Proinflammatory CXCL12-CXCR4/CXCR7 Signaling Axis Drives Myc-Induced Prostate Cancer in Obese Mice

Abstract

Obesity is a prognostic risk factor in the progression of prostate cancer; however, the molecular mechanisms involved are unclear. In this study, we provide preclinical proof of concept for the role of a proinflammatory CXCL12-CXCR4/CXCR7 signaling axis in an obesity-driven mouse model of myc-induced prostate cancer. Analysis of the stromal vascular fraction from periprostatic white adipose tissue from obese HiMyc mice at 6 months of age revealed a dramatic increase in mRNAs encoding various chemokines, cytokines, growth factors, and angiogenesis mediators, with CXCL12 among the most significantly upregulated genes. Immunofluorescence staining of ventral prostate tissue from obese HiMyc mice revealed high levels of CXCL12 in the stromal compartment as well as high staining for CXCR4 and CXCR7 in the epithelial compartment of tumors. Prostate cancer cell lines derived from HiMyc tumors (HMVP2 and derivative cell lines) displayed increased protein expression of both CXCR4 and CXCR7 compared with protein lysates from a nontumorigenic prostate epithelial cell line (NMVP cells). CXCL12 treatment stimulated migration and invasion of HMVP2 cells but not NMVP cells. These effects of CXCL12 on HMVP2 cells were inhibited by the CXCR4 antagonist AMD3100 as well as knockdown of either CXCR4 or CXCR7. CXCL12 treatment also produced rapid activation of STAT3, NFκB, and MAPK signaling in HMVP2 cells, which was again attenuated by either AMD3100 or knockdown of CXCR4 or CXCR7. Collectively, these data suggest that CXCL12 secreted by stromal cells activates invasiveness of prostate cancer cells and may play a role in driving tumor progression in obesity. Targeting the CXCL12-CXCR4/CXCR7 axis could lead to novel approaches for offsetting the effects of obesity on prostate cancer progression. Cancer Res; 77(18); 5158-68. ©2017 AACR.

Figure 1

Recruitment and infiltration of adipocytes and inflammatory cells in the ventral prostate of HiMyc mice as a function of dietary energy balance manipulation. Immunofluorescence analysis of representative prostate tissue sections from HiMyc mice fed on 30 %CR, control or DIO diets with antibodies against perilipin-1, CD3 (T-lymphocytes) and RM0029-11H3 (macrophages) with red and green fluorophore conjugated secondary antibodies. Representative images of tissue sections from 5 mice per group are shown. All tissue sections from each diet group gave similar results. Scale bar 200 μm (top row), 25 μm (middle and bottom rows).

Figure 2

A, Increased mRNA expression of CXCL12, CXCR4 and CXCR7 in the SVF of obese HiMyc mice. mRNA was isolated from the SVF of adipose tissue surrounding the ventral prostate of HiMyc mice at 6 months of age. Fold change in gene expression in obese mice fed the DIO (60kcal% fat) diet relative to mice fed the control (10kcal% fat) diet is shown. Quantitation of fold change of mRNA is presented as mean ± SEM of three independent experiments. Significantly different, *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001 (Student’s t-test). B,C, Immunofluorescence analysis of representative prostate tissue sections from age-matched HiMyc mice fed on CR, Control or DIO diet with antibodies against CXCL12, α-SMA, CXCR4 and CXCR7 with red and green fluorophore-conjugated secondary antibodies. Shown are representative images from one mouse out of a group of 5 mice. All tissue sections from each diet group gave similar results. Scale bar 100 μm.

Figure 3

Increased CXCR4 and CXCR7 protein expression in cancer cell lines. A, Protein lysates prepared from cultured cells were subjected to Western blotting. B, Relative protein levels were quantitated by densitometry; β-actin was used to control for protein loading. Representative blots of two separate experiments with similar results are shown. Note increased expression in PCa cells compared to normal mouse (NMVP) prostate epithelial cells.

Figure 4

CXCL12 induces migration and invasion of HMVP2 cells. A, B. cells in 12 well plates scratched with 200 μl pipette tips and treated with vehicle control, CXCL12 (100 ng/mL), IL-6 (10 ng/mL), AMD3100 (40 ug/mL) or CXCL12 + AMD3100 at 0 and 18 h post-treatment. Quantitation of migration levels are presented as mean ± SEM of three independent experiments. ***P<0.001; (One Way ANOVA). C, D. NMVP and HMVP2 cells were seeded in Matrigel in the upper transwell chambers and then placed into 24 well plates with medium that contained the vehicle (control), CXCL12 (100 ng/mL), IL-6 (10 ng/mL), AMD3100 (40 ug/ml) or CXCL12 + AMD3100. After 18h, the invading cells in the lower chamber were fixed, stained and visualized by light microscopy. Quantitation of invading cells presented as mean ± SEM of three independent experiments. ***P<0.001; (One Way ANOVA)

Figure 5

CXCR4 and CXCR7 knockdown decreases migration and invasion in HMVP2 cells. Cells (A, transfected with control or CXCR4 shRNA; C, transfected with control or CXCR7 shRNA) in 12 well plates scratched with 200 μl pipette tips and treated with vehicle control or CXCL12 (100 ng/mL). Quantitation of migration levels were presented as mean ± SEM of three independent experiments. ***P<0.001; (Student’s t-test). Cells (B, transfected with control or CXCR4 shRNA; D, transfected with control or CXCR7 shRNA) were seeded in Matrigel in the upper transwell chambers and then placed into 24 well plates with medium that contained the vehicle (control) or CXCL12 (100 ng/mL). After 18h, the invading cells in the lower chamber were fixed, stained and visualized by light microscopy. Quantitation of invading cells were presented as mean ± SEM of three independent experiments. ***P<0.001; (Student’s t-test)

Figure 6

Activation of NFκB, STAT3 and MAPK signaling by CXCL12 in HMVP2 cells. A,B. Protein lysates were prepared from cultured NMVP and HMVP2 cells treated with CXCL12 (100 ng/mL) for 5 to 30 min and subjected to Western blot analysis using antibodies for the indicated phospho and total proteins. Relative protein levels were quantitated by densitometry; β-actin was used to control for protein loading. Representative blots of three separate experiments with similar results are shown.