Cholesterol homeostasis in two commonly used human prostate cancer cell-lines, LNCaP and PC-3

Abstract

Background: Recently, there has been renewed interest in the link between cholesterol and prostate cancer. It has been previously reported that in vitro, prostate cancer cells lack sterol-mediated feedback regulation of the major transcription factor in cholesterol homeostasis, sterol-regulatory element binding protein 2 (SREBP-2). This could explain the accumulation of cholesterol observed in clinical prostate cancers. Consequently, perturbed feedback regulation to increased sterol levels has become a pervasive concept in the prostate cancer setting. Here, we aimed to explore this in greater depth.

Methodology/principal findings: After altering the cellular cholesterol status in LNCaP and PC-3 prostate cancer cells, we examined SREBP-2 processing, downstream effects on promoter activity and expression of SREBP-2 target genes, and functional activity (low-density lipoprotein uptake, cholesterol synthesis). In doing so, we observed that LNCaP and PC-3 cells were sensitive to increased sterol levels. In contrast, lowering cholesterol levels via statin treatment generated a greater response in LNCaP cells than PC-3 cells. This highlighted an important difference between these cell-lines: basal SREBP-2 activity appeared to be higher in PC-3 cells, reducing sensitivity to decreased cholesterol levels.

Conclusion/significance: Thus, prostate cancer cells are sensitive to changing sterol levels in vitro, but the extent of this regulation differs between prostate cancer cell-lines. These results shed new light on the regulation of cholesterol metabolism in two commonly used prostate cancer cell-lines, and emphasize the importance of establishing whether or not cholesterol homeostasis is perturbed in prostate cancer in vivo.

Figure 1. Sterol-mediated regulation of SREBP-2 target genes exists in PCa cells.

Cells were treated with the statin compactin (CPN, 5 µM), oxysterol 25-HC (1 µg/ml), or LDL (50 µg/ml) for 24 h. Cellular RNA was harvested and mRNA expression of the LDLR and HMGCR genes was analyzed by qRT-PCR. The mRNA levels were made relative to the vehicle condition as described in Materials and Methods. Data are mean + SEM, from 3 separate experiments for each cell-line. Each experiment was performed with triplicate wells per condition. (A) Data presented separately for each cell-line. * p<0.05, ** p<0.01, two-sample t-test versus vehicle condition. (B) Data from (A) has been overlaid for each gene, represented as mean±SEM for each datapoint. The PCa cell-lines are represented by solid lines, whilst the non-PCa cell-lines are represented by broken lines.

Figure 2. Responses to changing sterol status involve SREBP-2 activation in prostate cancer cells.

(A) Cells were transfected as described in Materials and Methods. Treatment included the statin compactin (CPN, 5 µM), oxysterol 25-HC (1 µg/ml), or LDL (50 µg/ml) for 24 h. SRE-specific luciferase activity was determined as described in Materials and Methods, and normalized to the vehicle condition. The wildtype and mutant promoter values are shown in Fig. S1. Data are mean + SEM, from 3 separate experiments for each cell-line. Each experiment was performed with triplicate wells per condition. * p<0.05, ** p<0.01, two-sample t-test versus vehicle condition. (B) Data from (A) has been overlaid, represented as mean±SEM for each datapoint. The PCa cell-lines are represented by solid lines, whilst the non-PCa cell-lines are represented by broken lines. (C) Cells were treated with CPN (5 µM) or 25-HC (1 µg/ml) for 24 h. Cell lysates were subjected to SDS-PAGE and Western blotted with the IgG-1C6 anti-SREBP-2 antibody. The C-terminal cleavage product of SREBP-2, SREBP-2(C), is labeled with an arrow – we assume that the band below is a non-specific band. Probing for α-tubulin served as an internal loading control. The blot shown is representative of at least 2 separate experiments for each cell-line.

Figure 3. Sterol feedback regulation has functional effects in PCa cells.

Cells were treated with the statin compactin (CPN, 5 µM) or the oxysterol 25-HC (1 µg/ml) for 24 h in Medium C. (A) Cells were prepared, treated, and assayed for DiI-LDL internalization as described in Materials and Methods. The amount of DiI-LDL internalized provides an indication of LDLR activity. Data are presented as mean + SEM, from at least 3 separate experiments for each cell-line. Each experiment was performed with triplicate wells per condition. (B) After treatment, cells were washed with PBS and radiolabeled with [1-¹⁴C]-acetic acid for 2 h. Radiolabeling was performed in the presence of treatment, with the exception of the CPN treatment (in which case the CPN was absent). Cells were then harvested and lipid extracts were subjected to thin layer chromatography and phosphorimaging as described in Materials and Methods. The phosphorimages shown are representative of at least 3 separate experiments for each cell-line. Densitometry was performed and data presented as mean+SEM for each cell-line. * p<0.05, ** p<0.01, two-sample t-test versus vehicle condition.

Figure 4. SRE-specific activity is saturated under lipoprotein-deficient conditions in PC-3 cells.

PC-3 cells were transfected as described in Materials and Methods. Treatment included the statin compactin (CPN, 5 µM) or oxysterol 25-HC (1 µg/ml) for 24 h, in either full-serum (Medium A) or lipoprotein-deficient serum (Medium C). SRE-specific luciferase activity was determined as described in Materials and Methods, and normalized to the vehicle condition. Data are mean + SD, representative of 2 separate experiments. Each experiment was performed with triplicate wells per condition.

Figure 5. Basal LDLR gene expression and activity is higher in PC-3 than LNCaP cells.

Data was pooled from experiments where LNCaP and PC-3 cells received vehicle treatment in Medium C for 24 h. (A) For each qRT-PCR experiment, the ΔC_t values (for threshold = 10^−1.5 normalized fluorescence units) for the LDLR gene, relative to the PBGD housekeeping gene, were considered. Expression is represented as a fold change (relative to PC-3 cells), whereby a one-unit increase in ΔC_t results in a two-fold decrease in mRNA expression. Data are presented as mean + SEM, from at least 4 separate experiments for each cell-line. (B) Raw LDL uptake, measured as fluorescence normalised by protein content in the LDL uptake assay, was averaged between experiments and made relative to the PC-3 cells. Data are presented as mean + SEM, from at least 3 separate experiments for each cell-line. (C) For each SRE-specific luciferase assay, the firefly/Renilla luciferase ratios generated from the LDLp-mutSRE plasmid were normalised to that of the LDLp-588luc plasmid. Data presented as mean+SEM, from at least 5 separate experiments for each cell-line. Each experiment in (A)–(C) was performed with triplicate wells per condition. * p<0.05, ** p<0.01, two-sample t-test versus PC-3 cells. (D) A model depicting the differences in cholesterol homeostasis between PC-3 and LNCaP cells. The threshold level of cellular cholesterol, at which SREBP-2 activity becomes dramatically reduced, may be higher in PC-3 cells. This would reduce the effect of statins, relative to the basal condition.