[CANCER RESEARCH 47, 4160-4164, August 1. 1987] Expression of the Gene Encoding a Prolactin-inducible Protein by Human Breast Cancers in Vivo: Correlation with Steroid Receptor Status1 Leigh C. Murphy,2 Matthew Lee-Wing,3 Gerald J. Goldenberg, and Robert P. C. Shiu4 Departments of Physiology [L. C. M., M. L-W, K. P. C. S.J and Medicine [G. J. GJ, University of Manitoba and the Manitoba institute of Cell Biology, Winnipeg. Manitoba, Canada, R3E OWi ABSTRACT We have previously reported the identification and characterization of a prolactin-inducible protein (PIP) as well as the cloning of the gene encoding PIP in cultured human breast cancer cells. We now present three lines of evidence that the gene encoding PIP is also expressed by some human breast cancers in vivo: (a) detection of PIP immunoreactivity in the serum of some breast cancer patients; (b) imntunohistochemical detection of PIP in breast cancer sections; and (<•) the presence of PIP mRNA, detected by complementary DNA hybridization in human breast biopsy samples. In a preliminary study using Western blot analysis authentic PIP was detected in the serum of some patients with breast cancer. Subsequently the sera of 234 unselected patients with breast cancer were assayed for the presence of PIP using a specific radioimmunoassay. Thirty-five % of these sera contained detectable PIP (i.e., > 3 ng/ml). As well we were able to show by immunohistochemical techniques that PIP immunoreac tivity was present in some human breast biopsy specimens. Levels of estrogen receptor, progesterone receptor, and PIP mRNA were deter mined in an unselected population of 51 human breast tumor biopsies. Sixty-one % of these tumors had detectable PIP mRNA; a positive correlation (r = 0.52; /' < 0.01) was found between PIP mRNA levels in breast biopsy samples and estrogen receptor content, a known prognostic indicator in human breast cancer. INTRODUCTION A PIP5 secreted by the HBC cell line T-47D has been de scribed previously (1), and recently its cDNA was cloned.6 Maximum expression of this gene occurs when T-47D cells are treated with human prolactin in the presence of hydrocortisone for 4-5 days (I).6 While not all HBC cell lines were found to express the PIP gene,5 those that did were both prolactin and estrogen receptor positive. The data suggested a potential value of PIP as a marker of prolactin action in HBC. However, before this question could be addressed it was first necessary to deter mine if the PIP gene was expressed in vivo and if so to investi gate the relationship of PIP expression to known prognostic indicators in HBC such as ER and PgR. Since PIP is a secreted protein (1) we approached the above questions by first investigating the presence of PIP in the serum of breast cancer patients by Western blot analysis and radioim munoassay. Next we investigated the expression of PIP by HBC Received 1/20/87; revised 4/14/87; accepted 5/1/87. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1The work was supported by the Medical Research Council of Canada, the Manitoba Health Research Council, the St. Boniface Hospital Research Foun dation, and the Faculty of Medicine of the University of Manitoba. 2 Medical Research Council (Canada) Fellow. To whom requests for reprints should be addressed. * Recipient of a Terry Fox Cancer Research Clerkship Award from the National Cancer Institute of Canada. ' Medical Research Council (Canada) Scientist. 9 The abbreviations used are: PIP. prolactin-inducible protein; HBC, human breast cancer; PBS, phosphate-buffered saline; TBS, Tris buffered saline; ER, estrogen receptor; PgR, progesterone receptor; cDNA, complementary DNA; BSA, bovine serum albumin; SDS, sodium dodecyl sulfate; SSC. standard saline citrate. 6 L. C. Murphy, D. Tsuyuki. and R. P. C. Shiu. Cloning and regulation of a prolactin-inducible gene in human breast cancer, submitted for publication. biopsies using both immunohistochemical techniques and spe cific cDNA hybridization techniques. MATERIALS AND METHODS Serum Samples. Coded blood samples were obtained from an unse lected population of breast cancer patients attending the Manitoba Cancer Treatment and Research Foundation. Normal human blood samples were obtained from healthy volunteers. Blood samples were clotted, centrifuged, and the serum stored at —20°C until assay. Human Tumor Samples. Breast tumor biopsies sent frozen to the laboratory for routine ER and PgR assay were the source of breast tumor tissue used for either immunohistochemical studies or RNA extraction and cDNA hybridization. Other human tumor samples were kindly provided by Dr. A. Alguacil of the Department of Pathology, Health Sciences Centre, Winnipeg, Canada. Dr. R. Matusik, Depart ment of Physiology, University of Manitoba, generously donated sam ples of RNA extracted from a range of human benign prostatic and prostaiic carcinoma specimens. All tissues had been frozen as quickly as possible following surgery and pathological examination and stored at -70°C. ER and PgR Assays. Samples weighing between 0.2 and 0.3 g were taken for ER and PgR assay. Available steroid receptors were assayed using a single saturating dose and a charcoal-dextran method (2) for separation of bound from free hormone. ER and PgR concentrations were expressed as fmol of steroid bound per mg of cytosol protein. RNA Isolation and Hybridization Analysis. RNA was isolated by the guanidinium thiocyanate/cesium chloride method (3). For Northern blot analysis RNA isolated as described above was denatured in 50% (v/v) formamide and 2.2 M formaldehyde, size fractionated on a 1% (w/v) agarose-2.2 M formaldehyde gel (4) and then blotted onto nitro cellulose paper (5). Filters were baked for 2 h at 80°Cunder vacuum and then prehybridized in hybridization solution for at least 3 h. The filters were then hybridized with nick-translated "P-labeled purified PIP cDNA insert, cPIP-8-3 (specific activity, 1-5x10* cpm/Mg DNA). Hybridizations, usually for 24 h, were performed at 42°C in the presence of 50% (v/v) deionized formamide, 5x Denhardt's solution (Ix Den- hardt's = 0.02% w/v each of BSA, Ficoll, and polyvinylpyrrolidine), 5x SSPE (Ix SSPE = 0.15 M NaCl-0.01 M NaH2PO4-l mM EDTA), 250 Mg/ml denatured salmon sperm DNA, and 0.1% SDS. At the end of the hybridization period the blots were washed twice in 0.1% SDS- 2x SSC (Ix SSC = 0.15 M NaCl-0.015 M sodium citrate) for 15-30 min at room temperature, followed by one wash in O.lx SSC-0.1% SDS for 45-60 min at 65°C.Blots were then exposed to Kodak XAR film at —70°C with an intensifying screen. For dot-blot analysis RNA extracted as described above was dena tured according to the protocol of White and Bancroft (6) and spotted in varying concentrations (30-0.25 Mg for tumors and MCF-7 HBC cells or 2-0.0625 Mgfor T-47D HBC cells) onto nitrocellulose filters using a BRL dot-blot manifold. Filters were treated as described above. Quantitation of serially diluted RNA samples was achieved by den- sitometric scanning of various exposures of the dot-blot autoradi- ographs. The line of best fit obtained by least squares regression analysis of the integrated peak areas of the signal intensity versus the amount of RNA loaded was compared to that obtained from a standard serial dilution of hormone treated T-47D RNA, arbitrarily given a value of 100, present on each dot-blot. A negative control of MCF-7 human breast cancer cell RNA was also included on each dot-blot. Consistently we have been unable to detect the presence of PIP mRNA in this cell line by Northern analysis.6 4160 Research. on November 18, 2015. © 1987 American Association for Cancer cancerres.aacrjournals.org Downloaded from