The Anti-Cancer Effects of Anti-Parasite Drug Ivermectin in Ovarian Cancer

Ivermectin is an old, common, and classic anti-parasite drug, which has been found to have a broad-spectrum anti-cancer effect on multiple human cancers. This chapter will focus on the anti-cancer effects of ivermectin on ovarian cancer. First, ivermectin was found to suppress cell proliferation and growth, block cell cycle progression, and promote cell apoptosis in ovarian cancer. Second, drug pathway network, qRT-PCR, and immunoaffinity blot analyses found that ivermectin acts through molecular networks to target the key molecules in energy metabolism pathways, including PFKP in glycolysis, IDH2 and IDH3B in Kreb ’ s cycle, ND2, ND5, CYTB, and UQCRH in oxidative phosphorylation, and MCT1 and MCT4 in lactate shuttle, to inhibit ovarian cancer growth. Third, the integrative analysis of TCGA transcriptomics and mitochondrial proteomics in ovarian cancer revealed that 16 survival-related lncRNAs were mediated by ivermectin, SILAC quantitative proteomics analysis revealed that ivermectin extensively inhibited the expressions of RNA-binding protein EIF4A3 and 116 EIF4A3-interacted genes including those key molecules in energy metabolism pathways, and also those lncRNAs regulated EIF4A3-mRNA axes. Thus, ivermectin mediated lncRNA-EIF4A3-mRNA axes in ovarian cancer to exert its anticancer capability. Further, lasso regression identified the prognostic model of ivermectin-related three-lncRNA signature (ZNRF3-AS1, SOS1-IT1, and LINC00565), which is significantly associated with overall survival and clinicopathologic characteristics in ovarian cancer patients. These ivermectin-related molecular pattern alterations benefit for prognostic assessment and personalized drug therapy toward 3P medicine practice in ovarian


Introduction
Ivermectin is chemically derived from avermectin that was discovered and isolated from soil in Jan by Omura in 1973 [1]. It was approved by Federal Drug Administration (FDA) to use for anti-parasite drug in 1987, which has significantly improved global public health as an antiparasite medicine [2]. In 2015, its discovers Drs. Omura and Campbell earned the Nobel Prize in physiology or medicine [2]. Recent years, many studies have demonstrated that ivermectin has extensive roles in anti-bacteria, anti-virus, and anticancer, except for its anti-parasite effects [3][4][5].
Its anticancer effect has been shown by many in vitro and in vivo experiments in multiple cancers, including ovarian cancer, breast cancer, triple-negative breast cancer, cervical cancer, lung cancer, gastric cancer, colon cancer, glioblastoma, melanoma, and leukemia [4,6], with a wide safe and clinically reachable drug concentration of anticancer according to its pharmacokinetic range in treatment of a parasite-infected patient [7]. It offers a promising opportunity to develop a new anticancer drug via drug repositioning of this existing compound with confirmed clinical safety [8].
Ovarian cancer, a very common cancer with high mortality and poor survival in women [9,10], are involved in multiple signaling pathway network changes [11,12]. Many intracellular molecules and signaling pathways would be the targets of ivermectin [13]. Ivermectin have shown a potential addition role for ovarian cancer treatment. For example, ivermectin can improve the chemosensitivity of overran cancer via targeting Akt/mTOR signaling pathway [14], and can inhibit PAK1-dependent growth of ovarian cancer cells via blocking the oncogenic kinase PAK1 [15]. Ivermectin also acts as a PAK1 inhibitor to induce autophagy in breast cancer [16]. Ivermectin can enhance p53 expression and cytochrome C release, and reduce the expression levels of CDK2, CDK4, CDK6, Bcl-2, cyclin E, and cyclin D1 in glioblastoma, those promoted the cancer cell apoptosis [17]. Ivermectin can inhibit cancer cell proliferation via decreasing YAP1 protein expression in the Hippo pathway [18]. Ivermectin represses WNT-TCF pathway in WNT-TCF-dependent disease [19]. Ivermectin can promote TFE3 (Ser321) dephosphorylation to block the binding between TFE3 and 14-3-3, and induce TFE3 accumulation in the nucleus of human melanoma cells [20]. Moreover, ivermectin also affects other signaling pathway network in human cancers, including oxidative stress, mitochondrial dysfunction, angiogenesis, epithelial-mesenchymal transition, drug resistance, and stemness in tumor [6]. Thereby, ivermectin demonstrates the potential therapeutic efficiency in multiple malignant tumors.
This book chapter discussed the anti-cancer effects of ivermectin on ovarian cancer in the following aspects: (i) ivermectin inhibited cell proliferation and growth, blocked cell cycle progression, and promoted cell apoptosis in ovarian cancer [4,21]; (ii) ivermectin inhibited ovarian cancer growth through molecular networks to target the key molecules in energy metabolism pathways, including glycolysis, Kreb's cycle, oxidative phosphorylation, and lactate shuttle pathways [21]; (iii) Integrated omics revealed that ivermectin mediated lncRNA-EIF4A3-mRNA axes in ovarian cancer to exert its anticancer capability [4,13]; and (iv) lasso regression identified the prognostic model of ivermectin-related three-lncRNA signature (ZNRF3-AS1, SOS1-IT1, and LINC00565) that is significantly related to overall survival and clinicopathologic characteristics of ovarian cancers [4].

Ovarian cancer cell biological behaviors affected by ivermectin
The normal ovarian cells IOSE80 and ovarian cancer cells TOV-21 and SKOV3 were treated with ivermectin to measure ivermectin-mediated ovarian cancer cell biological behavior changes. (i) IOSE80, TOV-21G, and SKOV3 were treated with ivermectin (0-60 μM) for 24 h, followed by the use of CCK8 to measure the IC50 of ivermectin in each cell. (ii) TOV-21G and SKOV3 were treated with ivermectin (0 μM, 10 μM, 20 μM, and 30 μM) for 24 h, followed by the use of EdU assay to measure DNA synthesis in each cell. (iii) TOV-21G and SKOV3 were treated with ivermectin (0 μM, 10 μM, 20 μM, and 30 μM) for 48 h, followed by clonogenic assay to measure the in vitro effects of ivermectin in each cell. (iv)TOV-21G and SKOV3 were treated with ivermectin (0 μM, 10 μM, 20 μM, and 30 μM) for 24 h, followed by flow cytometry to measure cell cycle and cell apoptosis changes in each cell. (v) When A2780 and TOV-21G seeded in 6-well plates were grown to approximately 90% confluency, followed by the use of 10-μl pipette tip to make an artificial wound, and then treated with ivermectin (0 μM, 10 μM, 20 μM, and 30 μM) for 24 h, and measure the wound healing. The relative percentage of wound healing = (the width of wound at 0 h À the width of wound at 24 h)/the width of wound at 0 h. The detailed procedure was described previously [4,21].

Ivermectin-mediated pathway network predicted by ingenuity pathway analysis
The classical pathway network analysis software, Ingenuity Pathway Analysis (IPA) (http://www.ingenuity.com) [5] was used to predict ivermectin-related potential target molecules in three energy metabolism pathways. For this analysis, ivermectin and target genes in three energy metabolism pathways are all input into the IPA tool. The detailed procedure was described previously [21]. The predicted ivermectin-mediated targets in energy metabolism pathways were the basis for further experiment verification.

Ivermectin-mediated proteome changes in ovarian cancer identified by SILAC-based quantitative proteomics
SILAC (stable isotope labeling with amino acids in cell culture)-based quantitative proteomics was used to identified differentially expressed proteins in ovarian cancer TOV-21G treated with and without 20 μM ivermectin [13]. The identified differentially expressed proteins were used for molecular network and signaling pathway analyses to obtain ivermectin-related signaling pathway networks [13]. The detailed procedure was described previously [13].

LncRNA-based prognostic signature optimized with lasso regression for ovarian cancers
Lasso regression means least absolute shrinkage and selection operator regression, which was used to optimize and construct lncRNA-based prognostic signature, and the glmnet R package was used to measure the association between survival risk and lncRNA signature in ovarian cancers. Moreover, univariate and multivariate Cox regression, and Kaplan-Meier method were used to identify overall survivalrelated clinical characteristics described above in ovarian cancers to confirm the established lncRNA-based prognostic model. The detailed procedure was described previously [4].

Statistical significance
Benjamini-Hochberg (FDR) for multiple testing was used to correct the p values of IPA, GO, and KEGG analyses. Student's t test was used for qRT-PCR and western blot data (p < 0.05) with data expression of mean AE SD (n = 3).

Effects of ivermectin on biological behaviors of ovarian cancers
First, CCK8 experiments were used to measure cell proliferation changes between ovarian cancer cells (SKOV3; TOV-21G) and control cells (IOSE80), treated with and without ivermectin (Figure 1). Each type of cells was significantly inhibited by ivermectin with a dose-dependent relationship. The IC50 (half maximal inhibitory concentration) was 29.46 μM for IOSE80 cells, 20.85 μM for SKOV3, and 22.54 μM for TOV-21G ( Figure 1A). The IC50 of ovarian cancers were significantly lower than the normal controls. Further, 20 μM ivermectin -slightly lower than IC50can effectively inhibit ovarian cancer proliferation ( Figure 1B and C) [21]. For in vivo human trial, the highest FDA-approved ivermectin dose was 200 μg/kg for human use in anti-parasite; however, a study on 68 human subjects found that the dose up to 2,000 μg/kg still worked well without CNS toxicity. The mean area under the curve ratios for the 30 and 60 mg doses were 1.24 and 1.40, indicating a minimal accumulation of ivermectin [5,22]. These data demonstrate that ivermectin was a well-tolerated safe drug. Second, EdU cell proliferation experiments also confirmed that ivermectin significantly suppressed cell proliferation of ovarian cancers with a time-dependent relationship ( Figure 1D-F) [21]. Third, Clonogenic survival experiments confirmed that ivermectin effectively inhibited the formation of cell clones with a time-dependent relationship ( Figure 1G-H) [21]. Moreover, 10 μM ivermectin cannot effectively inhibit cell proliferation of ovarian cancers, 30 μM ivermectin caused cell death of ovarian cancers, and 20 μM ivermectin was a suitable dose to significantly suppress growth and proliferation of ovarian cancer cells.

Effects of ivermectin on cell cycle and apoptosis in ovarian cancers
Flow cytometry was used to measure cell cycle and apoptosis of ovarian cancer cells treated with and without ivermectin (Figure 2) [21]. First, the cell proportion was significantly increased in G0/G phase, decreased in S phase, and no change in G2/M phase in the high concentration (20-and 30-μM) compared to the low concentration (0-and 10-μM) of ivermectin groups (Figure 2A-C). Second, compared to control group, the proportion of apoptosis cells was significantly increased  [21], with copyright permission from nature springer publisher, copyright 2020.
in different concentration of ivermectin groups, with a dose-dependent relationship ( Figure 2D and E).

Effect of ivermectin on cell migration in ovarian cancers
Wound healing experiment was used to test the effect of ivermectin on cell migration of ovarian cancer cells. The results showed that cell migration was significantly inhibited in cells A2780 and TOV-21G after treatment of 20 μM and 30 μM ivermectin (Figure 3) [4].

Pharmaceutic molecular network predicted the association of ivermectin with ROS and energy metabolism
Ingenuity Pathway Analysis (IPA) was used for pharmaceutic molecular network analysis of ivermectin. The results showed that ivermectin was significantly

RT-qPCR and Western blot confirmed the effects of ivermectin on the key molecules in energy metabolism pathways at the mRNA and protein levels
RT-qPCR analysis confirmed the mRNA expression alterations of key molecules in energy metabolism pathways in ovarian cancer cells treated with ivermectin (0 μM, 10 μM, 20 μM, and 30 μM) (Figure 5), and further western blot analysis confirmed the protein expression alterations of those corresponding key molecules ( Figure 6) [21]. These key molecules included PFKP, and PKM in glycolysis pathway, PDHB, CS, IDH2, IDH3A, IDH3B, and OGDHL in Kreb's cycle pathway, ND2, ND5, CYTB, and UQCRH in oxidative phosphorylation pathway, MCT1, and MCT4 in lactate shuttle. These results clearly showed that ivermectin regulated energy metabolism pathways in ovarian cancer cells.  [4], with copyright permission from nature springer publisher, copyright 2020.

Ivermectin regulated lncRNA-EIF4A3-mRNA axis in ovarian cancer cells
Our quantitative mitochondrial proteomics data identified 1198 differentially mitochondrial proteins (mtDEPs) in human ovarian cancer tissues relative to control ovary tissues [11,23]. Six RNA-binding proteins among those 1198 mtDEPs were identified, including EIF4A3, SFRS1, IGF2BP2, UPF1, C22ORF28, and EWSR1. Of them, only EIF4A3 was predicted to bind to the mRNA of key molecules in energy metabolism pathways. Further, Starbase predicted 3636 EIF4A3-biding mRNAs in various cancer; and of them, 306 EIF4A3-binding mRNAs was associated Reproduced from Li et al. [21], with copyright permission from nature springer publisher, copyright 2020.  [21], with copyright permission from nature springer publisher, copyright 2020.
These findings clearly demonstrated that ivermectin regulated lncRNA-EIF4A3-mRNA axis in ovarian cancer cells, and these mRNAs included the key molecules in energy metabolism pathways in ovarian cancer cells.

The prognostic model of ivermectin-related three-lncRNA signature for ovarian cancers identified and optimized by lasso regression
Based on those nine ivermectin-mediated lncRNAs in ovarian cancers, survival analysis and lasso regression were used to identify and optimize the prognostic model of ivermectin-related three-lncRNA signature (ZNRF3-AS1,  [21], with copyright permission from nature springer publisher, copyright 2020. SOS1-IT1, and LINC00565) (Figure 8) [4]. This prognostic model was significantly related to overall survival and clinicopathologic characteristics in ovarian cancer patients [4], which might benefit for prognostic assessment and personalized drug therapy toward 3P medicine practice in ovarian cancer.    Table 2.
The proteins of 116 EIF4A3-biding mRNAs and EIF4A3 were inhibited by ivermectin, identified with SILAC quantitative proteomics in ovarian cancer cells treated with (H) and without (L) ivermectin. Reproduced from Li et al. [4], with copyright permission from nature springer publisher, copyright 2020.

Conclusions
Ivermectin, as an old, common, and classic anti-parasite drug, has demonstrated its effective in vitro anti-cancer efficiency for ovarian cancer. Ivermectin significantly inhibited cell proliferation, growth and migration, blocked cell cycle progression, RT-qPCR analysis revealed the effects of ivermectin on lncRNAs in ovarian cancers relative to control cells. Reproduced from Li et al. [4], with copyright permission from nature springer publisher, copyright 2020.  A and B). Lasso regression complexity is controlled by lambda using the glmnet R package. (C). Overall survival analysis of three-lncRNA signature between high-risk and low-risk groups. Reproduced from Li et al. [4], with copyright permission from nature springer publisher, copyright 2020. and promoted cell apoptosis of human ovarian cancer cells. Drug pathway network analysis of ivermectin revealed that it was significantly related to the key molecules of four energy metabolism pathways, and RT-qPCR and immunoaffinity blot analyses found that ivermectin significantly regulated these key molecules for those energy metabolism pathways, including PFKP in glycolysis, IDH2 and IDH3B in Kreb's cycle, ND2, ND5, CYTB, and UQCRH in oxidative phosphorylation, and MCT1 and MCT4 in lactate shuttle. The integrative analysis of TCGA transcriptomics and mitochondrial proteomics in ovarian cancer revealed that 16 survival-related lncRNAs were mediated by ivermectin, which were further confirmed with RT-qPRC in human ovarian cancer cells. SILAC quantitative proteomics analysis revealed that the expressions of RNA-binding protein EIF4A3 and 116 EIF4A3-interacted genes were extensively inhibited by ivermectin. Those 116 EIF4A3-interacted proteins included those key molecules in four energy metabolism pathways, and those lncRNAs regulated EIF4A3-mRNA axes. Thus, ivermectin mediated lncRNA-EIF4A3-mRNA axes in ovarian cancer to exert its anticancer activities. Moreover, lasso regression identified the prognostic model of ivermectin-related three-lncRNA signature (ZNRF3-AS1, SOS1-IT1, and LINC00565), which was significantly associated with overall survival and clinicopathologic characteristics of ovarian cancer patients. These ivermectinrelated molecular pattern alterations benefit for prognostic assessment and personalized drug therapy in the context of 3P medicine practice in ovarian cancer.
Moreover, one must realize that these achieved data about the anti-cancer activities of ivermectin in ovarian cancers are derived from the in vitro cell models. It is necessary to expand it into the in vivo animal experiments and pre-clinical and clinical experiments for its real application in ovarian cancers.