MicroRNA miR-99b-5p Targets mTOR/AR axis, Induces Autophagy, and Inhibits Prostate Cancer Cell Proliferation

MicroRNAs (miRNAs) are the small non-coding regulatory RNA molecules involved in gene regulation via base-pairing with complementary sequences in mRNAs. The dysregulation of specic miRNAs, such as miR-99b-5p (miR-99b), is associated with prostate cancer (PCa) progression. However, the mechanistic role of miR-99b in PCa remains to be determined. In this study, we aimed to investigate the functional and clinical signicance of miR-99b in PCa. The expression of miR-99b and its downstream targets mTOR/AR in the PCa samples were analyzed by RT/qPCR. The effects of miR-99b overexpression/inhibition on PCa cell survival/proliferation, spheroid formation, and cell migration were examined by specic assays. Luciferase reporter assays were performed to determine the binding of miR-99b to 3’ UTR of the mTOR gene. The effect of miR-99b on the expression of mTOR, AR, and PSA proteins, as well as on AKT/mTOR signaling, autophagy, and neuroendocrine differentiation markers, was analyzed by western blotting. The expression of miR-99b, mTOR, AR, PSA in AR-negative PC3 and AR-positive LNCaP cells was analyzed by RT/qPCR. The effect of miR-99b on global gene expression in PC3 cells was analyzed by RNA-seq. inhibited androgen receptor apoptosis. of by (DHT) downregulated miR-99b promoted cell PCa cell growth/survival, whereas inactivation miR-99b mediated PCa cell growth. DESeq2 rst with accession number In conclusion, our study suggests that downregulation of miR-99b expression in PCa tumors activates mTOR and AR signaling. This is consistent with our studies that show that overexpression miR-99b in PCa cells suppressed PCa cell growth by targeting mTOR/AR axis. miR-99b directly binds with 3’UTR of mTOR gene and inhibits mTOR expression, resulting in the induction of cellular autophagy in PCa cells. Our data also suggests that miR-99b also inhibits AR signaling and induces cell apoptosis. In contrast, activation of AR signaling by dihydrotestosterone (DHT) downregulated miR-99b expression and promoted cell PCa cell growth/survival. Coupling these observations makes a strong case for a role for miR-99b in mTOR regulation and as a mechanism to bypass the AR pathway. The discovery of a role for miR-99b in mTOR/AR axis regulation suggests miR-99b is a promising molecule for PCa biomarker and therapeutic development.


Background
Prostate cancer (PCa) is the second most commonly occurring cancer in men worldwide and in the United States and according to the American Cancer Society (ACS), over 248,530 cases and 34,130 deaths are estimated in the United States in 2021(https://cancerstatisticscenter.cancer.org/) [1,2]. PCa displays a health disparity, affecting African American men disproportionally with a mortality rate almost 2.2 times higher compared to white men [3]. PCa incidence increases with age, and the disease may remain asymptomatic (indolent form) in the early stages, making early detection and prevention of progression to aggressive disease more di cult [3]. Aggressive PCa ultimately leads to metastatic disease and causes severe health consequences including death [4,5].
Therapies based on Androgen Receptor (AR) inhibition such as abiraterone or enzalutamide are initially effective but eventually inherent or acquired resistance to androgen-deprivation therapies develops, a condition termed Castration-Resistant Prostate Cancer (CRPC) [6]. The advancement to CRPC typically results in incurable disease [7]. Identi cation of key biomarkers is a critical step in detecting and monitoring the progression of aggressive PCa to CRPC [8,9] [10]. Both protein and nucleic acid-based biomarkers such as microRNAs are being sought from blood, serum/plasma, and urine of PCa patients for this purpose [11][12][13][14].
In the present study, we investigated the role of miR-99b-5p (henceforth called miR-99b) in PCa. Previous studies have shown that miR-99b plays a tumor suppressor role in gastric cancer (GC) and overexpression of miR-99b reduced GC cell proliferation and cell cycle progression by targeting insulinlike growth factor 1 receptor (IGF-1R) [25]. Similarly, expression of the miR-99 family (miR-99a, miR-99b, and miR-100) is downregulated in PCa tumors and miR-99a potentially targets mTOR/AR signaling to play a role in tumor suppression [26]. Previously we demonstrated that the expression of miR-99b was signi cantly downregulated in PCa tissues with signi cantly lower expression in PCa tissues from African American patients when compared to PCa tissue samples from Caucasian American patients [27]. The mechanistic role of miR-99b in PCa remains poorly understood. In the present study, we investigated the role of miR-99b in modulating PCa cell hallmarks such as proliferation, colony-formation, spheroid formation, apoptosis, and sensitivity to anti-cancer drugs. We demonstrate that miR-99b targets mammalian Target of Rapamycin (mTOR) and induces autophagy and inhibits PCa cell growth.

RT/qPCR
Total RNAs from RWPE1, PC3, LNCaP, DU145, and MDA-PCa-2b cells were isolated using TRIZOL reagent (Life Technology). In other experiments, PC3 and LNCaP cells were transfected with NC mimic or miR-99b mimic separately for 30h. Cells were harvested, and total RNAs were isolated using TRIZOL reagent.
For miR-99b expression analysis, total miRNAs from PCa cells were isolated using the mirVana microRNA Isolation Kit (Thermo Fisher Scienti c). Total miRNAs (10 ng) were reverse transcribed using primers speci c for miR-99b-5p and U44 (Assay ID 000436 and 001094, Applied Biosystems, Carlsbad, CA) and TaqMan Reverse Transcription reagents (Applied Biosystems). Expression of miR-99b and U44 was quanti ed by RT/qPCR using TaqMan PCR master mixture and Taqman expression assay primers. U44 expression was used as an internal control. The PCR reactions were run on a QuantStudio-3 PCR system (Applied Biosystems) and relative quantitation was analyzed according to the manufacturer's protocols.
PCa tumor RNA samples RNA samples from PCa tissues (n=24) and benign adjacent tissues (n=24) were obtained from Prostate Cancer Biorepository Network (PCBN), Johns Hopkins University School of Medicine Baltimore MD. RNA samples that include miRNAs were used for the analysis of miR-99b expression. RNA samples (10 ng) were reverse transcribed using primers speci c for miR-99b-5p and U44 (Assay ID 000436 and 001094, Applied Biosystems, Carlsbad, CA) and TaqMan Reverse Transcription reagents (Applied Biosystems). Expression of miR-99b and U44 was quanti ed by RT/qPCR as described above.

Cell survival/proliferation
PCa PC3 and LNCaP cells (1x10 4 cells/well) were grown in 96 plates and transfected with NC or different concentrations of miR-99b (25-150 nM) for 48h. The effect of miR-99b on cell survival was analyzed by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) reagent (MP Biochemicals, Santa Ana, CA). Similarly, cells also transfected with NC or miR-99b alone or transfected cells were treated with vehicles or docetaxel (DTX; 5nM), rapamycin (Rapa; 100ng/ml), enzalutamide (Enz; 10 µM), and dihydrotestosterone (DHT; 1-10 nM). After 48h of transfection and drug treatments, cells were incubated with 5µl/well of MTT reagent (5 mg/ml) for 1 h at 37°C in a cell culture incubator. Cells were washed with PBS, formazan crystals were dissolved in DMSO, and cell survival was quanti ed by reading the plates at 570 nm using a Fluostar Omega plate reader. In other experiments, cells (1 x10 4 ) were seeded onto 96-well plates and transfected with NC and miR-99b or miR-99b inhibitor for 48h. Cell proliferation was measured by adding 10 µl of WST-1 reagent according to the manufacturer's instructions (Roche Applied Science, Indianapolis, IN). Cell proliferation was measured by reading the plates at 450 nm using Fluostar Omega plate reader (BMG Lab tech, Cary, NC). All experiments were repeated three times with triplicates.
Cell Colony formation assay PC3 and LNCaP cells (1x10 6 cells/well) were grown in a 6-well plate for 18h and transfected with NC, miR-99b, or miR-99b inhibitor (100 nM). After 24h of transfection, cells were trypsinized and counted, and live cells (4000 cells/well) were re-plated in 6-well plates in triplicate and allowed to grow for 7-10 days. Cells were washed with PBS, xed with cold methanol, and stained with 0.1% crystal violet for 1h. Cells were washed with distilled water and allowed to dry. Cell blue colonies pictures were photographed, cell colonies were counted and plotted.

Cell migration assay
The effect of miR-99b on the migratory ability of PCa PC3 and LNCaP cells was determined by wound healing migration assay as described previously [28,29]. Cells (1x10 6 cells/well) were grown in a 6-well plate for 18h and transfected with NC or miR-99b mimic or miR-99b inhibitor mimic for 24h. After transfection, a cell monolayer was scraped using a micropipette tip (A 0 ). At 24h post-wounding (A 24 ), cells were photographed, and the migration gap length was calculated using ImageJ software (https://imagej.nih.gov/ij/). The percent wound closure was calculated using the formula [(A 0 -A 24 )/A 0 ] × 100 and plotted.
Luciferase assay PC3 and LNCaP cells (1x10 4 cells/well) were transfected with 0.5 µg of mTOR-3'UTR-Luciferase construct (OriGene) in 96 well plates. After 18h of transfection, cells were transfected with 100 nM of miR-99b and miR-99a wild-type mimics, or miR-99b and miR-99a mutant mimic (custom-designed from Integrated DNA Technologies) or NC mimic separately for an additional 24h. Transfected cells were washed with PBS and mixed with luciferase substrate (Promega). Plates were covered with aluminum foil to protect from light and incubated at room temperature for 15 min. The plates were read by using a Fluostar Omega plate reader (BMG Lab Tech, Cary, NC) and relative luciferase activity was measured and plotted.

Immuno uorescence
To determine the effect of miR-99b on LC3B mediated autophagy-related puncta formation, rst PC3 and LNCaP cells (1 × 10 5 ) cells were grown on coverslips in 6-well plates and transfected with LC3B-GFP construct (1µg/well). After 24h of transfection, cells were further transfected with NC or miR-99b mimic or miR-99b inhibitor mimic for 24h. After washing with PBS, cells were xed with paraformaldehyde (4%) for 15 min and after washing with PBS cells were mounted with Vectashield-DAPI mounting medium (Vector Laboratories, Burlingame, CA, Cat # H-1500-10). Slides were observed under a Nikon uorescence microscope type 120 and images were captured. The LC3B mediated puncta formation was quanti ed from cells (n=6) and plotted.

RNA-seq and analysis
PC3 cells were transfected with NC mimic or miR-99b mimic for 24h and total RNA was isolated using TRIZOL reagents (Sigma) from triplicate samples according to the manufacturer's instructions and sent to GENEWIZ, LLC. (South Plain eld, NJ, USA) for RNAseq analysis. RNA samples were quanti ed using Qubit 2.0 Fluorometer (Life Technologies, Carlsbad, CA, USA) and RNA integrity was checked using Agilent TapeStation 4200 (Agilent Technologies, Palo Alto, CA, USA). RNA sequencing libraries were prepared using the NEBNext Ultra RNA Library Prep Kit for Illumina following the manufacturer's instructions (NEB, Ipswich, MA, USA). Genes with adjusted p-values < 0.05 and absolute log2 fold changes > 1 were called differentially expressed genes for each comparison. A gene ontology analysis was performed on the statistically signi cant set of genes by implementing the software GeneSCF. The goa_human GO list was used to cluster the set of genes based on their biological process and determine their statistical signi cance. A PCA analysis was performed using the "plotPCA" function within the DESeq2 R package. The plot shows the samples in a 2D plane spanned by their rst two principal components. The top 500 genes, selected by highest row variance, were used to generate the plot. The array data were submitted to Array Express with accession number E-MTAB-10350.

Statistical analysis
Results from duplicate or triplicate independent experiments are presented as mean ± SEM. Differences between groups were analyzed using a two-tailed Student's t-test. A P value of < 0.05 was considered statistically signi cant. Statistical signi cance was determined by Graph Pad Prism 9 software (GraphPad Software Inc., La Jolla, CA).

Expression of miR-99b in PCa
In earlier studies, we and others have demonstrated that miR-99b is signi cantly downregulated in PCa tumor tissues compared with their paired adjacent non-tumor tissues [26,27]. We have recently analyzed the expression of miR-99b-5p and miR-99b-3p in prostate adenocarcinoma (PRAD) as reported in The Cancer Genome Atlas (TCGA) data set and as presented by MIR-TV (http://mirtv.ibms.sinica.edu.tw/analysis.php; access date 10 September 2021) ( Supplementary Fig. S1).
In the current study, we mainly focus on miR-99b-5p (hereafter called miR-99b) and it's biological signi cance in PCa. Since TCGA data set and earlier studies suggest the dysregulation of miR-99b expression in PCa tissues. To clarify the expression of miR-99b, we now extend these studies using RT/qPCR to further quantify the expression levels of miR-99b in prostate cancer patient tissues and benign normal/control tissues. Our data show that the expression of miR-99b is signi cantly downregulated in PCa tissues (n=24) compared with their paired adjacent normal tissues (n=24) (p<0.001) (Fig. 1A). Our RT/qPCR data indicated that in the same PCa tissues where there was downregulation of miR-99b, there was also increased expression of mTOR (p<0.05), AR (p<0.05), and AR downstream PSA gene (P<0.069) compared with paired adjacent normal tissues ( Fig. 1B & Fig. 1C left and right panels) suggesting that miR-99b regulates the expression of mTOR and AR expression (and AR target genes such as PSA) in PCa tissues, consistent with an earlier report [26].
To investigate the role of miR-99b in regulating mTOR and AR/PSA expression, we rst analyzed the expression of miR-99b in RWPE1 cells (derived from normal prostate tissue) versus PCa cell lines PC3 and DU145 (AR-negative) and LNCaP and MDA-PCa-2b (AR-positive) (Fig. 1D). RT/qPCR data indicated that the expression of miR-99b is higher in PC3 cells (~12.5 fold) and LNCaP cells (~2.23 fold) compared with RWPE1 cells (Fig. 1D). No signi cant differences were observed in DU145 and MDA-PCa-2b cells. Since miR-99b has been previously reported to target mTOR mRNA [26], we also evaluated the relationship between miR-99b expression and mTOR protein expression in PCa cell lines. Immunoblotting data demonstrated that in PC3 cells, where miR-99b expression was highest, the expression of mTOR protein was lower compared with other cell lines, indicating that the endogenous higher expression of miR-99b is inversely associated with the expression of mTOR (Fig. 1E). In addition, the lower expression on miR-99b in LNCaP cells was correlated with upregulation of mTOR protein expression (Fig. 1D &E).
To investigate the role of miR-99b in mTOR regulation, we overexpressed miR-99b in PC3 and LNCaP cells (Fig. 1F). Overexpression of miR-99b resulted in signi cantly decreased mTOR mRNA expression in both cell lines compared with the negative control (NC) transfected PC3 and LNCaP cells (Fig. 1G). Immunoblotting data showed that overexpression of miR-99b in PC3 and LNCaP cells downregulated total mTOR protein expression including the activated form of phosphorylated serine-2448 mTOR (pS2448-mTOR) in both cell lines We also assessed the impact of miR-99b overexpression on AR expression and AR downstream PSA protein expression by immunoblotting. Overexpression of miR-99b inhibited AR expression and AR downstream PSA expression in LNCaP cells (Fig. 1H, upper and lower panels). In contrast, transfection of miR-99b inhibitor inactivated endogenous miR-99b and activated pS2448-mTOR and mTOR levels in both cell lines and AR/PSA levels in LNCaP cells (Fig. 1H, upper and lower panels). Collectively our data suggest that miR-99b controls mTOR and AR expression in PCa cells.

miR-99b inhibits PCa cell proliferation/migration and increases sensitivity towards docetaxel (DTX) by inducing cell apoptosis
To investigate the role of miR-99b in PCa cell proliferation, we transfected PC3 and LNCaP cells with NC or miR-99b mimic (three different concentrations) for 48h and analyzed the effect on PCa cell proliferation by WST-1 assay ( Fig. 2A). A dose-dependent expression of miR-99b in PC3 and LNCaP cells decreased cell proliferation (~50%) compared with NC transfected cells ( Fig. 2A). Also, overexpression of miR-99b signi cantly decreased cell survival activity (MTT assay) in PC3 (32 %) and LNCaP (22%) cells compared with NC transfected cells (Fig. 2B). In addition, the inactivation of miR-99b by miR-99b inhibitor restored PCa cell survival (Fig. 2B). We next analyzed the effect of miR-99b mimic and miR-99b inhibitor on cell colony formation (Fig. 2C). Overexpression of miR-99b in PC3 and LNCaP cells decreased (~55%) cell colony formation ability compared with NC transfected cells and inactivation of miR-99b by miR-99b inhibitor restored PCa cell colony-forming ability (Fig. 2C, left and right panels). To test the effect of miR-99b on PCa cell migration, we performed a wound-healing assay (Fig. 2D). Overexpression of miR-99b signi cantly decreased cell migration compared with NC transfected PCa cells and inactivation of miR-99b restored cell migration (Fig. 2D, left and right panels). Overexpression of miR-99b also signi cantly decreased PCa cell spheroid formation in vitro (Fig. 2E, left and right panels). Finally, we analyzed the effect of miR-99b on cell apoptosis. PCa cells were transfected with two concentrations of miR-99b and lysates were immunoblotted with apoptotic markers (Caspase 3 and Cleaved-PARP). The immunoblotting data demonstrated that expression of miR-99b decreased mTOR expression as expected and also induced cleaved caspase 3 expression in PC3 cells and cleaved PARP expression in LNCaP cells, suggesting that miR-99b induces cell apoptosis and inhibits cell proliferation (Fig. 2F).
We analyzed the sensitivity of PCa cell lines PC3 and LNCaP towards the anti-PCa drug docetaxel (DTX) after transfection of miR-99b. DTX-resistance is a major challenge in PCa treatment and the mechanism of the resistance is not fully understood, however, a previous study demonstrated that PC3 cells are signi cantly more resistant to DTX compared with LNCaP or DU145 cells [30]. DTX treatment reduced cell survival in PC3 and LNCaP cells and LNCaP cells showed signi cantly higher susceptibility towards DTX as expected (Fig. 2G, left and right panels). Expression of miR-99b and DTX treatment signi cantly decreased cell survival compared with NC transfected and DTX treated PCa cells suggesting that overexpression of miR-99b increased sensitivity of PC3 and LNCaP cells towards the anti-PCa drug DTX (Fig. 2G, left and right panels). In addition, the overexpression of miR-99b did not alter the expression of E-cadherin and N-cadherin expression in PC3 and LNCaP cells, suggesting that enforced expression of miR-99b did not play any role in epithelial-mesenchymal transition (EMT) (Fig. 2H). Collectively our data suggest that miR-99b inhibits PCa cell proliferation/spheroid formation, cell migration, and increases sensitivity to DTX by inducing cell apoptosis.
We next analyzed the effect of miR-99a and miR-99b on mTOR expression. RT/qPCR data showed that both miR-99a and miR-99b overexpression downregulated mTOR expression in PC3 and LNCaP cells and downregulated AR and PSA expression in LNCaP cells (Fig. 3C & D). In addition, we analyzed the effect of miR-99a and miR-99b on mTOR, AR, and PSA protein expression in PC3 and LNCaP cells (Fig. 3E). Immunoblotting data showed that overexpression of miR-99a or miR-99b decreased mTOR expression in both PC3 and LNCaP cells and AR and PSA expression in LNCaP cells as expected (Fig. 3E), suggesting that miR-99a or miR-99b targets the mTOR-3'UTR and inhibits mTOR protein expression. However, the mechanistic role of miR-99a and miR-99b on the downregulation of AR and its downstream PSA is unknown and needs further investigation.

miR-99b inactivates mTOR, induces autophagy and cell death in PCa cells
To determine the impact of miR-99b on global gene expression in PCa cells, we transiently overexpressed miR-99b in PC3 cells (Supplementary gure 2A), and the effect of miR-99b on global gene expression was analyzed by RNA-seq. RNA-seq heat maps and volcano plots of differential gene expression, and gene ontology analysis revealed that overexpression of miR-99b in PC3 cells modulated global gene expression pathways involved in PCa cell signaling, cell proliferation, angiogenesis, and drug-response pathways (Supplementary gure 2B-D). We next speci cally analyzed the impact of miR-99b on mTOR signaling and related gene expression (Fig. 4A). The RNA-seq data also indicated that overexpression of miR-99b downregulated mTOR expression as expected, and also downregulated the mTOR complex regulatory genes DEPTOR and RHEB and upregulated ULK1 expression. Moreover, miR-99b downregulated mTOR downstream genes including RPS6KA1, RPS6KL1, RPS6KA5, and upregulated HIF1A-AS2 and HIF1AN and GRB10, genes which are directly or indirectly involved in the cellular autophagic process (Fig. 4A).
Since the induction of autophagy is controlled by the mTOR signaling pathways [32] and the expression of mTOR is controlled by miR-99b, we analyzed the role of miR-99b on the regulation of autophagy through mTOR. PC3 and LNCaP cells were transfected with two concentrations of miR-99b, and the expression of autophagic markers was analyzed by immunoblotting (Fig. 4B). Overexpression of miR-99b in PC3 and LNCAP cells decreased mTOR expression as expected and increased expression of autophagic biomarkers including Beclin1, ATG3, 4-EBP1, and LC3B I/II (Fig. 4B, left two panels). In contrast, inactivation of miR-99b by miR-99b inhibitor stabilized endogenous mTOR and decreased expression of Beclin1, ATG3, 4-EBP1, and LC3B I/II autophagy-related biomarkers (Fig. 4B, right two panels). To visualize the role of miR-99b in the induction of autophagy, we analyzed LC3B-related puncta formation in PC3 and LNCaP cells after co-transfection of LC3B-GFP plasmid and miR-99b mimic or miR-99b inhibitor (Fig. 4C). Immuno uorescence data showed that overexpression of miR-99b induced LC3Brelated puncta formation in PC3 and LNCaP cells, whereas inactivation of miR-99b decreased LC3Brelated puncta formation (Fig. 4C, left and right panels) suggesting that miR-99b induced cellular autophagy by inactivation of mTOR.
We also analyzed the effect of miR-99b and the mTOR inhibitor rapamycin on AKT/mTOR signaling, autophagy biomarkers, and PCa cell death (Fig. 4D-G). Treatment with rapamycin and overexpression of miR-99b inactivated phosphorylation of serine-2448 of mTOR (pS2448-mTOR) and phosphorylation of serine-473 of AKT (pS473-AKT) (Fig. 4D, left panel) and induced expression of Beclin1, ATG3, 4-EBP1, and LC3B I/II autophagy markers in PC3 and LNCAP cells compared with NC transfected and rapamycintreated cells (Fig. 4D, right panels). Moreover, treatment with rapamycin and overexpression of miR-99b inhibited AR and PSA expression and induced cleaved PARP expression compared with NC transfected or miR-99b transfected LNCaP cells (Fig. 4E & F). Interestingly, treatment of rapamycin and overexpression of miR-99b signi cantly reduced cell survival in PC3 and LNCAP cells compared with NC transfected and rapamycin-treated or miR-99b transfected PCa cells (Fig. 4G). These data suggest that miR-99b targets mTOR as well as AR signaling pathways and induces autophagy and cell death in PCa cells. miR-99b/mTOR/AR axis modulates autophagy and cell death in PCa cells Enzalutamide (ENZ) is an AR inhibitor that improves PCa patient survival in metastatic and hormonesensitive prostate cancer [33]. Our data suggest that miR-99b downregulates AR expression in LNCaP cells (Fig. 1H, 3D, and 4E) and thus may affect PCa cell ENZ sensitivity. We rst analyzed the effect of ENZ alone on AR, AKT/mTOR levels, and cell death in PCa cell lines. Treatment of two concentrations of ENZ for 24h in LNCaP cells decreased AR expression and its downstream target PSA as expected (Fig.  5A). Interestingly no effect on AKT/mTOR signaling was observed when PC3 or LNCaP cells were treated with ENZ for 24h. However, increased expression of LC3B II and cPARP was detected in PC3 and LNCaP cells (Fig. 5A). A time-dependent (24-72h) treatment of ENZ (10 µM) decreased expression of AR and PSA. Treatments of ENZ do not show any effect on AKT/mTOR signaling. However, ENZ treatments increased the expression of LC3B II and cPARP in a time-dependent manner (Fig. 5B). Further, a timedependent treatment of ENZ affects miR-99b expression in PC3 cells but did not show any signi cant effect on miR-99b expression in LNCaP cell lines (Fig. 5C, Upper and lower panels). Interestingly, immunoblotting data demonstrated that overexpression of miR-99b decreased AR and PSA levels and increased c-PARP expression in LNCaP cells (Fig. 5D, right panel) and inactivation of miR-99b by miR-99b inhibitor in LNCaP cells speci cally activated AR and PSA expression and decreased c-PARP expression (Fig. 5D, right panel). Since miR-99b induced c-PARP expression, we further analyzed the effect of miR-99b and ENZ on PCa cell survival. Compared with NC-transfected and ENZ treatment, overexpression of miR-99b and ENZ treatment decreased cell survival more e ciently in LNCaP cells compared with PC3 cells (Fig. 5E, right and left panels) suggesting that ENZ treatment and enforced expression of miR-99b suppresses LNCaP cell growth by inducing autophagy and cell death.
Dysregulation of the androgen dihydrotestosterone (DHT) was reported in PCa and DHT is known to regulate AR signaling [34]. We investigated the role of DHT in the regulation of miR-99b. PC3 and LNCaP cells were treated with DHT for 30h and the expression of miR-99b was analyzed. RT/qPCR data demonstrated that DHT treatments downregulated miR-99b expression in LNCaP cells, but no signi cant effect on miR-99b expression was observed in PC3 cells (Fig. 6A). DHT treatments also upregulated mTOR and PSA gene expression in LNCaP cells, but no signi cant change in mTOR expression was observed in PC3 cells (Fig. 6B, right and left panels). Similarly, immunoblotting data demonstrated that DHT treatment increased AR and PSA expression as well as pS2448-mTOR expression in LNCaP cells (Fig. 6C). Interestingly, DHT treatment decreased cleaved PARP expression in LNCaP cells (Fig. 6C), and a dose-dependent DHT treatment increased cell survival in LNCaP compared with PC3 cells (Fig. 6D) suggesting that DHT downregulates miR-99b expression, activates mTOR/AR expression, and increases LNCaP cell survival. miR-99b regulates neuroendocrine differentiation (NED) in PCa cells miRNAs are known to regulate neuroendocrine differentiation (NED). For example, previous studies have reported that miR-663 upregulates the expression of chromogranin A (CHGA), synaptophysin (SYP), neuron-speci c enolase (ENO2), dehydrogenase/reductase 7 (DHRS7), NK3 homeobox 1 (NKX3.1), 24dehydrocholesterol reductase (DHCR24), and proteasome 20S subunit alpha 7(PSMA7) and induces NED in PC3, LNCaP and DU145 cell lines [35][36][37]. Since our RNA-seq data suggested that miR-99b modulates cell proliferation and drug-response pathways, we asked whether miR-99b is involved in NED. Using RNAseq microarray data, we analyzed the effect of miR-99b on the expression of NED-related markers. Our data showed that overexpression of miR-99b signi cantly up-regulated enolase 2 (ENO2) and neuronal growth regulator 1 (NEGR1) but had no signi cant effect on synaptophysin and neural cell adhesion molecule 2 (NCAM2), dehydrogenase/reductase 7 (DHRS7), NK3 homeobox 1 (NKX3.1), and proteasome 20S subunit alpha 7(PSMA7) expression (Fig. 7A). To test the effect of miR-99b on NED in PCa cells, we grew PC3 and LNCaP cells in normal serum or charcoal-stripped serum and, after transfection of NC and miR-99b for 48h, the cells were observed under a microscope for neurite development (Fig. 7B). Our data show that expression of miR-99b induced neurite development in PC3 and LNCaP cells grown in charcoal-stripped serum compared with normal serum grown cells and transfected with miR-99b (Fig. 7B). Furthermore, we immunoblotted the cell lysates with NED markers and observed that overexpression of miR-99b upregulated the expression of synaptophysin, chromogranin A, and NCAM-1/CD56 in cells transfected with miR-99b and grown in charcoal-stripped serum media (Fig. 7C). These results suggest that miR-99b may be involved in the neuroendocrine differentiation in PCa cells.

Discussion
In the current study, we investigated the role of miR-99b in PCa cells. Our data demonstrated that miR-99b targets the mTOR/AR axis and inhibits PCa cell proliferation. Multiple previous studies have demonstrated that miR-99b targets mTOR. For example, miR-99b targets mTOR and modulates neuroregeneration in mice after spinal cord injury [38], and downregulation of miR-99b in pancreatic cancer cells upregulates mTOR expression which leads to increased irradiation resistance [31]. Moreover, reports have demonstrated that miR-99b and miR-203a target mTOR and suppress gastric cancer growth [25]. Other studies have shown that the miR-99b family miRNAs (miR-99a, miR-99b, and miR-100) are downregulated in PCa tumors and miR-99a potentially targets mTOR and AR signaling and inhibits PCa growth [26]. Our data further support these studies as we have shown that miR-99b expression was downregulated in PCa tumor samples and leads to up-regulation of mTOR and AR expression.
We also investigated the role of miR-99b in PCa cell growth/suppression and our data showed that miR-99b inhibits mTOR expression by binding with the 3'UTR of the mTOR gene and suppresses mTOR activity in both AR-negative PC3 and AR-positive LNCaP cells and miR-99b also inhibits AR signaling in LNCaP cells. Our data further suggest that through targeting mTOR and AR signaling, miR-99b inhibits cell proliferation, cell survival, cell migration, cell colony formation, and spheroid formation in PCa cells.
Enforced expression of miR-99b also induced cell apoptosis in both PC3 and LNCaP cells. Interestingly, we found that miR-99b induced cellular autophagy by inactivation of mTOR and also suppressed PCa cell growth (Fig. 6E).
Autophagy is an endogenous catabolic process that degrades damaged cell organelles and unfolded proteins through lysosomal degradation and thus maintain cellular homeostasis [39]. In cancer, autophagy regulates cell proliferation, differentiation, survival, and apoptosis and plays a dual role in tumor development as well as suppression [40]. Since mTOR signaling downregulates autophagy [41], here we demonstrated that by inactivation of mTOR, miR-99b induces expression of autophagy-related biomarkers including Beclin1, ATG3, 4EBP1, and LC3B. RNA-seq data further showed that enforced miR-99b expression dysregulates mTOR complex 1 and mTOR complex 2 gene expression, as well as affects mTOR downstream gene signaling cascades. Interestingly, reports also suggest that activation of cellular autophagy contributes to neuroendocrine differentiation (NED) in prostate cancer cells [42,43]. In our studies, we have demonstrated that enforced expression of miR-99b increased expression of the NED biomarkers chromogranin A, CD56, and synaptophysin in prostate cancer cells. The exact mechanism by which miR-99b leads to increased NED in prostate cancer cells remains unclear.
Enzalutamide (ENZ) was approved in 2012 by the FDA for clinical use. Due to its robust effectiveness against PCa, it remains among the most prevalent rst-line treatment for Castrate-Resistant Prostate Cancer (CRPC) [44]. Drug resistance mechanisms of CRPC in advanced-stage prostate cancer, are now better understood, but the mechanisms are not completely elucidated. Enzalutamide resistance is believed to occur through one of three major paths; 1. Persistent activation of the androgen receptor, 2. Bypass of the AR pathway, and 3. Development of androgen receptor independence [44]. In our study, we demonstrated that exposure ENZ, particularly in AR-positive LNCaP cells, did not show any effect of miR-99b expression. However, ENZ inactivated AR signaling and increased LC3B and c-PARP expression in LNCaP cells, suggesting that ENZ up-regulates autophagy and cell apoptosis. In contrast, overexpression of miR-99b and ENZ treatment reduced LNCaP cell survival. As suggested earlier, the expression of miR-99 is downregulated in hormone native tissue, hormone-refractory prostate cancer (HRPC) tissue, and prostatic hyperplasia (BHP) tissue [45]. Our data suggest that exposure of LNCaP cells to dihydrotestosterone (DHT) downregulates miR-99b expression and DHT treatments also activate AR/mTOR signaling, downregulate autophagy and cell death, and increase cell proliferation.
Upregulation of the AKT/mTOR pathway can play a dominant role over AR signaling and drug resistance in PCa cells [46]. The constitutive activation of AR through mTOR upregulation makes it a critical pathway of interest in studying the circumvention of AR resistance [46]. More aggressive PC3 cells demonstrate a relatively rapid change in mTOR pathway overregulation, strongly indicating the importance of controlled exposure of enzalutamide in CRPC [47]. Prostate cancer cells already sensitized to enzalutamide and AR dysfunction may rapidly bypass AR through upregulation of mTOR pathway expression [48]. A recent study demonstrated that exposure to the AR agonist R1881 activates AR and enhanced the nuclear localization of mTOR [49]. The study indicated a paradigm shift from a focus on the central role of AR to an emphasis on mTOR/nuclear mTOR as the master transcriptional regulator of PCa cell metabolism [49]. Overall, these results suggest that the mTOR/AR axis plays an important role in PCa progression and miR-99b could be used as a therapeutic target to control AR/mTOR over-activation in PCa.

Conclusions
In conclusion, our study suggests that downregulation of miR-99b expression in PCa tumors activates mTOR and AR signaling. This is consistent with our studies that show that overexpression miR-99b in PCa cells suppressed PCa cell growth by targeting mTOR/AR axis. miR-99b directly binds with 3'UTR of mTOR gene and inhibits mTOR expression, resulting in the induction of cellular autophagy in PCa cells.
Our data also suggests that miR-99b also inhibits AR signaling and induces cell apoptosis. In contrast, activation of AR signaling by dihydrotestosterone (DHT) downregulated miR-99b expression and promoted cell PCa cell growth/survival. Coupling these observations makes a strong case for a role for miR-99b in mTOR regulation and as a mechanism to bypass the AR pathway. The discovery of a role for miR-99b in mTOR/AR axis regulation suggests miR-99b is a promising molecule for PCa biomarker and therapeutic development.

Consent for publication
Not applicable.

Availability of data and materials
All data generated during the current study are included in this article and its additional les. The datasets used and/or analyzed during the current study are available from the corresponding author on request.
The mTOR, AR, and PSA protein levels were quanti ed using ImageJ software (https://imagej.nih.gov/ij/) and presented below. ***P<0.001 compared to NC transfected cells. Figure 2 miR-99b inhibits PCa cell survival by inducing cell apoptosis. (A) PC3 and LNCaP cells were transfected with NC (150 nM) or increasing concentrations of miR-99b mimic as indicated for 48h. The effect of miR-99b on PCa cell proliferation was determined by WST1 assay as described in the materials and methods section. *** P<0.001, ### P<0.001 compared to NC transfected cells. (B) PC3 and LNCaP cells were transfected with NC or miR-99b mimic or miR-99b inhibitor for 48h and PCa cell survival was determined by MTT assay as described in the materials and methods section. ***P<0.00, ### P<0.001 compared to NC or miR-99b inhibitor transfected cells. (C) The effect of miR-99b mimic and miR-99b inhibitor on PCa cell colony formation was analyzed as described in the materials and methods section. The number of colonies was quanti ed and plotted (right panel). ***P<0.00, ### P<0.001 compared to NC or miR-99b inhibitor transfected cells. (D) The effect of miR-99b mimic and miR-99b inhibitor on PCa cell migration was analyzed by wound healing assay, percentages of wound closure were determined by ImageJ software and plotted. Representative images of the wound healing assay (left panels) and the calculated scratch area (right panels) are shown. **P<0.01 compared to NC transfected cells. ns-not signi cant. (E) PC3 and LNCaP cells were transfected with NC or miR-99b mimic for 24h, transfected cells were trypsinized, counted and cells (1x103) were further grown in ultra-low attachment surface 96 well plate (Corning) for 72h and cell spheroid formation was monitored and images were captured under a light microscope using 4X objective and presented. The effect of miR-99b on PCa cell spheroid formation was analyzed (left panels) and the spheroid area (n=6) was calculated using ImageJ software  and presented. We also generated a wild type and mutated binding nucleotides of miR-99b and miR-99a of the mTOR-3′UTR of the mTOR gene. Wild-type-miR-99b, wild-type-miR-99a, mutant-miR-99b and mutant-miR-99a custom-designed miRNAs are shown. (B) Luciferase reporter assay: PCa PC3 and LNCaP cells were transfected with mTOR-3′UTR-Luciferase reporter construct (0.5 µg DNA) for 18h and then cells were transfected with NC mimic or wild-type-miR-99b and miR-99a mimics or mutant-miR-99b and miR-99a mimic for an additional 24h. Transfected cells were lysed, and luciferase activity was measured.
(E) The schematic model represents the role of miR-99b in the regulation of mTOR/AR signaling and PCa cell apoptosis and drug resistance.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.