Hypoxia-inducible factor 1α (HIF-1α) mediates the epithelial-mesenchymal transition in benign prostatic hyperplasia

Abstract

Background: Epithelial-mesenchymal transition (EMT) based cancer cell invasion and metastasis has been thoroughly studied in prostate cancer. It was well known that EMT markers which have been found in benign prostatic hyperplasia (BPH) tissues, but system descriptions have not been described. Methods: First, in order to construct the epithelial cells to mesenchymal cell transformation model, BPH-1 cells were cultured with supernatant of prostate matrix normal prostate stromal WPMY-1 cells, after obtaining the culture medium through a filter. After that, we observed the morphology of cells cultured for a period of time by microscopy, detected cell invasion ability by transwell assay, detected cell proliferation ability by MTT, and detected EMT marker expression by western. Finally, we treated the cells with anti-HIF-1α drugs to study their effects on EMT, and then tested several related proteins simultaneously. Results: The results showed that the morphology of BPH-1 cells gradually changed to fusiform after cultured with WSCM. At the same time, E-cadherin and cytokeratin levels were significantly lower than those in normal medium. Simultaneous detection of vimentin (SMA) and Snail was positive compared to normal cultured cells. At the same time, the cells were cultured with WSCM and the invasive ability was up-regulated. After treatment with anti-HIF-1α drug, E-cadherin and CK5/8 protein expression was up-regulated, but vimentin, α-SMA, and Snail expression was down-regulated, and in addition, p-Smad3 protein expression was also down-regulated after anti-HIF-1α drug was added. Conclusion: The above results indicated that WSCM-1 stromal cell supernatant WSCM can induced BPH-1 cell interstitialization, and at the same time, by inducing EMT, secreting HIF-1α activates Smad3 signaling. Our study shows that inhibition of HIF-1α expression provides a new reference for clinical treatment of BPH.

Introduction

Benign prostatic hyperplasia (BPH) is a noncancerous enlargement of the prostate gland often resulting in lower urinary tract symptoms (LUTS). Prevalence of BPH increases with age, and BPH is a major disease among older males which significantly reduces the quality of their life [1-3]. It is well known that epithelial-mesenchymal transition (EMT) is a key inducing factor of aberrant prostate growth. EMT and its reverse mechanism mesenchymal-epithelial transition (MET) play important roles in the embryonic development, especially in the formation of the prostate, chronic inflammation, tissue reconstitution and cancer metastasis [4]. When epithelial cells undergo EMT, they lose their original markers such as E-cadherin and CK8, in addition, acquiring the phenotypes of mesenchymal cells, for instance, expression of vimentin and α-smooth muscle actin [5,6]. Until now, some clear triggers of EMT have been studied. First, cadherin switching: transitioning epithelial cells lose epithelial cadherin (E-cadherin) and express neural cadherin (N-cadherin) [7]. Secondly, apoptosis deregulation: cells lose apoptosis control and increase proliferation-to-apoptosis ratio [8]. Additionally, epigenetic regulation is also extremely significant to EMT [9-11]. Finally, the reactive stroma in the prostate microenvironment is a crucial contributor driving the EMT, it regulates androgen status and levels of multiple factors including transforming growth factor-β (TGF-β) via paracrine mechanisms [12,13]. Recently, the microenvironment, as an important regulator of EMT has been studied extensively, and plenty of factors have been found that are involved in EMT process.

The hypoxia inducible factor 1 (HIF 1) signaling pathway promotes angiogenesis, and it contributes to the progression of numerous hyperplastic diseases, including BPH. HIF 1 is a crucial signaling factor which contributes to the activation of the ‘angiogenic switch’ and results in increasing expression of numerous angiogenic growth factors, including vascular endothelial growth factor (VEGF) and basic fibroblast growth factors (bFGF) [14-17]. HIF 1α and HIF 1β form the heterodimeric transcription factor HIF-1, and as a master regulator of oxygen homeostasis, HIF 1 accumulates in the cytoplasmic matrix [18]. Along with translocation to the nucleus, HIF-1 activates hypoxia-sensitive genes and induces the expression of genes encoding angiogenic cytokines, such as VEGF and bFGF [19,20]. VEGF or bFGF binds to the specific receptors on the cytomembrane of vascular endothelial cells (ECs), which regulates ECs survival, proliferation, migration, and extends the primitive vascular network [21-25]. However, so far, there is no study that shows the role of HIF-1 in the stromal microenvironment of prostate.

We hypothesized that HIF 1α may play a significant role in inducing EMT of prostate epithelial cells through affecting stromal microenvironment, which causes BPH. To test our conjecture, we used the WPMY-1, which is a prostate stromal cell line, to simulate a stromal microenvironment for BPH-1 cells in vitro, and then we used this system to explore the influence of HIF-1α on BPH-1 cells’ EMT.

Materials and methods

Cell culture

The human BPH-1 cells were purchased from American Type Culture Collection (ATCC) (USA), the WPMY-1 cell is a human prostatic cell line, which was purchased from The Global Bioresource Center (USA). The BPH-1 cells were cultivated in RPMI-1640 fundamental medium (Invitrogen, Canada) with 10% fetal bovine serum (FBS) (Gibco, DC), but the WPMY-1 cells were maintained culture in RPMI-1640 with 5% FBS (Gibco, USA) medium. Those cells were cultured at 37 degree in temperature incubator with 5% CO₂.

Experiment reagents

All of antibody was purchased from Abcam, including polyclone anti-HIF-1α antibody and mouse or ribbit IgG isotype control antibody. And all of antibody was dissolved in PBS contain 5% BSA. The antibody HIF-1α was adapted to concentration of 2 ug/ml, and the WSCM solution was added to anti-HIF-1α (2 ug/ml).WPMY-1 supernatant-conditioned medium (WSCM) was collected from WPMY-1 cell culture supernatant. When WPMY-1 cells were cultured at 80% density, WSCM was obtained, and WPMY-1 cells need replaced to fresh complete medium per two days. In the first place, the supernatant was spun in centrifuge tubes at 2000 rpm and filtered through a 0.45 mm filter for concentration to a proper state. In the end, the fresh WSCM was storage at -80°.

Cell grouping and treatment

The cells were plated per 1×10⁵ in 6 cm dish and maintained for 24 h, then we divided those dishes into seven groups and treatment with different stimulation: a: BPH-1 cells cultured in ordinary complete medium; b: BPH-1 cells cultured in WSCM; c: BPH-1 cells cultured in WSCM treatment with anti-HIF-1α; d: BPH-1 cells cultured in WSCM treatment with mouse or rabbit IgG control; e: BPH-1 cells cultured in ordinary complete medium treatment with anti-HIF-1α antibody; f: BPH-1 cells cultured in ordinary complete medium treatment with mouse or rabbit IgG; g: WPMY-1 cells cultured in special complete medium of WPMY-1.

Immunofluorescence analysis

BPH-1 cells were collected in a 1.5 ml tube and were washed by phosphate-buffered saline (PBS), and the cells were fixed with 4% paraformaldehyde; we next used the 0.5% Triton X-100 permeabilized cells for 10 min, and washed with PBS. Finally, the cells were blocked with 5% Bovine Serum Albumin (BSA) or 2% FBS for 20 minutes at RT. We used the primary antibody rabbit anti-E-cadherin or mouse anti-vimentin (1:100; Abcam); rabbit anti-CK5/8 (1:100; Cell Signaling Technology); and mouse anti-α-SMA and anti-Snail (1:100; Abcam). Cells were incubated overnight at 4°C. Next day, we used PE-labeled secondary antibody incubated with the cells (1:1000; Cell Signaling Technology) for 2 h or the cells were incubated with DAPI (Sigma) for 20 min at RT, and washed twice by PBS. Finally, the cells were observed by fluorescence microscope (PerkinElmer Life Sciences).

Western blotting assay

The cells were harvested, washed twice with cold PBS, and the total proteins were prepared from lysed cells using cell lysis buffer (Amresco, Cochran, GA) (50 mM Tris-HCl [pH 8.0], 150 mM NaCl, 0.5% sodium deoxycholate (SDS), 1% NP-40, 1× protease inhibitor cocktail, 1 mM EDTA) on ice, and supplemented with a protease inhibitor cocktail tablet (Roche, USA). We use the BCA protein assay kit detected protein concentration (Thermo fisher, USA). The cell lysates were sonicated, and resolved by 10%-12% SDS-PAGE Gels were transferred to PVDF membranes (Millipore, USA), blocked with 6% BSA, and probed using Abs directed against anti-E-cadherin and anti-vimentin (1:3,000; Cell Signaling Technology), rabbit anti-CK5/8, mouse anti-alpha-SMA, anti-Snail (1:4,000; Cell Signaling Technology), rabbit anti-p-Smad3 (1:1000; Abcam, USA) and GAPDH (1:1000; Abcam, USA). Secondary Abs conjugated to HRP that were used for detection included anti-rabbit (1:5,000) and anti-mouse (1:5,000; Cell Signaling Technology, Beverly, MA). In the end, the protein signals were detected by chemiluminiscence systems (GE, USA), and the western blot figure was quantified using PS software.

Real-time PCR

Total RNA of cells was extracted by the RNeasy extraction kit (Qiagen, Valencia, CA), conferring to the manufacturer’s instructions. CDNAs were synthesized using Reverse Transcriptase kit (Thermofisher, USA), the genes of smad3 were detected by qPCR with SYBR® Green RT-PCR Master Mix (Roche, Switzerland) using 7500 Fast (ABI, USA). The PCR amplification process (95°C, 30 s; and 60°C, 1 min) was done for 40 cycles. The levels of gene expression relative to GAPDH were calculated with Delta-Delta-Ct (ddCt) algorithm method. All of the primers were quality checked for a single curve on melting curve, and efficiencies were 90-110%. Smad3, Forward, 5-GAGTAGAGACGCCAGTTCTACC-3, Reverse, 5-GGTTTGGAGAACCTGCGTCCAT-3; GAPDH, Forward, 5-ATCTTCTCGAACCCCGAGTGA, Reverse, CGGTT CAGCC ACTGG AGCTT.

Transwell migration assays in vitro

The BPH-1 cells were grown up to ~ 80% density and were treated with WSCM of covering with/without anti-HIF-1α in the 24-well transwell and cultured with ordinary but FBS-free medium for 24 h. The BPH-1 cells were collected and carried through cell migration and transference assay after 24 h. In both experiments, BPH-1 cells cultured with/without WSCM were added to the lower chamber and then the cells were migrated the upper chamber. After 24 h of incubation, the migration of cells was measured through the pores using calculating under the microscope (Zeiss, Germany) using a 20× microscope.

The MTT assay

To detect the cell viability/proliferation curves when the cells were treated with inhibitor, BPH-1 cells were plated into 24-well plates with 5000 cells/well and grown in RPMI-1640 media or WSCM Cells. Survival was detected by MTT assay. On this measurement day, medium was carefully changed to MTT and set at 37°C for 3-5 h. After removing MTT mediums, and 200 μl solution of DMSO was added. MTT reduction was detected absorbance at 570 nm using the EThermo Scientific Microplate Reader (BioTek, USA).

Statistical analyses

Two groups data were compared using Student’s t test, and P < 0.05 was considered significant. All of the statistical significance calculations were implemented in Graghpad prism 5 software.

Results

EMT was induced by WSCM in BPH-1 cells

Noncancerous prostatic cell WPMY-1 showed a spindle shaped myofibroblast morphology. As previously researched [26], BPH-1 cells were cultured by the WPMY-1 cell culture supernatant in vitro. In Figure 1, BPH-1 cells were cultured in ordinary medium (left panel), which show the representative feature of epithelial cells; however, BPH-1 cells that were maintained with WSCM medium, which presented a spindle-shaped morphology (right panel), and BPH-1 cells had a typical spindle shaped morphology (middle panel) after maintenance in WSCM 5 days, which is similar to the prostate normal WPMY-1 (Figure 1). To be sure of this phenotype, we detected E-cadherin and CK5/8 of the epithelial markers using western blot, and we found the protein levels of E-cadherin and CK5/8 expression were decresedin BPH-1 cells with WSCM medium in comparison to the control medium (Figure 1B, 1C). Notably, we also respectively detected the mesenchymal markers vimentin and α-SMA or the EMT marker Snail expression in the BPH-1 cell with WSCM medium. We found that the expression of E-cadherin and CK5/8 were decreased in WSCM cultured BPH-1 cells, but the BPH-1 cells unexpectedly acquired Vimentin, α,-SMA and Snail expression by Immunofluorescent staining (Figure 1D). Migration assay showed that the BPH-1 cells’ migratory capacity was also up-regulated by treatment with WSCM medium (Figure 1E, right panel). Therefore, to further ensure the role of WSCM in BPH-1 cell proliferation, we implemented the MTT assay. As in Figure 1F, we did not detect a difference in growth rates with ordinary medium or WSCM medium cultured the BPH-1 cells, respectively. From the above results, we say that EMT can be induced in BPH-1 cells with WSCM medium.

Figure 1.

EMT was induced by WSCM in BPH-1 cells. BPH-1 was cultured in ordinary RPMI 1640 medium and WSCM-BPH-1 was cultured WPMY-1 supernatant. WPMY-1 cells were photographed as a control. A. Cells were photographed after cultured in different media for 5 days. B, C. Western blot analysis of EMT-related proteins in BPH cells after 7 days’ culture period and relative protein levels to GAPDH. D, E. Migration assay results showed WSCM can promote BPH-1 cell migration. F. MTT assay shows the growth of BPH-1 did not alter cells after 1, 3 and 5 days of WSCM culture.

Anti-HIF-1α inhibit the WSCM-induced EMT

HIF-1α is a transcription factor and belongs to Hypoxia-inducible factor 1 (HIF-1) family, and previous research reported that HIF-1α was an EMT inducer in diverse normal or cancer cell types [27,28]. To detect whether HIF-1α is referred to changing of the EMT markers expression in BHP-1 cells, we evaluated its transcription factors’ change on the WSCM-induced EMT by adding anti-HIF-1α antibody with WSCM medium. As shown in Figure 2, BPH-1 cells had a spindle form in WSCM media; however, this morphology was reversed by anti- HIF-1α treatment to culture medium. To further determine anti-HIF-1α role in WSCM-induced EMT, we carried out western blotting, and we found the E-cadherin and CK5/8 expression were strongly enhanced after treatment with anti-HIF-1α. Vimentin, α-SMA and Snail protein expression were decreased by western blot. In addition to this, we got similar results from immunofluorescence, when the BPH-1 cells was added to the HIF-1α antibody. Results showed that vimentin, α-SMA and Snail were inhibited, but E-cadherin and CK5/8 were upregulated (Figure 2D). In the end, we also proved that anti-HIF-1α could reverse the WSCM induced the BPH-1 migratory potential by using a migration assay (Figure 2E). These results with anti-HIF-1α indicated that it may play a role in WSCM cultured BPH-1 cells.

Figure 2.

Anti-HIF-1α inhibits the WSCM-induced EMT. A. Morphologic changes under different culture conditions. B, C. Western blotting analysis of EMT marker protein expression in different conditions and relative protein levels were scanned and relative density of each band was calculated as ratio to GAPDH. D, E. Migration assay. Anti-TGF-b1 Ab was added into the culture system and we detected the migration capability of BPH-1 cells induced by WSCM.

HIF-1α affects its downstream mediator Smad3

Smad3, as a HIF-1α signal-pathway receptor protein, has a significant role in the WSCM-induced EMT in BPH-1 cells. Hence, we investigated the function of Smad3 in the BPH-1 cells. To further investigate the interaction between HIF-1α and Smad3, we performed the qPCR and western blotting to determine the expression of Smad3 in Figure 3. As shown in Figure 3A, we found Smad3 was phosphorylated and was up-regulated in WSCM medium cultured BPH-1 cells by comparison to the ordinary control medium. However, when we added HIF-1α antibody into WSCM-cultured BPH-1 cells, we found that the embellishment of p-Smad3 was decreased after 7 days, and the gene of Smad3 expression was increased in mRNA level in WSCM medium cultured BPH-1 cells. We next found that if we add anti- HIF-1α antibody into BPH-1 cells, this phenotype will be partly reversed (Figure 3B). These data further show that HIF-1α may play an important role in WSCM-induced EMT in BPH-1 cells, and HIF-1α may be coordinated by a Smad3 signal pathway.

Figure 3.

HIF-1α affects its downstream mediator Smad3. A. Western blot analysis of protein level of phosphorylated-Smad3 after 1 day treatment under different culture conditions. B. RT-PCR analysis of total Smad3 mRNA level of after 7 days’ treatment under different culture conditions. C. Smad3 expression in the cells treated with 5 ng/ml or 10 ng/ml HIF-1α.

Discussion

BPH is a well known age-related disease, in men over sixty years old [29,30]. The main clinical manifestation of BPH is LUTS, which include increased frequency of urination, weak urine stream, urgency, hesitancy and nocturia [31,32]. Thus, BPH reduces life quality. Several processes are involved in BPH, and the EMT process of prostate epithelial cells plays a very crucial role in BPH [33]. Numerous factors contribute to the EMT of prostate epithelial cells, and among these complicated regulators, the microenvironment provided by stromal cells is an extremely important one [7,34]. It is already known that stromal cells play a key role in age-related tissue remodeling [35]. Compared to young men, older men have more reactive stromal cells that are characterized by up-regulation of the amount of myofibroblasts and loss of smooth muscle cells. Since myofibroblasts have more active function of secreting aberrant factors than normal fibroblasts this induces changes in nearby epithelial cells [36,37]. Hence, these years, studies of microenvironment in BPH have attracted more and more attention. In our research, we used a BPH-1 cell line, derived from an aged male, and it has been used for the similar studies of microenvironment in BPH.

The HIF-1α and VEGF were highly expressed with prostate hyperplasia; prostate epithelial rats cells, on the contrary, rarely have HIF-1α expressed in normal environments [38-40]. Plenty of previous reports showed the conclusion that HIF-1α has an active effect on the process of prostate hyperplasia; therefore, HIF-1α is identified as an underlying therapeutic target for BPH [38,39]. Until now, numerous molecules are prepared for clinical use for inhibiting HIF-1α [41]. However, a large proportion of HIF-1α inhibitors perform their functions by targeting genes that regulate HIF-1α expression. In addition, a HIF-1α specific inhibitor has not been developed; thus, up to now, targeted therapy against HIF-1α has not been established. Therefore, it is meaningful to establish a specific method to impair the function of HIF-1α.

After treatment with WSCM, BPH-1 cells presented a stretched contour and the markers of epithelia, namely E-cadherin and CK5/8, were decreased (Figure 1). Additionally, the WSCM-treated BPH-1 cells expressed α-SMA, and Vimentin, both mesenchymal markers, and the Snail EMT marker (Figure 1). Snail is a zinc-finger transcription factor and is an EMT inducer. Snail can bind to an E-cadherin promoter E-box and represses its transcription; additionally, Snail also suppress endocytosis of transmembrane E-cadherin. In cancer cells, the high expression of Snail causes impairment of E-cadherin function and often induces EMT. In our results, we found that the microenvironment provided by stromal cells could boost EMT in BPH-1 cells by secreting EMT factors. And we discovered that WSCM could boost the BPH-1 cell migratory potential by migration assay (Figure 1). We also found that WSCM influence on the proliferation of BPH-1 cells was unchanged by the MTT results (Figure 1).

To examine the function of HIF-1α in stromal microenvironment-induced EMT in BPH cells, we neutralized HIF-1α by treating WSCM culture medium with HIF-1α antibody. Then we found that anti-HIF-1α treatment partially reversed morphologic changes induced by WSCM in comparison with the control medium. Inaddition, CK5/8 and E-cadherin expression were markedly increased; on the contrary, the expression of vimentin, α-SMA and Snail declined. Those results also can be supported by migration assays. Anti-HIF-1α treatment reversed the migratory potential of BPH cells induced by WSCM.

Conclusions

In conclusion, our research shows that HIF-1α may play a significant role in the stromal microenvironment-induced EMT process in BPH-1 cells; in addition, we established a specific method to suppress the function of HIF-1α. Therefore, we found a novel therapeutic target for BPH.

Disclosure of conflict of interest

None.