,” “mirtarbase”, and “tarbase”. Seven genes were screened out, including GALNT10, BCL2L2, UBAP2, AACS, SLC7A1, ENTPD4, and AKT3, whose expressions were highly negatively correlated with that of miR-122-5p [Supplementary Figure 5E,F, https://links.lww.com/CM9/C112]. qRT-PCR and WB assays proved a significantly reduced expression of GALNT10 after Gk treatment [Supplementary Figure 1 and 5G, https://links.lww.com/CM9/C112]. Next, we discovered that transfecting MDA-MB-231 cells with the 122i can rescue the decreased expression of GALNT10 induced by Gk [Supplementary Figure 5H, https://links.lww.com/CM9/C112]. Dual-luciferase reporter gene assay was performed to determine whether the 3′-untranslated regions (UTR) of GALNT10 can be targeted by miR-122-5p. As expected, co-transfection with mimics of miR-122-5p (mimic 122) and wild-type GALNT10 (GALNT10-WT) displayed the lowest luciferase activity, suggesting the strong interaction between miR-122-5p and GALNT10 [Supplementary Figure 1H, https://links.lww.com/CM9/C112]. Therefore, GALNT10 is a downstream target for miR-122-5p in Gk-induced cell death. This is consistent with a previous study reporting that miR-122 regulated GALNT10 expression in chronic hepatitis B virus-associated hepatocarcinogenesis.[4] GALNT10 is an oncoprotein that promotes malignant behaviors of cancer cells like proliferation, apoptosis resistance, and metastatic features.[4,5] We next investigated the function of GALNT10 in breast cancer using its specific small interfering RNA (siRNA). The siRNA significantly knocked down the expression of GALNT10 in MDA-MB-231 cells by over 70% [Supplementary Figure 5I, https://links.lww.com/CM9/C112]. Knocking down GALNT10 expression suppressed the proliferation of MDA-MB-231 cells, increased the proportion of apoptotic cells, and attenuated the invasion and migration of MDA-MB-231 cells [Supplementary Figure 1I and 5J,K, https://links.lww.com/CM9/C112]. Therefore, GALNT10 confers the malignant phenotype of breast cancer cells, whereas Gk treatment can reduce GALNT10 expression by enhancing the expression levels of its upstream regulator miR-122-5p. To find out the downstream effector molecule of GALNT10, pathways associated with GALNT10 gene expression were analyzed in the TCGA breast cancer dataset. The results of GSEA showed that GALNT10 expression was associated with the Wnt signaling pathway regulation [Supplementary Figure 6A, https://links.lww.com/CM9/C112], of which the Smad3 gene in the Wnt signaling pathway was up-regulated after Gk treatment [Supplementary Figure 6B, https://links.lww.com/CM9/C112]. In line with the downregulation of GALNT10, we discovered a dose-dependent reduction in the expression of Smad3 in MDA-MB-231 cells after Gk treatment, including total Smad3 and phosphorylated Smad3 (p-Smad3, Supplementary Figure 6C,D, https://links.lww.com/CM9/C112). Knocking down GALNT10 with siRNA resulted in decreased expression of Smad3, suggesting that Smad3 is a downstream signaling molecule of GALNT10 [Supplementary Figure 6E, https://links.lww.com/CM9/C112]. Since miR-122-5p can down-regulate GALNT10, we speculated that 122i transfection can impact Smad3 expression by regulating GALNT10. As expected, the expression of Smad3 in MDA-MB-231 cells is elevated in the presence of 122i after Gk treatment compared to pre-transfection with NCi [Supplementary Figure 6F, https://links.lww.com/CM9/C112]. These results indicated that Gk could inhibit breast cancer growth by regulating miR-122-5p/GALNT10/Smad3 signaling. To confirm the anti-tumor activity of Gk in vivo, MDA-MB-231 tumor-bearing SCID mice were administered with Gk (30 mg/kg) or vehicle intraperitoneally. As shown in Supplementary Figure 1J and 7A,B, https://links.lww.com/CM9/C112, Gk treatment significantly suppressed tumor growth compared to the vehicle group. No statistically significant changes between groups in body weight were observed after treatment [Supplementary Figure 7C, https://links.lww.com/CM9/C112], indicating the acceptable safety of Gk. To validate the involvement of the miR-122-5p/GALNT10 axis in vivo, we isolated the total miRNAs from the resected tumors after Gk treatment and assessed the expression of miR-122-5p. The expression of miR-122-5p was significantly elevated after Gk treatment [Supplementary Figure 1K, https://links.lww.com/CM9/C112]. At the same time, a downward trend was observed in the expression of GALNT10 mRNA in the tumor tissue from Gk-treated mice [Supplementary Figure 7D, https://links.lww.com/CM9/C112]. The immunofluorescence assay also revealed a dramatic decrease in the expression of GALNT10 protein in the tumor tissues in the Gk group compared to the control group [Supplementary Figure 1L, https://links.lww.com/CM9/C112]. Conclusively, Gk is a promising anti-cancer agent that inhibits tumor growth both in vitro and in vivo. High-throughput sequencing and bioinformatic analysis revealed gene alterations after Gk treatment and uncovered the critical role of miRNA-122-5p/GALNT10/Smad3 signaling, which were further supported by the results from various experiments. These results added great value for future work on Gk as a candidate agent in breast cancer therapy, including but not limited to searching for combination strategies to improve the efficacy. Funding This study was supported by the Natural Science Foundation of Sichuan Province (No. 2022NSFSC0774) and the National Natural Science Foundation of China (No. 82102896). Conflicts of interest None.