Gastrin-releasing peptide receptor silencing suppresses the tumorigenesis and metastatic potential of neuroblastoma

Abstract

Neuroblastoma accounts for nearly 15% of all pediatric cancer-related deaths. We have previously shown that gastrin-releasing peptide (GRP) stimulates neuroblastoma growth, and that its cell surface receptor, GRP-R, is overexpressed in advanced-stage human neuroblastomas; however, the effects of GRP/GRP-R on tumorigenesis and metastasis in vivo are not clearly elucidated. In the present study, we found that GRP-R knockdown in the aggressive cell line BE(2)-C induced cell morphology changes, reduced cell size, decreased cell proliferation, and inhibited DNA synthesis, corresponding to cell cycle arrest at G(2)/M phase. Activated Akt, a crucial regulator of cell survival and metastasis, was down-regulated by GRP-R silencing. In addition, expression of p-p70S6K and its downstream target molecule S6, key regulators of protein synthesis and cell metabolism, were also significantly decreased by GRP-R silencing. GRP-R knockdown also up-regulated the expression of tumor suppressor PTEN, the inhibitor of the PI3K/Akt pathway. Furthermore, silencing GRP-R as well as GRP in BE(2)-C cells suppressed anchorage-independent growth in vitro. Conversely, overexpression of GRP-R in less aggressive SK-N-SH neuroblastoma cells resulted in soft agar colony formation, which was inhibited by a GRP-blocking antibody. Moreover, GRP-R deficiency significantly delayed tumor growth and diminished liver metastases in vivo. Our findings demonstrate that GRP and GRP-R have important oncogenic properties beyond their established mitogenic functions. Therefore, GRP-R may be an ideal therapeutic target for the treatment of aggressive neuroblastomas.

Fig. 1.

GRP-R silencing inhibits BE(2)-C cell proliferation and down-regulates the PI3-K/Akt pathway. (A) BE(2)-C cells expressing either shGRP-R or shCON were plated 1 × 10⁴ cells per well and cell proliferation was measured (mean ± SEM; *, P < 0.004 vs. shCON). (B) DNA synthesis was analyzed by measuring the BrdU incorporation (mean ± SEM; *, P < 0.0001 vs. shCON). (C) Stably transfected cells (1 × 10⁶ cells/well) were plated and analyzed for the percentage of cells in different cell cycle phases. (D) Western blot analysis was performed with the indicated antibodies in BE(2)-C cells after stable transfection. (E) GRP-R knockdown was confirmed with RT-PCR and Western blot analyses.

Fig. 2.

GRP-R silencing induces changes in cell morphology, reduces cell size and decreases p-p70S6K and S6 expression. (A) The morphology of BE(2)-C cells transfected with either shGRP-R or shCON were revealed by phase contrast microscopy (magnification 100×). (B) The size of cells expressing shGRP-R and shCON was determined as described in Materials and Methods. The average size is represented as the mean FSC-H ± SD of the counted cells (*, P < 0.0001 vs. shCON). (C) Western blot analysis of p-p70S6K, p-S6, and S6.

Fig. 3.

Knockdown of GRP-R or GRP inhibits soft agar colony formation. (A) BE(2)-C cells expressing either shGRP-R or shCON were plated in soft agar (2.5 × 10³ cells per well) for 3 weeks, and colony formation was quantitatively assessed (*, P < 0.0001 vs. shCON). (B) BE(2)-C cells were transfected with siGRP-R or siNTC for 48 h and then plated in soft agar (2.5 × 10³ cells/well) for 3 weeks (live cells, magnification 40×; Left). Bar graph represents the quantitative assessment (*, P < 0.05 vs. siNTC; Center). Western blotting confirmed inhibition of GRP-R protein levels by siRNA (Right). (C) BE(2)-C cells were transfected with siGRP or siNTC for 48 h, incubated in soft agar at 2.5 × 10³ cells/well for 3 weeks, and then photographed after staining (Left). Bar graph represents the quantitative assessment of colony growth (*, P < 0.05 vs. siNTC; Center). The knockdown of GRP mRNA by siRNA was confirmed with RT-PCR (Right).

Fig. 4.

Constitutive expressions of GRP-R correlate to anchorage-independent growth in human neuroblastoma cells. (A) BE(2)-C and SK-N-SH cells were incubated in soft agar at 2.5 × 10³ cells/well in a six-well plate for 3 weeks and 5 × 10³ cells per well for 4 weeks, respectively, and then photographed after staining (Left). Western blot analysis of endogenous levels of GRP-R in BE(2)-C and SK-N-SH cells (Right). (B) Soft agar analysis of SK-N-SH cells stably transfected to overexpress GFP or GFP-tagged GRP-R; arrows indicate colonies (live cells, magnification 40×; Left). Quantitative analysis of soft agar assay (*, P < 0.05 vs. GFP control cells; Center). GFP and GFP-tagged GRP-R were expressed in SK-N-SH cells stably transfected with pEGFP and pEGFP-GRP-R plasmids (Right). (C) GRP-R stably transfected SK-N-SH cells (5 × 10³ cells per well) were incubated in soft agar for 4 weeks. GRP-neutralizing antibody (1 ng/ml) was applied to the top of the soft agar and colonies were photographed after staining (Left). Quantitative analysis of soft agar assay (*, P < 0.05 vs. without antibody; Right).

Fig. 5.

GRP-R silencing blocks tumorigenesis of BE(2)-C cells in vivo. (A) BE(2)-C cells stably transduced with shGRP-R or shCON were grown in subscapular region of athymic nude mice. Pictures were taken at 23 and 61 days postinjection for shCON group (Upper) and shGRP-R group (Lower), respectively. (B) Tumor volumes of transduced BE(2)-C xenografts of mice killed on day 23 (shCON) or 61 (shGRP-R) postinjection (n = 3 per group (*, P < 0.05 vs. shCON; †, P < 0.05 vs. shCON, day 1). (C) Tumor volumes (Left) of transduced xenografts of mice killed on day 19 (n = 5 per group; *, P < 0.05 vs. shCON; †, P < 0.05 vs. shCON, day 1). Tumor weights (Right) of mice killed on day 19 (n = 5 per group; *, P < 0.05 vs. shCON).

Fig. 6.

Knockdown of GRP-R inhibits tumor cell metastasis in vivo. (A) Representative GFP and gross images of primary tumor (arrowheads) in spleen and secondary liver metastases (arrows) from mice injected with BE(2)-C/GFP/shCON (Upper) and BE(2)-C/GFP/shGRP-R (Lower) cells. (B) Representative images of livers from mice injected with BE(2)-C/GFP/shGRP-R and BE(2)-C/GFP/shCON cells (Upper). Quantitative analysis of liver weight relative to body weight (*, P < 0.05 vs. shCON; Lower). (C) Representative H&E-stained sections of livers from mice injected with BE(2)-C/GFP/shCON (Upper) and BE(2)-C/GFP/shGRP-R (Lower) cells.