Pascale Rio

Quantification of BRCA1 protein in sporadic breast carcinoma with or without loss of heterozygosity of the BRCA1 gene.

In sporadic breast cancer, no mutations of the BRCA1 gene have been reported so far, whereas BRCA1 mRNA is markedly decreased in invasive breast cancer. To elucidate the contribution of the BRCA1 gene in sporadic breast cancer, we quantified the BRCA1 protein, using [125I] labeling of whole‐cell proteins, lentil‐lectin affinity chromatography, immunoprecitation by anti‐BRCA1 antibodies (C‐20, D‐20, I‐20 and K‐18), purification of the immune complex by protein A affinity chromatography and chromato‐focusing. As loss of 1 allele may lead to a decreased expression of the gene, 10 tumors were previously checked for loss of heterozygosity (LOH) of the BRCA1 gene, using 3 intragenic microsatellite markers. Our results indicated that the BRCA1 gene product was decreased in the 4 tumors with LOH compared with matched normal breast tissues. Reduced amounts of BRCA1 protein were also detected in 3 of 6 tumors without LOH. Our quantitative method allowed us to demonstrate that the BRCA1 protein level was decreased in sporadic invasive breast carcinomas with or without LOH of the BRCA1 gene, implying multiple mechanisms of BRCA1 expression down‐regulation in these tumors. Our data suggest that the amount of BRCA1 protein present may play an important role in human sporadic breast carcinoma. Int. J. Cancer 80:823–826, 1999.

Differential effects of n -3 and n -6 polyunsaturated fatty acids on BRCA1 and BRCA2 gene expression in breast cell lines

Current evidence strongly supports a role for the breast tumour suppressor genes, BRCA1 and BRCA2, in both normal development and carcinogenesis. In vitro observations reported that BRCA1 and BRCA2 are expressed in a cell cycle-dependent manner. Interestingly, differences in the actions of n-3 and n-6 polyunsaturated fatty acids have been observed: while the n-3 polyunsaturated fatty acids have been described to reduce pathological cell growth, the n-6 polyunsaturated fatty acids have been found to induce tumour proliferation. Here, we examined the expression of BRCA1 and BRCA2 in breast cell lines after treatment with polyunsaturated fatty acids. Real-time quantitative polymerase chain reaction determinations conclusively demonstrated increases in BRCA1 and BRCA2 mRNA expressions in MCF7 and MDA-MB 231 tumour cell lines after treatment with n-3 polyunsaturated fatty acids (eicosapentaenoic acid and docosahexaenoic acid), but no variation was noticed with the n-6 polyunsaturated fatty acid (arachidonic acid). On the other hand, no variation of the expression of BRCA1 and BRCA2 mRNA was detected in MCF10a normal breast cell line treated by polyunsaturated fatty acids. The level of BRCA1 and BRCA2 proteins quantified by affinity chromatography remained unchanged in tumour (MCF7, MDA-MB 231) and normal (MCF10a) breast cell lines. We suggest the presence of a possible transcriptional or post-transcriptional regulation of BRCA1 and BRCA2 after n-3 polyunsaturated fatty acid treatment in breast tumour cells.

Detection of heterozygous carriers of the ataxia-telangiectasia (ATM) gene by G2 phase chromosomal radiosensitivity of peripheral blood lymphocytes

Abstract In ataxia-telangiectasia (A-T) patients, mutations in a single gene, ATM, result in an autosomal recessive syndrome that embraces a variety of clinical features and manifests extreme radiosensitivity and a strong predisposition to malignancy. Heterozygotes for the ATM gene have no clinical expression of A-T but may be cancer prone with a moderate increase in in vitro radiosensitivity. We performed a blind chromosomal analysis on G2-phase lymphocytes from 7 unrelated A-T patients, 13 obligate A-T heterozygotes (parents of the patients), and 14 normal controls following X-irradiation with 1 Gy in order to evaluate this cytogenetic method as a tool for detection of ATM carriers. Both A-T homozygotes and heterozygotes showed significantly increased levels of radiation-induced chromatid damage relative to that of normal controls. These results show that the G2-phase chromosomal radiosensitivity assay can be used for the detection of A-T heterozygotes. In combination with molecular genetic analyses, this test may be of value in studies of familial and sporadic cancers aimed at determination of the potential involvement of ATM mutations in tumor risk or development.

Design, synthesis, and biological activity of pyridopyrimidine scaffolds as novel PI3K/mTOR dual inhibitors.

The design, synthesis, and screening of dual PI3K/mTOR inhibitors that gave nanomolar enzymatic and cellular activities on both targets with an acceptable kinase selectivity profile are described. A docking study was performed to understand the binding mode of the compounds and to explain the differences in biological activity. In addition, cellular effects of the best dual inhibitors were determined on six cancer cell lines and compared to those on a healthy diploid cell line for cellular cytotoxicity. Two compounds are highly potent on cancer cells in the submicromolar range without any toxicity on healthy cells. A more detailed analysis of the cellular effect of these PI3K/mTOR dual inhibitors demonstrated that they induce G1-phase cell cycle arrest in breast cancer cells and trigger apoptosis. These compounds show an interesting kinase profile as dual PI3K/mTOR tool compounds or as a chemical series for further optimization to progress into in vivo experiments.

Loss of heterozygosity at 11q23.1 and survival in breast cancer: Results of a large European study

Among the chromosomal regions commonly undergoing deletions in breast tumors is 11q23.1. The genes that are targets for loss of heterozygosity (LOH) in this region is not yet established. One of the candidate genes located in this region is ATM, responsible for the rare autosomal recessive disorder ataxia‐telangiectasia (A‐T). Interestingly, A‐T heterozygotes may have an increased risk of cancer, in particular breast cancer, although this is still controversial. A common assumption has been that the target for the LOH at 11q23.1 in breast carcinoma is the ATM gene, but the area studied has been too large, the density of markers too low, and the number of tumors studied has been too small to draw any firm conclusions. The present study is a multicenter study including 918 breast cancer patients with clinical information and survival data available for most of them. Primary breast tumors were investigated for LOH using a high density of microsatellite markers spanning approximately 6 Mb around the ATM gene. Survival analyses showed that there are most likely one or more candidate genes in a 3–4 Mb region between the markers D11S1819 and D11S927 including the ATM gene. Cancer‐specific survival was significantly reduced in patients whose tumors exhibited LOH of markers D11S2179 (within the ATM gene), D11S1778, D11S1294, and D11S1818. The highest survival hazard ratios were 1.8 (CI 1.2–2.8, P = 0.010) and 2.1 (CI 1.4–3.0, P = 0.0004) for markers D11S2179 and D11S1818, respectively. One or more of these markers are therefore most likely to be located close to or within genes associated with breast cancer survival. Genes Chromosomes Cancer 25:212–221, 1999.

European multicenter study on LOH of APOC3 at 11q23 in 766 breast cancer patients: relation to clinical variables

SummaryHigh frequencies of loss of heterozygosity (LOH) in chromosome 11q22-qter have been observed in various malignancies, including breast cancer. Previous studies on breast carcinomas by Winqvist et al (Cancer Res 55: 2660–2664) have indicated that a survival factor gene is located in band 11q23, and that the highly informative microsatellite polymorphism at the APOC3 locus would be a suitable tool to perform more extensive LOH studies. In this European multicentre study, we have examined the occurrence of APOC3 LOH and evaluated the effect of LOH of this chromosomal subregion on the clinical behaviour of the disease in a cohort of 766 breast cancer patients in more detail. LOH for APOC3 was found in 42% of the studied tumours, but it was not found to be significantly associated with any of the studied clinical variables, including cancer-specific survival time or survival time after recurrent/metastatic disease. According to the present findings, the putative survival factor gene on 11q23 is not located close enough to the APOC3 gene, but apparently at a more proximal location.

Isolation, purification and quantification of BRCA1 protein from tumour cells by affinity perfusion chromatography

A new procedure for the isolation, purification and quantification of the product of the oncosuppressor gene brca1 in breast tissues, was carried out. It involves internal cell protein [35S]methionine labelling followed by two perfusion chromatographies. The first one is heparin affinity chromatography, to purify all of the cell DNA-binding proteins. A subsequent specific immunoprecipitation of BRCA1 protein was performed with an antibody raised against BRCA1. The immune complex was isolated using the second chromatographic step, Protein A affinity chromatography. The amount of BRCA1 expressed by cells was expressed as a ratio, in percent, calculated as follows: 100x amount of labelled DNA-binding proteins (dpm) that bound specifically to the anti-BRCA1 polyclonal antibodies (K-18)/amount of whole labelled DNA-binding protein (dpm) purified on a heparin column. Applications to MCF7 and T-47D human breast tumour cell lines, which were treated or not using 2 mM sodium butyrate demonstrated an increase in BRCA1 protein expression.

BRCA2 protein expression in sporadic breast carcinoma with or without allelic loss of BRCA2

To elucidate the cellular role of BRCA2 in sporadic breast tumors, we studied the cellular localization and the expression of BRCA2 in carcinomas presenting or not allelic loss of BRCA2. The breast tumors were first classified with or without allelic loss of BRCA2 and then immunohistochemical staining was performed on tumors and matched normal tissues using antibodies raised against BRCA2. We showed that BRCA2 is found either in the nucleus or in perinuclear compartments such as the endoplasmic reticulum and the Golgi vesicles. We have earlier demonstrated the presence of BRCA1 as an exocrine secretion in the lumen of ductules in the normal mammary gland, as well as BRCA1 and BRCA2 in milk‐fat globules inside mammary‐gland ductules during lactation. Here we show that BRCA2 is present within the lumina of breast ductules, indicating that BRCA2 protein may also be secreted in the mammary gland. No correlation was found in breast tumors between the expression of BRCA2 protein and allelic loss of BRCA2. Int. J. Cancer 86:453–456, 2000.

BRCA1 and BRCA2 proteins are expressed in milk fat globules

Two genes, BRCA1 (Miki et al.,1994) andBRCA2 (Wooster et al.,1995), have been implicated in inherited predisposition to female breast cancer. Mutations in BRCA1 and BRCA2 also predispose to ovarian cancer, whereas mutations in BRCA2are involved in male breast cancer and apparently in pancreatic, cervical and laryngeal cancers. However, mutation of neither gene appears to play an important role in sporadic breast cancer, implying that sporadic tumors have different etiologies of mutagenesis than those most frequently found in inherited cancers (Stratton, 1996). Jensenet al.(1996) provided conflicting evidence that BRCA1 is a protein localized in the cytoplasm and cell membrane, and is also present as a secreted protein in cell culture supernatants. The BRCA1 protein shares sequence homology and biochemical analogy with the granin family of secreted proteins, located in secretory granules, and some granin-type proteins are regulated by estrogen (Thompson et al.,1995). Moreover, the breast is an exocrine gland whose primary role is secretion. Using K-18 antibodies against BRCA1, we observed BRCA1 staining in the lumina of the ductules in normal breast demonstrating that BRCA1 proteins were present in secretions of the breast. A patient suffering from apocrine metaplasia also presented apical cytoplasmic staining and apocrine secretion staining in the lumen of the tubes with K-18 antibodies (BernardGallonet al.,1998). Coeneet al. (1997) reported localization of BRCA1 in the perinuclear compartment of the endoplasmic reticulum-Golgi complex and in tubes invaginating the nucleus in breast tissues. These findings account for BRCA1 being a member of the family of secreted proteins. We have detected BRCA1 and BRCA2 in the cytoplasm and/or the nucleus of cells of normal tissues, carcinomas in situ and invasive carcinomas, which is in accordance with the proposed localization of BRCA1 and BRCA2. Nevertheless, it is noteworthy that 2 patterns of nuclear staining were detected. In some breast carcinomas, the nucleus was completely stained and in others, the staining was more specifically perinuclear and more marked in Golgi vesicles located close to the nucleus, or in plaques in the cytoplasm corresponding to the endoplasmic reticulum. These 2 different patterns were found inside one same carcinoma, indicating different cell cycle steps. BRCA2 staining was also found in secretions in mammary gland ducts and in tufts of invasive carcinoma (data not shown). Based on these results, we further investigated the presence of BRCA1 and BRCA2 in milk fat globules (MFG) using immunochemical analysis with a large panel of antibodies against BRCA1 and BRCA2 (Table I). MFGs are formed by exocytosis of lipid from epithelial cells of the mammary gland and are enveloped by plasma membrane from epithelial cells. Analyses were performed on snap-frozen mammary gland samples collected from 2 children. The first (child I) died suddenly at the age of 4 weeks. At that age, the mammary gland is active and may continue to grow and secrete milk (Fig. 1 a–f). The second (child II) died at the age of 5 months, when newborn breast development and milk secretion stop (McKiernan et al., 1998) (Fig. 2a–e). The last sample was from a lactating hamartoma of the breast from an 8-month term pregnant woman (Fig. 3a–g). Using different antibodies against BRCA1 and BRCA2, we found that lobular ducts in the mammary gland of the 4-weekold child contained milk secretion with BRCA1 (Fig. 1 c,d) and BRCA2 expression in the cytoplasm of MFG (Fig. 1 e,f) as well as in the lactating hamartoma of the breast (Fig. 3 d–g), for which the presence of MFG secretion is clearly visible in Figure 3b. In contrast, tissue sections of the older child’s mammary gland, whose milk secretion had ceased, were devoid of MFG expressing BRCA1 (Fig. 2 b–d) and BRCA2 (Fig. 2 e) proteins. An exclusively nuclear staining pattern was more frequently found in mammary gland sections using the monoclonal antiBRCA1 antibodies 8F7 (Fig. 2 c) and 17F8 (Fig. 3d) raised against a glutathione-S-transferase (GST)-BRCA1 fusion prot in containing amino acids encoded by a 3 8 portion of BRCA1 exon 11 and by a 5 8 portion of BRCA1 exon 11, respectively (Chenet al.,1995, 1996a,b). Using these antibodies, Chent al. (1995) concluded that BRCA1 is a nuclear protein of normal breast epithelial cells. We also observed nuclear staining, when TABLE I – SPECIFICITY OF THE PRIMARY ANTIBODIES USED

Expression of human BRCA1 and BRCA2 proteins in lung from a fetus at 19 weeks' gestation

Germ-line mutations in the tumor-suppressor gene BRCA1 predispose women to an elevated risk of breast cancer, while germ-line mutations in the tumor-suppressor gene BRCA2 predispose both men and women to breast cancer. Many studies have suggested an important developmental role for the murine homologues of BRCA1 (Marquis et al., 1995) and ofBRCA2 (Rajan et al., 1997) in the regulation of proliferation and differentiation. The spatial localization of Brca1 (Marquis et al., 1995) and Brca2 mRNA (Rajan et al., 1997) has been determined on frozen sections of murine embryos by in situ hybridization, revealing intense organ-specific expression in multiple tissues, notably the lung. We have determined here, using immuno-histochemistry and a panel of antibodies (Table I), the pattern of expression of BRCA1 and BRCA2 in human lung tissue during development from a fetus at 19 weeks of gestation, following therapeutic abortion. The staining specificity of K-18 antibodies for BRCA1 had been ascertained by Western blotting, and a 220-kDa band detected in both HBL100 and MCF7 breast cell lines, with other major bands appearing around 100 kDa, possibly corresponding to BRCA1 variants (Bernard-Gallon et al., 1998). Mouse monoclonal antibodies against BRCA1 (8F7 and 17F8) were purchased from GeneTex (San Antonio, TX) and, upon immunoblotting, showed single-band specificity for BRCA1, a 220-kDa protein (Y. Chenet al., 1995, 1996; C. Chenet al.,1996). The anti-BRCA1 antibody (66046E), raised against amino acids 768–793 of human BRCA1, purchased from PharMingen (San Diego, CA), recognized a 220-kDa band on Western blots of HBL100 cells, and several smaller immunoreactive species representing degradation products or BRCA1 forms arising from alternative RNA splicing or processing (Ruffner and Verma, 1997) . The anti-BRCA2 antibody (66066E), raised against amino acids 1323–1346 of human BRCA2, and the 66076E antibody, raised against amino acids 2586–2600 of human BRCA2, were obtained from PharMingen and are known to recognize a 350to 380-kDa BRCA2 protein in HBL100 human breast cell lysates by Western blot analysis. These BRCA2 antibodies also cross-react with smaller proteins, which might be degradation products. The pattern of staining of fetal lung with the different antibodies is summarized in Table I and illustrated in Figure 1. Nuclear staining of the BRCA1 protein with K-18 (Fig. 1 b), 8F7 (Fig. 1c) and 17F8 (Fig. 1d) antibodies was observed in bronchiolar cells and terminal air spaces. Nuclear staining was rather strong with antibodies against BRCA1, except the 8F7 antibody, with which staining was faint but specific. Intensive cytoplasmic staining was also seen with 17F8 (Fig. 1 d) and 66046N (Fig. 1e) antibodies, but it was less intense with K-18 (Fig. 1b) in bronchiolar cells and terminal air spaces. Moreover, with the 17F8 antibody, cytoplasmic staining was also apical (Fig. 1d). Staining of BRCA1 and BRCA2 thus was detected in normal cells that were not of breast or ovarian origin. In addition, BRCA1 and BRCA2 were expressed in cells undergoing rapid proliferation prior to differentiation, as reported in mice by Marquiset al.(1995), Laneet al.(1995) and Rajanet al.(1997). We explain the difference in the location of BRCA1 protein between nucleus and cytoplasm by the fact that different portions and/or alternatively spliced variants of BRCA1 are present in different cellular compartments, like in breast tissue. Many spliced variants of BRCA1 have been described. The most notable one is an exon 11 splice variant which encodes a protein that is exclusively present in the cytoplasm (Thakuret al., 1997). Another splice variant, missing exons 9 and 10, is also commonly found in normal cells (Wilson et al., 1997). The significance of these splice variants is not known. They may participate in the regulatory role of the full-length BRCA1 protein. We also observed a strong staining pattern of BRCA2 protein in the apical cytoplasm of bronchiolar ciliated and non-ciliated cells with the 66066E antibody (Fig. 1 f). Using the 66076E antibody, strong apical cytoplasmic staining and occasional intense nuclear staining of bronchiolar cells were also observed TABLE I – SPECIFICITY OF THE PRIMARY ANTIBODIES USED AND CORRESPONDING SUBCELLULAR LOCALIZATION AND EXTENT OF STAINING OBSERVED IN FETAL LUNG