MYB Oncogene Amplification in Hereditary BRCA1 Breast Cancer1

[CANCER RESEARCH 60, 5323–5328, October 1, 2000]

Advances in Brief

MYB Oncogene Amplification in Hereditary BRCA1 Breast Cancer1

Paivikki Kauraniemi, 2 Ingrid Hedenfalk, 2 Karin Persson, David J. Duggan, Minna Tanner, Oskar Johannsson, HåkanOlsson, Jeffrey M. Trent, Jorma Isola, and Åke Borg3

Laboratory of Cancer Genetics, Institute of Medical Technology, University of Tampere and Tampere University Hospital, FIN-33101 Tampere, Finland [P. K., M. T., J. I.];Department of Oncology, University Hospital, SE-22185 Lund, Sweden [I. H., K. P., M. T., O. J., H. O., J. I., A. B.]; Cancer Genetics Branch, National Human Genome ResearchInstitute, NIH, Bethesda, Maryland 20892 [I. H., D. J. D., J. M. T.]

Abstract

Comparative genomic hybridization analysis has demonstrated thatbreast tumors from BRCA1 and BRCA2 germ-line mutation carrierscontain a large number of chromosomal copy number gains and losses. Ahigh regional copy number gain at 6q22-q24 was observed in oneBRCA1tumor, and fluorescencein situ hybridization analysis indicated a strongamplification of the MYB oncogene (15 copies ofMYB compared with 1copy of chromosome 6 centromere). Fluorescencein situ hybridizationanalysis revealed amplification ofMYB in 5 (29%) of 17 BRCA1 breasttumors, whereas none of 8BRCA2 tumors and 13 breast cancer cell lines,and only 2 of 100 sporadic breast tumors exhibited alteredMYB copynumbers. Gene amplification resulted in mRNA overexpression as deter-mined by Northern blot and cDNA microarray analysis, and proteinoverexpression by immunohistochemical staining. We conclude thatMYBamplification is infrequent in sporadic breast cancer but common inbreast tumors from BRCA1mutation carriers, suggesting a role of this cellcycle regulator and transcription factor in the progression of someBRCA1tumors. However, we cannot rule out the significance of other genes in the6q22-q24 amplicon.

Introduction

Breast tumors fromBRCA1 and BRCA2 mutation carriers arecharacterized by a large number of chromosomal copy number gainsand losses, as seen by CGH4 (1). The apparent genomic instability inthese tumors is consistent with the proposed role of BRCA proteins inrecombination-mediated double-strand DNA break repair, transcrip-tion-coupled repair, centrosome amplification control, and G2-M cellcycle checkpoint (reviewed in Ref. 2). CGH analysis has also dem-onstrated thatBRCA1- andBRCA2-associated tumors display specificpatterns of genetic changes, clearly distinguished from sporadic breastcancer (1). This suggests that tumor evolution progresses along dis-tinct pathways in these genetically predisposed individuals. Indeed,inactivation of the p53 checkpoint function has been proposed as aprerequisite for development of these tumors (3). Similarly, the highfrequency of 4q and 5q deletions inBRCA1tumors, as well as 17q23and 20q gain inBRCA2tumors, may signal the sites of other geneswith selective roles in keeping or overriding cell cycle control andgenome integrity (1). Gene amplification is a common mechanism of

oncogene activation in solid tumors such as breast cancer. AmplifiedDNA typically retained in autonomously replicating extrachromo-somal bodies may rapidly increase in number in an evolving cell cloneby unequal distribution at mitosis, or be successively lost if notproviding a growth advantage. CGH analysis of fresh-frozen breasttumor tissue fromBRCA1mutation carriers, identified through genetictesting of Scandinavian breast cancer families in Lund, Sweden,demonstrated an exceptionally distinct copy number gain at chromo-some 6q22-q24 in one tumor. This is a novel amplified region inbreast cancer and justified further studies in hereditary and sporadicbreast tumors, as well as attempts to identify the putative targetgene(s) of the amplicon.

Materials and Methods

Tumors and Cell Lines. Primary tumors were received from pathologydepartments in the southern Sweden health care region and stored freshlyfrozen at270°C. The present study included 17 tumors fromBRCA1mutationcarriers, 8 fromBRCA2mutation carriers, and 100 sporadic breast tumors.BRCA1and BRCA2mutation analyses have been described earlier (4), andconsisted of the protein truncation test, single-strand conformational polymor-phism analysis, and direct sequencing. Thirteen established breast cancer celllines (BT-474, DU4475, MDA134, MDA157, MDA361, MDA436, MCF-7,MPE600, SKBR3, T47D, UACC812, UACC893, and ZR75-1) were grown inrecommended conditions and harvested at confluency to obtain interphasenuclei from cells that were predominantly in the G1-phase of the cell cycle (5).Trypsinized cells were cytocentrifuged and air-dried at room temperature andfixed in Carnoy’s fluid [methanol:acetic acid (3:1)].

Probes for FISH. PAC probes forMYB, MYBL1 (A-MYB), MYBL2 (B-MYB), and the ERa gene were obtained by screening a PAC library by PCRusing primers specific for each gene. The specificity of the probes wasconfirmed by FISH to normal metaphase chromosomes, which showed thepresumed chromosomal localization for each probe. The probes were labeledwith digoxigenin by standard nick translation. A spectrum green-labeled peri-centromeric probe (Vysis, Inc., Downers Grove, IL) was used as a referenceprobe to determine the copy number of chromosomes 6, 8, and 20 forMYB,MYBL1, andMYBL2, respectively.

FISH. Touch imprint preparations were made for FISH analysis by lightlypressing a semi-thawed frozen tumor onto Superfrost Plus microscope slides(Menzel, Braunschweig, Germany) microscope slides and air-dried. Prior tohybridization, imprint preparations were fixed with 50, 70, and 100% Carnoy’ssolution [methanol:acetic acid (3:1)] for 10 min each. Dual-color FISH exper-iments were performed as described previously (5).MYB, MYBL1, MYBL2,andERprobes were hybridized together with chromosome 6 centromere probe.The hybridization was carried out overnight at 42°C in a mixture containing 5ng of pericentromeric probes, 20 ng of gene-specific probes, and 10mg ofhuman placental DNA. After hybridization, excess probes were washed with0.43 SSC (2 min at 74°C) and 23SSC (1 min at room temperature), anddetected immunohistochemically with antidigoxigenin rhodamine. Slides werecounterstained with 0.2 mM 4,6-diamidino-2-phenylindole in an antifade so-lution (Vectashield; Vector Laboratories, Burlingame, CA). Hybridizationsignals were evaluated using an Olympus BX50 epifluorescence microscopeequipped with a363 oil-immersion objective (numeric aperture, 1.4). A dualband-pass fluorescence filter (Chromotechnology; Brattleboro, VT) was usedto visualize the FITC and rhodamine signals simultaneously.

Received 2/17/00; accepted 8/17/00.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby markedadvertisementin accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1 The present study was supported by grants from the Swedish Cancer Society, theNordic Cancer Union, Mrs. Berta Kamprads Foundation, the Gunnar Arvid & ElisabethNilsson Foundation, the Crafoord Foundation, the Hospital of Lund Foundations, the F &M Bergqvist Foundation, the King Gustav V:s Jubilee Foundation, the Finnish CulturalFoundation, the Maud Kuistila Foundation, the Sigrid Juselius Foundation, and theFinnish Cancer Society.

2 Contributed equally to this work.3 To whom requests for reprints should be addressed, at Department of Oncology,

University Hospital, SE-22185 Lund, Sweden. Phone: 46 46 177569; Fax: 46 46 147327;E-mail: [email protected].

4 The abbreviations used are: CGH, comparative genomic hybridization; ER, estrogenreceptor; FISH, fluorescencein situ hybridization; PAC, P1 artificial chromosome; PgR,progesterone receptor; EST, expressed sequence tag.

5323

Research. on August 26, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

At least 80 nonoverlapping nuclei with intact morphology based on 4,6-diamidino-2-phenylindole counterstaining were scored to determine the num-ber of hybridization signals forMYB and centromere probes. Control hybrid-izations to normal lymphocytes were done to ascertain that the probesrecognized a single copy target and that the hybridization efficiencies weresufficient. Both absolute copy numbers and the copy number ratio (betweenaverage ofMYB, MYBL1, andMYBL2 and the respective centromere copynumbers) were determined. Amplification ofMYBgene was defined as a copynumber ratio of$2.0, whereas a copy number gain was assigned to tumorshaving a ratio.1.5 and,2.0.

CGH. CGH was performed according to a published protocol (6). Briefly,tumor DNA and normal female reference DNA was extracted using a standardprotocol and labeled with FITC-dCTP and Texas-Red-dUTP (DuPont, Boston,MA), respectively, using standard nick translation. Labeled DNAs (400–800ng of each) and 10mg of unlabeled Cot-1 DNA (Life Technologies, Gaith-ersburg, MD) were hybridized onto commercially available normal metaphasechromosomes (Vysis). The hybridizations were evaluated by a commercialdigital image analysis system (Vysis).

Northern Blot. Total RNA was extracted from the tumors with the RNeasykit (Qiagen), followed by TRIzol (Life Technologies). Tenmg of RNA fromeach tumor was size-fractionated on a 1% agarose gel containing formalde-hyde. After capillary transfer to Hybond-N membranes, the RNA was hybrid-ized with a32P-labeledMYBprobe. Following hybridization overnight at 42°C,the membranes were washed twice in 23SSC containing 0.1% SDS at 42°Cand twice in 0.13SSC containing 0.5% SDS at 65°C before exposure toautoradiographic film at270°C.

cDNA Microarray Preparation, Hybridization, and Analysis. Micro-arrays were prepared by PCR amplification and arrayed on poly-L-lysine-coated glass slides using a custom, high-speed robotic printer as describedpreviously (7). Total RNA was extracted from the biopsy specimen and thereference MCF-10A cells using RNeasy (Qiagen) and TRIzol in successionaccording to the manufacturers’ recommendations. Fluorescent-labeled cDNA

targets were prepared by a single round of reverse transcription in the presenceof Cye3- or Cye5-dUTP (Amersham) and hybridized to the probes as describedpreviously (8). Fluorescence intensities were measured using a custom-de-signed scanning laser confocal microscope with appropriate excitation andemission filters, and a photomultiplier tube. ArraySuite software (9) was thenused to identify probe sites, extract fluorescence intensities, and merge the twoimages. Color images were formed by arbitrarily assigning the red (Cye3) andgreen (Cye5) channels. Local background was calculated for each probelocation. The ratios for all of the probes and confidence intervals for eachexperiment were determined with the aid of 88 “housekeeping” genes whosetheoretical expression ratios are expected not to deviate significantly from 1.0(7, 9).

Immunohistochemistry. A tumor xenograft from tumor 13996 was grownin nude mice as described elsewhere.5 Freshly prepared xenograft tissue wasfixed overnight in 4% buffered formalin and processed to a paraffin blockaccording to a standard procedure. Sections (3–4mm) were cut onto poly-L-lysine-coated slides. Prior to immunostaining, antigen retrieval was performedby immersing the dewaxed sections in 10 mM EDTA (pH 8.0) at 94°C for 20min in a temperature-controlled microwave oven. A standard avidin-biotin-peroxidase technique was used for visualization with diaminobenzidine as achromogen (Histostain Plus-kit; Zymed Laboratories, San Francisco, CA). Apolyclonal antibody to c-Myb was obtained from Santa Cruz Technologies andwas used in a 1:500 dilution.

Results

Identification of MYB Amplification. Analysis of relative copynumber gains and losses by CGH indicated that tumor 13996 from agerm-lineBRCA1mutation (1806C3T) carrier showed a high-levelcopy number gain at 6q22-q24 (Fig. 1A). A search in the gene bank

5 Johannssonet al., manuscript in preparation.

Fig. 1.A, comparative genomic hybridization show-ing a high-level copy number gain at 6q22–24 in aprimary breast cancer (13996) from a germ-lineBRCA1mutation carrier. The pseudocolored image of chromo-some 6 is shown on theleft; the corresponding green-to-red fluorescence ratio profile is shown on theright.B, FISH demonstrating high-level amplification ofMYBin aBRCA1tumor (13996).C, FISH of aBRCA1tumor(14510) with low-level amplification ofMYB.D, FISHof a BRCA1tumor (12224) with noMYBamplification.Note that not all amplified gene copies (B–D, in red) arein the focal plane. Chromosome 6 centromere is shownin greenfluorescence.

5324

MYB AMPLIFICATION IN BRCA1 BREAST CANCER

Research. on August 26, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

and a recent study of pancreatic carcinoma (10) suggestedMYBoncogene as a possible target gene for the amplification. A large-insertsize PAC probe was prepared forMYBand used in FISH. A high-levelamplification was verified in the imprint touch preparation of thesame tumor (mean of 15 copies ofMYB and 1 copy of chromosome6 centromere; Fig. 1B). As could be estimated from CGH, the ampli-con is small and did not include theERa gene, located at 6q25 (FISHdata not shown).

Amplification of MYB in Breast Tumors by FISH. FISH anal-ysis of 17 tumors from 16BRCA1germ-line mutation carriers showedthatMYBamplification is common in this tumor entity. Amplificationwas found in 5 (29%) of the 17 tumors, with aMYB/centromere copynumber ratio ranging from 2.27 to 11.2. The meanMYB gene copynumber in amplified tumors varied between 6.6 and 21.2, reflecting ahigh degree of aneuploidy in some tumors (Table 1). One tumorshowed a borderline copy number gain (mean, 10.5 copies ofMYBversus5.8 copies of chromosome 6 centromere, giving a ratio of1.80). Highest amplification was seen in tumor 13996. Interestingly, abilateral tumor (8571) from this patient had noMYB amplification(Table 1).MYBamplification was found in none of the 8 tumors fromBRCA2mutation carriers, in 2 of 100 sporadic breast cancers, and in

none of the 13 breast cancer cell lines. The ratio ofMYB/chromosome6 centromere copy number in the two amplified sporadic tumors wasin the same range as in amplifiedBRCA1 tumors (5–10; data notshown). Examples of tumors with a moderate level ofMYB amplifi-cation and noMYB amplification are shown in Fig. 1,C and D,respectively. No evidence ofMYBL1 (A-MYB) or MYBL2(B-MYB)gene amplification was seen in the present series ofBRCA1,BRCA2,and sporadic breast tumors (data not shown).

Overexpression ofMYB Transcripts, and Correlations to Hor-mone Receptor Status.Northern blot analysis showed thatBRCA1tumors (13996-xeno, 12421, 8822, and 14510) withMYB gene am-plification manifested elevatedMYB expression compared with tu-mors withoutMYBamplification (Fig. 2). Tissue for RNA extractionwas not available from the remaining amplified tumors (BRCA1tumor10359 and two sporadic breast tumors). Two of fiveBRCA1tumorswithout MYB gene amplification manifestedMYB overexpression,whereas the remaining three showed low, borderline, and absentMYBexpression, respectively.BRCA1tumors were hormone receptor-neg-ative, the only exception being tumor 12421, which showed border-line ER expression (12 fmol/mg of protein). Five sporadic tumorswithout MYB amplification where analyzed by Northern blot.MYB

Table 1 MYB copy numbers in primary breast cancers from germ-line BRCA1 and BRACA2 mutation carriers

Tumor number Family BRCAmutationMYB copies

(mean)Chr 6 copies

(mean) Ratio Interpretation

BRCA113996 Lund 56 1806C3T 13.1 1.17 11.2 High amplification12421 Lund 141 2594delC 21.2 5.56 3.81 Amplification8822 Lund 66 1201del11 8.18 3.04 2.69 Amplification10359 Lund 134 3172ins5 6.60 2.62 2.52 Amplification14510 Lund 30 Cys61Gly 15.1 6.66 2.27 Amplification9252 Lund 289 2594delC 10.5 5.82 1.80 Gain14007 Lund 221 3172ins5 5.58 3.14 1.78 Gain12224 Lund 264 1806C3T 4.88 3.53 1.38 No change13812 Lund 331 4808C3G 5.04 4.20 1.20 No change11394 Lund 133 1177G3A 2.71 2.28 1.19 No change11808 Lund 263 3829delT 2.34 2.00 1.17 No change8517a Lund 56 1806C3T 6.18 5.62 1.10 No change10697 Lund 1 Exon 22del 2.26 2.10 1.08 No change10581 Lund 212 1806C3T 3.12 3.08 1.01 No change14090 Lund 221 3172ins5 2.12 2.12 1.00 No change14970 Lund 85 2594delC 2.02 2.18 0.93 No change13714 Lund 241 5382insC 4.52 5.06 0.89 No change

BRCA210588 Lund 10 4486delG 3.64 2.12 1.72 Gain13816 Lund 46 3058A3T 3.32 2.00 1.72 Gain11787 Lund 46 3058A3T 3.30 2.04 1.62 Gain11506 Lund 225 6293C3G 3.84 2.72 1.41 No change7936 Lund 225 6293C3G 2.86 2.16 1.32 No change11721 Lund 119 5445del5 2.62 2.26 1.16 No change11900b Lund 29 2024del5 4.28 3.86 1.11 No change14486 Lund 29 2024del5 3.34 3.30 1.01 No change

a Bilateral tumor from the same patient as 13996.b Bilateral tumor from the same patient as 14486.

Fig. 2. Northern blot analysis ofMYB mRNAexpression inBRCA1,BRCA2, and sporadic breastcancer. The same membranes were hybridized withan actin probe, and based on densitometry evalua-tion of MYB/actin band intensity ratios,MYB ex-pression was scored as negative (2), borderline[(1)], or positive (1, 1 1, 1 1 1, 1 1 1 1,1 1 1 1 1). MYBgene copy number was scoredaccording to Table 1 as normal (2), amplified (1),or highly amplified (1 1). ER and PgR expressionwas scored as negative (2;,10 fmol/mg of pro-tein), positive (1; 10–200 fmol/mg of protein), orstrongly positive (1 1; .200 fmol/mg of protein).nr, number.

5325

MYB AMPLIFICATION IN BRCA1 BREAST CANCER

Research. on August 26, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

expression was seen in the three hormone receptor-positive sporadictumors, whereasMYBtranscripts were absent in two ER/PgR-negativetumors. AllBRCA2tumors were positive (.10 fmol/mg of protein) orstrongly positive (.200 fmol/mg of protein) for either ER or PgR andmanifested varying levels ofMYB expression, two tumors being ofborderline positivity.

Overexpression of Myb Protein. Amplification of MYB was ac-companied not only by overexpression of its mRNA but its proteinproduct as well. Immunohistochemical staining of a xenograft fromtumor 13996 (Fig. 1,A and B) indicated strong nuclear immuno-staining. A corresponding xenograft from breast cancer cell line T47D(non-BRCA1mutation-positive) withoutMYB amplification showedno nuclear immunostaining (data not shown).

cDNA Microarrays. The microarrays consisted of 4688 cDNAclones selected from the UniGene collection. The clones included2629 homologous to known genes and 2059 clones with no knownhomology. Analysis of theBRCA1tumor 13996 (with a high level ofMYB amplification, 11.2-fold; Table 1) showedMYB (v-myb avianmyeloblastosis viral oncogene homologue) expression to be greatlyincreased as well (15-fold; Table 2), relative to theMYBexpression inthe reference MCF-10A cells. Furthermore, several putativeMYB-orcell cycle-regulated genes [i.e., cdc2, cyclin B1, aurora, and retino-blastoma-like 1 (p107) protein] were determined to be overexpressedas well (Table 2). Other putative candidate genes localized to the6q22-q24 region, including MAP/ERK kinase 5, interferon-g receptora, phosphodiesterase I/nucleotide pyrophosphatase 1, gap junctionprotein a1, myristylated alanine-rich protein kinase C substrate(MARCKS, 80K-L), connective tissue growth factor, and severalanonymous ESTs, were not found to be overexpressed. However, agene designated pleomorphic adenoma gene-like 1 and mapped to6q24-q25, was found to be 8-fold overexpressed in tumor 13996. Acomplete list of clones and hybridization results from tumor 13996 is

available on request, and can also be found at the National HumanGenome Research Institute Web site.6

Discussion

Results from the present study suggest theMYB oncogene as apossible target gene in the 6q22-q24 amplicon found by CGH analysisof a breast tumor from aBRCA1 germ-line mutation carrier andsubsequently by FISH analysis in additionalBRCA1tumors (5 of 17;29%). In contrast,MYB amplification was not found in 8BRCA2tumors or in 13 breast cancer cell lines, and in only 2 of 100 sporadicbreast tumors. One earlier study reports a similar low incidence (1 of59) of MYB amplification in sporadic breast cancer (11). Moreover,infrequent observations ofMYB amplification have been made inacute myelogenous leukemia, and colon and pancreas carcinoma (10,12, 13).

TheMYBgene was found to be overexpressed in amplified tumorsusing Northern blot and cDNA microarray analysis, and c-Myb pro-tein overexpression was detected by immunohistochemistry. The am-plicon did not extend to the 6q25 region as evidenced by a normalcopy number of the ERa gene in these tumors. Other 6q23 candidategenes and ESTs present in the array were not found to be overex-pressed. We cannot rule out the importance of other target genes in the6q23 region, including IGFBP4, a DNA-binding protein; A20, anelongation factor homologue; and additional ESTs not present on themicroarray. However, the fact thatMYBappeared as one of the mosthighly overexpressed genes among the;5000 clones analyzed sup-ports the assumption that 6q23 amplification inBRCA1 tumors isdriven by this oncogene.

The Myb proteins are ancient regulators of gene expression, con-

6 http://wwwdev.nhgri.nih.gov/MYB/.

Table 2 cDNA microarray analysis of BRCA1 tumor 13996 with high-level MYB gene amplification

Shown are a subset of the genes found to have high expression in the tumor 13996 relative to the reference MCF-10A cells. In the lower part of the Table are genes selected basedon their known importance in cell cycle control, DNA repair, or suggested as being regulated by MYB. A complete list of clones and hybridization results is available on request.

UniGene no. Description Expression ratio

Hs. 119571 Collagen, type III,a1 (Ehlers-Danlos syndrome type IV, autosomal dominant) 73Hs. 76391 Myxovirus (influenza) resistance 1, homologue of murine (interferon-inducible protein p78) 52Hs. 35120 Replication factor C (activator 1) 4 (37 kDa) 45Hs. 77695 KIAA0008 gene product 45Hs. 83164 Collagen, type XV,a1 43Hs. 814 MHC, class II, DPb1 32Hs. 93002 Human cyclin-selective ubiquitin carrier protein mRNA 29Hs. 76753 Endoglin (Osler-Rendu-Weber syndrome 1) 26Hs. 94953 ESTs, highly similar to complement C1Q subcomponent, C-chain precursor 24Hs. 3068 SWI/SNF-related, matrix-associated, actin-dependent regulator of chromatin, subfamily a, member 3 24Hs. 206503 Homo sapiensmRNA for CDC2 T, complete cds 22Hs. 173035 Human mRNA for KIAA0300 gene 18Hs. 198005 ESTs, highly similar to mosaic protein LR11 18Hs. 107573 Human sialyltransferase SThM 16Hs. 73932 Human MHC class II DQ-b associated with DR2. DQw1 protein, complete cds 16Hs. 199147 Human mRNA for aurora/IPL-1-related serine/threonine kinase 16Hs. 1334 v-mybavian myeloblastosis viral oncogene homologue 15Hs. 85482 Ubiquitin-specific protease 13 (isopeptidase T-3) 15Hs. 170121 Protein tyrosine phosphatase, receptor type, C polypeptide 15Hs. 79914 Lumican 14Hs. 2533 Aldehyde dehydrogenase 9 (g-aminobutyraldehyde dehydrogenase, E3 isozyme) 14Hs. 82891 GlutathioneS-transferase M4 14Hs. 20191 Seven in absentia (Drosophila) homologue 2 14Hs. 763 Fc fragment of IgG, low-affinity IIIa, receptor for CD16 14Hs. 82502 Human mRNA for KIAA0039 gene 14Hs. 198620 STs, highly similar to tyrosine-protein kinase receptor HEK-2 precursor (H. sapiens) 13

Hs. 79078 MAD2 (mitotic arrest-deficient, yeast, homologue)-like 1 9.1Hs. 83758 CDC28 protein kinase 2 6.8Hs. 78934 MutS (Escherichia coli) homologue 2 (colon cancer, nonpolyposis type 1) 6.5Hs. 200542/87 ESTs/retinoblastoma-like 1 (p107) 6.0Hs. 1846 Tumor protein p53 5.8Hs. 23960/168383 Cyclin B1/Intercellular adhesion molecule 1 (CD54), human rhinovirus receptor 5.6Hs. 156346 ESTs, highly similar to DNA topoisomerase II,a isozyme (H. sapiens) 5.1

5326

MYB AMPLIFICATION IN BRCA1 BREAST CANCER

Research. on August 26, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

sisting of three discrete functional domains responsible for sequence-specific DNA binding, transcriptional activation, and negative regu-lation of the protein, and they function as key regulators of cell growthand differentiation (14). TheMYB oncogene was first discovered inavian retroviruses that cause acute leukemia; it encodes v-Myb pro-teins that are truncated and deregulated versions of the cellular c-Mybprotein. The normal c-Myb protein is highly expressed in immaturehematopoietic cells, and its expression decreases dramatically duringcell differentiation (15). B-Myb, one of two closely relatedMYBgenes present in vertebrates, is a ubiquitously expressed key regulatorof cell cycle progression, and ectopic expression of B-Myb proteinwas shown to override p53-induced G1 arrest, allowing cell cycleprogression under circumstances when DNA repair or apoptosis nor-mally would prevail (16). Like B-Myb, A-Myb is also regulated bycyclin-dependent phosphorylation during the G1-S and S-phases, but(similar to c-Myb) it exhibits a more restricted tissue pattern ofexpression (14). It was recently shown that mice homozygous for anA-MYB germ-line mutation develop to term but show defects ingrowth of certain tissues (17). MaleA-MYB2 2/2 2 mice wereinfertile because of a block in spermatogenesis and the pachytenephase of meiosis, whereas females manifested poor mammary glandmorphogenesis, mainly during pregnancy, and progesterone-inducedductal branching and alveolar outgrowth (17). A similar lack ofpregnancy-induced breast epithelial cell proliferation, duct branching,and alveoli development was observed in mice with conditionalBRCA1knockout in mammary tissue (18). Thus, bothBRCAandMYBproteins are crucially involved in the processes of DNA recombina-tion and mammary gland development (2, 17, 18).

The singleMYB homologue present inDrosophila (Dm myb) andin Schizosaccharomyces pombe(cdc5) has been implicated as havinga role in G2 progression and G2-M transition (19, 20). It was foundthat wing cells in Dm myb mutants enter and complete DNA synthesisbut are blocked in G2 or at G2-M transition where they continue toreplicate their DNA, resulting in polyploidy. This suggests that Dmmyb possesses two cell cycle checkpoint functions, one in regulationof the G2-M transition and a second in prevention of endoreplicationand maintenance of diploidy (19). In all likelihood, Dm myb repre-sents the progenitor of all threeMYBgenes evolved in vertebrates andencompasses their combined functions. Thus, in addition to tissue-specific expression, the different vertebrateMYB genes may possessboth distinct and overlapping functions in cell cycle G1-S progressionand G2-M control, as well as in cell differentiation. A potential role ofBRCA1 in a G2-M checkpoint has also emerged (2), and its seemsreasonable to suspect an interplay betweenMYB and BRCAproteinfunctions.

Homozygous disruption of theMYB gene in mice is lethal andcauses death from severe anemia at day 15 of embryogenesis, whichdemonstrates the critical role of c-Myb during hematopoiesis anderythroid cell differentiation (21). However, c-Myb expression hasalso been noted in nonhematopoietic cells, including normal andtumorigenic human breast epithelial cells and breast tumors (22, 23).Surprisingly, c-Myb expression in breast cells was found to be asso-ciated with estrogen stimulation and the presence of ER, possiblybecause of a posttranscriptional stabilization of theMYB transcript.Results from the present study support a relationship between c-Mybexpression and ER-positive phenotype. ER-positive sporadic andBRCA2tumors withoutMYBamplification displayed moderate c-Mybexpression in Northern blot analysis, whereas ER-negative sporadictumors exhibited low or no c-Myb expression. Thus, c-Myb may, likeA-Myb, play a role in hormone-regulated growth and differentiationof breast epithelial cells and ER-positive breast cancer. However,becauseBRCA1tumors usually are ER-negative (24), the activity ofMYB (single or amplified copy number) inBRCA1-deficient cells

would be regulated by other pathways. Possibly,MYBamplification inBRCA1 tumors may reflect a compensatory mechanism to executevital cellular functions or to override a cell cycle block caused byDNA damage. Providing further suggestive evidence for a link be-tween MYB and BRCA1 regulation, several potentialMYB bindingsites (PyAACG/TG) are found in theBRCA1promotor region.7

cDNA microarray analysis ofBRCA1tumor 13996 with high-levelMYB amplification disclosed high expression of a multitude of cellcycle regulators, some of which have been suggested previously asbeing activated byMYB, including cdc2/cdc28, topoisomerase IIa,cyclin B1, p53,MSH2, mitotic feedback control protein MAD2-like 1,aurora/IPL-1-related kinase, and retinoblastoma protein homologuep107 (14). This deregulated gene expression may be a consequence ofother cellular processes and should not be taken as evidence for aMYB-related activity. For example, tumor 13996 carries a somatic p53mutation (Ser215Ile; data not shown). However, it provides a possiblelink between a transcription factor and its target genes and demon-strates the usefulness of microarray analysis in depicting cellularsignaling pathways.

In conclusion, amplification of theMYB gene at 6q22-q24 wasunexpectedly found to be a prevalent genetic aberration inBRCA1-associated breast tumors. This is an infrequent finding in sporadicbreast cancer, and indeed, both sporadic and hereditary breast cancerusually are characterized by deletions on chromosome 6q (1, 25).Other oncogenes such as theERBB2 oncogene, commonly foundamplified in (ER-negative) breast cancer, are not associated withBRCA1tumor development (24). This illustrates that gene amplifica-tion is not merely a result of genomic instability and suggests afunctional role of the c-Myb transcription factor inBRCA1breasttumor progression. However,MYB is not essential for progression ofBRCA1-related breast tumors because the majority of tumors exam-ined lackedMYB amplification or overexpression. Moreover, wecannot rule out the existence of another target gene in the 6q22-q24amplicon.

References

1. Tirkkonen, M., Johannsson, O., Agnarsson, B. A., Olsson, H., Ingvarsson, S., Karhu,R., Tanner, M., Isola, J., Barkardottir, R. B., Borg, Å., and Kallioniemi, O-P. Distinctsomatic genetic changes associated with tumor progression in carriers of BRCA1 andBRCA2 germ-line mutations. Cancer Res.,57: 1222–1227, 1997.

2. Welcsh, P. L., Owens, K. N., and King, M. C. Insights into the functions of BRCA1and BRCA2. Trends Genet.,16: 69–74, 2000.

3. Kinzler, K. W., and Vogelstein, B. Cancer-susceptibility genes. Gatekeepers andcaretakers. Nature (Lond.),386: 761–763, 1997.

4. Håkansson, S., Johannsson, O., Johansson, U., Gerdes, A-M., Holmberg, E., Dahl, N.,Pandis, N., Kristofferson, U., Olsson, H., and Borg, Å. Moderate frequency ofBRCA1 and BRCA2 germ-line mutations in Scandinavian familial breast cancer.Am. J. Hum. Genet.,60: 1068–1078, 1997.

5. Tanner, M. M., Tirkkonen, M., Kallioniemi, A., Isola, J., Kuukasjarvi, T., Collins, C.,Kowbel, D., Guan, X. Y., Trent, J., Gray, J. W., Meltzer, P., and Kallioniemi, O. P.Independent amplification and frequent co-amplification of three nonsyntenic regionson the long arm of chromosome 20 in human breast cancer. Cancer Res.,56:3441–3445, 1996.

6. Kallioniemi, O. P., Kallioniemi, A., Piper, J., Isola, J., Waldman, F. M., Gray, J. W.,and Pinkel, D. Optimizing comparative genomic hybridization for analysis of DNAsequence copy number changes in solid tumors. Genes Chromosomes Cancer,10:231–243, 1994.

7. DeRisi, J. L., Penland, L., Brown, P. O., Bittner, M. L., Meltzer, P. S., Ray, M., Chen,Y., Su, Y. A., and Trent, J. M. Use of a cDNA microarray to analyse gene expressionpatterns in human cancer. Nat. Genet.,14: 457–460, 1996.

8. Kahn, J., Simon, R., Bittner, M., Chen, Y., Leighton, S. B., Pohida, T., Smith, P. D.,Jiang, Y., Gooden, G. C., Trent, J. M., and Meltzer, P. S. Gene expression profilingof alveolar rhabdomyosarcoma with cDNA microarrays. Cancer Res.,58: 5009–5013, 1998.

9. Chen, Y., Dougherty, E. R., and Bittner, M. L. Ratio-based decisions and thequantitative analysis of cDNA microarray images. J. Biomed. Optics,2: 364–374,1997.

10. Wallrapp, C., Muller-Pillasch, F., Solinas-Toldo, S., Lichter, P., Friess, H., Buchler,M., Fink, T., Adler, G., and Gress, T. M. Characterization of a high copy number

7 http://www2.ncbi.nlm.nih.gov/genbank.

5327

MYB AMPLIFICATION IN BRCA1 BREAST CANCER

Research. on August 26, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

amplification at 6q24 in pancreatic cancer identifies c-mybas a candidate oncogene.Cancer Res.,57: 3135–3139, 1997.

11. Masuda, H., Battifora, H., Yokota, J., Meltzer, S., and Cline, M. J. Specificity ofproto-oncogene amplification in human malignant diseases. Mol. Biol. Med.,4:213–227, 1987.

12. Pelicci, P. G., Lanfrancone, L., Brathwaite, M. D., Wolman, S. R., and Dalla-Favera,R. Amplification of the c-myb oncogene in a case of human acute myelogenousleukemia. Science (Washington DC),224: 1117–1121, 1984.

13. Alitalo, K., Winqvist, R., Lin, C. C., de la Chapelle, A., Schwab, M., and Bishop,J. M. Aberrant expression of an amplified c-myb oncogene in two cell lines from acolon carcinoma. Proc. Natl. Acad. Sci. USA,81: 4534–4538, 1984.

14. Ness, S. A. The Myb oncoprotein: regulating a regulator. Biochim. Biophys. Acta,1288: F123–F139, 1996.

15. Introna, M., Luchetti, M., Castellano, M., Arsura, M., and Golay, J. The myboncogene family of transcription factors: potent regulators of hematopoietic cellproliferation and differentiation. Semin. Cancer Biol.,5: 113–124, 1994.

16. Lin, D., Fiscella, M., O’Connor, P. M., Jackman, J., Chen, M., Luo, L. L., Sala, A.,Travali, S., Apella, E., and Mercer, W. E. Constitutive expression of B-myb canbypass p53-induced Waf1/Cip1-mediated G1 arrest. Proc. Natl. Acad. Sci. USA,91:10079–10083, 1994.

17. Toscani, A., Mettus, R. V., Coupland, R., Simpkins, H., Litvin, J., Orth, J., Hatton,K. S., and Reddy, E. P. Arrest of spermatogenesis and defective breast developmentin mice lacking A-myb. Science (Washington DC),386: 713–717, 1997.

18. Xu, X., Wagner, K-U., Larson, D., Weaver, Z., Li, C., Ried, T., Henninghausen, L.,Wynshaw-Boris, A., and Deng, C-X. Conditional mutation of Brca1 in mammary

epithelial cells results in blunted ductal morphogenesis and tumour formation. Nat.Genet.,22: 37–43, 1996.

19. Katzen, A. L., Jackson, J., Harmon, B. P., Fung, S. M., Ramsay, G., and Bishop, J. M.Drosophila myb is required for the G2/M transition and maintenance of diploidy.Genes Dev.,12: 831–843, 1998.

20. Ohi, R., McCollum, D., Hirani, B., Den Haese, G. J., Zhang, X., Burke, J. D., Turner,K., and Gould, K. L. TheSchizosaccharomyces pombecdc51 gene encodes anessential protein with homology to c-Myb. EMBO J.,13: 471–483, 1994.

21. Mucenski, M. L., McLain, K., Kier, A. B., Swerdlow, S. H., Schreiner, C. M., Miller,T. A., Pietryga, D. W., Scott, J., and Potter, S. S. A functional c-myb gene is requiredfor normal murine fetal hepatic hematopoiesis. Cell,65: 677–689, 1991.

22. Gudas, J. M., Klein, R. C., Oka, M., and Cowan, K. H. Posttranscriptional regulationof the c-mybproto-oncogene in estrogen receptor-positive breast cancer cells. Clin.Cancer Res.,1: 235–243, 1995.

23. Guerin, M., Sheng, Z. M., Andrieu, N., and Riou, G. Strong association betweenc-myb and oestrogen-receptor expression in human breast cancer. Oncogene,5:131–135, 1990.

24. Johannsson, O., Idvall, I., Anderson, C., Borg, Å., Barkardottir, R. B., Egilsson, V.,and Olsson, H. Tumour biological features of BRCA1-induced breast and ovariancancer. Eur. J. Cancer,33: 362–371, 1997.

25. Devilee, P., van Vliet, M., van Sloun, P., Kuipers Dijkshoorn, N., Hermans, J.,Pearson, P. L., and Cornelisse, C. J. Allelotype of human breast carcinoma: a secondmajor site for loss of heterozygosity is on chromosome 6q. Oncogene,6: 1705–1711,1991.

5328

MYB AMPLIFICATION IN BRCA1 BREAST CANCER

Research. on August 26, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from

2000;60:5323-5328. Cancer Res   Päivikki Kauraniemi, Ingrid Hedenfalk, Karin Persson, et al.   Cancer

BreastBRCA1 Oncogene Amplification in Hereditary MYB

  Updated version

  http://cancerres.aacrjournals.org/content/60/19/5323

Access the most recent version of this article at:

  Cited articles

  http://cancerres.aacrjournals.org/content/60/19/5323.full.html#ref-list-1

This article cites 23 articles, 9 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/60/19/5323.full.html#related-urls

This article has been cited by 17 HighWire-hosted articles. Access the articles at:

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

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  [email protected] at

To request permission to re-use all or part of this article, contact the AACR Publications

Research. on August 26, 2015. © 2000 American Association for Cancercancerres.aacrjournals.org Downloaded from