Snail transcription factor regulates neuroendocrine differentiation in LNCaP prostate cancer cells

Abstract

Background: Snail transcription factor induces epithelial-mesenchymal transition (EMT) via decreased cell adhesion-associated molecules like E-cadherin, and increased mesenchymal markers like vimentin. We previously established Snail-mediated EMT model utilizing androgen-dependent LNCaP cells. These cells express increased vimentin protein and relocalization of E-cadherin from the cell membrane to the cytosol. Interestingly, Snail transfection in LNCaP cells resulted in cells acquiring a neuroendocrine (NE)-like morphology with long neurite-like processes.

Methods: We tested for expression of NE markers neuron-specific enolase (NSE) and chromogranin A (CgA) by Western blot analysis, and performed proliferation assays to test for paracrine cell proliferation.

Results: LNCaP cells transfected with Snail displayed increase in the NE markers, NSE and CgA as well as translocation of androgen receptor (AR) to the nucleus. LNCaP C-33 cells that have been previously published as a neuroendocrine differentiation (NED) model exhibited increased expression levels of Snail protein as compared to LNCaP parental cells. Functionally, conditioned medium from the LNCaP-Snail transfected cells increased proliferation of parental LNCaP and PC-3 cells, which could be abrogated by NSE/CgA siRNA. Additionally, NED in LNCaP-C33 cells or that induced in parental LNCaP cells by serum starvation could be inhibited by knockdown of Snail with siRNA.

Conclusion: Overall our data provide evidence that Snail transcription factor may promote tumor aggressiveness in the LNCaP cells through multiple processes; induction of EMT may be required to promote migration, while NED may promote tumor proliferation by a paracrine mechanism. Therefore, therapeutic targeting of Snail may prove beneficial in not only abrogating EMT but also NED.

Figure 1. Snail transcription factor induces EMT in LNCaP cells

(A) LNCaP protate cancer cells stably transfected with Snail (Snail1 clone) displayed a neuroendocrine morphology when compared with LNCaP cells transfected with empty vector (LNCaP-Neo1). (B) LNCaP-Neo1 or –Snail1 were plated overnight onto 16-well chamber slide, fixed with methanol/acetone and examined by immunofluorescent analysis for E-cad, Vimentin or secondary antibody alone negative control. Mag. X200. (C) LNCaP prostate cancer cells stably transfected with either empty vector (LNCaP-Neo1) or Snail transcription factor (Snail mix or Snail1 clone) were examined for expression of Snail EMT markers, E-cadherin (E-cad) and vimentin by Western blot analysis. Actin was utilized as a loading control. Quantification of the Western blot results by densitometry with normalization to actin levels was done using the Quantity One quantification software (BioRad). All the experiments were repeated at least three times.

Figure 2. Induction of NED and AR nuclear translocation by Snail transcription factor

(A) Total RNA or cell lysate was isolated from the LNCaP-Neo1 and LNCaP-Snail1 cells, and analyzed by RT-PCR or Western blot using primers or antibodies, respectively, for NED markers Neuron Specific Enolase (NSE), chromogranin A (CgA), as well as AR and PSA. 18SrRNA or actin was used as a loading control. Quantification of the Western blot results by densitometry with normalization to actin levels was done using the Quantity One quantification software (BioRad). (B) Immunofluorescence with specific antibodies revealed increased expression of NSE and CgA and AR translocation to the nucleus in LNCaP-Snail1 cells when compared to LNCaP-Neo1 cells. Mag. X100. All results shown are representative of three independent experiments.

Figure 3. Snail inhibits cell proliferation in transfected LNCaP cells, but conditioned media from Snail-transfected cells can increase cell proliferation of parental LNCaP and PC3 cells by a paracrine mechanism

(A) LNCaP-Neo1 and LNCaP-Snail1 or LNCaP-Neo5 and LNCaP-Snail5 clones were plated onto 96-well plates and MTS assay performed for 0–3 days. Cells transfected with Snail showed decreased cell proliferation. (B) Parental LNCaP or PC3 cells were plated overnight onto 96-well plates, followed by replacement of media with 25% conditioned media from either LNCaP-Neo1 or LNCaP-Snail1 cells. The MTS assay was then performed for 0–3 days. All experiment were done in triplicate and repeated twice independently. Bars, SD *, P < 0.05, Student's t test compared with LNCaP-Neo1 CM.

Figure 4. NED induced in LNCaP cells by androgen deprivation can be abrogated by Snail siRNA

(A) LNCaP cells were plated in 10% FBS overnight then cultured in either 10 % FBS or androgen-deprived media (5% DCC) for 7 days. The cells in DCC were then cultured for an additional 3 days in either 5% DCC plus 200 nM control siRNA or 5% DCC plus 200 nM Snail siRNA. Total RNA and cell lysates were prepared and analyzed by RT-PCR and western blot analyses for Snail, NSE and CgA. GAPDH or actin were utilized as loading controls for RT-PCR or western blot, respectively. (B) Quantification of the Western blot results by densitometry with normalization to actin levels was done using the Quantity One quantification software (BioRad).

Figure 5. NED in the LNCaP-c33 cell model is mediated by Snail transcription factor

(A) LNCaP-C33 cells that are NED cells derived by prolonged androgen deprivation, were analyzed for expression of Snail and the NED markers NSE and CgA utilizing western blot analysis. As compared to LNCaP cells, there was higher expression of Snail, NSE and CgA protein. (B) LNCaP-C33 cells were treated with control of Snail siRNA for 7 days, which resulted in approximately 50% knockdown of Snail, NSE and CgA protein as seen by Western blot analysis. Quantification of the Western blot results by densitometry with normalization to actin levels was done using the Quantity One quantification software (BioRad). The conditioned media from LNCaP-C33 cells treated for 7 days with either control or Snail siRNA was collected and added to (C) parental LNCaP, or (D) PC3 cells whose cell proliferation was monitored for up to 3 days. All experiments were done in triplicate and repeated twice independently. Bars, SD *, P < 0.05, Student's t test compared with control siRNA.

Figure 6. Paracrine cell proliferation induced in LNCaP and PC3 parental cells by LNCaP-Snail conditioned media can be abrogated by NSE and CgA siRNA

(A) LNCaP-Snail5 cells were plated in 10% FBS overnight then transfected with either 200 nM control siRNA or 200 nM NSE and/or CgA siRNA for 7 days. Untreated LNCaP-Neo5 and LNCaP-Snail5 were included as controls. Total cell lysates were prepared and analyzed by western blot analyses for Snail, NSE and CgA. Actin was utilized as a loading control for western blot. (B) Quantification of the Western blot results by densitometry with normalization to actin levels was done using the Quantity One quantification software (BioRad). The conditioned media from untreated LNCaP-Neo5, untreated LNCaP-Snail5 and LNCaP-Snail5 cells treated for 7 days with either control or NSE/CgA siRNA was collected and added to (C) parental LNCaP, or (D) PC3 cells whose cell proliferation was monitored for up to 4 days. All experiments were done in triplicate and repeated twice independently. Bars, SD, P = 0.069 for LNCaP treated with NSE+CgA siRNA CM for 4 days, P = 0.058 for PC3 treated with, NSE+CgACM for 4 days; Student's t test compared with control siRNA CM.