Proteomic Analysis to Identify Breast Cancer Biomarkers in Nipple Aspirate Fluid
Total Page:16
File Type:pdf, Size:1020Kb
7500 Vol. 10, 7500–7510, November 15, 2004 Clinical Cancer Research Featured Article Proteomic Analysis to Identify Breast Cancer Biomarkers in Nipple Aspirate Fluid Hannah Alexander,1 Andrew L. Stegner,1 influenced GCDFP-15 and AAG more in women without Colette Wagner-Mann,2 Garrett C. Du Bois,3 breast cancer than in women with breast cancer. apoD levels Stephen Alexander,1 and Edward R. Sauter2 did not correlate significantly with breast cancer. 1 2 Conclusions: Our study revealed that the NAF pro- Division of Biological Sciences, and Department of Surgery, teome, as defined by two-dimensional PAGE, consists of a University of Missouri, Columbia, Missouri; and 3Kimmel Cancer Center, Thomas Jefferson University, Philadelphia, Pennsylvania limited number of proteins, and that the expression of AAG and GCDFP-15 correlates with disease presence and stage. ABSTRACT INTRODUCTION Purpose: Proteomic analysis of breast nipple aspirate fluid (NAF) holds promise as a noninvasive method to iden- Early detection is a major factor contributing to the 3.2% tify markers of breast cancer. The objectives of the study annual decline in breast cancer death rates over the past 5 years were to: (a) describe the NAF proteome, (b) identify candi- (1, 2). Unfortunately, currently available breast cancer screening date markers of breast cancer in NAF by using proteomic tools such as mammography and breast examination miss up to analysis, and (c) validate the markers identified by using a 40% of early breast cancers and are least effective in detecting quantitative, high-throughput ELISA analysis. cancer in young women, whose tumors are often more aggres- Experimental Design: For proteome analysis, NAF pro- sive. An invasive needle or surgical biopsy must be performed teins from a single subject without breast cancer were sep- when an area of suspicion is identified to confirm, by cytologic arated by two-dimensional PAGE and were subjected to or histologic evaluation, the presence of malignant disease. The matrix-assisted laser desorption ionization time-of-flight development of noninvasive techniques that would distinguish mass spectometry identification. A total of 41 different pro- between women with and women without breast cancer, as well teins were identified, 25 of which were known to be secreted. as between different disease stages, is, therefore, of crucial To identify breast cancer markers, we separated 20 NAF importance. samples (10 normal, 10 cancer) by two-dimensional PAGE. Recent advances in comprehensive molecular technologies Three protein spots were detected that were up-regulated in have allowed the analysis of global gene expression or protein three or more cancer samples. These spots were identified to profiles in cancerous versus normal tissues with the goal of be gross cystic disease fluid protein (GCDFP)-15, apoli- identifying RNA or protein markers that are differentially ex- poprotein D (apoD), and ␣1-acid glycoprotein (AAG). To pressed between benign and malignant tissues. One such study validate these three potential biomarkers, 105 samples (53 (3, 4) analyzed RNA by using serial analysis of gene expression from benign breasts and 52 from breasts with cancer) were to identify molecular alterations involved in breast cancer pro- analyzed using ELISA. gression. The authors concluded that very few genes were Results: Among all of the subjects, GCDFP-15 levels up-regulated in all tumors, presumably reflecting the high de- gree of diversity among tumors at the molecular level. The most ؍ were lower (P < 0.001) and AAG levels were higher (P 0.001) in breasts with cancer. This was also true in prem- dramatic changes occurred at the transition from benign breast -but not tissue to in situ carcinoma, in which many of the highly ex (0.002 ؍ AAG, P ;0.011 ؍ enopausal (GCDFP-15, P in postmenopausal women. GCDFP-15 levels were lowest pressed genes encoded secreted proteins, mainly cytokines and -and AAG levels highest (P < 0.001) in women chemokines, implicating abnormal paracrine and autocrine sig (0.003 ؍ P) with ductal carcinoma in situ (DCIS). Menopausal status naling in the initiation of breast cancer. Other nucleic acid-based studies have examined samples of normal versus cancerous tissues, established cell lines, and samples from before and after treatment with chemotherapeutic agents for differences in expression of a set of Ͼ400 genes, by Received 5/24/04; revised 7/30/04; accepted 8/16/04. using hierarchical clustering to analyze the results (5, 6). Al- Grant support: Supported by NIH grants CA95484 (E. Sauter), though the authors report variation among different tumor sam- GM53929 (S. Alexander), and CA95872 (S. Alexander); and by Depart- ples, these studies have shown that cancers can be classified into ment of Defense grant DAMD17-02-1-0594 (E. Sauter). The costs of publication of this article were defrayed in part by the subgroups based on patterns of gene expression, and that this payment of page charges. This article must therefore be hereby marked classification may prove useful for prognosis. advertisement in accordance with 18 U.S.C. Section 1734 solely to Several proteomic studies have compared normal and can- indicate this fact. cerous breast cells [reviewed in detail by Hondermarck et al. Requests for reprints: Edward Sauter, Department of Surgery, Uni- versity of Missouri-Columbia, Columbia, MO 65212. Phone: (573) 882- (7)]. A hierarchical clustering analysis of proteomes of normal 4471; Fax: (573) 884-4585; E-mail: [email protected]. and different stages of disease have shown that it is possible to ©2004 American Association for Cancer Research. distinguish between normal, benign, and cancerous breast tis- Downloaded from clincancerres.aacrjournals.org on September 25, 2021. © 2004 American Association for Cancer Research. Clinical Cancer Research 7501 sues on the basis of the protein profile (8). Page et al. (9) matrix-assisted laser desorption ionization–time-of-flight mass established an extensive map of the normal human luminal and spectrometry (MALDI-TOF MS) with different expression lev- basal (myoepithelial) breast cell proteome as the basis for future els based on whether a subject did or did not have breast comparisons with breast cancer cells. The authors observed 170 cancer, and (c) the subsequent validation of these differentially differentially displayed protein spots, and identified 51 of them. expressed proteins by using an ELISA on a large set of A subsequent proteomic study of breast ductal carcinoma iden- NAF specimens from women with and without breast cancer. tified 57 proteins that were differentially expressed based on Three candidate proteins: gross cystic disease fluid protein disease stage, 10 of which were validated by immunohistochem- (GCDFP)-15 (also know as prolactin-induced protein or ical studies (10). Many were proteins involved in the regulation BRST2), lipoprotein D [apolipoprotein D (apoD)], and ␣1-acid of intracellular trafficking, cytoskeletal architecture, chaperone glycoprotein (AAG), were differentially expressed in a small set function, apoptosis, and genome instability. In a more focused of cancer versus benign samples analyzed by two-dimensional study, the molecular chaperone 14-3-3 was shown to be in- PAGE and subsequently examined by ELISA in a large set of volved in the transition of breast epithelial cells to neoplasia. By NAF samples comparing women with or without breast cancer, virtue of its role as a tumor suppressor, it was suggested that as well as women with in situ (early stage) versus invasive 14-3-3 may be a useful marker to identify cells that have (more advanced stage) disease. We demonstrate that the levels undergone this transition (11). of GCDFP-15 are substantially lower and AAG are substantially Interestingly, many proteins that have been identified by higher in samples from women with breast cancer, whereas proteomic studies are different from those found by nucleic levels of apoD in NAF were not associated with disease. This acid-based studies. This underscores the importance of perform- stepwise approach to biomarker detection and validation sug- ing biomarker screens at the protein level and suggests that gests that proteomic analysis of bodily fluids such as NAF are many of the differences between normal and cancer samples are appropriate for biomarker detection, can be readily validated by due to posttranslational modifications such as glycosylation or using high-throughput, quantitative technologies and may prove truncation (12). However, although promising, proteomic anal- useful for early breast cancer detection and/or prognosis predic- yses of breast tissues, to date, suffer from the fact that the tissues tion. are not homogeneous and more than one cell type is likely to be present in the samples. For this reason some of the above studies were done on cells in culture, although the relevance of the in MATERIALS AND METHODS vitro physiology and biochemistry in these studies to that of the Subjects. NAF specimens from subjects, 17–83 years normal or diseased human breast is unclear. Many of the studies old, were collected after Institutional Review Board approval that use cells have found few proteins with substantial differ- and informed consent were obtained. Some subjects had spec- ential expression. Moreover, studies requiring a large cellular imens analyzed from one breast only, whereas others had spec- sample require an invasive procedure to obtain the samples for imens analyzed from both breasts. The presence or absence of analysis. cancer was based on whether or not disease was present in a Nipple aspiration is a noninvasive, low-cost procedure that given breast. Fifty-three specimens were collected from 38 provides a relatively small set of breast-specific proteins. Be- patients without evidence of cancer in the analyzed breast. The cause the proteins are secreted, they represent the final pro- remaining 52 specimens were obtained from 51 subjects that cessed form of the protein, which makes proteomic analyses less had pathologically confirmed ductal carcinoma in situ (DCIS) or ambiguous and can provide clues to changes in protein transla- invasive breast cancer.