Morphogenic and tumorigenic potentials of the mammary growth hormone/growth hormone receptor system

Abstract

Due to the characteristics of the luteal phase of the ovarian cycle in the dog, which spans a prolonged time period, this species is a suitable model to study the role of progestins in both normal morphogenic and abnormal tumorigenic processes in the mammary gland. It has been convincingly shown that progestins, including endogenous progesterone, induce the synthesis of growth hormone (GH) in the normal and the tumorous canine mammary gland. The growth hormone receptor (GHR) is also expressed in normal and tumorous canine mammary tissues and in this concise overview we highlight recent advances in our understanding of the significance of the GH/GHR system for mammary gland (patho)biology. In an attempt to unravel the cellular and molecular mechanisms associated with the GH/GHR system, we were able to show that both GH and GHR are differentially expressed in normal canine mammary tissues. Maximum expression of both GH and GHR occurs during the proliferation phase of the tissue, which links the progestin-induced mammary GH synthesis to the progestin-associated proliferation of epithelial cells in the mammary gland. Expression of the GH/GHR system is also present in most canine mammary tumors, albeit that GHR expression may be downregulated in undifferentiated mammary carcinomas. Upon GH stimulation of the GHR-positive CMT-U335 canine mammary tumor cell line, the transcription factors STAT5A and STAT5B become phosphorylated on their tyrosine residues, which is likely to reflect the significance of mammary GH in vivo. Molecular analysis of the canine mammary GHR transcripts by RT-PCR provided evidence for normal and alternative processing of the GHR primary transcript encoding the full-length plasma membrane GHR and at least four putative GH binding proteins (GHBPs), respectively. The translation products from the alternatively spliced GHR transcripts indicate an intact N-terminal ligand binding domain and an unique C-terminal portion, lacking the transmembrane domain and cytoplasmic tail. Thus, these proteins are considered to be able to bind GH, but have lost their signaling potential. The exact biological role of these GHBPs remains to be established, but GHBPs may have a transport function in the endocrine route, regulate the level of biologically available GH locally, or dominant-negatively influence the full-length plasma membrane GHR. In dog mammary cancer specimens strongly reduced levels of alternatively spliced GHR transcripts were found compared to the non-malignant mammary tissue. Notably, expression of both GH and GHR in mammary cancer cells is not restricted to dogs. Recent experiments generated evidence for GH and GHR expression in human breast cancer cells, and also in human prostate cancer cells, which represents another highly prevalent hormone-sensitive human malignancy. In agreement with our findings in the dog, the expression of the hGH-N gene in human mammary cancer cells seemed to correlate positively with their progesterone receptor status, which warrants, in our opinion, a reconsideration of the role of progestins in breast cancer of women. In human prostate cancer cells four different hGH-N transcripts were detected, which encode classical 22 kDa GH and GH-related proteins. Consistent with the findings on the canine GHR, different GHR transcripts in human mammary cancer cells and prostate cancer cells were detected encoding the full-length plasma membrane GHR and putative GHBPs.