Anti-cancer effect of pristimerin by inhibition of HIF-1α involves the SPHK-1 pathway in hypoxic prostate cancer cells

Hypoxia is a typical character of locally advanced solid tumours. The transcription factor hypoxia-inducible factor 1α (HIF-1α) is the main regulator under the hypoxic environment. HIF-1α regulates various genes to enhance tumour progression, angiogenesis, and metastasis. Sphingosine kinase 1 (SPHK-1) is a modulator of HIF-1α. To investigate the molecular mechanisms of pristimerin in association with SPHK-1 pathways in hypoxic PC-3 cancer cells. Vascular endothelial growth factor (VEGF) production, cell cycles, and SPHK-1 activity were measured, and western blotting, an MTT assay, and an RNA interference assay were performed. Pristimerin inhibited HIF-1α accumulation in a concentration- and-time-dependent manner in hypoxic PC-3 cells. Pristimerin suppressed the expression of HIF-1α by inhibiting SPHK-1. Moreover, inhibiting SPHK-1 with a sphingosine kinase inhibitor enhanced the suppression of HIF-1α, phosphorylation AKT, and glycogen synthase kinase-3β (GSK-3β) by pristimerin under hypoxia. Furthermore, a reactive oxygen species (ROS) scavenger enhanced the inhibition of HIF-1α and SPHK-1 by pristimerin. Taken together, these findings suggest that pristimerin can exert an anti-cancer activity by inhibiting HIF-1α through the SPHK-1 pathway.


Background
Hypoxia is a common characteristic of locally advanced solid tumours [1] and up to 50-60 % of solid tumours include areas of hypoxic tissues [2]. The hypoxic tumour contributes to aggressive and metastatic cancer phenotypes that are associated with resistance to radiation therapy, chemotherapy, and a poor treatment outcome [3,4].
Here, we demonstrate that pristimerin inhibits HIF-1α via the SPHK-1 signaling pathway in a prostate cancer cell lines. The results we have yielded provide the mechanism for inhibitory action of HIF-1α and angiogenesis by pristimerin in hypoxic prostate cancer cell lines.

Western blot analysis
The cells were lysed in RIPA buffer (Cell signaling, USA). The protein extract were separated on SDS-polyacrylamide gels and were electrotransferred to a nitrocellulose membrane (GE healthcare life sciences, UK).

Sphingosine kinase assay
To measure sphingosine kinase activity, sphingosine kinase activity assay kit (Echelon, Salt Lake City, UT, USA) was used. The Sphingosine kinase activity assay method was previously described in our other study [37,38]. Protein extracts (30 μg) were reacted in reaction buffers, 100 μM of sphingosine, and 10 μM of ATP, for 1 h at 37°C , and then to stop the kinase reaction, a luminescence attached ATP detector was added. Lumistar Optima luminometer (BMG LABTECH, Offenburg, Germany) was used to measure kinase activity. All samples were prepared in triplicates and the assay was repeated at least three times.

Measurement of VEGF production
VEGF ELISA kit (Invitrogen, Carlsbad, CA, USA) was used to assess VEGF levels in pristimerin and/or SKI exposed PC-3 cells. The VEGF production level measurement methods was previously described in our other study [39]. The culture supernatants was added in a 96well plate, and reacted with dilution buffer and incubation buffer at room temperature for 2 h. The wells were then washed four times with washing buffer, and then biotin conjugate was added to each well at room temperature for 1 h. After washing, the stabilized chromogen was added into each well and reacted for 30 min at room temperature. The density was measured at 450 nm using a microplate reader (Molecular Devices Co., Sunnyvale, CA, USA) after adding 100 μl of the stop solution.

Cell cycle assay
The cell cycle was determined according to the protocol described previously [40]. Cells were fixed with 75 % ethanol and resuspended in PBS with RNase (1 mg/mL) at 37°C for 1 h and stained with propidium iodide (PI). The stained cells were analyzed for DNA content by FACS Calibur containing Cell-Quest Software (Becton-Dickinson, Heidelberg, Germany).

RNA interference experiments
The siRNA transfection method was previously described in our other study [37,38]. A polyplus siRNA transfection reagent (Illkirch, France) was used to transfect siRNA for the control or SPHK-1 into PC-3 cells. In brief, siRNA (80 pmol) was mixed with a transfection reagent in serum-free media and reacted for 10 min at room temperature. The siRNA/transfection reagent mixture was added to the cells and incubated for 48 h. The medium was changed before the treatment with pristimerin and/or SKI under hypoxia.

Statistical analysis
The data showed as means ± S.D. (standard deviation) of three replications each experiment in this study. Analysis of variance (ANOVA) was used to assess the significance of differences between groups. P <0.05 was considered to indicate statistical significance.

Pristimerin decreases cell viability under hypoxia
To measure whether pristimerin affects cell viability under hypoxic and normoxic conditions, cells were treated with various concentrations of pristimerin in PC-3 cells under hypoxia or normoxia for 24 h. Pristimerin significantly decreased cell viability under hypoxia than it did under normoxia (Fig. 1a). As shown in Fig. 1b and c, pristimerin treatment for 48 h reduced cell growth in hypoxic PC-3 cells. Similar to the 24 h data, pristimerin significantly decreased cell growth under hypoxia more than normoxia.

Pristimerin decreases HIF-1α abundance and VEGF secretion
To examine whether pristimerin inhibits hypoxiainduced HIF-1α, pristimerin was treated into PC-3 cells under hypoxia for 4 h. As shown in Fig. 1d and e, pristimerin decreased HIF-1α abundance. To examine whether hypoxia-induced VEGF secretion is decreased by pristimerin, the VEGF secretion level was measured on a hypoxia-induced PC-3 cell medium, with pristimerin treatment for 24 h. As shown in Fig. 1f, the VEGF secretion level under  (HIF-1α). The results are expressed as means ± SD for the duplicate. * p <0.05 and ** p <0.01 compared with hypoxia control at each time point hypoxia was higher than under normoxia control. Pristimerin reduced the hypoxia-induced VEGF secretion.
Pristimerin exerts significant inhibition of SPHK-1 in hypoxic PC-3 cells To investigate whether pristimerin affects SPHK-1 in PC-3 cells, the cells were incubated under hypoxia for 4 h with 0.5 or 1 μM of pristimerin. Pristimerin at 1 μM reduced SPHK-1 to 55 % under hypoxia compared with the control (Fig. 2a and b). As SPHK-1 is one of the regulators of HIF-1α, the effect of hypoxia was assessed with the HIF-1α expression. Both the SPHK-1 and HIF-1α accumulation reached the peak 4 h after hypoxia exposure and then decreased in a time-dependent manner. The SPHK-1 and HIF-1α expressions were effectively inhibited by pristimerin (Fig. 2c, d and e).

SPHK-1 mediates the activation of HIF-1α under hypoxia
To confirm the involvement of SPHK-1 in the pristimerin-mediated inhibition of HIF-1α during hypoxia, the effects of pristimerin was evaluated by using SPHK-1 siRNA and an SPHK-1 inhibitor, on SPHK-1 activity and the phosphorylation of AKT and GSK-3β. This is because the SPHK-1 dependent stabilization of HIF-1α is known to be mediated by AKT/GSK-3β, downstream of SPHK-1. The phosphorylation of AKT and GSK-3β was induced under hypoxia (Fig. 3a). Pristimerin suppressed the phosphorylation of GSK-3β and AKT in hypoxic PC-3 cells (Fig. 3a). SKI, an SPHK-1 inhibitor, blocked the expression of HIF-1α and the phosphorylation of AKT and GSK-3β (Fig. 3a). The SPHK-1 activity was significantly decreased by pristimerin and SKI (Fig. 3b). Consistently, SPHK-1 siRNA transfection suppressed pristimerin-mediated inhibition of SPHK-1 in PC-3 cells under hypoxia ( Fig. 3c and d). As shown in Fig. 4a, we assessed whether pristimerin suppresses The activity of SPHK-1 in pristimerin treated PC-3 cells. SPHK-1 activity was measured by using SPHK-1 activity kit. Data are presented as means ± SD. * p <0.05 and ** p <0.01 compared with hypoxia control. c PC-3 prostate cancer cells were transfected with control vector or SPHK-1 siRNA for 48 h to decrease the expression of SPHK-1. Then PC-3 cells were treated with 1 μM of pristimerin for 4 h. Western blotting was performed to determine the expression of SPHK-1, HIF-1α, p-AKT, AKT, pGSK-3β, GSK-3β, and β-actin in hypoxic PC-3 cells. d The activity of SPHK-1 in pristimerin treated PC-3 cells. SPHK-1 activity was measured by using SPHK-1 activity kit. Data are presented as means ± SD. * p <0.05 and ** p <0.01 compared with hypoxia control hypoxia-induced HIF-1α and SPHK-1 in several prostate cancer cell lines (PC-3, DU145, and LNCaP). Pristimerin inhibited HIF-1α and the phosphorylation of AKT and GSK-3β in all cell lines tested, which is similar to the results from PC-3 cells (Fig. 4a).

Pristimerin inhibits VEGF production via SPHK-1 inhibition in Hypoxic PC-3 cells
As shown in Fig. 1c, pristimerin significantly reduced VEGF production. To exam the role of SPHK-1 on the secretion of VEGF, an angiogenic factor, PC-3 cells were treated with pristimerin and SKI under hypoxia for 24 h and VEGF levels were then measured by an ELISA and Western blot. VEGF levels elevated significantly in the hypoxia control group while pristimerin and SKI treatment reduced VEGF secretion (Fig. 5a). In addition, combination treatment with pristimerin and SKI significantly diminished VEGF secretion in PC-3 cells under hypoxia (Fig. 5a).

SPHK-1 mediates pristimerin-induced G1 arrest in hypoxia-induced PC-3 cells
As shown in Fig. 1b and c, pristimerin significantly decreased cell viability under hypoxia as opposed to normoxia and decreased cell proliferation. Therefore, the effect of SKI and pristimerin on cell proliferation during hypoxia was evaluated by FACS analysis and western blotting.
PC-3 cells were treated with SKI and pristimerin for 48 h under hypoxic conditions. Treatment with pristimerin and SKI significantly increased G1-arrest and decreased the expression of G1 regulatory proteins, such as cylinD1 and CDK4, in hypoxic PC-3 cells (Fig. 5b and  c). PCNA is essential for DNA replication. The PCNA level under normoxia was similar to that under hypoxia. Combination treatment with pristimerin and SKI reduced PCNA under hypoxia (Fig. 5c).

Discussion
Most solid tumours are more aggressive and resistant to chemotherapy or radiation under hypoxic conditions [37,41]. Hypoxia is a typical characteristic of locally advanced solid tumours [42]. The transcription factor HIF-1α, which targets 60 genes to enhance the tumour progression, angiogenesis, and metastasis, is regarded as the master regulator under the hypoxic environment [12,43]. Our previous study showed that the accumulation of HIF-1α is mediated by the AKT/GSK-3β pathway, and related to HIF-1α stabilization through the activation of SPHK-1 [44]. SPHK-1 is a decisive regulator of this sphingolipid rheostat and as such, a potent therapeutic target for cancer treatment [45,46]. Furthermore, the activity and expression of SPHK-1 are significantly induced under hypoxia and by HIF-1α, and thus is a critical therapeutic target through pVHLdependent proteasomal degradation for cancer treatment [23,[47][48][49]. Pristimerin, a triterpenoid quinone methide compound, is involved in a variety of activities, which includes anti-inflammatory and anti-cancer action [27,[29][30][31][32][33][34][35][36]. A recent study reported that pristimerin suppressed HIF-1α and hypoxia-induced metastasis in prostate cancer PC-3 cells [4]. However, the mechanisms of the inhibition of hypoxia-induced HIF-1α by pristimerin are not fully comprehended. In this study, pristimerin significantly decreased cell viability under hypoxia more than it did under normoxia, which connotes the potential of pristimerin treatmentresistant cancer cells, given that HIF-1α promotes cancer resistance. Our study showed that SPHK-1 and HIF-1α accumulations began to increase after 30 min of hypoxia exposure in PC-3 prostate cancer cells compared with the normoxia, which is consistent with previous studies [37,38]. Moreover, the hypoxia-induced HIF-1α accumulation was suppressed in the presence of pristimerin. In addition, we found that pristimerin suppressed hypoxia-induced SPHK-1. To further confirm the involvement of SPHK-1 in pristimerinmediated inhibition of HIF-1α under hypoxia, we tested the effects of pristimerin on the phosphorylation of AKT and GSK-3β since AKT/GSK-3β is downstream of SPHK-1 and mediates HIF-1α stabilization [23]. Furthermore, co-treatment of pristimerin and SKI suppressed the phosphorylation of AKT and GSK-3β. Likewise, SPHK-1 siRNA transfection suppressed the phosphorylation of AKT and GSK-3β.
Hypoxia leads to an increase in mitochondrial production of ROS, [50] and ROS production is required for hypoxia-mediated HIF stabilization [51][52][53][54]. Several a Cells were treated with pristimerin (1 μM) and/or SPHK-1 inhibitor (SKI) (10 μM) for 24 h under hypoxia. VEGF level was measured by ELISA. Data are presented as means ± SD. ** p <0.01 and *** p <0.001 compared with hypoxia control. Western blotting was performed to determine the expression of VEGF and β-actin in hypoxic PC-3 cells. b Cells were treated with pristimerin (1 μM) and/or SPHK-1 inhibitor (SKI) (10 μM) for 48 h under hypoxia. Cell cycle distribution was analyzed by flow cytometry. Bar graphs represent the percentage of sub-G1, G1, S, and G2-M phase cells. Data represent mean ± SD of three independent experiments. * p <0.05 compared with untreated control. c Western blotting was performed to determine the expression of SPHK-1, PCNA, CyclinD1, CDK4, and β-actin in hypoxic PC-3 cells recent studies have shown that SPHK-1 activity and HIF-1α are stimulated by ROS production [44,55]. In addition, our previous studies showed that N-acetylcysteine (NAC), an ROS scavenger, suppresses HIF-1α by blocking SPHK-1 under hypoxia.
To confirm whether pristimerin suppresses hypoxiainduced HIF-1α accumulation via the inhibition of SPHK-1 and ROS generation in prostate cancer cells, we evaluated the effect of NAC on HIF-1α and SPHK-1 abundance in hypoxic PC-3 cells, treated with pristimerin. The co-treatment of pristimerin with NAC affected HIF-1α and SPHK-1 abundance.
PI3K is necessary for cell growth and survival, and PI3K can be activated by growth factors binding to cell surface receptor and hypoxia. PI3K induces the accumulation, activation, and stabilization of HIF-1α proteins during hypoxia in cancer cells [56]. To confirm whether pristimerin inhibits hypoxia-induced HIF-1α accumulation by the inhibition of PI3K, PC-3 cells were treated with pristimerin and SKI under normoxic and hypoxic conditions for 24 h. PI3K levels did not change (Additional file 1: Figure S1).
There is evidence that HIF-1α can regulate VEGF secretion in cancer cells [57,58]. In the present study, the inhibition of SPHK-1 activity using SKI prevented VEGF production in PC-3 cells. Similarly, studies have demonstrated that SPHK-1 plays a critical role in HIF-1αmediated VEGF secretion under hypoxia [37,38]. Pristimerin significantly inhibited cell proliferation for 48 h (Fig. 1c). It is well known that SPHK-1 mediates cancer cell proliferation and progression. Thus, to confirm the involvement of SPHK-1 in pristimerin-mediated inhibition of cell proliferation, hypoxic PC-3 cells were treated with SKI and pristimerin for 48 h. Interestingly, SKI and pristimerin cotreatment induced G1 arrest and decreased G1 regulatory factors in hypoxic PC-3 cells.

Conclusions
Our study shows that pristimerin inhibits HIF-1α, SPHK-1 expression or activity, and phospho-AKT/GSK-3β and decreases VEGF production in hypoxic PC-3 cells. These results suggest that pristimerin may inhibit HIF-1α accumulation by inactivation of SPHK-1 including the free radical scavenging effect in PC-3 cells under hypoxia.

Additional file
Additional file 1: Figure S1. Pristimerin does not affect PI3K in PC-3 cells under hypoxia. PC-3 cells were treated with pristimerin (1 μM) and or SPHK-1 inhibitor (SKI) (10 μM) for 4 h under hypoxia. Effect of pristimerin on the expression of PI3K in hypoxic PC-3 cells. Western blotting was performed to determine the expression of PI3K and β-actin in hypoxic PC-3 cells. (TIF 66 kb)