A reduction in Pten tumor suppressor activity promotes ErbB-2-induced mouse prostate adenocarcinoma formation through the activation of signaling cascades downstream of PDK1

Abstract

Loss of function at the Pten tumor-suppressor locus is a common genetic modification found in human prostate cancer. While recent in vivo and in vitro data support an important role of aberrant ErbB-2 signaling to clinically relevant prostate target genes, such as cyclin D1, the role of Pten in ErbB-2-induced prostate epithelial proliferation is not well understood. In the Pten-deficient prostate cancer cell line, LNCaP, restoration of Pten was able to inhibit ErbB-2- and heregulin-induced cell cycle progression, as well as cyclin D1 protein levels and promoter activity. Previously, we established that probasin-driven ErbB-2 transgenic mice presented with high-grade prostate intraepithelial neoplasia and increased nuclear cyclin D1 levels. We show that mono-allelic loss of pten in the probasin-driven-ErbB-2 model resulted in increased nuclear cyclin D1 and proliferating cell nuclear antigen levels and decreased disease latency compared to either individual genetic model and, unlike the probasin-driven-ErbB-2 mice, progression to adenocarcinoma. Activated 3-phosphoinositide-dependent protein kinase-1 was observed during cancer initiation combined with the activation of p70S6K (phospho-T389) and inactivation of the 4E-binding protein-1 (phosphorylated on T37/46) and was primarily restricted to those cases of prostate cancer that had progressed to adenocarcinoma. Activation of mTOR was not seen. Our data demonstrates that Pten functions downstream of ErbB-2 to restrict prostate epithelial transformation by blocking full activation of the PDK1 signaling cascade.

Figure 1

Pathology of dorsolateral prostate (DLP) and ventral prostate (VP) sections from (A, DLP) non-transgenic and (B, VP; C, VP) genetically modified mouse models.

Figure 2

Pten regulates ErbB-2-induced proliferation markers in vivo and in vitro. A: Immunohistochemical staining for either cyclin D1 (left) or PCNA (right), performed on normal non-transgenic or PB-ErbB-2 × pten^+/− PCa ventral prostate tissue. B: Analysis of the effect of Pten rescue on activated ErbB-2 signaling to the cyclin D1 promoter in the Pten deficient cell line, LNCaP. The average fold change (±SD) in promoter activity versus CMV control transfections for N ≥ 3 separate experiments is shown *P ≤ 0.05, **P ≤ 0.01.

Figure 3

Pten levels are reduced but not eliminated in adenocarcinomas. A: Bright field image (top) and unmixed pseudo-colored (bottom) images of Pten staining performed on normal and cancerous ventral prostate tissue. Blue staining in the unmixed images represent nuclei, red is Pten staining. B: Multispectral imaging data showing the individual spectral profiles of DAB and hematoxylin. C: Average (±SD) Pten signal obtained using Nuance spectroscopy. PB-Cre × pten^PC1 PCa tissue was used as a negative control for Pten staining.

Figure 4

PDK1 levels are increased during PCa initiation and progression. A: Total PDK1 and (B) phospho-PDK1 immunostaining of non-transgenic and PB-ErbB-2 × Pten^+/− DLP (B top and bottom right) and VP (B bottom left) sections.

Figure 5

p70^S6K and 4E-BP1 immunostaining. Left panels, phospho-p70^S6K staining in (A) Normal tissue from non-transgenic prostate. B: Normal or (C) PB-ErbB-2 PIN. D: PCa lesions from PB-ErbB-2 × pten^+/− mice. Right panels, phospho-4E-BP1 staining in (E) normal tissue from non-transgenic mice, (F) normal, or (G) PB-ErbB-2PIN. H: PCa lesions from PB-ErbB-2 × pten^+/− prostate. Both DLP (D, E and H) and VP (A, B, C, F, and G) sections are shown.

Figure 6

mTOR activity in PCa. A, B: Immunostaining adenocarcinomas of the DLP (A, B, C, and D) and VP showing lack of mTOR activity. Immunostaining for phospho-mTOR in (C) mouse pheochromocytoma and (D) PB-Cre × pten^PC1 DLP PCa tissue are shown as a positive controls.

Figure 7

Pathways of ErbB-2-induced signaling in human PCa cells. A: Short-term (30 minutes) effect of chemical inhibitors on protein phosphorylation by Western blotting. B: Effect of prolonged exposure (16 hours) to inhibitors on cell cycle progression in randomly cycling prostate caner cells. Data are average ±SD ≥3 separate experiments *P < 0.05, **P < 0.01.

Figure 8

Mechanisms of prevention of growth factor-induced prostate cancer by Pten, and loss there of, in vivo. In the normal prostate epithelium, expression of ErbB-2 induces proliferation and hypertrophy, which over time results in PIN. Transformation is blocked in part through an inhibition of signaling downstream of PDK1. Since Pten functions in part to inhibit signaling downstream of ErbB-2, a reduction in Pten anti-tumor surveillance function allows for engaged PI3Kinase signaling by ErbB-2, inducing PDK1 and p70S6K signaling and inhibiting 4E-BP1, thereby driving transformation. We propose therefore that p70S6K and 4E-BP1 are key prostate-disease-inducing proteins whose activities are highly sensitive to modest changes in Pten. Phosphorylated proteins associated with either (*) PIN induction or (**) PCa transformation.