Supplementalmaterialsandmethods

K6a-tvaBACtransgenicmice. ThisBACtransgeniclinewasgeneratedusingtherecombineeringmethod(Lee et al 2001, Muyrers et al 2001). Briefly,theRP23-473O22BACclone(200-kb)containingtheK6agenewaspurchasedfromBACPACResourcesCenter(Children’sHospitalOaklandResearchInstitute,Oakland,California). TheCounterSelectionBACModificationKit(GeneBridgesGmbH,Dresden,Germany)wasusedtomodifytheBACfollowingthemanufacturer’sinstructions:TheplasmidpSC101-BAD-gbaAtetwaselectroporatedintoE. colicontainingRP23-473O22BACtoprovidetherecombinase. Togenerateselecting/counter-selectingmarkersforreplacingexon1ofK6a,PCRwasperformedusingpRpsL-neoastemplateandthefollowingprimers:5’tcttcagctctgccctgccgtttctctacttcccagccttctcatctccaggaaccGGCCTGGTGATGATGGCGGGATCGand5’Tctgttcagcattccctgtagacacacaggctgctctagggaggaccatgtctcactcagaagaactcgtcaagaaggcg. Togeneratetvaforreplacingtheselecting-/counter-selecting-markers,PCRwasconductedusingtvacDNAasthetemplateandthefollowingprimers:5’tcttcagctctgccctgccgtttctctacttcccagccttctcatctccaggaaccATGGCGCGGCTGCTGCCCGCGand5’tctgttcagcattccctgtagaCACACAGGCTGCTCTAGGGAGGACCATGTCTCACTCAGTCCCATCTCACCAGCTC. CandidaterecombinantcolonieswerescreenedwithPCRusingprimers5’CCATATATAAGCGGCTGCCCand5’cagtatcagagctgagactg,andfurtherconfirmedbyDNAsequencing. Thefinalconstruct(K6a-tva)hastvacDNAimmediatelydownstreamofthetranslationinitiationcodoninexon1ofK6a. Whilethe3’endoftvadidnotrecombinewiththehomologysequenceinintron1ofK6a,aswedesigned,itrecombinedwiththehighlyhomologoussequenceinintron1ofthedownstreamK6bgene. Thus,thetvaexpressionintheresultingBACconstruct(K6a-tva)wasinitiatedbytheK6apromoterandterminatedbytheK6bpoly(A)(SupplementalFigure1A). ThisBACconstructwasfurtherconfirmednottohaveaberrantrecombinationbyNotIandXhoIfingerprintingusingpulsefieldelectrophoresis(SupplementalFigure1B). ThisconstructwaspurifiedusingtheNucleoBondBACMaxiKit(635941;BDBiosciences,SanJose,CA),digestedbyPI-SceI,andintroducedintoone-cellmouseembryosoftheFVB/NstrainbypronuclearmicroinjectionintheBCMTransgenicMouseCore. Theresultingpupsweregenotypedusingprimers5’CCATATATAAGCGGCTGCCCand5’GGAGCCTCTGTGCCGTTGTC.

Animalexperiments.RCAS-PyMT,RCAS-GFP,viruspreparation,andmammaryintraductalinfectionhavebeenpreviouslydescribed(Du et al 2006). Mammaryfatpadclearanceandtransplantationhavebeendescribed(DeOme et al 1959, Siwko et al 2008);3-week-oldMMTV-tvamicewereusedastherecipientstoavoidpotentialimmuneresponsetoTVA. Immediatelyafterclearing,donorcellsin10µlofDMEM/F12mixedwithMatrigel(356234;BDBioscience)(1:1)wereinjected. Forhormonetreatment,100μlofsesameoil(S-3547,Sigma-Aldrich,St. Louis,MO)containing1μgβ-estradiol3-benzoate(E-8515,Sigma-Aldrich)and1mgprogesterone(P-0130,Sigma-Aldrich)wasinjectedsubcutaneouslybetweentheshoulderbladesdailyforfivedays. AllmicewereontheFVB/Nbackground,andwerekepton2920XTekladGlobalExtrudedRodentDiet(SoyProtein-Free)(HarlanLaboratories,Indianapolis,IN)intheTransgenicMouseFacilityatBCM. AllproceduresusingmicewereperformedincompliancewithanInstitutionalAnimalCareandUseCommittee-approvedanimalprotocol.

Preparationofsinglecellsuspensionsfrommammaryglandsandtumors. Mammaryglandswerecollectedavoidinglymphnodes. Thesenormaltissuesandmammarytumorsweremadeintosinglecellsuspensionsusingapreviouslyreportedmethodwithmodifications(Shackleton et al 2006, Stingl et al 2006). Briefly,threemammaryglandsoranequivalentamountofmammarytumorswerechoppedintosmallpieces,anddigestedin10mlofDMEM/F12(11330;Invitrogen,Carlsbad,CA)supplementedwith5%FBS,5µg/mlinsulin(350-020;Biosource,Rockville,MD),500ng/mlhydrocortisone(H0888;Sigma-Aldrich),10ng/mlEGF(13247-051,Invitrogen),20ng/mlcholeratoxin(C-3012;Sigma-Aldrich),300U/mlcollagenaseIII(S4M7602S,Worthington,Lakewood,NJ),and100U/mlhyaluronidase(H3506;Sigma-Aldrich)inahumidified37°Cincubatorsupplementedwith5%CO2for1hour,withintermittentpipetting. Thedigestionmediumwasremovedbycentrifugation,andthecellpelletwasresuspendedin2mlof5mg/mldispaseII(10295825001;RocheDiagnostics,Indianapolis,IN)and0.1mg/mldeoxyribonuclease(58C10349;Worthington)forfiveminutes. ThedigestedcellswerethenwashedinHankssolutionanddigestedfor1-2minuteswith1mlof0.25%trypsin,whichwasstoppedwith10mlof5%FBSin DMEM/F12. Aftercentrifugation,thecellpelletwasresuspendedin2mloftheRBClysisbuffer(00433-57;eBioscience,SanDiego,CA),whichwasstopped2minuteslaterwith12mlofHankssolution. Thiscellsuspensionwasfilteredthrougha40µmfiltertogetsinglecellsuspensions.

FlowcytometryanalysisandFACS. FlowcytometryanalysisandFACSwereperformedintheBCMCytometryandCellSortingCoreusingBDLSRII(BDBioscience),and FACSAriaw/UV(BDBioscience)withanozzlesizeof130µm,respectively. FACSDivaV6.1.2software(BDBioscience)wasusedfordataanalysis. Antibodiesusedforflowcytometryinclude:ratanti-mouseCD16/CD32(553141,BDBioscience), FITC-anti-CD49f(555735,BDBioscience),FITC-ratIgG2bκisotypecontrol(553988,BDBioscience),PE-anti-CD24(553262,BDBioscience),PE-ratIgG2bκisotypecontrol(553989,BDBioscience),PE-anti-CD61(104307,BioLegend,SanDiego,CA),PEArmenlanhamsterIgGisotypecontrol(400907,BioLegend),biotin-anti-CD45R/B220(51-01122J,BDBioscience),biotin-anti-TER-119(51-09082J,BDBioscience),biotin-anti-CD31(553371,BDBioscience),biotin-anti-CD140a(13-1401-80,eBioscience), streptavidin-FITC(554060,BDBioscience), rabbitpolyclonalantibodyagainstTVA(madebyA. Leavitt),andAPC-conjugatedgoatanti-rabbitIgG(A10931,Invitrogen). DeadcellswerealwaysgatedoutbasedonPIstaininginallantibody-basedFACS.

Mammosphereassayandcolony-formingassay. Themammosphereassaywasperformedaspreviouslydescribed(Dontu et al 2003)withafewmodifications. Inbrief,10,000singlecellswereseededineachwellofultra-lowattachment6-welldishes(3471;CorningInc.,Corning,NY)with2mlofDMEM/F12mediumcontaining20µl/mlB27(17504-044;Invitrogen),20ng/mlbFGF(13256-029;Invitrogen),20ng/mlEGF(13247-051;Invitrogen),and2xAntibiotic/Antimycotic(15240-062;Invitrogen). Insomeexperiments,RCAS-GFPwasalsoaddedat5x106IUs/ml. Thecellswerefedwith0.5mloffreshmediumevery5days. Thecolony-formingassaywascarriedoutinregular6-welltissuecultureplates(353046;BDBioscience). 1,000cellswereseededineachwellwith2mlofDMEMmedium(11965;Invitrogen)containing5%FBS,5µg/mlinsulin,500ng/mlhydrocortisone,10ng/mlEGF,and20ng/mlcholeratoxin. Mediumwaschangedtwiceaweek. Insomeexperiments,RCAS-GFPwasaddedat5x106IUs/ml,andthevirus-containingmediumwasreplaced16hourslaterwithvirus-freeculturemedium. Aninvertedmicroscope(Olympus1x70,OlympusAmerica,Inc.,Melville,NY)equippedwithaRetiga1300camera(QImaging,Surrey,BC,Canada)andtheImage-ProPlus5.0.2.9software(MediaCybernectics,Inc.,Bethesda,MD)wasusedforimagingthesecolonies.

Tissueprocessingandimmunostaining. Tissuesandtumorswerefixed,processedintoparaffinsections,andstainedbyimmunohistochemistryorimmunofluorescenceasdescribed(Chen et al 2008, Du et al 2006). Co-immunofluorescentstainingforTVAandK6wascarriedoutonfrozensections.Thefollowingantibodieswereused:FITC-labeledandunconjugatedrabbitIgGsagainstmouseK6(FITC-169LandPRB-169P,Covance,Princeton,NJ);rabbitIgGsspecificforK5(PRB-160P,Covance),K14(PRB-155P,Covance),HA(PRB-101P,Covance),TVA,ERα(sc-542,SantaCruz),andE-cadherin(4065,CellSignaling,Danvers,MA);mousemonoclonalantibodiesagainstαSMA(M0851,Dako,Glostrup,Denmark),andβ-catenin(610153,BDBioscience);ratIgGagainstK8(TROMA-I)andK19(TROMA-III)(TheDevelopmentalStudiesHybridomaBank,IowaCity,organizedundertheauspicesoftheNationalInstituteofChildHealthandHumanDevelopmentandmaintainedbytheUniversityofIowa);goatIgGsagainstvimentin(sc-7557,SantaCruz);FITC-conjugatedhorseanti-mouseIgG(FI-2001,VectorLaboratories,Burlingame,CA);FITC-conjugatedgoatanti-rabbitIgG(FI-1000,VectorLaboratories);andAlexaFluor568-conjugatedgoatanti-rabbit,-mouse,and-ratIgGs(A11011,A1104,A11077,Invitrogen).

RNAextractionandAffymetrixGeneChiparrayhybridization.Alltissuesampleswerecollectedfreshandsnap-frozeninliquidnitrogen. RNAextraction,cDNAsynthesis,cRNAsynthesisandlabeling,andarrayhybridizationwereperformedasdescribed(Huang et al 2006). Briefly,frozensamplesweregroundintofinepowderinliquidnitrogen,andRNAwasextractedbyTrizol(LiftTechnologies)withtwocyclesofisopropanolprecipitation. TheOne-CycleTargetLabelingandControlReagentskitfromAffymetrix(Part#900493)(SantaClara,CA,USA)wasusedforcDNAsynthesisandbiotin-labeling. M430A2.0GeneChipswereusedandscannedbytheAffymetrixGeneChipscanner3000usingtheGeneChipoperatingsystem(GCOS)v1.2. ImageanalysisandprobelevelquantificationwerecarriedoutusingGCOS1.2,andtheprobeleveldata(CELfiles)weresavedforeacharrayforsubsequentstatisticalanalyses. ThesearraydatahavebeendepositedatGeneExpressionOmnibus(GEO)withtheaccessionnumberGSE20614.

Statisticalanalysisforexpressionarraydata.Medianintensitywasusedforthenormalizationoftheexpressionarrays,andtheperfectmatch(PMonly)modelingalgorithmwasused(Li and Hung Wong 2001). AnalysisofVariance(ANOVA)wasusedtoselectprobesetsdifferentiallyexpressedamongthethreegroups. ArrayswereclusteredbyCLUSTER,anddatavisualizedbyTREEVIEW(Eisen et al 1998). Forcomputinginter-profilecorrelationsbetweenthemousetumorsinourowndatasetandthemousetumorsfromHerschkowitzetal. (Herschkowitz et al 2007),wefirstcenteredthegeneexpressionvaluesineachdatasetonthemeancentroidofthecomparisongroups,thencomputedthePearson’scorrelationbetweeneachprofile(usingthe12,307uniquegenescommontobothplatforms). ForeachgenecommontoourarrayplatformandtheHoadleyarrayplatform(representingthehumantumordataset)(Hoadley et al 2007),wefirstcomputedintheHoadleydatasetthemeancentroidofthefourmajorhumantumorsubtypes(basal-like,ErbB2+,luminalA,luminalB,andnormal-like)andcenteredeachgroupaverageonthecentroid;wethentookthePearson'scorrelation(usingall5,343geneorthologscommontobothdatasets)betweentheHoadleycenteredaveragesandtheexpressionvaluesofeachofourmousetumors. ThemappingofgenesbetweenarraydatasetswasmadeontheEntrezGeneidentifier. Wheremultiplegenetranscriptprobesreferredtothesamegene,theprobewiththehighestvariationwastakentorepresentthegene.

Statistical analysis of transplantation experiments. Analysis of data from our small transplantation experiments (Figure 2D) using standard limiting dilution analysis(Hu and Smyth 2009) indicates that the single hit model assumption is violated. In particular, the slope for log(dose) in the generalized linear model (with log(-log) link) is greater than 1.0. We have fit the generalized linear model allowing slope1, and estimated predicted values (and 95% confidence intervals) at dose=200 in order to obtain approximate estimates of the frequency of outgrowth-generating cells in a sample of 200 cells. Frequencies are estimated as exp (predicted ± 2 x se predicted) x 200, where predicted values and standard errors are on the link scale. The dose of 200 cells was used because all cell subsets were tested at this dose. Modeling and computations were carried out with R version 2.12.2 (R Development Core Team 2011)and the statmod package(Smyth et al 2010).

References

Chen J, Jaeger K, Den Z, Koch PJ, Sundberg JP, Roop DR (2008). Mice expressing a mutant Krt75 (K6hf) allele develop hair and nail defects resembling pachyonychia congenita. J Invest Dermatol128: 270-279.

DeOme KB, Faulkin LJ, Jr., Bern HA, Blair PB (1959). Development of mammary tumors from hyperplastic alveolar nodules transplanted into gland-free mammary fat pads of female C3H mice. Cancer Res19: 515-520.

Dontu G, Abdallah WM, Foley JM, Jackson KW, Clarke MF, Kawamura MJ et al (2003). In vitro propagation and transcriptional profiling of human mammary stem/progenitor cells. Genes Dev17: 1253-1270.

Du Z, Podsypanina K, Huang H, McGrath A, Toneff MJ, Bogoslovskaia E et al (2006). Introduction of oncogenes into mammary glands in vivo with an avian retroviral vector initiates and promotes carcinogenesis in mouse models. Proc Natl Acad Sci U S A103: 17396-17401.

Eisen MB, Spellman PT, Brown PO, Botstein D (1998). Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci U S A95: 14863-14868.

Herschkowitz JI, Simin K, Weigman VJ, Mikaelian I, Usary J, Hu Z et al (2007). Identification of conserved gene expression features between murine mammary carcinoma models and human breast tumors. Genome Biol8: R76.

Hoadley KA, Weigman VJ, Fan C, Sawyer LR, He X, Troester MA et al (2007). EGFR associated expression profiles vary with breast tumor subtype. BMC Genomics8: 258.

Hu Y, Smyth GK (2009). ELDA: extreme limiting dilution analysis for comparing depleted and enriched populations in stem cell and other assays. J Immunol Methods347: 70-78.

Huang S, Podsypanina K, Chen Y, Cai W, Tsimelzon A, Hilsenbeck S et al (2006). Wnt-1 is dominant over neu in specifying mammary tumor expression profiles. Technol Cancer Res Treat5: 565-571.

Lee EC, Yu D, Martinez de Velasco J, Tessarollo L, Swing DA, Court DL et al (2001). A highly efficient Escherichia coli-based chromosome engineering system adapted for recombinogenic targeting and subcloning of BAC DNA. Genomics73: 56-65.

Li C, Hung Wong W (2001). Model-based analysis of oligonucleotide arrays: model validation, design issues and standard error application. Genome Biol2: RESEARCH0032.

Muyrers JP, Zhang Y, Stewart AF (2001). Techniques: Recombinogenic engineering--new options for cloning and manipulating DNA. Trends Biochem Sci26: 325-331.

R Development Core Team (2011). R: A language and environment for statistical computing. R Foundation for Statistical Computing: Vienna, Austria.

Shackleton M, Vaillant F, Simpson KJ, Stingl J, Smyth GK, Asselin-Labat ML et al (2006). Generation of a functional mammary gland from a single stem cell. Nature439: 84-88.

Siwko S, Bu W, Gutierrez C, Lewis BC, Jechlinger M, Schaffhausen B et al (2008). Lentivirus-mediated oncogene introduction into mammary cells in vivo induces tumors. Neoplasia11: 653-662.

Smyth G, Hu Y, Dunn P, Phipson B (2010). Statmod: Statistical Modeling. R package 1.4.8.

Stingl J, Eirew P, Ricketson I, Shackleton M, Vaillant F, Choi D et al (2006). Purification and unique properties of mammary epithelial stem cells. Nature439: 993-997.

1