Role of JNK in mammary gland development and breast cancer

Abstract

cJun NH(2)-terminal kinase (JNK) signaling has been implicated in the developmental morphogenesis of epithelial organs. In this study, we employed a compound deletion of the murine Jnk1 and Jnk2 genes in the mammary gland to evaluate the requirement for these ubiquitously expressed genes in breast development and tumorigenesis. JNK1/2 was not required for breast epithelial cell proliferation or motility. However, JNK1/2 deficiency caused increased branching morphogenesis and defects in the clearance of lumenal epithelial cells. In the setting of breast cancer development, JNK1/2 deficiency significantly increased tumor formation. Together, these findings established that JNK signaling is required for normal mammary gland development and that it has a suppressive role in mammary tumorigenesis.

Figure 1. JNK-deficient mammary epithelial cells

A,B) Jnk1^LoxP/LoxP Jnk2^−/− Cre^ERT mice treated without or with tamoxifen in vivo were used to prepare mammary gland extracts and primary mammary gland epithelial cells. Genomic DNA was examined by PCR to detect Jnk1^LoxP and Jnk1^Δ alleles (A). The expression of JNK and α-Tubulin was examined by immunoblot analysis (B). C) Primary cultures of Cre^ERT (Control) and Jnk1^Δ/Δ Jnk2^−/− Cre^ERT mammary epithelial cells prepared from tamoxifen-treated mice were examined by immunofluorescence analysis by probing with antibodies to pan-cytokeratin (red) and E-cadherin (green). DNA was stained with DAPI (blue). Representative images are shown.

Figure 2. Effect of JNK-deficiency on mammary epithelial cell proliferation

A) Primary cultures of Cre^ERT (Control) and Jnk1^Δ/Δ Jnk2^−/− Cre^ERT mammary epithelial cells prepared from tamoxifen-treated mice were examined by phase contrast microscopy. Representative images are shown. B) The cells were pulse-labeled with BrdU and examined by flow cytometry. The number of BrdU positive cells (%) is presented (mean ± SD; n = 5). Significant differences between Control cells and JNK-deficient cells are indicated with an asterisk (p < 0.05). C) Relative cell proliferation was measured using the WST-1 assay (mean ± SD; n = 3). Significant differences between Control cells and JNK-deficient cells are indicated with an asterisk (p < 0.05).

Figure 3. Effect of JNK-deficiency on mammary epithelial cell motility and invasion

A,B) Primary cultures of Cre^ERT (Control) and Jnk1^Δ/Δ Jnk2^−/− Cre^ERT mammary epithelial cells prepared from tamoxifen-treated mice were examined using Boyden chambers coated with Collagen I (A) or with a Matrigel layer (B). The relative number of cells that moved from the upper chamber to the lower chamber is presented (mean ± SD; n = 5). Significant differences between Control cells and JNK-deficient cells are indicated with an asterisk (*, p < 0.05; **, p < 0.001).

Figure 4. Effect of JNK-deficiency on mammary branching morphogenesis

A) Primary organoid Cre^ERT (Control) and Jnk1^Δ/Δ Jnk2^−/− Cre^ERT cultures prepared from tamoxifen-treated mice were examined by immunofluorescence microscopy by staining with an antibody to smooth muscle actin (SMA, green) and with phalloidin to stain F-actin (red). DNA was stained with DAPI (blue). Branching morphogenesis was initiated by treatment of the cultures with FGF2. The organoids were examined using a Leica TCS SP2 confocal microscope by acquiring 10 optical sections that were collapsed to a single image. Representative images are shown. B) The number of branches per organoid was measured (mean ± SD, n = 20). Significant differences between Control cells and JNK-deficient cells are indicated with an asterisk (*, p < 0.01).

Figure 5. Effect of JNK-deficiency on mammary gland development

Transplantation assays were performed using mammary tissue from tamoxifen-treated female donor mice and female recipient nude mice. Control (Cre^ERT) tissue was transplanted in one cleared mammary gland and Jnk1^Δ/Δ Jnk2^−/− Cre^ERT tissue was transplanted in the contralateral gland of the same recipient mouse. A–C) Whole mount mammary glands stained with carmine alum at 8 weeks following transplantation are shown (A,C). The number of branches per unit length (mm) of duct is presented as the mean ± SD; n = 3 (B). Statistically significant differences are indicated with an asterisk (*, p < 0.05). Scale bars: 5 mm (A); 200 µm (C). D–F) Terminal end buds in whole mount mammary glands stained with carmine alum at 2 weeks following transplantation are shown (D). Stained sections of the end buds in whole mount mammary glands are presented (E). The number of branches per unit length (mm) of duct is presented as the mean ± SD; n = 3 (F). Statistically significant differences are indicated with an asterisk (*, p < 0.05). Scale bars: 200 µm (D); 100 µm (E).

Figure 6. Effect of JNK-deficiency on mammary tumor development

Transplantation assays were performed using mammary tissue from female donor mice and female recipient nude mice. Control (KRas^LSL-G12D/+ Trp53^LoxP/LoxP Cre^ERT) tissue was transplanted in one cleared mammary gland and KRas^LSL-G12D/+ Trp53^LoxP/LoxP Jnk1^LoxP̃/LoxP Jnk2^−/− Cre^ERT tissue was transplanted in the contralateral gland of the same recipient mouse. The transplanted mice were treated with tamoxifen at two weeks post-surgery. A) Sections of transplanted breast mammary glands were stained with H&E. Scale bar = 50 µm. B) Sections of breast tumors were stained with an antibody to PCNA (red). DNA was stained with DAPI. Scale bar = 75 µm. C) Genomic DNA isolated from Control and JNK-deficient breast tumors was examined by PCR using amplimers to detect the wild-type Jnk1 allele (1.5 kb), the Jnk1^LoxP allele (1.1 kb), and the ablated allele ΔJnk1 (0.4kb). D) Kaplan-Meier analysis of tumor-free survival of transplanted mice (n = 8). The development of JNK-deficient tumors was significantly more rapid than Control tumors (p < 0.02; Log-rank test).

Figure 7. JNK-deficiency causes increased basal-like mammary carcinogenesis

Transplantation assays were performed using Control and JNK-deficient tissue (Figure 6). A) Sections showing the periphery of Control and JNK-deficient mammary tumors were stained with H&E. Scale bar = 150 µm. B) Sections of Control and JNK-deficient mammary tumors were stained with an antibody to cytokeratin 5 (green) and cytokeratin 8 (red). DNA was stained with DAPI. Scale bar = 150 µm.