BMP suppresses PTEN expression via RAS/ERK signaling

Abstract

Bone morphogenetic protein (BMP), a member of the transforming growth factor beta family, classically utilizes the SMAD signaling pathway for its growth suppressive effects,and loss of this signaling cascade may accelerate cell growth. In the colon cancer predisposition syndrome Juvenile Polyposis, as well as in the late progression stages of nonsyndromic colorectal cancers, SMAD4 function is typically abrogated. Here, we utilized the SMAD4-null SW480 colon cancer cell line to examine BMPs effect on a potential target gene, PTEN, and how its expression might be regulated. Initial treatment of the SMAD4-null cells with BMP resulted in mild growth suppression, but with prolonged exposure to BMP, the cells become growth stimulatory, which coincided with observed decreases in transcription and translation of PTEN, and with corresponding increases in phospho-AKT protein levels. BMP-induced PTEN suppression was mediated via the RAS/ERK pathway, as pharmacologic inhibition of RAS/ERK, or interference with protein function in the cytosol by DN-RAS prevented BMP-induced growth promotion and changes in PTEN levels, as did treatment with noggin, a BMP ligand inhibitor. Thus, BMP downregulates PTEN via RAS/ERK in a SMAD4-null environment that contributes to cell growth, and constitutes a SMAD4-independent but BMP-responsive signaling pathway.

Figure 1

BMP treatment induces a biphasic growth pattern in SW480 cells over eight days of culture. Cells were treated with 100 ng/mL of BMP2 with or without PD98059 (25μM), and cells were counted. Results are expressed as percent change in growth with BMP2 treatment compared to untreated controls. The MEK1/2 inhibitor PD980059 blocked the growth proliferative capability of BMP2 by day eight of growth. Results are representative of three independent experiments.

Figure 2

BMP2 reduces PTEN mRNA in SW480 cells at 36 hours after treatment, which is blocked by noggin or ERK inhibition. (A) Semi-quantitative RT-PCR of PTEN after 36 and 48 hours of BMP2 treatment, with relative amounts of PTEN:GAPDH shown in the bar graph. (B) Semi-quantitative RT-PCR of PTEN was performed in the presence or absence of the BMP ligand inhibitor noggin or RAS/ERK inhibition with PD98059 (25 μM). GAPDH was used as a loading control. Results are representative of three independent experiments.

Figure 3

BMP2 reduces the transcription of PTEN in SW480 cells at 36 hours post-treatment, which is reversed with ERK inhibition. The luciferase construct, PTEN-luc, was transfected into SW480 cells, and cells were treated with or without BMP2 and/or PD98059 for 36 hours, and compared to untreated controls. Luciferase activity was normalized to Renilla luciferase activity as detected by the cotransfected pRL-TK vector. Results are representative of three independent experiments.

Figure 4

BMP suppresses PTEN protein in SW480 cells after 84 hours of treatment, which is blocked with ERK inhibition. (A) Representative Western blots of PTEN, phosphor-Akt, total Akt, and GAPDH, which was used as a loading control. Cells were treated with increasing dosages of BMP2 (0 to 200 ng/mL), with or without PD98059. (B) Densitometry of PTEN from Western blot without PD98059 treatment. (C) Densitometry of phosphor-Akt from Western blot without PD98059 treatment. Results are representative of 3 independent experiments.

Figure 5

RAS inhibition blocks BMP-induced PTEN suppression in SW480 cells after 84 hours of treatment. Representative Western blots of PTEN, phospho Akt, total Akt, phosphor-ERK, total ERK, and GAPDH (used as a loading control) from SW480 cells treated with or without BMP2, and with transfection of mock or DN K-RAS vector. Note the lack of PTEN suppression by BMP when DN K-RAS is present, and the corresponding lack of phosphor-Akt changes that are observed when the mock vector is transfected. Results are representative of 3 independent experiments.