Leelamine suppresses cMyc expression in prostate cancer cells in vitro and inhibits prostate carcinogenesis in vivo

Aim: Leelamine (LLM) inhibits growth of human prostate cancer cells but the underlying mechanism is not fully understood. The present study was undertaken to determine the effect of LLM on cMyc, which is overexpressed in a subset of human prostate cancers. Methods: The effect of LLM on cMyc expression and activity was determined by western blotting/confocal microscopy and luciferase reporter assay, respectively. A transgenic mouse model of prostate cancer (Hi-Myc) was used to determine chemopreventive efficacy of LLM. Results: Exposure of androgen sensitive (LNCaP) and castration-resistant (22Rv1) human prostate cancer cells to LLM resulted in downregulation of protein and mRNA levels of cMyc. Overexpression of cMyc partially attenuated LLM-mediated inhibition of colony formation, cell viability, and cell migration in 22Rv1 and/or PC-3 cells. LLM treatment decreased protein levels of cMyc targets (e.g., lactate dehydrogenase), however, overexpression of cMyc did not attenuate these effects. A trend for a decrease in expression level of cMyc protein was discernible in 22Rv1 xenografts from LLM-treated mice compared with control mice. The LLM treatment (10 mg/kg body weight, 5 times/week) was well-tolerated by Hi-Myc transgenic mice. The incidence of high-grade prostatic intraepithelial neoplasia, adenocarcinoma in situ, and microinvasion was lower in LLM-treated Hi-Myc mice but the difference was not statistically significant. Conclusion: The present study reveals that LLM inhibits cMyc expression in human prostate cancer cells in vitro but concentrations higher than 10 mg/kg may be required to achieve chemoprevention of prostate cancer.


Introduction
Prostate cancer is a leading cause of cancer-related deaths in the United States [1] . Several oncogenic drivers of prostate cancer have been identified, including androgen receptor (AR) and cMyc transcription factors [2] . Studies using transgenic mice and human specimens clearly indicate an oncogenic role for cMyc in prostate cancer [3][4][5][6][7][8][9][10][11] . For example, transgenic expression of cMyc in prostate epithelial cells resulted in transformation to prostatic intraepithelial neoplasia (PIN) at 2-4 weeks of age that progressed to adenocarcinoma by 6 months [3] . About 67% of human prostate adenocarcinoma (n=9) exhibited high cMyc level when compared to benign hypertrophied prostate tissue (n=19) [4] . An androgen responsive cell line (LNCaP) derived from lymph node metastasis of a human prostate cancer patient also showed gene amplification, rearrangement, and overexpression of cMyc [5] . Nuclear level of cMyc protein was shown to be an early event in the pathogenesis of human prostate cancer [7] . The cMyc overexpression was reported in about 70% of early stage prostate cancer and up to 37% of metastatic prostate cancer patients [8,10] . The 8q gain leading to cMyc overexpression was associated with poor survival in prostate cancer patients [11] . Moreover, the cMyc overexpression is suggested to be an important mechanism in the resistance of prostate cancers to antiandrogen therapy [12] . The cMyc overexpression causes increased cell proliferation and metabolic deregulation, a hallmark of different cancers including prostate cancer, including increased glycolysis and de novo synthesis of fatty acids [13][14][15] . It is not surprising that considerable efforts have been devoted to therapeutic targeting of cMyc in prostate cancer [16] . Chemoprevention using natural products or synthetic chemicals is attractive for blunting the death and suffering from prostate cancer, but a clinically effective chemopreventive intervention for this malignancy is still lacking [17][18][19][20][21][22] . Leelamine (LLM) is one such phytochemical whose effectiveness has been tested against prostate and other cancers [23][24][25] . The antitumor effect of LLM was initially reported in melanoma but our laboratory was the first to document its effectiveness against prostate cancer cells [24,25] . In melanoma model, the antitumor effect of LLM was attributed to suppression of multiple oncogenic signalling pathways, including phosphatidylinositol 3-kinase, mitogen-activated protein kinase, and signal transducer and activator of transcription 3 [24] . In prostate cancer cells, LLM was shown to inhibit cellular growth by suppressing expression and activity of AR [25] . Intraperitoneal administration of about 9 mg LLM/kg body weight (5 times/week) to 22Rv1 xenograft bearing mice resulted in a significant decrease in Ki-67 expression, mitotic fraction, expression of full-length AR and ARV7 splice variant, and prostate-specific antigen [25] .
In the present study, we investigated the effect of LLM treatment on cMyc expression and activity using human prostate cancer cell lines and Hi-Myc transgenic mouse model. The foundation for this study was based on the following observations: (a) LLM suppresses AR activity that is a ligand-independent regulator of cMyc expression in prostate cancer [25,26] ; and (b) cMyc promotes fatty acid synthesis in prostate cancer that is inhibited by LLM treatment in vitro and in vivo [15,27] .

Ethics statement
The prostate tumors from control and LLM-treated mice from our previously published study were used for western blotting [25] . The care of Hi-Myc mice was consistent with the University of Pittsburgh Animal Care and Use Committee guidelines.

Reagents
The LLM was purchased from Cayman Chemical Company (Ann Arbor, MI). The reagent for cell culture were purchased from Life Technologies-Thermo Fisher Scientific (Waltham, MA). Antibodies against cMyc, hexokinase II (HKII), pyruvate kinase M2 (PKM2), and lactate dehydrogenase A (LDH-A) were purchased from Cell Signaling Technology (Danvers, MA). The antibodies against glucose transporter 1 (Glut1) and Cyclin D1 were purchased from Abcam (Cambridge, MA) and Santa Cruz Biotechnology (Dallas, TX), respectively. An antibody against glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was purchased from GeneTex (Irvine, CA). The anti-β-Actin antibody was purchased from Sigma-Aldrich (St. Louis, MO). Alexa Fluor 488-conjugated goat anti-rabbit antibody was from Life Technologies. A kit for colorimetric measurement of lactate was purchased from BioVision (Milpitas, CA).

Cell lines
The LNCaP, 22Rv1, and PC-3 human prostate cancer cell lines were purchased from the American Type Culture Collection (Manassas, VA) and cultured by following the supplier's recommendations. The above listed cell lines were last authenticated by us in March of 2017. The Myc-CaP cell line derived from the prostate of a Hi-Myc mouse was a kind gift from Dr. Charles L. Sawyers (Memorial Sloan Kettering Cancer Center, New York, NY, USA). The Myc-CaP cells were cultured in Dulbecco's modified essential medium supplemented with 4.5 g/L glucose, L-glutamine, sodium pyruvate, 10% fetal bovine serum, and antibiotic mixture. The PC-3 and 22Rv1 cells were stably transfected with empty pcDNA3 vector or the same vector encoding cMyc (Addgene, #16011) as described by us previously [28] .

Western blotting
Details of cell lysate preparation and western blotting have been described by us previously [29,30] .

Immunocytochemistry
LNCaP (2×10 4 ), 22Rv1 (3×10 4 ), and Myc-Cap (3×10 4 ) cells were plated on coverslips in 24-well plates. After overnight incubation to allow attachment, the cells were treated with ethanol or LLM (2.5 or 5 μM) for 24 h. The cells were then fixed and permeabilized with 2% paraformaldehyde and 0.5% Triton X-100, respectively. After blocking with phosphate buffered saline (PBS) containing 0.5% bovine serum albumin and 0.15% glycine, the cells were treated overnight at 4°C with the cMyc primary antibody followed by Alexa Fluor 488-conjugated goat anti-rabbit antibody (1:2000 dilution) for 1 h and then counterstained with DRAQ5 (nuclear stain) for 5 min at room temperature in the dark.

Luciferase reporter assay
The LNCaP or 22Rv1 cells were co-transfected with 2 μg pBV-Luc wt MBS1-4 plasmid and 0.2 μg of pCMV-RL using FuGENE6. Twenty-four hours after co-transfection, the cells were treated with ethanol or the indicated doses of LLM for 12 h. Luciferase activity was determined using Dual-Luciferase ® Reporter Assay kit from Promega, and by following the manufacturer's instructions. The values were corrected for protein concentration and renilla luciferase.

Clonogenic assay
Empty vector transfected (hereafter abbreviated as EV) control cells and cMyc overexpressing 22Rv1 (1000 cells) were seeded in 6-well plates in triplicate and allowed to attach by overnight incubation. The cells were then treated with ethanol or desired doses of LLM. The medium containing ethanol or LLM was replaced every third day. After 10 days, the cells were fixed with 100% methanol for 5 min at room temperature and stained with 0.5% crystal violet solution in 20% methanol for 30 min at room temperature. The colonies from control and LLM treatment groups were counted under an inverted microscope.

Cell viability and cell migration assay
Trypan blue dye exclusion assay was performed to determine the effect of LLM on cell viability essentially as described by us previously [32] . Cell migration assay was performed as described by us previously [33] .

Determination of lactate
Lactate level of plasma and tumors archived from our previous study [25] was measured using a colorimetric assay kit. Plasma specimens and tumor lysates were filtered using a 10 kDa filter. Assays were performed according to manufacturer's instructions.

The in vivo study using Hi-Myc mice
Female transgenic FVB-Tg(ARR2/Pbsn-MYC) mice were crossed with male wild-type FVB/N mice, both purchased from the NCI Mouse Repository, to establish a colony. Following transgene verification, 5-week old male transgenic Hi-Myc mice were divided into two groups.
Mice of group 1 were treated with 100 μL vehicle while group 2 mice received 10 mg LLM/kg body weight by intraperitoneal injection 5 times/week (Monday through Friday). The vehicle consisted of 10% ethanol, 10% dimethyl sulfoxide, 30% kolliphor EL (Sigma Aldrich, St. Louis, MO) and 50% PBS. The study was terminated after 5 weeks of treatment and the animals were euthanized by CO 2 overdose (supplied via compressed gas cylinder) and blood and prostate tissue were collected.

Statistical tests
Student's t-test was used for two group comparisons. One-way analysis of variance (ANOVA) followed by Bonferroni's test was used for multiple comparisons. ANOVA followed by Dunnett's test was used for dose-response comparisons. Statistical significance for the effect of LLM on incidence of hyperplasia (HYP), low-grade PIN (LG-PIN), high-grade PIN (HG-PIN) adenocarcinoma (AC) in situ, and microinvasion (Mic-inv) was determined by Fisher's test. GraphPad Prism (version 8.0.0) was used to perform statistical analyses.

Effect of LLM on expression of cMyc
Pharmacokinetics of LLM has been investigated using male ICR mice after a single oral administration with 10 mg/kg body weight dose [33] . The peak plasma concentration of LLM was shown to be about 2.8 μM [33] . Thus, LLM concentrations of 2.5 and 5 μM were used to determine its effect on expression of cMyc. As can be seen in Figure 1A, the expression of cMyc protein was decreased in a concentration-dependent manner. The LLM-mediated downregulation of cMyc protein was confirmed by confocal microscopy ( Figure 1B). The LLM treatment also inhibited mRNA levels of cMyc ( Figure 1C) and suppressed its transcriptional activity ( Figure 1D). Collectively, these results indicated suppression of cMyc expression and activity following LLM treatment in LNCaP and 22Rv1 human prostate cancer cell lines. Figure 2A shows overexpression of cMyc protein in 22Rv1 and PC-3 cells stably transfected with the cMyc plasmid (hereafter abbreviated as cMyc) in comparison with EV control cells. The clonogenicity of EV cells was inhibited by LLM treatment in a dose-dependent manner (Figure 2A). The inhibition of clonogenicity resulting from LLM exposure was partly attenuated by cMyc overexpression (Figure 2A). Similarly, the cMyc overexpression conferred partial protection against LLM-mediated inhibition of cell viability ( Figure  2B) and cell migration ( Figure 2C) in 22Rv1 and/or PC-3 cells. These results indicated functional significance of cMyc suppression in anticancer effects of LLM.

Effect of LLM treatment on protein levels of cMyc regulated proteins
Next, we determined the effect of LLM treatment on levels of cMyc target proteins. Expression of several glycolysis-related proteins was decreased by LLM treatment in 22Rv1 and LNCaP cells ( Figure 3A). There was a decrease in circulating level of lactic acid, the end product of glycolysis, in LLM-treated mice than control mice but the difference was not significant. The 22Rv1 xenograft level of lactate was also not suppressed by LLM treatment ( Figure 3B). The expression of cMyc and some its target proteins was also not suppressed to a significant extent by LLM administration to 22Rv1 xenograft bearing mice ( Figure 4A and B).

Effect of LLM treatment on cancer incidence in Hi-Myc mice
Initially, we used a cell line (Myc-CaP) derived from the prostate tumor of a Hi-Myc transgenic mouse to determine the effect of LLM treatment on cMyc protein level. Similar to human prostate cancer cells ( Figure 1A and B), western blot analysis revealed a decrease in protein level of cMyc following LLM treatment in the Myc-CaP cell line ( Figure  5A). Confocal microscopy confirmed LLM-mediated downregulation of cMyc protein level ( Figure 5B). The LLM administration was well-tolerated by the Hi-Myc mice ( Figure 5C). We have shown previously that LLM is well-tolerated by mice (27). Figure 5D shows the H&E stained prostate tissue sections from control and LLM-treated Hi-Myc mice. The incidence of HG-PIN, AC in situ, and Mic-inv was lower in LLM-treated mice compared to vehicle-treated control mice but the difference was not statistically significant ( Figure 5E).

Discussion
The present study shows inhibitory effect of LLM on cMyc expression and activity in prostate cancer cells. The cMyc inhibition is functionally significant because the anticancer effects of LLM, including clonogenicity, cell viability, and cell migration, are partially attenuated by cMyc overexpression. It is also clear that cMyc suppression alone is not responsible for growth inhibitory effects of LLM on prostate cancer cells. Several other oncogenes have been shown as targets of LLM in other cancer cell types like melanoma, including phosphatidylinositol 3-kinase, mitogen-activated protein kinase, and STAT3 [24] . All these oncogenic signalling pathways are known to contribute to the pathogenesis of human prostate cancer [34][35][36] . For example, analysis of the Oncomine database reveals upregulation of c-Jun N-terminal kinases in prostate cancer [35] . Likewise, STAT3 plays an important role in progression of prostate cancer to incurable metastatic castration-resistant disease by integrating multiple signalling pathways involved in the reactivation of androgen receptor pathway, stem like cells and the epithelial to mesenchymal transition [36] . It is conceivable that LLM inhibits these signalling pathways to inhibit growth of prostate cancer cells. However, additional work is necessary to systematically explore this possibility.
Chemoprevention represents a worthwhile strategy for reducing the mortality and morbidity associated with prostate cancer [20][21][22] . The feasibility of prostate cancer chemoprevention has been explored clinically with 5α-reductase inhibitors (PCPT and REDUCE trials) as well as a novel combination of selenium and vitamin E (e.g., SELECT trial) [17][18][19] . Whereas administration of 5α-reductase inhibitors resulted in ~23-25% decrease in overall relative risk, these agents were not approved by the FDA for chemoprevention of prostate cancer due to high-grade tumors in the treatment arm [17,18] . In an 18-years of follow-up study, high-grade prostate cancer was still more common in the finasteride treatment group [37] . Similarly, the SELECT trial did not show any preventative benefit of vitamin E and selenium supplementation [18] . Therefore, a safe and efficacious intervention for chemoprevention of prostate cancer is still a clinically unmet need. In this study, we explored the possibility of chemoprevention by LLM using a transgenic mouse model in which prostate cancer is driven by prostate epithelial-specific overexpression of cMyc. There was trend for a decrease in the incidence of HG-PIN, AC in situ, and Mic-inv but the difference was not significant at the 10 mg/kg body weight dose of LLM. These results are surprising because in melanoma models, LLM doses of 5 and 7.5 mg/kg body weight were effective for in vivo inhibition of tumor growth [24] .
In conclusion, the present study reveals in vitro inhibitory effect of LLM on cMyc expression and activity in prostate cancer cells. However, LLM doses higher than 10 mg/kg body weight might be required to achieve chemoprevention of prostate cancer by LLM in vivo. Thus, similar to certain other phytochemicals like curcumin, resveratrol, and neem component, LLM may be useful for prevention of prostate cancer [38][39][40] . In one such study, supercritical extract of neem leaves (SENL) was shown to inhibit integrin β1, calreticulin, and focal adhesion kinase activation in LNCaP-luc2 and PC3 prostate cancer cells [40] . The growth of LNCaP-luc2 xenograft was also reduced significantly by SENL administration that was associated with formation of hyalinized fibrous tumor tissue, reduction in the prostate-specific antigen, and increase in AKR1C2 levels [40] . of LLM treatment on activity of cMyc as determined by luciferase reporter assay. The results shown are mean ± SD (n=3). *Significantly different (P<0.05) compared with ethanol-treated control by one-way ANOVA followed by Dunnett's test. Each experiment was performed twice, and the results were generally consistent. Cell migration assay using EV and cMyc cells following treatment with vehicle or LLM. The results shown are mean ± SD (n=3). Significantly different (P<0.05) compared with *ethanol-treated control, and #between EV and cMyc groups by one-way ANOVA followed by Bonferroni's test. Each experiment was performed twice, and the results were generally consistent.