Janus kinase 2 is required for the initiation but not maintenance of prolactin-induced mammary cancer

Abstract

The prolactin receptor (PRLR), its associated Janus kinase 2 (Jak2) and the signal transducer and activator of transcription 5 (Stat5) are essential for normal mammary gland development. Owing to the upregulation of the PRLR and the local synthesis of its ligand in neoplastic cells, it has been proposed that PRL can act as a local growth factor in human breast cancers. This notion is supported by experimental evidence in transgenic mice, which showed that the mammary-specific expression of PRL contributes to carcinogenesis in vivo. To assess the importance of Jak2/Stat5 signaling during mammary cancer initiation and progression, we generated a PRL-induced mammary cancer model that allows the functional ablation of the Jak2 gene in the mammary epithelium before and after neoplastic transformation. Collectively, the results of this study show that the functional ablation of Jak2 protects against the onset of PRL-induced mammary tumorigenesis, suggesting that targeting this kinase is a relevant strategy for mammary cancer prevention. Surprisingly, Jak2 deficiency did not affect the growth and survival of PRL-induced mammary cancer cells in culture and in vivo. Consequently, Jak2 cannot be a sole therapeutic target to treat the established disease. PRL-induced mammary cancers exhibited an upregulation of ErbB2 and other ErbB receptor tyrosine kinases that may supersede the functionality of PRLR signaling through Jak2.

Fig. 1. Inhibition of PRL-induced mammary tumorigenesis through deletion of Jak2

Mammary gland whole-mounts from two NRL-PRL transgenic lines (1655-8 and 1647-13) that are Jak2-deficient (b, d) and their littermate wildtype controls (a, c). Arrows in panels a and c indicate the location of atypical hyperplastic lesions; LN lymph node; bar = 1mm.

Fig. 2. A subset of PRL-induced mammary cancers expresses the estrogen receptor α

Immunohistochemistry of ERα in various mammary tumors (a-c) and the normal mammary epithelium (d); a, ERα-positive tumor, b, mixed lesion containing a subset of ERα-positive cells, c, ERα-negative cancer. Panel e shows a serial section of the mammary duct shown in panel d that was stained without the primary antibody against ERα. All slides were counterstained with hematoxylin (bar, 50 μm).

Fig. 3. PRL-induced mammary cancer cells that are deficient in Jak2 lack activation of Stat5

(a) Experimental design for the conditional deletion of Jak2 in PRL-induced mammary cancer cells. (b) PCR assay to verify the loss of both Jak2 floxed alleles (Jak2^fl/fl) and the presence of Jak2 knockout alleles (Jak2^-/-) following Cre-mediated recombination. (c) Expression analysis of Jak2, Estrogen Receptor α (ERα), and Cyclin D1 (Ccnd1) in Jak2-deficient cancer cells and their isogenic control cells. Beta-Actin (ActB) served as a loading control. (d) Immunoprecipitation (IP) and western blot (WB) analysis to assess the activation of Stat5 and Stat3 following PRL stimulation.

Fig. 4. Jak2 deficiency does not affect the growth of PRL-induced mammary cancer cells in vivo

(a) Growth of PRL-induced mammary cancer cells lacking Jak2 and their isogenic controls in an orthotropic transplant of Athymic nude mice. (b) PCR assay to verify the presence of Jak2 floxed alleles or Jak2 null alleles in resulting mammary tumors following orthotropic transplantation. Note that the Jak2 wildtype allele (Jak2^wt), which was present in the PCR assay in both experimental groups, originated from stromal cells and blood vessels of the wildtype host. This allele was not present in the engrafted Jak2^fl/fl or Jak2^-/- cancer cells. NC: negative control.

Fig. 5. Jak2 deficiency leads to lack of Stat5 activation but not expression of ERα and Cyclin D1

(a) Immunohistochemistry to assess the PRL-mediated activation of Stat5 in control tumors and cancers lacking Jak2. Insets show the PRL-induced activation of Stat5 in normal #3 mammary glands derived from the tumor-bearing, wildtype recipient females. (b) Immunohistochemistry to determine the continuous expression of ERα and Cyclin D1 (Ccnd1) in PRL-induced mammary cancers that lack Jak2 and their wildtype controls. All slides were counterstained with hematoxylin (bars, 50 μm).

Fig. 6. PRL-induced mammary cancers exhibit an upregulation of ErbB receptor tyrosine kinases

(a) Western blot analysis to determine the expression of all four ErbB family members in NRL-PRL-induced mammary cancers as well as normal mammary glands (MG) from lactating NRL-PRL females. (b) Immunoprecipitation (IP) and western blot (WB) analysis to assess the tyrosine phosphorylation and heterodimer formation of ErbB2 and ErbB3. (c) Western blot analysis to assess the expression of ErbB receptors in untransformed and neoplastic mammary epithelial cells that lack Jak2 and their isogenic controls expressing Jak2. Beta-Actin (ActB) served as a loading control in both panels.