Steroid Sulfatase Stimulates Intracrine Androgen Synthesis and is a Therapeutic Target for Advanced Prostate Cancer

Targeting androgen signaling with androgen deprivation therapy (ADT) has been the mainstay of clinical interventions in prostate cancer. While initially effective, most men experience only transient benefit from these interventions before developing castration-resistant prostate cancer (CRPC), which is currently incurable.

The second-generation antiandrogen drugs enzalutamide and abiraterone, and most recently apalutamide and darolutamide, are approved for the treatment of CRPC (1–4). Even though these treatments are initially effective, resistance frequently occurs. This is the most common cause of treatment failure. Better understanding of the mechanisms that confer resistance is urgently needed to design more effective treatments to overcome resistance and improve therapies.

While androgen receptor (AR) mutations, amplifications, and overexpression of its variants, such as AR-V7, have been suggested to play important roles in promoting CRPC progression and induction of resistance to enzalutamide and abiraterone therapy, the mechanisms associated with resistance to these treatments are still largely unknown. Intratumoral androgen biosynthesis is well-characterized as a mechanism of CRPC (5–8). Many enzymes are involved in androgen synthesis, including steroid sulfatase (STS; refs. 3, 9). STS converts the abundantly circulated adrenal dehydroepiandrosterone sulfate (DHEAS) to biologically active DHEA and its expression in CRPC could be contributing to increased androgen levels (10). DHEAS is present at plasma concentrations up to 500 times higher than testosterone and can potentially be transported into prostate cancer cells via organic anion transporters (11–13). Once into the cells, DHEAS can be desulfated via endogenous STS into DHEA. DHEA can then be used as a precursor for the production of testosterone through the activity of hydroxysteroid dehydrogenases and further into DHT via 5α reductase, as discussed previously at European Association of Urology Annual Congress EAU20 (14).

STS is present in most prostate cancer specimens and its expression is increased in CRPC (10, 15–17). Furthermore, DHEA is a major source of tumor androgens for prostate cancer and DHEAS is the most abundant adrenal steroid in male circulation. Therefore, conversion of circulated adrenal DHEAS to DHEA by STS is believed to be an alternative source of intracrine androgen, which cannot be fully inhibited by abiraterone. While the nonspecific CYP17A1 inhibitor, ketoconazole, or the specific CYP17A1 inhibitor, abiraterone, greatly reduces circulating levels of DHEAS by up to 95%, significant circulating DHEAS concentrations are still present (∼20 μg/dL, which is about 2,000 times higher than castrate levels of testosterone; ref. 13). This suggests that DHEAS could act as a depot for further downstream androgen formation via desulfation. These clinical data highlight the importance for exploring STS inhibition to prevent DHEAS conversion to DHEA.

While the focus of prostate cancer treatment is often on reducing androgen signaling, the importance of adrenal androgens produced from DHEA is overlooked. DHEAS is present 400- to 1,000-fold higher than DHEA (DHEAS/DHEA: 2–10 μmol/L/10–25 nmol/L; refs. 11–13). Numerous STS inhibitors (STSi) have been tested for potential activity both in animal and clinical studies in breast cancer (18–25). However, while STS has been widely investigated in breast cancer, there are limited studies in prostate cancer.

In this study, we demonstrate that STS is overexpressed in patients with CRPC and CRPC cells. We found that STS controls CRPC cell growth through regulation of AR signaling and knockdown of STS suppresses cell growth and AR transcriptional activity in CRPC cells. Furthermore, we characterize two novel STS small-molecule inhibitors that demonstrated efficacy in inhibition of prostate cancer cell proliferation, AR transcriptional activity, and castration-resistant tumor growth. In addition, we found that overexpression of STS conferred resistance to enzalutamide in C4-2B cells and combination of STSi with the leading antiandrogen drug, enzalutamide, had increased efficacy for inhibiting cell growth over either drug on its own in vitro and in vivo.

LNCaP and VCaP cells were obtained from the ATCC. All experiments with cell lines were performed within 6 months of receipt from the ATCC or resuscitation from cryopreservation. C4-2B cells were kindly provided and authenticated by Dr. Leland Chung (Cedars-Sinai Medical Center, Los Angeles, CA). VCaP cells were maintained in DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 0.1 mg/mL streptomycin. PZ-HPV7 cells were maintained in keratinocyte serum-free medium with required supplements (Invitrogen). Other cell lines were maintained in RPMI1640 supplemented with 10% FBS, 100 U/mL penicillin, and 0.1 mg/mL streptomycin. All cells were maintained at 37°C in a humidified incubator with 5% carbon dioxide. All drugs were dissolved in DMSO and stored at −20°C.

STS siRNA (HSS100722) was purchased from Invitrogen. For luciferase assays, VCaP cells (1 × 10⁵ cells/well of 12-well plate) were transfected with 0.5 μg of pGL3-PSA6.0-Luc reporter plasmid with or without 20 pmol STS siRNA. The luciferase activity was determined 72 hours after transfection using a Dual-Luciferase Reporter Assay System (Promega) as described previously (26).

For overexpression of STS, STS-expressing plasmids were obtained from GenScript and transfected into C4-2B or LNCaP cells alongside vector control using Lipofectamine as described previously. STS overexpression was confirmed via Western blot analysis as described below.

C4-2B STS, C4-2B Neo, LNCaP STS, and LNCaP Neo cells supplemented with 100 nmol/L DHEAS in charcoal stripped FBS media were collected and whole-cell pellets (10 × 10⁶ cells/pellet) were flash-frozen in liquid nitrogen. Mass spectrometry for steroid metabolites was performed on the Thermo Fisher Vanquish/SciEx QTrap 6500 Spectrometer.

Cellular protein extracts were resolved on SDS-PAGE and proteins were transferred to nitrocellulose membranes. After blocking for 1 hour at room temperature in 5% milk in PBS/0.1% Tween-20, membranes were incubated overnight at 4°C with the indicated primary antibodies: STS (catalog no.: 17870-1-AP, Proteintech) and Tubulin (T5168, Sigma-Aldrich). Tubulin was used as loading control. Following secondary antibody incubation, immunoreactive proteins were visualized with an enhanced Chemiluminescence Detection System (Millipore).

VCaP cells were seeded on 12-well plates at a density of 2 × 10⁵ cells/well in media containing 10% FBS and treated with various concentrations of SI's for 24 hours. Cell lysates were harvested and subjected to STS activity assay by mixing with 0.5 mmol/L of 4-methylumbelliferyl sulfate in a 96-well plate. While 4-methylumbelliferyl sulfate is also a substrate for arylsulfatases A and B, these enzymes are inactive at pH 7.5 wherein STS has optimal activity, which makes this assay specific to STS under normal cell culture conditions (27). The plate was read in a fluorescence microtiter plate reader at excitation and emission wavelengths of 355 and 460 nm, respectively.

PZ-HPV7, LNCaP, C4-2B, VCaP, C4-2B Neo, C4-2B STS, LNCaP Neo, or LNCaP STS cells were seeded at a density of 1 × 10⁵ cells/well or 2 × 10⁵ cells/well (VCaP cells only) in 6-well plates and STS activity was determined using the Sulfatase Activity Assay Kit (BioVision) according to the manufacturer's instructions. Sulfatase activity was normalized to total protein. A drawback of this method over the 4-methylumbelliferyl sulfate assay is that it is not as specific to STS. However, due to the fact that endogenous expression levels of STS correlate to increased overall sulfatase activity when comparing the PZ-HPV7, LNCaP, C4-2B, and VCaP cells, and that the only difference between the C4-2B Neo and C4-2B STS or LNCaP Neo and LNCaP STS cells is STS overexpression, we are confident that this assay is reflective of STS activity in these cells.

C4-2B and VCaP cells were seeded in 12-well plates at a density of 0.5 × 10⁵ and 1 × 10⁵ cells/well, respectively, in media containing 10% FBS and transiently transfected with STS siRNA or control siRNA. Total cell numbers were counted after 3 and 5 days. C4-2B and VCaP cells were treated with different concentrations of STSis and total cell numbers were counted on days 3 and 5. VCaP cells were seeded in 12-well plates at a density of 1 × 10⁵ cells/well in DMEM containing 10% FBS and treated with DMSO, 20 μmol/L enzalutamide, STSis, or their combination for 3 days. Total cell numbers were counted and the cell survival rate (%) was calculated. Cell survival rate (%) = (treatment group cell number/control group cell number) × 100.

C4-2B cells were plated in 12-well plates at a density of 0.5 × 10⁵ cells/well and transiently transfected with STS-overexpressing plasmids or vector control. The following day, cells were treated with 20 μmol/L enzalutamide, 10 μmol/L abiraterone, or DMSO for 3 days. Cell number was determined using Cell Counting Kit 8 (Dojindo Laboratories) per the manufacturer's instructions.

Stably transfected C4-2B Neo, C4-2B STS, LNCaP Neo, and LNCaP STS cells were plated in 12-well plates at a density of 0.5 × 10⁵ cells/well and treated with 20 μmol/L enzalutamide or DMSO for 3 days. Cell number was determined using cell counting kit 8 per the manufacturer's instructions.

C4-2B or VCaP cells were treated with DMSO or different concentrations of STSis in media containing 10% FBS. Cells were plated at equal density (2,000 cells/dish for C4-2B and 20,000 cells for VCaP) in 100-mm dishes for 2–4 weeks. Colonies were rinsed with PBS before staining with 0.5% crystal violet/4% formaldehyde for 30 minutes and the number of colonies was counted.

Total RNAs were extracted using TRizol Reagent (Invitrogen). cDNAs were prepared after digestion with RNase-free RQ1 DNase (Promega). The cDNAs were subjected to real-time reverse transcription-PCR (RT-PCR) using Sso Fast Eva Green Supermix (Bio-Rad) according to the manufacturer's instructions and as described previously (28). Each reaction was normalized by coamplification of actin. Samples were run in triplicate on default settings of a Bio-Rad CFX-96 Real-Time Cycler. Primers used for real-time PCR are: KLK3, 5′-ccctgcccgaaagg-3′ (forward) and 5′-gatccacttccggtaatgca-3′ (reverse); NKX3.1, 5′-ccgagacgctggcagagacc-3′ (forward) and 5′-gcttaggggtttggggaag-3′ (reverse); FKBP5, 5′-gggaagatagtgtcctggttag-3′ (forward) and 5′-gcagtcttgcagccttattc-3′ (reverse); and Actin, 5′-agaactggcccttcttggagg-3′ (forward) and 5′-gtttttatgttcctctatggg-3′ (reverse).

For determination of intratumoral testosterone, sections of tumors from the VCaP model mice were homogenized in a buffer consisting of 20 mmol/L EDTA, 20 mmol/L Tris, and 20 mmol/L NaCl for 30 seconds. The homogenized tumor was extracted 2× using ethyl acetate and desiccated using a vacuum centrifuge. The resulting desiccate was reconstituted in 300 μL of calibrator diluent provided in the assay. From there, testosterone was measured using the Parameter Testosterone Assay (R&D Systems, Inc.) according to the manufacturer's directions using 100 μL of reconstituted tumor extract per well.

Blood was collected from the VCaP tumor model mice and the serum was isolated. The PSA level was measured using a PSA ELISA Kit (KA0208, Abnova, Inc.) according to the manufacturer's instructions as described previously (29).

VCaP cells (4 × 10⁶) were mixed with Matrigel (1:1) and injected subcutaneously into the flanks of 6- to 7-week-old male SCID mice. After PSA levels were detectable (5–10 ng/mL), mice were castrated and the tumor volume was monitored, once the tumors grew back to 50–100 mm³, tumor-bearing mice were randomized into three groups (5 mice in each group) for the first animal study: (i) vehicle control (5% Tween 80 and 5% ethanol in PBS, i.p.), (ii) SI-1 (25 mg/kg, i.p.), and (iii) SI-2 (25 mg/kg, i.p.). Tumors were measured using calipers twice a week and tumor volumes were calculated using the formula: length × width²/2. Tumor tissues were harvested after 3 weeks of treatment.

For the second animal experiment, VCaP cells (4 × 10⁶) were mixed with Matrigel (1:1) and injected subcutaneously into the flanks of 6- to 7-week-old male SCID mice. After PSA levels were detectable (5–10 ng/mL), mice were castrated and the tumor volume was monitored, once the tumors grew back to 50–100 mm³, tumor-bearing mice were randomized into four groups (5 mice in each group) and treated as follows: (i) vehicle control (5% Tween 80 and 5% ethanol in PBS, i.p.), (ii) enzalutamide (20 mg/kg, p.o.), (iii) SI-1 (25 mg/kg, i.p.), and (iv) enzalutamide (20 mg/kg, p.o.) + SI-1 (25 mg/kg, i.p.). Tumors were measured using calipers twice a week and tumor volumes were calculated using the formula: length × width²/2. Tumor tissues were harvested after 3 weeks of treatment.

Tumors were fixed by formalin. Paraffin-embedded tissue blocks were dewaxed, rehydrated, and blocked for endogenous peroxidase activity. Antigen retrieval was performed in sodium citrate buffer (0.01 mol/L, pH 6.0) in a microwave oven at 1,000 W for 3 minutes and then at 100 W for 20 minutes. Nonspecific antibody binding was blocked by incubating with 10% FBS in PBS for 30 minutes at room temperature. Slides were then incubated with anti-Ki-67 (at 1:500; NeoMarker) at 4°C overnight. Slides were then washed and incubated with biotin-conjugated secondary antibodies for 30 minutes, followed by incubation with avidin DH-biotinylated horseradish peroxidase complex for 30 minutes (Vectastain ABC Elite Kit, Vector Laboratories). The sections were developed with the Diaminobenzidine Substrate Kit (Vector Laboratories) and counterstained with hematoxylin. Nuclear staining of cells was scored and counted in five different vision fields. Images were taken on an Olympus BX51 Microscope equipped with a DP72 camera.

Data are presented as means ± SD of the mean. Statistical analyses were performed with Microsoft Excel analysis tools or JMP. Differences between individual groups were analyzed by one-way ANOVA followed by the Scheffé procedure for comparison of means. P < 0.05 was considered statistically significant.

Adrenal-derived steroids such as DHEAS act as precursors for the production of more potent androgens in peripheral tissues including prostate. DHEA synthesized from cholesterol can be sulfonated by DHEA sulfotransferase into DHEAS. The conversion of DHEAS back to biologically active DHEA is mediated by STS. To determine the role of STS in prostate cancer, we first interrogated two independent Gene Expression Omnibus (GEO) databases and examined STS expression in patients with primary prostate cancer and metastatic castration-resistant prostate cancer (mCRPC). As shown in Fig. 1A, STS expression was significantly elevated in mCRPC compared with primary prostate cancer tissues in two independent databases. We next determined the expression levels of STS in benign prostate epithelial cells and cancer cells. As shown in Fig. 1B, STS expression was relatively low in PZ-HPV7 benign epithelial cells, and moderate in LNCaP and C4-2B cells. The level of STS is substantially upregulated in VCaP cells and LuCaP 35CR tumors. STS expression correlated to increased overall sulfatase activity (Fig. 1B) as well as STS activity (Supplementary Fig. S1A). We next determined whether STS affects cell proliferation in CRPC cells, as shown in Fig. 1C left, knockdown of STS significantly inhibited C4-2B cell growth in a time-dependent manner. These results were also confirmed in VCaP cells, where knockdown of STS also significantly reduced cell number (Fig. 1D). Furthermore, transient overexpression of STS in C4-2B cells promotes resistance to treatment with enzalutamide and abiraterone (Fig. 1E).

In LNCaP and C4-2B cells stably overexpressing STS, we found that overexpression of STS resulted in increased sulfatase activity (Fig. 1F) and STS activity (Supplementary Fig. S1B). Intracellular levels of androgens were increased in the STS-overexpressing cells when the cells were supplemented with DHEAS (Fig. 1G). Unexpectedly, DHEAS was increased in the C4-2B STS-overexpressing cells compared with the C4-2B Neo cells. This observation is currently under further investigation, and one possible explanation could be incomplete rinsing of these samples prior to mass spectrometry analysis to cleanse them of the supplementary DHEAS in the media. In addition, we found that the stable STS-overexpressing clones had a marked resistance to treatment with enzalutamide compared with vector-expressing controls (Fig. 1H). Collectively, these results suggest that STS is overexpressed in CRPC and overexpression of STS increases intracrine androgens and regulates cell growth.

Having demonstrated that inhibition of STS expression using specific siRNA suppressed prostate cancer cell growth, we next designed and synthesized 11 small-molecular STSis (SI-1 to SI-11) with molecular weights ranging from 300 to 460 kDa. These STSis were assessed for their efficacy on preventing sulfatase activity. As shown in Fig. 2A, eight novel inhibitors were able to inhibit STS activity. Among them, SI-1, SI-2, SI-3, SI-6, SI-7, and SI-10 inhibited STS activity by 50%–60%, and SI-4 and SI-5 suppressed 30%–40% STS activity. We also determined whether these inhibitors suppressed prostate cancer cell growth. As shown in Fig. 1B, SI-1, SI-2, SI-3, SI-4, SI-6, SI-7, and SI-11 impeded VCaP cell growth. In general, the extent of cell growth inhibition by these inhibitors was correlated with their ability to decrease STS activity (Fig. 2A and B). We then selected SI-1 and SI-2, which had slightly improved overall ability to both inhibit STS activity and reduce cell number, to be further characterized in subsequent studies (Fig. 2C). As shown in Fig. 3A and B, both SI-1 and SI-2 significantly suppressed cell growth in a time-dependent manner. These results were further confirmed by clonogenic assays; both STSis significantly suppressed colony formation in a dose-dependent manner in C4-2B and VCaP cells (Fig. 3C). Collectively, these results suggest that inhibition of STS activity using selective STSis suppresses cell growth.

Knowing that STS overexpression results in increased androgen levels, we next wanted to determine the effects of STS on AR signaling. We analyzed RNA-sequencing transcriptomes of VCaP cells treated with STSis to identify gene programs affected by STS inhibition (GEO accession number: GSE153245). Genes that are characterized as AR targets or androgen-induced genes were significantly inhibited by both inhibitors (Fig. 4A). Conversely, androgen-repressed genes were strongly induced by the inhibitors (Fig. 4A). Further examination by gene set enrichment analysis (GSEA) using the AR activity signature revealed a highly robust disruption of AR gene programs by STS inhibition (Fig. 4B). Significantly, the STSis also strongly inhibited the expression of genes preferentially upregulated by AR-V7 (Fig. 4C). qRT-PCR verified that well-characterized AR targets, including FKBP5, KLK3, and NKX3-1, were strongly inhibited by STSis in C4-2B and VCaP cells (Fig. 4D). The effects of STSis on AR signaling in C4-2B and VCaP cells were further analyzed by assessing PSA luciferase activity in C4-2B and VCaP cells treated with SI-1 and SI-2. PSA luciferase activity was significantly inhibited in a dose-dependent manner in both cell lines (Fig. 4E). Similarly, knockdown of STS expression using STS-specific siRNA suppressed the expression of AR target genes and inhibited PSA luciferase activity (Fig. 4F and G). These results suggest that the selected STSis can downregulate AR signaling in CRPC cells.

We next examined the effects of STS inhibition on tumor growth in vivo. The VCaP tumor model was selected because it expresses higher levels of endogenous STS. VCaP tumors were allowed to develop in intact SCID mice and the animals were castrated after PSA levels reached around 5–10 ng/mL. One to 2 weeks later, when the tumors grew back to 80–100 mm³ (indicating relapse), mice were treated with SI-1 or SI-2 (25 mg/kg, i.p.) for 3 weeks. Both SI-1 and SI-2 significantly inhibited tumor growth compared with the untreated controls (P = 0.00155 and P = 0.00061, respectively; Fig. 5A–C). Mouse body weights were not affected by either treatment compared with the control group (Fig. 5D). Serum PSA level was significantly decreased after 3 weeks of drug treatment. As shown in Fig. 5E, the control group expressed 300 ng/mL PSA at 3 weeks of treatment timepoint, SI-1 decreased PSA by about 50% (P < 0.01) and SI-2 by about 55% (P < 0.01). Furthermore, intratumoral testosterone levels were also decreased by SI-1 and SI-2 treatment (Fig. 5F). Tumor proliferation was appraised by IHC. As shown in Fig. 5G, both SI-1 and SI-2 treatments had less Ki67 staining in tumors. Taken together, these results further demonstrate that inhibition of STS by SI-1 or SI-2 has great antitumor efficacy in vivo.

Previous studies demonstrate that uncontrolled intraprostatic androgen synthesis likely plays an important role in enzalutamide and abiraterone resistance (28, 30). To test whether inhibition of STS improved enzalutamide treatment, SI-1 and SI-2 were used in combination with enzalutamide in VCaP cells. As shown in Fig. 6A, enzalutamide reduced VCaP cell growth in vitro, however, SI-1 and SI-2 were more effective. Combination of SI-1 or SI-2 with enzalutamide further reduced VCaP cell growth in vitro. In addition, siRNA specific to STS alone or in combination with enzalutamide significantly diminished AR transcriptional activity (Fig. 6B). We also confirmed the growth inhibition ability of the combination treatment in C4-2B cells (Fig. 6C).

To test the efficacy of the combination of SI inhibitors with enzalutamide in vivo, we next used the relapsed VCaP tumor model. As shown in Fig. 6D and E, enzalutamide treatment alone only marginally inhibited tumor growth and produced a tumor growth curve very similar to the vehicle control group. SI-1 treatment alone significantly suppressed tumor growth and combination of SI-1 with enzalutamide further suppressed tumor growth more than either compound alone. IHC staining confirmed that enzalutamide did not affect Ki67 expression in the relapsed VCaP tumors. SI-1 significantly decreased Ki67 expression, while combination treatment further lowered Ki67 expression (Fig. 6F). These data demonstrate that inhibition of STS activity by SI-1 improved enzalutamide treatment in vivo, providing preclinical data to support further clinical development of combination therapy of STSis and enzalutamide to treat advanced prostate cancer.

STS catalyzes the conversion of DHEAS into biologically active DHEA. Emerging clinical evidence shows that serum levels of steroids, particularly DHEAS and DHEA, remain in the high range and may serve as an ample pool of precursors for androgen synthesis, even when CYP17A1-mediated testosterone biosynthesis is inhibited with abiraterone in CRPC cells. STS is present in 85% of prostate cancer specimens (11) with increased expression in CRPC (11). Conversion of the abundantly circulated adrenal androgen DHEAS into DHEA by STS is believed to be an alternative source of androgens, which cannot be inhibited by abiraterone. In this study, we found that STS is overexpressed in patients with CRPC and in resistant cells. Overexpression of STS stimulates intracrine androgen synthesis, increases cell growth, and confers resistance to enzalutamide. Targeting STS activity using selected small-molecule inhibitors of STS inhibits STS activity, suppresses AR transcriptional activity, and reduces the growth of resistant C4-2B and VCaP prostate cancer cells in vitro and in vivo. Furthermore, STSis enhance enzalutamide treatment in vitro and in vivo.

Numerous STSis have been tested for potential activity both in animal and clinical settings in breast cancer (19, 21, 31). TZS-8478 has potent STS inhibitory activity and exhibits no estrogenicity in vitro or in vivo (19). It can completely inhibit STS activity in tumor, uterus, and liver, and also lowers the concentrations of estrone and estradiol in plasma (19). STX64 and STX 213 are new-generation STSis that can suppress hormone-dependent human breast cancer growth (20, 21). Noticeably, a phase II trial using the STSi, irosustat (STX-64), in combination with aromatase inhibitor for patients with estrogen receptor (ER)-positive breast cancer (IRIS Study) has been launched (25). Addition of irosustat to aromatase inhibitor further suppressed estradiol levels in patients with advanced breast cancer. KW-2581 can irreversibly inhibit STS activity in breast cancer cells and cause tumor shrinkage in a mouse xenograft model (32, 33). In addition, SR 16157 is a dual STS and ERα modulators inhibitor. It not only suppresses STS activity but also inhibits ERα binding affinity and has antiestrogenic effects in breast cancer (34). Although STSis have been widely investigated in breast cancer, there are limited reports in prostate cancer. In this study, we report that STS is more abundant in CRPC cell lines, such as C4-2B and VCaP cells, compared with benign prostate cells. We also identified several potent inhibitors of STS, and demonstrated that inhibition of STS activity using the selected novel inhibitors suppresses androgen signaling, and reduces cell growth in vitro and tumor growth in vivo, suggesting STS could be a valuable therapeutic target for CRPC treatment.

As expected, the androgen signaling pathway is one of the main targets for STS inhibition in our study. Both siRNA-mediated knockdown of STS and pharmacologic inhibition of STS activity decrease AR transcriptional activity. Inhibition of STS expression and activity reduces AR transactivation indicated by both the luciferase assay and RT-PCR detection of downstream molecules such as FKBP5, KLK3, and NKX3-1. Adrenal androgens in circulation, such as DHEA, contribute to CRPC tumor growth by promoting androgen synthesis. Data from our study demonstrate that overexpression of STS in both LNCaP cells and C4-2B cells results in a significant increase in intracellular testosterone. This is likely the result of upregulated STS activity producing increased amounts of DHEA from DHEAS, which is then further converted to testosterone by other enzymes. A retrospective cohort study showed that rise of serum levels of androsterone, DHEA, estrone, and estradiol correlated with shorter time to development of CRPC in men receiving ADT (35). Both DHEA and estrone are the products of STS activity and, therefore, inhibition of STS activity could reduce the levels of active DHEA and thus, suppress growth and tumor progression. In addition, STS has been implicated in epithelial-to-mesenchymal transition (36). Overexpression of STS upregulates β-catenin and its nuclear translocation. Inhibition of STS activity suppresses cyclin D1, c-Myc, and MMP-7 and expression of the mesenchymal markers vimentin, N-cadherin, and twist (36). In addition to the predicted effects of the novel STSi achieved through blocking DHEAS desulfation, there is the potential for off-target effects. Because of the fact that SI-1 and SI-2 have minimal effects in vitro in charcoal stripped media in the absence of ligand (data not shown), we feel that any off-target activities are likely minor. Ongoing studies in our laboratories continue to investigate SI-1 and SI-2 to identify alternative effects these compounds may have, and for further clinical development to treat advanced prostate cancer.

In summary, we found that STS is overexpressed in CRPC cells. Overexpression of STS increases intracrine androgen synthesis, promotes cell proliferation, and confers resistance to enzalutamide. Targeting STS activity using selected small-molecule inhibitors of STS inhibits STS activity, suppresses AR transcriptional activity, and reduces the growth of resistant C4-2B and VCaP prostate cancer cells in vitro and in vivo. Furthermore, STSis enhance enzalutamide treatment in vitro and in vivo. These studies suggest that STS plays a critical role in prostate cancer proliferation and targeting this enzyme represents a therapeutic strategy to treat CRPC and improve second-generation antiandrogen therapy.

A.C. Gao is listed as a coinventor on a patent regarding how steroid sulfatase inhibitor compounds can be useful for the treatment of cancers such as prostate and breast cancer co-owned by the University of California and Ohio State University. C.P. Evans reports personal fees from Astellas (honorarium for speaking engagements and advisory board) and Janssen (honorarium for speaker lectures and chair advisory board) outside the submitted work. No potential conflicts of interest were disclosed by the other authors.

C.M. Armstrong: Data curation, formal analysis, validation, investigation, visualization, methodology, writing-original draft, writing-review and editing. C. Liu: Data curation, formal analysis, validation, investigation, visualization, methodology, writing-original draft, writing-review and editing. L. Liu: Data curation, investigation, methodology, writing-original draft. J.C. Yang: Methodology. W. Lou: Resources, methodology. R. Zhao: Methodology. S. Ning: Methodology. A.P. Lombard: Methodology. J. Zhao: Methodology. L.S. D’Abronzo: Methodology. C.P. Evans: Resources, writing-review and editing. P.-K. Li: Resources, writing-review and editing. A.C. Gao: Conceptualization, data curation, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing-original draft, project administration, writing-review and editing.

This work was supported, in part, by grants CA179970, CA 225836, DOD PC150229, and the U.S. Department of Veterans Affairs, Office of Research & Development BL&D grant number I01BX0002653, and a Research Career Scientist Award (all to A.C. Gao). A.C. Gao is also a research career scientist at VA Northern California Health Care System (Mather, CA).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.