Cytogenetic and Cdna Microarray Expression Analysis of MCF10 Human Breast Cancer Progression Cell Lines

Published OnlineFirst July 7, 2009; DOI: 10.1158/0008-5472.CAN-09-0420

Molecular Biology, Pathology, and Genetics

Cytogenetic and cDNA Microarray Expression Analysis of MCF10 Human Breast Cancer Progression Cell Lines

Narasimharao V. Marella,1 Kishore S. Malyavantham,1 Jianmin Wang,2 Sei-ichi Matsui,3 Ping Liang,2 andRonaldBerezney 1

1Department of Biological Sciences, University at Buffalo, State University of New York; 2Department of Cancer Genetics and 3SKY Core Resource Facility, Roswell Park Cancer Institute, Buffalo, New York

Abstract derived cell lines: (a) the spontaneously immortalizedcell line MCF10A (5), which do not show any characteristics of invasiveness We used a combination of spectral karyotyping, array comparative genomic hybridization, and cDNA microarrays to gain or tumor formation andhence are consideredto be a normal-like breast epithelial cell line (6); (b) MCF10A cells were transformedwith insights into the structural and functional changes of the c genome in the MCF10 human breast cancer progression model c-Ha-Ras to yieldthe premalignant MCF10AT1 cell line (7, 8); and( ) cell lines. Spectral karyotyping data showed several chromo- MCF10AT1 cells were xenograftedonto mice anda thirdcell line, somal aberrations and array comparative genomic hybridiza- MCF10CA1a, was selected, which shows all the characteristics of a tion analysis identified numerous genomic gains and losses that fully malignant breast cancer cell type (9). Karyotypic analysis of might be involved in the progression toward cancer. Analysis of these three cell lines showeda characteristic (3;9) translocation that the expression levels of genes located within these genomic confirmeda common lineage (9). regions revealed a lack of correlation between chromosomal Spectral karyotyping (10) has been employedpreviously to study chromosomal aberrations in breast cancer cell lines andpatient gains and losses and corresponding up-regulation or down- regulation for the majority of the f1,000 genes analyzed in this tumors andledto the identification of numerous chromosomal study. We conclude that other mechanisms of gene regulation translocations, deletions, and rearrangements (11). Array compara- that are not directly related to chromosomal gains and losses tive genomic hybridization (aCGH) is useful in the detection of copy number variations that arise due to the genomic instability in cancer play a major role in breast cancer progression. [Cancer Res cells or tissues (12) andhas ledto the detection of gains within 2009;69(14):5946–53] chromosome arms 1q, 3p, 4q, 8q, 11q, 17q, and20q andlosses in 6q,11q, 8p, 9p, 13q,16q, and17p (13) in breast cancer. In most of Introduction these studies, the altered regions have been shown to harbor essential gene(s), the deregulation of which leads to the establishment of Breast cancer is the most common cancer andaccounts for the tumorigenesis (3). cDNA microarray analysis makes it feasible to secondhighest mortality rate worldwideamong cancer-related study the expression patterns of thousands of genes in cancer (14). deaths in women (1). Numerous studies have indicated that This technique has resultedin the identification of numerous initiation andprogression towarda breast cancer phenotype is a changes in gene expression associatedwith breast cancer (15). multistep process involving accumulation of genomic aberrations In the present study, we attempted to correlate the changes (2). In particular, amplifications, deletions, and rearrangements have observedin the immortalizednormal human breast epithelial cell been observedin breast cancer of critical genes involvedin cell line MCF10A andits malignant counterpart, MCF10CA1a, by growth, differentiation, and cell death (3). Breast cancers can be combinedspectral karyotyping, aCGH andcDNA microarray either hereditary or sporadic. Germ-line mutations in BRCA1 and analysis. We observednumerous changes in gene expression BRCA2 (4) genes have been linkedto the developmentof familial patterns that were indicative of tumor progression. Thus, combining breast cancer. these three approaches has provided insights into how DNA copy Although the genomic andmolecular changes in breast cancer number variations affect the gene expression patterns andfor the have been studied in detail, the initial events leading to tumor identification of new candidate genes that might be associated with formation remain to be elucidated. Cancer progression models have breast cancer andits progression. become an invaluable tool in studying the precise genetic aberrations that correlate with a shift from a normal to a disease phenotype. In this investigation, we usedthe MCF10 human breast Materials and Methods cancer progression model cell lines to study the genetic changes that Cell culture. The breast cancer progression model cell lines (MCF10A, occur during the transformation of breast epithelial cells into breast MCF10AT1, andMCF10CA1a) usedin these studieswere obtainedfrom cancer. The MCF10 progression model consists of three directly the Barbara Ann Karmanos Center. The MCF10A cell line was culturedin DMEM supplementedwith horse serum (5%), insulin (2.5 mg/mL), epidermal growth factor (50 Ag/mL), cholera enterotoxin (150 Ag/mL), hydrocortisone Note: Current address for N.V. Marella: Cancer Genetics, Inc., 201 Meadows Office (2.5 mg/mL), andHEPES (5 mmol/L), whereas MCF10AT1 andMCF10CA1a Complex, Route 17 North, Rutherford, NJ 07070. Current address for K.S. Malyavantham: cell lines were culturedin DMEM supplementedwith only horse serum (5%). IMMCO Diagnostics Ltd., 60 Pine View Drive, Buffalo, NY 14228. Current address for j P. Liang: Department of Biological Sciences, Brock University, St. Catharines, Ontario, All three cell lines were grown at 37 C in a 5% CO2 substitutedincubator. Cells Canada L2S 3A1. were collectedat the same passage andusedfor each of the analyses (spectral Requests for reprints: RonaldBerezney, Department of Biological Sciences, karyotyping, CGH, andgene expression microarray) donein this study. University at Buffalo, State University of New York, Buffalo, NY 14260. Phone: 716-645- Spectral karyotyping. Preparations of metaphase chromosome spreads 2363, ext. 154; Fax: 716-645-2975; E-mail: [email protected]. I2009 American Association for Cancer Research. were subjectedto the spectral karyotyping procedurerecommendedby doi:10.1158/0008-5472.CAN-09-0420 AppliedSpectral Imaging. The images were capturedusing a combination of

Cancer Res 2009; 69: (14). July 15, 2009 5946 www.aacrjournals.org

Breast Cancer Tumor Progression rhodamine, Texas red, Cy5, FITC, and Cy5.5 filter sets mounted on a Nikon elutedin 30 AL RNase-free water. The cRNA quality andconcentration were fluorescence microscope equippedwith a spectral cube andan interferom- assessedusing a NanoDrop ND-1000 UV-visible spectrophotometer eter module. Karyotypic analysis of the images was carried out using the (ThermoScientific).cRNAsampleswererequiredtohaveayield>750ng Spectral Karyotyping View software. anda specific activity >8.0 pmol Cy3 or Cy5/A g cRNA to proceedto the aCGH array analysis. DNA printing solutions were preparedfrom hybridization step. sequence connected RPCI-11 BAC by ligation-mediated PCR as described Before hybridization, 0.75 Ag Cy5-labeled, linearly amplified cRNA (test) previously (16). The array contains f19,000 BAC clones that were chosen by and0.75 Ag Cy3-labeled, linearly amplified cRNA (control) were combined virtue of their STS content, endsequence, andassociation with heritable andfragmentedby incubating at 60 jC for 30 min in fragmentation buffer. The disorders and cancer (17). Each clone is printed in duplicate on amino- hybridization reaction mixture was prepared by combining 2Â hybridization silanatedglass slides(Schott Nexterion type A+) using a MicroGridll TAS buffer to the fragmentedcRNA to a final volume of 210 AL. The hybridization arrayer (Genomic Solutions). The BAC DNA products have f80 Am diameter reaction mixture was placedon ice andthen loadedslowly onto the 44K array, spots with 150 Am center to center spacing creating an array of f39,000 allowing even flow and distribution of hybridization cocktail across the elements. The printed slides dry overnight and are UV crosslinked (350 mJ) in surface. The arrays were securedin a SureHyb chamber cover andthen placed a Stratalinker 2400 (Stratagene) immediately before hybridization. in a rotisserie hybridization oven at 65jC for 17 h. After hybridization, the Reference andtest sample genomic DNA (1 Ag each) were individually slides were removed from the hybridization oven and washed with Gene fluorescently labeledusing the BioPrime DNA labeling kit (Invitrogen) for 18 h Expression Wash Buffers 1 and2. Following the wash steps, the array images at 37jC with the appropriate Cy dye (Cy3 or Cy5). After ethanol pre- were capturedby scanning at 532 and635 nm with an Agilent scanner and cipitation, the probes are resuspended in H2O, combined, and purified of analyzedwith Agilent feature extraction software version 8.5. unincorporateddye by passage over a Qiagen spin column. Before hy- Data analysis for aCGH and cDNA microarrays. The 19K aCGH array bridization, the test and reference probes were resuspended in 110 AL was analyzedusing DNAcopy R package, which uses the circular binary SlideHyb Buffer 3 (Ambion) containing 5 ALof20Ag/AL Cot-1 and5 ALof segmentation algorithm (19). A segment was identified as gain or loss if the 100 Ag/AL yeast tRNA (Invitrogen), heatedto 95 jC for 5 min, andplacedon estimatedcopy number ratio is more than 0.2 or less than À0.2 ice. Hybridization to the 6K BAC arrays was done for 16 h at 55jC using a (corresponding to a genomic fold increase of 1.15 and decrease of 1.15), GeneTAC hybridization station (Genomic Solutions) as described (18). The which is roughly eight times the SE log ratio on each chip. For Agilent gene hybridized aCGH slides are scanned using a GenePix 4200A scanner expression arrays using the dye-swap design, significantly changed genes are (Molecular Devices) to generate high-resolution (5 Am) images for both Cy3 identified using limma R package (20), which uses a modified t test with (test) andCy5 (control) channels. Image analysis was doneusing the ImaGene multiple tests in consideration. The gene ontology analysis for the significant changedgenes was doneby GOstat software (21). To better plot the aCGH version 6.1.0 (Bio Discovery). The log2 test/control ratios were normalized using a subgrid Loess correction. Mapping information was added to the results, an in-house ideogram and aCGH data plotting program was developed and used. resulting log2 test/control values. cDNA microarray analysis. cDNA microarray analysis was done on an Agilent 44K whole human genome oligo microarray. Total RNA from the cell Results cultures (MCF10A andMCF10CA1a) were preparedusing the RNeasy mini kits (Qiagen) following the manufacturer’s instructions. After elution, RNA Expression analysis of normal and malignant breast cell lines. samples were concentratedby ethanol precipitation at À20jC overnight and Microarray expression analysis of immortalizednormal MCF10A resuspended in nuclease-free water. Before labeling, RNA samples were and malignant MCF10CA1a cell lines was done to determine the quantifiedusing a Genequant spectrophotometer (GE Healthcare) and changes in gene expression patterns. A total of 42,000 genes yielded evaluated for degradation using a 2100 Bioanalyzer (Agilent Technologies). hybridization signals for both samples (Supplementary Table S1). To screen the samples for gene expression, cRNA was synthesizedby Analysis of the data yielded a total of f7,000 genes that showedat in vitro transcription anddirectly labeledwith Cy3-CTP or Cy5-CTP using the least 2-foldexpression level changes ( f3,044 genes were up- low RNA input linear amplification kit as per the manufacturer’s instructions regulatedandf 3,829 genes were down-regulated) in the (Agilent Technologies). Initially, first-strandcDNA was synthesizedby MCF10CA1a cell line (Supplementary Table S2). incubating 0.2 to 2 Ag total RNA andspiking controls andT7 oligo(dT)for 10 min at 65jC. Following the addition of first-strand reaction components, A group of genes that were either highly up-regulatedor greatly the reaction continuedfor 2 h at 40 jC. cRNA was synthesizedby incubating repressedin the MCF10CA1a cell line is listedin Table 1. All these the resuspended double-stranded cDNA with the transcription master mix genes were chosen basedon their functional involvement in breast containing cyanine-CTP, ribonucleotides, and T7 RNA polymerase for 2 h at cancer. Up-regulatedgenes included SEPP1 and DCN, which have an 40jC. The cyanine-labeledcRNA was recoveredby column purification and antimetastatic role in breast cancer (22–24); FBN1, AOX1, and

Table 1. Highly up-regulated and down-regulated genes in MCF10CA1a involved in breast cancer

Gene Fold change (log fold change MCF10CA1a/MCF10A)Chromosome location

Selenoprotein plasma protein (SEPP1) 7.37 5p12 Angiopoietin 1 (ANGPT1) 6.77 8q23.1 Decorin (DCN) 6.70 12q21.33 Fibrillin 1 (FBN1) 5.59 15q21.1 Prostaglandin E receptor 2 (PTGER2) 5.29 14q22.1 Aldehyde oxidase 1 (AOX1) 5.24 2q33.1 Aldehyde dehydrogenase 1 family, member A3 (ALDH1A3) À8.72 15q26.3 E-cadherin (epithelial) (CDH1) À8.65 16q22.1 Interleukin-1h (IL1B) À8.11 16q22.1 S100 calcium binding protein A14 (S100A14) À6.15 1q21.3 Bradykinin receptor B2 (BDKRB2) À6.07 14q32.2

www.aacrjournals.org 5947 Cancer Res 2009; 69: (14). July 15, 2009

Cancer Research

Table 2. Selected down-regulated genes in MCF10CA1a coding for functionally relevant proteins

Genes Expression Chromosome location (log fold change MCF10CA1a vs MCF10A)

Extracellular matrix Keratin 6 (KRT6 ) À6.36 12q13.13 Keratin 8 (KRT8 ) À1.26 12q13.13 Keratin 13 (KRT13) À5.47 17q21.2 Keratin 14 (KRT14) À3.82 17q21.2 Keratin 15 (KRT15) À5.14 17q21.2 Keratin 16 (KRT16) À3.99 17q21.2 Keratin 17 (KRT17) À3.79 17q21.2 Keratin 18 (KRT18) À2.22 12q13.13 Keratin 19 (KRT19) À2.31 17q21.2 Keratin 23 (KRT23) À5.70 17q21.2 Collagen XIII, a1(COL13A1) À6.84 10q22.1 Cell communication Connexin 26 (GJB2) À4.81 13q12.11 Connexin 43 (GJA1) À6.05 6q22.31 Claudin 1 (CLDN1) À5.05 3q28 Claudin 4 (CLDN4) À4.23 7q11.23 Claudin 7 (CLDN7) À3.20 17p13.1 Claudin 12 (CLDN12) À1.13 7q21.13 Signal transduction Vascular endothelial growth factor (VEGF) À1.20 6p21.1 V-raf murine sarcoma viral oncogene homologue B1 (BRAF) À1.24 7q34 Erythroblastic leukemia viral oncogene homologue 2 (ERBB2) À1.57 17q12 Epidermal growth factor receptor (EGFR) À1.61 7p11.2 Harvey rat sarcoma viral oncogene homologue (HRAS) À2.64 11p15.5 Oncogenes andtumor suppressors Breast cancer 2 (BRCA2) À1.42 13q13.1 Myelocytomatosis viral oncogene (MYC) À1.55 8q24.21 Phosphatase andtensin (PTEN ) À4.28 10q23.31 Cytokines Interleukin-A (IL1A) À6.67 2q13 Interleukin-B (IL1B) À8.06 2q13 Interleukin-6 (IL6) À4.20 7p15.3 Interleukin-8 (IL8) À4.71 4q13.3 Interleukin-11 (IL11) À2.50 19q13.42

PTERG2, which are known to promote tumor growth (25–27); and, expressedat higher levels. Finally, many genes that are involvedin surprisingly, ANGPT1, which when overexpressedhas antitumor the formation of the extracellular matrix such as COL6A1, KRT7, properties (28). FBN1, MMP2, MUC1, and FN1 showedhigher levels of expression in Our results showeda down-regulation of CDH1. Deregulation of MCF10CA1a (Table 3). this gene was observedpreviously in breast andother types of cancer Spectral karyotyping analysis of MCF10 breast cancer prog- (29). The IL1B and S100A14 genes were both greatly repressed, which ression model cell lines. Spectral karyotyping identified several is contrary to the reportedhigh levels of protein expression of these chromosomal rearrangements in each of the three breast cancer cell two proteins in breast andother cancers (30). Moreover, the BDKRB2 lines (Fig. 1). A karyotype of 47 chromosomes was foundfor normal gene is known to induce mitosis in breast cells and enhance MCF10A andthe premalignant MCF10AT1 cell lines (Fig. 1 A and B; progression of cancer (31). Yet, our study shows that this gene was Supplementary Table S3), whereas the malignant MCF10CA1a cell significantly repressed. line had50 chromosomes (Fig. 1 C; Supplementary Table S3). Four Further analyses of the microarray data were done to establish the marker chromosomes were identified in MCF10A and MCF10AT1 relationships between expression levels andgene function (Tables 2 andnine in the malignant MCF10CA1a cell line (Supplementary and3). Several keratin, connexin, andclaudingenes involvedin the Table S3). MCF10A showedthe characteristic (3;9) reciprocal transformation of the extracellular matrix andcell-cell communication location that has been described as the single most important event were down-regulated. Genes involved in signal transduction path- in the immortalization andtransformation of the MCF10 breast ways such as ERBB2, HRAS, BRAF, VEGF, and EGFR, were also down- epithelial cell line (32). Other karyotypic changes including a gain of regulatedin the cancer cell line. Oncogenes andtumor suppressors, chromosome 8 (+8) andtranslocations involving chromosomes 5, 3, including MYC, PTEN, and BRCA2, were down-regulated, whereas and9 [t(5;3;9)] andchromosomes 6 and19 [t(6;19); Supplementary RB1, CDKNB1, and CCND3 were overexpressed. Table S3]. Karyotypic changes in the MCF10AT1 included a reciprocal Several interleukin genes such as IL6, IL8, IL1A, and IL1B were translocation involving chromosomes 3 and17 [rept(3;17)] in addition repressedin MCF10CA1a cell line, whereas IL7 andIL18 genes were to those foundin MCF10A (Supplementary Table S3). K aryotypic

Cancer Res 2009; 69: (14). July 15, 2009 5948 www.aacrjournals.org

Breast Cancer Tumor Progression alterations in MCF10CA1a also included a gain of chromosome 20 are locatedwithin the chromosome bands8q12.1, 8q13.2, 8p23.1, (+20) andtranslocation between chromosomes 5 and9 [t(5;9)] and 2p14, 2p21, 20q13.3, and8p11.21, respectively. Among the genes chromosomes 2 and10 [t(2;10)] in addition to the changes that are foundwithin genomic gain regions, 13 were up-regulatedand18 already present in MCF10A (Supplementary Table S3). were down-regulated, whereas, in the loss regions, 3 genes were aCGH analyses of chromosomal gains and losses in the breast down-regulatedand6 genes were up-regulated(Supplementary cancer progression model cell lines. Analysis of aCGH data of the Table S4). This indicated that in many cases there was no direct three cell lines revealedchromosomal breakpoints that were reported relationship between gains or losses of chromosomal regions and previously (33) as well as several new chromosomal aberrations. Three gene expression in those regions. independent aCGH experiments were done in the following manner: We further investigatedthis question at the global level of gene MCF10A versus a normal diploid cell line, MCF10A versus MCF10AT1, expression. About 1,000 genes were identified in MCF10CA1a andMCF10A versus MCF10CA1a. Ideograms of the complete analysis that showedboth changes in genomic gains or losses andat least a are presentedin Supplementary Fig. S1. 2-foldchange in their expression levels (Supplementary Table S5). The MCF10A cell line showedgains at 5q23.1-35.3, 19q13.11q13.43, To our surprise, we foundthat all 6 regions showing chromosomal and13q32.1-p32.2 (Table 4; Supplementary Fig. S1 A). A gain of 8p23.3- gains hada much higher number of down-regulatedgenes and3 of q24.3 was seen in MCF10A as evidenced by the presence of an extra 5 regions with genomic losses showedhigher numbers of up- copy of chromosome 8 in this cell line comparedwith the normal regulatedgenes (Table 5). Of the 701 genes identifiedinthe diploid karyotype. Losses were observed at 9p21.3, 3p26.3, 16p11.2, genomic gains regions, 76% were down-regulated, whereas 56% of 21p11-q11.2, and22q11.1 (Table 4; Supplementary Fig. S1 A). The the 301 genes identified in the loss regions were up-regulated premalignant MCF10AT1 cell line showedgains at 3p14.3, 3q13.3, (Table 5; Supplementary Table S5). 19q12-q34.3, 10q22.1-q22.2, 16q23.3, and17p11.2 andlosses of 5q12.1, 5q14.3-15, and15q21.1 (Table 4; Supplementary Fig. S1 B). The additional chromosome 8 copy found in the normal MCF10A cell line is deleted in this cell line. The malignant MCF10CA1a cell line showed Discussion several more genomic aberrations comparedwith the normal and Cancer progression models provide an approach to elucidate the premalignant MCF10 cell lines. Gains include 2p25.3-q21.2, 3p14.1- intermediate molecular and genetic changes that ultimately lead to q29, 9p24.3-p11.2, 9p34.13-p34.3, 10q11.1-q26.3, 17p11.2, and20p13- the transformation of a normal cell into a metastatic cell type. In this q13.3 (Table 4; Supplementary Fig. S1C). Moreover, loss of genomic present study, we used a combination of spectral karyotyping and regions 2q21.2-q22.1, 5q12.1, 5q14.3-q15, À8p23.3-q24.3, and16q23.1 aCGH methods to identify regions of chromosomal aberrations in were observed(Table 4; Supplementary Fig. S1 C). These cytogenetic the MCF10 human breast cancer progression model cell lines (9). gains andlosses were concordant with the loss andgain of genes that Gene expression analysis was also done in the immortalized normal have been reportedpreviously (33) in these cell lines. MCF10A andmalignant MCF10CA1a cell lines as a step toward Correlation of chromosomal gains and losses with gene identifying novel candidate protein factors that potentially correlate expression. Gene expression changes for 40 genes foundin 36 with cancer metastasis. In addition, because cancer is characterized different genomic regions showing gains or losses in MCF10CA1a are by numerous alterations in the genome comparedwith normal cells, shown in Supplementary Table S4. The ratio of genomic gains range we have attemptedto correlate changes in gene expression with from 1.18 to 1.49, whereas losses range from 0.71 to 0.77. We also genomic gains or losses of the individual genes in the malignant identified 7 genes that are important in breast cancer within the cell line. regions of genomic loss/gain in MCF10CA1a (Supplementary About 7,000 genes were identified in MCF10CA1a that showed a Table S4, double asterisks). These genes include CA8, SULF1, 2-folddifference in expression level (Supplementary Table S2). We C8orf13, C2orf32, TACSTD1, COL9A3, andtissue PLAT. These genes further classifiedgenes basedon their relevance to breast cancer

Table 3. Selected up-regulated genes in MCF10CA1a coding for functionally relevant proteins

Genes Expression Chromosome location (log fold change MCF10CA1a vs MCF10A)

Extracellular matrix Collagen type VI, a1(COL6A1) 1.54 12q13.13 Fibrillin 1 (FBN1) 2.07 15q21.1 Mucin 1 (MUC1) 3.35 1q21.3 Matrix metalloproteinase-2 (MMP2) 3.62 16q12.2 Fibronectin 1 (FN1) 3.96 2q35 Keratin 7 (KRT7 ) 5.59 21q22.3 Oncogenes andtumor suppressors Retinoblastoma 1 (RB1) 1.14 13q14.2 Cyclin-dependent kinase inhibitor (CDKNB1) 1.18 12p13.1 Cyclin D3 (CCND3) 1.36 6p21.1 Cytokines Interleukin-7 (IL7) 1.13 8q21.12 Interleukin-18 (IL18) 1.67 11q23.1

www.aacrjournals.org 5949 Cancer Res 2009; 69: (14). July 15, 2009

Cancer Research

(Table 1). Consistent with previous reports in breast cancer, we Genes such as MYC, PTEN,andBRCA2 (Table 2), which are involved detected up-regulation in MCF10CA1a of FN1 and MMP2 (34, 35). in cell cycle control andDNA repair, are down-regulated.Germ-line In contrast, although overexpression of SEPP1 and DCN has been mutations in BRCA2 predisposes humans toward development of shown to have an antimetastatic role in breast cancer (23, 24), these inheritedbreast cancer (4) andstudies on PTEN showedthat genes are highly expressedin the malignant MCF10CA1a line deregulation of this gene is associated with increased cell proliferation comparedwith MCF10A (Table 1). Whereas numerous studies andtumorigenicity (37). Cytokines including IL1A and IL1B are highly showedthat overexpression or amplification of the genes ERBB2 down-regulated in MCF10CA1a (Table 2) despite their reported and EGFR promotes breast cancer tumorigenesis (36), we findthat contribution to the invasiveness andaggressive phenotype of breast they are significantly down-regulated in the MCF10CA1a cells cancer (38). In contrast, IL7, which has also been implicatedin cancer (Tables 2 and3). Repression of the HRAS gene in MCF10CA1a is progression (39), is significantly up-regulatedin MCF10CA1a (Table 3). consistent with the overexpression of a mutant form of this protein Previous spectral karyotyping showeda variety of chromosomal engineeredinto this cell line (7). rearrangements in breast cancer (11). In particular, the reciprocal

Figure 1. Spectral karyotyping analysis of the MCF10 breast cancer progression model cell lines. A, MCF10A. B, MCF10AT1. C, MCF10CA1a.

Cancer Res 2009; 69: (14). July 15, 2009 5950 www.aacrjournals.org

Breast Cancer Tumor Progression

Table 4. aCGH analysis of the MCF10 breast cancer progression model cell lines

Cell line aCGH

Gains Losses

MCF10 A +5q23.1-35.3, +8p23.3-q24.3, +13q32.1-p32.2, +19q13.11q13.43 À3p26.3, À9p21.3, À16p11.2, À21p11-q11.2, À22q11.1 MCF10AT1 +3p14.3, +3q13.31, +9p24.3-11.2, +9q12-q34.3, +10q22.1-q22.2, À5q12.1, À5q14.3-15, À8p23.3-q24.3, À15q21.1 +16q23.3, +17p11.2 MCF10CA1a +2p25.3-q21.2, +3p14.1-q29, +9p24.3-p11.2, +9q34.3-q34.13, À2q21.2-q22, À5q12.1, À5q14.3-q15, À8p23.3-q24.3,-16q23.1 +10q11.1-q26.3, +17p11.2, +20p13-q13.33

translocation involving chromosomes 3 and9 [t(3;9)(p14;p21)] karyotyping method. Thus, combining both these approaches in identified in our analysis has been described as the single most our cell system has allowedus to identify large-scale genomic prominent event in the transformation of this cell type (32, 33). We aberrations andploidychanges along with regions of subchromo- also observedseveral other translocations involving chromosomes 2, somal abnormalities that might occur in the transformation of a 3, 5, 10, and17. Aberrations in chromosomes 3 and17 were reported normal cell to a cancerous cell. Overall, our spectral karyotyping and previously in breast cancer (40). Chromosome 17 has several putative aCGH results indicate that a hallmark of progression toward breast breakpoints that might be associatedwith higher frequency of cancer involves gradual accumulation of chromosomal aberrations translocations of this chromosome. A gain of chromosome 8 was in normal breast cells. observedin MCF10A cells, which was lost in the subsequent We further correlatedgenomic alterations with gene expression premalignant (MCF10AT1) andmalignant (MCF10CA1a) progres- profiles of normal andmalignant MCF10 cell lines. Regions within sion cell lines. We also identified a gain of an entire chromosome 20 the MCF10CA1a cell line identified for gains and losses by aCGH in the malignant breast cell line. Consistent with these findings, were examinedfor changes in levels of gene expression (Table 5; chromosomal rearrangements andamplification of chromosomes Supplementary Table S4). The majority of the f1,000 genes in- 8 (41) and20 (42) have been observedfrequently in breast cancer. vestigatedshoweda lack of direct correlation between the Using aCGH, the parental MCF10A breast cell line showedgains in chromosomal aberrations andits expression (Table 5). Previous 5q23.1-35.3, 19q13.11q13.43, and13q32.1-p32.2 regions andloss of studies also reported a weak, or in some cases, a complete lack of 9p21.3, 3p26.3, 16p11.2, 21p11-q11.2, and22q11.1 regions in com- correlation between genomic gains andlosses andgene expression parison with a normal diploid cell line (Supplementary Fig. S1A; levels in cancer cells (47). Despite this, we identified several genes Table 4). The loss of the 9p21.3 region is consistent with the findings in these regions that are relatedto breast cancer andshow large that a deletion in this region was associated with the t(3;9) changes in gene expression in the malignant MCF10CA1a versus the translocation event that leads to immortalization of this cell line MCF10A breast epithelial cells (Supplementary Table S4). (32). A gain was also detected in the 8p23.3-q24.3 region, which is Our results predict that progression to cancer is a com- consistent with the gain of an additional copy of chromosome 8 in bination of genomic instability as well as gene deregulation. the spectral karyotyping analysis. Gain of 3p14.3, 3q13.3, 19q12-q34.3, Chromosomal aberrations such as amplifications, deletions, and 10q22.1-q22.2, 16q23.3, and17p11.2 andloss of 5q12.1, 5q14.3-15, and complex chromosomal translocations are a hallmark of solidtumors 15q21.1 were observedin the premalignant MCF10AT1 cell line andare generally believedto occur via telomere dysfunction and (Table 4; Supplementary Fig. S1B). Cytogenetic changes in chromo- the breakage-fusion-bridge mechanism (48). Changes in gene copy somes 17 (43) and5 (44) have been reportedin other studies.In particular, losses in the 5q13-q23.3 region have been associatedwith Table 5. Relationships between genomic gains and breast cancers showing a mutation for the BRCA1 gene (44). This losses and gene expression in MCF10CA1a region has also been shown to harbor several putative tumor suppressor genes. Region No. up-regulated No. down-regulated Additional changes in MCF10CA1a include gains of 2p25.3-q21.2, genes genes 3p14.1-q29, 9p24.3-p11.2, 9p34.13-p34.3, 10q11.1-q26.3, and20p13- q13.3 andlosses of genomic regions 2q21.2-q22.1, 5q12.1, 8p23.3- Gains q24.21, and16q23.1 (Table 4; Supplementary Fig. S1C ). The +2p25.3-q21.2 44 122 additional chromosome 8 seen in MCF10A was deleted in both +3p4.1-q29 60 142 premalignant andmalignant cell lines. Deletions of certain genomic +9p24.3-p11.2 11 43 regions or the entire chromosome 8 are in agreement with previous +9q34.13-q34.3 3 24 studies on chromosome 8 abnormalities in breast cancer (45). The +10q11.1-q26.3 48 170 region 8p11-12 harbors f21 potential oncogenes andis frequently +7p11.2 3 31 amplifiedin breast cancer (46). The results of spectral karyotyping Losses À2q21.2-q22 10 13 andaCGH were in concurrence in nearly all instances thereby À5q12.1 3 2 substantiating the results from both approaches. À5q14.3-q15 16 9 A careful analyses of the aCGH findings revealed that almost all À8p23.3-q24.3 150 85 the subchromosomal genomic gains andlosses were in those À16q23.1 4 9 chromosomes that were identified as aberrant by the spectral www.aacrjournals.org 5951 Cancer Res 2009; 69: (14). July 15, 2009

Cancer Research number that occur due to the chromosomal aberrations lead to a metastatic. In cancer cells, methylation of cytosine in the CpG disruption of the normal expression pattern of the genes, therefore dinucleotides occurs in the promoter regions, which leads to gene causing transformation of a normal cell towarda malignant inactivation (50). Several genes that are important in cell cycle phenotype. control, DNA repair, andapoptosis have been foundto undergo In conclusion, our combinedapproach employing spectral hypermethylation, thereby promoting the development of a karyotyping, aCGH, andmicroarray gene expression analysis cancerous phenotype (50). Modification of histone tails such as identified several important chromosomal regions and genes with acetylation, deacetylation, and methylation are also known to structural andfunctional alterations that might be involvedin the influence gene expression (50), andin cancer, abnormal histone progression anddevelopmentof breast cancer. Importantly, several modifications that occur have shown to have an effect on gene of these genomic regions that have been shown previously to be transcription (50). Recent investigations also show a role of amplifiedor deletedin breast cancer (49). Although the expression microRNAs in repression of certain important genes in cancer (50). levels of many genes agreedwith previously publishedresults in Our studies, therefore, add to the growing evidence that amplifica- breast cancer, there were some significant differences. This could tion or deletions of genomic regions are not the sole mechanism for relate to the multiple changes andlikely variability in gene altering gene expression in progression towardcancer. expression that occur during cancer progression. Thus, a variety of combinations of expression changes in key regulatory factors may leadto the same endof a highly malignant tumor. Disclosure of Potential Conflicts of Interest We also report a lack of correlation between chromosomal No potential conflicts of interest were disclosed. aberrations in cancer andexpression patterns of genes within those regions. Other mechanisms of gene regulation are likely involved including epigenetic processes such as methylation (50), histone Acknowledgments modification (50), and expression of noncoding RNAs (50). Received2/3/09; revised5/12/09; accepted5/19/09; publishedOnlineFirst 7/7/09. Methylation of CpG residues and histone tail modifications are Grant support: NIH grant GM-072131 (R. Berezney). The costs of publication of this article were defrayed in part by the payment of page faithfully propagatedfrom one cell generation to another under charges. This article must therefore be hereby marked advertisement in accordance with normal circumstances andany change might leadthe cell to become 18 U.S.C. Section 1734 solely to indicate this fact.

12. Kallioniemi A, Kallioniemi OP, Sudar D, et al. 24. Goldoni S, Seidler DG, Heath J, et al. An antimeta- References Comparative genomic hybridization for molecular static role for decorin in breast cancer. Am J Pathol 1. Bray F, McCarron P, Parkin DM. The changing global cytogenetic analysis of solidtumors. Science 1992;258: 2008;173:844–55. patterns of female breast cancer incidence and mortal- 818–21. 25. Tseleni-Balafouta S, Gakiopoulou H, Fanourakis G, ity. Breast Cancer Res 2004;6:229–39. 13. Jonsson G, Staaf J, Olsson E, et al. High- et al. Fibrillin expression andlocalization in various 2. Ellsworth RE, Vertrees A, Love B, Hooke JA, Ellsworth resolution genomic profiles of breast cancer cell types of carcinomas of the thyroidgland.ModPathol DL, Shriver CD. Chromosomal alterations associated lines assessedby tiling BAC array comparative 2006;19:695–700. with the transition from in situ to invasive breast genomic hybridization. Genes Chromosomes Cancer 26. Wright RM, McManaman JL, Repine JE. Alcohol- cancer. Ann Surg Oncol 2008;15:2519–25. 2007;46:543–58. induced breast cancer: a proposed mechanism. Free 3. Reis-Filho JS, Savage K, Lambros MB, et al. Cyclin D1 14. DeRisi J, PenlandL, Brown PO, et al. Use of a cDNA Radic Biol Med 1999;26:348–54. protein overexpression andCCND1 amplification in microarray to analyse gene expression patterns in 27. Wang D, Dubois RN. Prostaglandins and cancer. Gut breast carcinomas: an immunohistochemical andchro- human cancer. Nat Genet 1996;14:457–60. 2006;55:115–22. mogenic in situ hybridisation analysis. Mod Pathol 2006; 15. Nuyten DS, van de Vijver MJ. Using microarray 28. Hayes AJ, Huang WQ, Yu J, et al. Expression and 19:999–1009. analysis as a prognostic andpredictivetool in oncology: function of angiopoietin-1 in breast cancer. Br J Cancer 4. Teng LS, Zheng Y, Wang HH. BRCA1/2 associated focus on breast cancer andnormal tissue toxicity. Semin 2000;83:1154–60. hereditary breast cancer. J Zhejiang Univ Sci B 2008;9: Radiat Oncol 2008;18:105–14. 29. Nollet F, Berx G, van Roy F. The role of the E- 85–9. 16. Nowak NJ, Snijders AM, Conroy JM, Albertson DG. cadherin/catenin adhesion complex in the development 5. Pauley RJ, Soule HD, Tait L, et al. The MCF10 family of The BAC resource: tools for array CGH andFISH. Curr andprogression of cancer. Mol Cell Biol Res Commun spontaneously immortalizedhuman breast epithelial Protoc Hum Genet 2005;Chapter 4:Unit 4 13. 1999;2:77–85. cell lines: models of neoplastic progression. Eur J Cancer 17. Nowak NJ, Gaile D, Conroy JM, et al. Genome-wide 30. Pietas A, Schluns K, Marenholz I, Schafer BW, Heizmann Prev 1993;2:67–76. aberrations in pancreatic adenocarcinoma. Cancer CW, Petersen I. Molecular cloning andcharacterization of the 6. Soule HD, Maloney TM, Wolman SR, et al. Isolation Genet Cytogenet 2005;161:36–50. human S100A14 gene encoding a novel member of the S100 andcharacterization of a spontaneously immortalized 18. Cowell JK, Wang YD, HeadK, Conroy J, McQuaidD, family. Genomics 2002;79:513–22. human breast epithelial cell line, MCF-10. Cancer Res Nowak NJ. Identification and characterisation of con- 31. Greco S, Elia MG, Muscella A, Romano S, Storelli 1990;50:6075–86. stitutional chromosome abnormalities using arrays of C, Marsigliante S. Bradykinin stimulates cell 7. Basolo F, Elliott J, Tait L, et al. Transformation of bacterial artificial chromosomes. Br J Cancer 2004;90: proliferation through an extracellular-regulated human breast epithelial cells by c-Ha-ras oncogene. Mol 860–5. kinase 1 and2-dependent mechanism in breast Carcinog 1991;4:25–35. 19. Olshen AB, Venkatraman ES, Lucito R, Wigler M. cancer cells in primary culture. J Endocrinol 2005;186: 8. Dawson PJ, Wolman SR, Tait L, Heppner GH, Miller Circular binary segmentation for the analysis of array- 291–301. FR. MCF10AT: a model for the evolution of cancer basedDNA copy number data.Biostatistics 2004;5: 32. Cowell JK, LaDuca J, Rossi MR, Burkhardt T, Nowak from proliferative breast disease. Am J Pathol 1996; 557–72. NJ, Matsui S. Molecular characterization of the t(3;9) 148:313–9. 20. Smyth GK. Linear models and empirical Bayes associatedwith immortalization in the MCF10A cell 9. Santner SJ, Dawson PJ, Tait L, et al. Malignant methods for assessing differential expression in micro- line. Cancer Genet Cytogenet 2005;163:23–9. MCF10CA1 cell lines derived from premalignant human array experiments. Stat Appl Genet Mol Biol 2004;3: 33. Worsham MJ, Pals G, Schouten JP, et al. High- breast epithelial MCF10AT cells. Breast Cancer Res Article3. resolution mapping of molecular events associatedwith Treat 2001;65:101–10. 21. Beissbarth T, SpeedTP. GOstat: findstatistically immortalization, transformation, andprogression to 10. Macville M, Veldman T, Padilla-Nash H, et al. Spectral overrepresentedgene ontologies within a group of breast cancer in the MCF10 model. Breast Cancer Res karyotyping, a 24-colour FISH technique for the genes. Bioinformatics 2004;20:1464–5. Treat 2006;96:177–86. identification of chromosomal rearrangements. Histo- 22. Squires J, Berry MJ. Selenium, selenoproteins, and 34. Christensen L. The distribution of fibronectin, laminin chem Cell Biol 1997;108:299–305. cancer. Hawaii MedJ 2006;65:239–40. andtetranectin in human breast cancer with special 11. Goodison S, Viars C, Urquidi V. Molecular cytoge- 23. Baliga MS, Wang H, Zhuo P, Schwartz JL, Diamond attention to the extracellular matrix. APMIS Suppl 1992; netic analysis of a human breast metastasis model: AM. Selenium andGPx-1 overexpression protect mam- 26:1–39. identification of phenotype-specific chromosomal rear- malian cells against UV-induced DNA damage. Biol 35. Mendes O, Kim HT, Lungu G, Stoica G. MMP2 role in rangements. Cancer Genet Cytogenet 2005;156:37–48. Trace Elem Res 2007;115:227–42. breast cancer brain metastasis development and its

Cancer Res 2009; 69: (14). July 15, 2009 5952 www.aacrjournals.org

Breast Cancer Tumor Progression

regulation by TIMP2 andERK1/2. Clin Exp Metastasis 41. Mark HF, Taylor W, Afify A, et al. Stage I andstage II 46. Yang ZQ, Streicher KL, Ray ME, Abrams J, Ethier SP. 2007;24:341–51. infiltrating ductal carcinoma of the breast analyzed for Multiple interacting oncogenes on the 8p11-p12 am- 36. Moasser MM. The oncogene HER2: its signaling and chromosome 8 copy number using fluorescent in situ plicon in human breast cancer. Cancer Res 2006;66: transforming functions andits role in human cancer hybridization. Pathobiology 1997;65:184–9. 11632–43. pathogenesis. Oncogene 2007;26:6469–87. 42. Hodgson JG, Chin K, Collins C, Gray JW. Genome 47. JiangM,LiM,FuX,etal.Simultaneouslyde- 37. Maehama T. PTEN: its deregulation and tumorigen- amplification of chromosome 20 in breast cancer. Breast tection of genomic andexpression alterations in pros- esis. Biol Pharm Bull 2007;30:1624–7. Cancer Res Treat 2003;78:337–45. tate cancer using cDNA microarray. Prostate 2008;68: 38. Balasubramanian SP, Azmy IA, Higham SE, et al. 43. Andersen CL, Monni O, Wagner U, et al. High- 1496–509. Interleukin gene polymorphisms andbreast cancer: a throughput copy number analysis of 17q23 in 3520 48. Chin K, de Solorzano CO, Knowles D, et al. In situ case control study and systematic literature review. tissue specimens by fluorescence in situ hybridization to analyses of genome instability in breast cancer. Nat BMC Cancer 2006;6:188. tissue microarrays. Am J Pathol 2002;161:73–9. Genet 2004;36:984–8. 39. Al-Rawi MA, Mansel RE, Jiang WG. Interleukin-7 (IL-7) and 44. Johannsdottir HK, Jonsson G, Johannesdottir G, et al. 49. Hampton OA, Den Hollander P, Miller CA, et al. A IL-7 receptor (IL-7R) signalling complex in human solid Chromosome 5 imbalance mapping in breast tumors sequence-level map of chromosomal breakpoints in tumours. Histol Histopathol 2003;18:911–23. from BRCA1 andBRCA2 mutation carriers andsporadic the MCF-7 breast cancer cell line yields insights 40. Anamthawat-Jonsson K, EyfjordJE, Ogmundsdottir breast tumors. Int J Cancer 2006;119:1052–60. into the evolution of a cancer genome. Genome Res HM, Petursdottir I, Steinarsdottir M. Instability of 45. Forozan F, Mahlamaki EH, Monni O, et al. Compar- 2009;19:167–77. Epub 2008 Dec 3. chromosomes 1, 3, 16, and17 in primary breast ative genomic hybridization analysis of 38 breast cancer 50. Clark SJ. Action at a distance: epigenetic silencing of carcinomas inferredby fluorescence in situ hybridiza- cell lines: a basis for interpreting complementary DNA large chromosomal regions in carcinogenesis. Hum Mol tion. Cancer Genet Cytogenet 1996;88:1–7. microarray data. Cancer Res 2000;60:4519–25. Genet 2007;16:R88–R95.

www.aacrjournals.org 5953 Cancer Res 2009; 69: (14). July 15, 2009

Correction: Article on Breast Cancer Tumor Progression

In the article on breast cancer tumor progression in the July 15, 2009 issue of Cancer Research (1), the following footnote should have been included:. Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

1. Marella NV, Malyavantham KS, Wang J, Matsui S, Liang P, Berezney R. Cytogenetic and cDNA microarray expression analysis of MCF10 human breast cancer progression cell lines. Cancer Res 2009;69:5946–53.

I2009 American Association for Cancer Research. doi:10.1158/0008-5472.CAN-69-19-COR1

Cancer Res 2009; 69: (19). October 1, 2009 7894 www.aacrjournals.org Published OnlineFirst July 7, 2009; DOI: 10.1158/0008-5472.CAN-09-0420

Cytogenetic and cDNA Microarray Expression Analysis of MCF10 Human Breast Cancer Progression Cell Lines

Narasimharao V. Marella, Kishore S. Malyavantham, Jianmin Wang, et al.

Cancer Res 2009;69:5946-5953. Published OnlineFirst July 7, 2009.

Updated version Access the most recent version of this article at: doi:10.1158/0008-5472.CAN-09-0420

Supplementary Access the most recent supplemental material at: Material http://cancerres.aacrjournals.org/content/suppl/2009/08/18/0008-5472.CAN-09-0420.DC1

Cited articles This article cites 49 articles, 7 of which you can access for free at: http://cancerres.aacrjournals.org/content/69/14/5946.full#ref-list-1

Citing articles This article has been cited by 6 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/69/14/5946.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/69/14/5946. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.