TRANSFECfION OF THE BREAST CANCER CELL LINE MDA-468 WITH ANTISENSE RNA TO P21

(1)

(2)

(3)

(4)

(5)

1.1

The author has granted a non- exclusive licence allowing the National Library ofCanadato reproduce, loan. distribute or sell copies of this thesisinmicroform, paper or electronic formats.

The author retains ownership of the copyrightinthis thesis. Neither the thesis nor substantial extracts fromit maybe printed or otbenvise reproduced without the author's permission.

L'auteur a accorde une licence non exclusive pennettant a 13 Bibliotheque natiooaJe du Canada de reproduire. peeler, distribuee ou vendre des copies de cette these sous laforme de micro6chelf:ilm. de reproduction sur papier ou sur format electronique.

L'auteur conserve 13 propriete du droit d'auteur qui protege cettc these.

Ni la thesenides extraits substantiels de ceUe-ci ne deiventetre imprimes ou autrement reproduits sans son autorisation.

0-612-55536-4

Canada

(6)

TRANSFECfION OF THE BREAST CANCER CELL LINE MDA-468 WITH ANTISENSE RNA TO P21

^CIPl

IN

ORDER TO INVESTIGATE THE MECHANISM OF EGF-MEDIATED G, ARREST IN THESE CELLS

by

A.ndriPaquin

A Thesis submitted 10 the School ofGmduate Studies in partial fulfilmeDt oftbe requirements for the degree of

Master's of Science

Faculty of Medicine Memorial University of Newfoundland

Submitted: August, 2000

SLJohn's Newfoundland

(7)

ABSfRACf

The human breast cancer cell line MDA-468 is characterizedbyan ovcrexpression ofthe epidennalgrowthfactor(EGF)receptor. As a result oftrus overexpression,tl.{DA-468cells Ilavethe unique characteristic of undergoing arrest of cell cycle in OJ when treatedwith ptwmacological ooncentrations of EGF. Whilethe ability of EOF 10 cause arrest of this cancer cell lineiswell documented,the effectorsthatmediate this arrest remainunknown.

Since:thetumoursuppressorgene, p21Cipl.isapotentinlnbitorofeyetin dependentkinases withawell-definedabilityto conferGtarreston tumour cells,itisproposedthatp21CipI may playa roleinthe EGF-mediatedarrest ofthesecells. An antisense against p2ICipI expressionwastransfected into MDA-468 cells to test this hypothesis. Single-clone cell lines werederivedand examined for(i) an elimination ofp2ICipI mRNA and protein, and (ii) a reduction oCtheability of these cells to undergo Ol arrest when treatedwithEGF.Three antisense-transfected clones demonstrated that antisense expression caused a reduction of p21CipIwitha significantlydecreasedcapacity toundergoGtarrest.In thesecell lines an abscocc:ofp2ICipI reduced the ability of EGF-treatrnent to causegrowtharrest. This c::orn:lationmayindicatethatp2ICipI is an important player in the EGF-mediatedgrowth arrestofMDA-468 ccUs.

i i

(8)

ACKNOWLEDGEMENTS:

{wouldlike togiwmy thanJrs and appreciation to Dr. JonChurch whose paliena and guidance I could always COIDIJ on. A big lhanks to Dr Ken Kao, Dr. Laura Gillespie, and Dr. Gary Paterno for their endless technical support, andfor a roofover my head

Tomy many colleagues at the Terry Fox Lab.1was always grateful jOr our collaborative andfriendly workplace. andfwish you continued success in all yow endl!QV()T$.. I am forever iNkbted to Ms. AnnLodwoodfor herpa1iOl{%and technical training. Iamonly hope /picked upQfracrion of her Jrard+work and dedication. A sincere thank-you(0Paula who war always uvai/able10help in any way.

Tomy family whose support and love Ihavealways cofUidered myself luclry to hC1\le. Thanla Mom andDad for being a ray of sunshine and a tremendous source of strength. And a special thanks to Isabelle, my iove.for kicking my butt to wrap this1Lp.

i i i

(9)

~ ~E:===~:::=:=:~:=~::::=~==::=::=:==:=:=:=::~

d. lderl!ificalioa ofdooesbeariDg Iiga:cdinxrt-.._ _ _.__._ _.__. .23 e.. IdelttificWoaoforienwion ofligatediuselu_ __..__.._ .. ._..._.24 iv)StabletnllSfettion ofMDA-46& CelIs._...._ •.•..•...•_ .._•.__.._ .•... _ .24

L Calcium. phosphate lrWISfeaion _ 2-4

b. FrcWngc:ellsu.ndc:rtiquiclnitrogen.. _._ 25

c. GenenDon ohi.n,g.le done oelIlirw::s...._._ .•. 25

v)NorthernBlot Analysis _ 26

L Isola1iooofRNA _ _ 26

b. Prep&mion ofNorthe:m blot... . 27

e. Venemion of probes for Northern blot analysis. . .. 21

d. Random primer labelling. . 28

e. Hybridiwionof probe to RNA blot 29

vi)Westem blot analysis... . ._._ 29

a. WholeeelllYlllte(denaturing)... . 29

b. Immunopreeipir.ation(nondenaturing).... . 30

e. S05-PAGEand.electroblottingofproIeins _ 31

d.lmmllnodetec;tionofproleinbloU 31

vii) EGF-mediatedgrowtharrest .. .. •..•...•...•.••...••...• . _ _ 32

3.RESULTS ..._.

i)EGFi~~lalioaofp21C~1...•...•..__.

ii)AMlysisofp21^0f01proleiainMDA-46&oeIls._..

iil)Consuuc:tiooor.p21~1antisense expra.sionveaor.._ _ iv) SeReningofliptedplasmicb_._..._.__

v) Der:enniningthe orienwion ofligated insen .._._.. __

. 34

.•...•....•.. _._. . __ }4

. _.. 31

..._.. 42 ... 42 ... 44

VI)Transfectioa of plasmid into MDA-46& cells 4S

vii) lsolationofsiogleeionesoftransfectedMDA-'l6I._ 48

viii)Cellgrowtbexperimenu 49

ix) Growthofsense-lransfectecleells ... . _ SS

x) NortbemAnalysisoftransfectedc:ells _ SS

(11)

1U1Ana1ysisofp21C1p, pn:lteiDiDsinaJedone$

lrii} TImeCQW"5eofEGFeffea:songrow!b. __

••._.•.. S9

... _.. 61

4. DISCUSSION 70 i)Upreguluionofp2I~'m.RNAinMDA-468cellspriol'toG,atT$. ...•.•..._70 ii} Westananalysisofp2ICip'inMDA-468g:Us.._..•. iii} AnliseDsetop2ICip,_._._.. __._ ••••••.. _•.• ..•••• ... 10

iY)Effectofantixnse expres1ioo on p21Cip' mRNA •..__ .•__.._ _ __._.__..__.._ LShonleog1haDlisemcmol«:uIes._ _.._ __.. _... . 10

b.TltJditlgaDtisensetotheuntransJ.tcdregion$oftargetmRJ.lA._... . 10

v) Effect ofanti_ expression on p21Cip'pmIemlevels. ...•...•... 10

vi) Gmwtbanalysisofarrtisense-tnlUfectcdceJls _ ...•... 70

vii) Gmwthofsenseuansfeetedsinaleclones ...•.••...•.•.... _ _.•.• viii}fnducibleexpressionofarttisense. ix)Tune CQW"5e ofEGF on tbegrowth ofMDA0468 and linaJecloDel It)EGF-lbedialed arrestinA431 celb: A comparuive am.I)'Sis. __ .__ Jd)Alltiseme to p21C1p, ill.A431 cellJ ...•._..• ltii} lnvolvemellt of Olher Cdk inhibitory prolcins.... Itiii}Proposedmechanisms for EGF·med,iatcd upregulation ofp21CipI ...

~ =\~::=~~:~;~i:&~:~o,:.;;:-ei~·:

SoCONQ.USION •.••_ ,. RJ:FERENCES .... vi . 10 ... 10

.._.._.._.•.. 10

._.10 . 70 ... 70

... 10

...._.. 70

(12)

LIST OF ABREVIATIONS

BSA BovineSenlm Albumin

CMV Cytomegalovirus

DMSO Dimethyl 5Ulpboxide EDTA EthylenediarniDetertraaeeticacid FACS Flourescnce-aetivated cell sorter fBS Fetal Bovine Serum GTE Glucose-Tris-EDTA buffo:!"

HBS Hcpc:s-bufferedsaliDC HRP Horse radish peroxidase

mRNA messengerRNA

MOPS 3-[N-mofllboline}-propane sulfonic acid NaOH Sodium hydroxide

PAGE polyacrylamidegelelectropboresis PMSF Phenyl methyl sulfonyl fluoride 50S Sodium doceyl sulphate

sse

St.a.ndard salillC citrate SSB SDS-sample loading buffer SSPE Standard saline buffered EDTA STE Sodium-Tris-EDTA buffer TAE Tris-aeew&-EDTA buffer TBE Tris-borale-EDTAbuffer T8ST Tris-buffem:lsalincwithTween d.ATP OcoxyadcoQsinctriphospbate dTJl> Ocoxytbimidinelripbospluue dGTP Dcoxyguanine aipbospbate dCTP Dcoxycytosinetripbosphale

vii

(13)

LIST OF TABLES

Table I: EffeaofEGFonthegrowtbofMDA-468aodtraosfeaedcd1Iines..__ _._ __.._. __.. _.50 Table 2: Tnnec:oursc: of tileeffectofEGF 00growth

owr'

days. ••••••••• .._ _._._ ...•.65

LIST OF FIGURES

Figure1; Nonban bioi:analysisoftiledoserapome&nd10118 tam effect ofEGFOftp21Cipi apcession

in PomA-468edIs... . _ __.35

Figure2:NortherublotanaJysisoftbeeffea ofEGf on p21a,.1 expression during onset ofG1arrest. ••...•37 Figure 3:Nottbem blot ana.Iysisorp2I~1express:iOD \Ipon shon·tcnn EGF ueatment and removal from

EGF.... . 39

Figure4; Westem blO1 analysisofp21ClplproteininMDA.-468 cells 41 Figure S: Diagram of the plasmids pcDNA-neQ and pBS-eipl(T7) and l'e$Ult ofXhoIdigest....• , 4) Fiaure6: Elhidillm bromide stain oflwo gels used to screen cloning of the anlisellSe and sensepllCipl

expression vectotS...•...•... _ 46 Figurc 7:DiagrnnofliS'tion procluets and diagnostie digestwithBgllJlXhoJ 47 Figure 8:Pcrocnl:change inJTOwtho(EGF-trel1lcd MDA-468 and .oli_tra.os(eeted cells. .:52 Figure9: Uotn:ItcdgrowthofMOA-46IandantisensccloncsovcrSdays.._ _ .:54 Figure10:Pcrcemcbangc inJTO"'1ho(EGF-trelItcd MDA-.468.saue-UIll$fecrcd.and ami$e.-

lIUlSf'eetcd.._... . _._ _ .._...•._ _.._ _.._ ....•...:56 Figure II:Nonben:Iblotanalysiso(thecff«:tofEGFonp2lo.t~ainMDA-468.antiscrtseand sensetnDtfcctedcelb .._._ _._.__ _ ...•..._. .:5S Figure 12:Hisl:ognmofinteasities ofbaDd in Figure IIA. _ _ ._.._ _._ __.__.. __ 60 Figure I): Wcstcm blot aoa.Iysis aa the cxpression ofp2lo.t aftern bolntreatm=ntfOfMDA-468.the

twoantismse-trantfecud cdllinc:s: 6-25 and 6-26, and the sense-uaosfcaed cdllinc 7-6. __ 62 Figurel":Wcstenlblotana.lysisoothecxpres$ionofp2Io.tafternboln~for-MDA-468.the

&rJtiscnse...moeetcdcdllinc6-l9 63

Figurc IS; TuoccowweffeetofEGFtrealJnCI\tongrowth..(I) Figure 16: TuneooutSCcffeet ofEGF lRatmeot ongrowth.(2)•....

Figure 17: Tunecourse cffeet ofEGF treatment 00growth.(3)•.•...

viii

..•..•...•...••... 67 . .... _ ..•... 68

. 69

(14)

1.INTRODUCITON

Thediseaseof cancer is characterizedby cells which haveacquiredtheability to proliferate indefinitely. UDder nonnaIgrowthconditions. cellsdivide:whenpropergrowth signals exist and likewise cease proliferation when inhlbitory signals prevent division.

Tumour celis, on the other hand, are cells which have undergone mutations such that their proliferation is no longerreguIaledleading to neoplastic or uncontrolledgrowth. With further mutational events, a neoplastic cell can adopt an aggressive or malignant phenotype which is u1timatc:ly fatal to theOt"ganism(Levine.1993).

To betterunderstandtheprocessof cdl division investigators have sought 10 determine what oornpoocnts of the ocll triggersDNAsynthesisand eventual cell division, what signaJs cause the cells tobemitotic attimesand quiescent at others. and most importantly. what causes the loss of this control, ultimately leading to the fonnation of deadlytumours.

;)TJr.~CdJ Cycle

Toacquirea better undc:I'Slanding of theprocessof twnourigcnesis. investigators focused00the centralmachineryof cell division todeterminewhatfacton governoormaJ.

growthand what ew:nts foster de-n::gulatioo to allow malignant cells to develop.Theresult of this search has developed whatisrefened to as tbecell cycle.Thecell cycle is a sequence of evenu or steps a cell must completeinorder to divide inlOtwonew daughter cells. TIle critical dements of the cycle involve duplication of the chromosomes, assembly of the

(15)

mitotic spindle, andthe segregation ofcbromosome:s intotwonew cells (Earnshaw, 1988;

Nicklas, 1988).

Toprqmefor a mitotic event,. a great deal of coordination isrequiredto ensure that each step inthe cycle is initiated and oompl.etod sequentially. For example theprocessof chromosomeduplication consists of DNA synthesis.. foUo~ by chromosomal condensation and then attachment tothe mitotic spindle (Wolniak, 1988). Completion of eachstep involves the timely synthesis. assembly and activation of ahostofproteins to mediatethesefunctionsin propersequence. Equally imponant is that key enzymes are inactivatedinproper sequence forthe processof division toproceedfrom one stage to the

1be coordination oftheseevents is referred to as ceU cycle control. Normal dividing cells can be thought ofasthosecells that make their waythroughthe cycle only whenproper growthpromotion signals exiSl (Murray 0 01.. 1989a). Tumour cells, on the otherhand.pass through stages ofthe cyclecootinually and are unresponsive togrowthinhibitory signals.In otherwordsthey haveescapedcdJ cycle cootrol (HartwelleJoJ.,1994)..

iJ)Slqnofth cdlq-dl!

Thecell cycle isdefinedby 5 different stages inthe preparation ofamitoticevent.A new daughter cell is usually quiescent orinGophase. At this stage the cellis dormantin termsof proliferation and depending onthe cell type.Gomay last a few hours, a few days or an infinite amount of time. When stimulated to divide by growth-promoting signals. such as growth factors. cells exit Go and enter a preparn.tory stage of the cycle known as 01stage in

(16)

3

whichthe cell activates gene productsrequiredforthe initiation of ONAreplication.At the end OCGI. the celllnitiatesthe replication ofcbromosomal DNA andissaidto have entered S phase. Following the complete replication ofthe DNAinS phase. the cycle enters a second preparatory stage, G1•in which the components required for theprocessof cytOkinesis are assembled and activated. At the end of 0 1• the cell cycle culminates in M phase when the cell segregates the chromosomes and physically separates into two new daugbterceUs (Nicldas, 1988).

The last decade basseen the emergence of a theory ofcell cycle controllhat involves regulationbypbospborylation.Thephosphorylation of a host of regulatory proteinsmodify thciractivity. assembly, stability. orsubceUular1ocati0ll.to mediate progressthrougha given stage ofthe cycle(Ducommun.1991).Theclass ofkinascs most responsible for cell-cycle specific phosphorylation are the cyclin-dependent lcinases(Cdks) (Hooteret aI.,1994).

iii)Cye/in Dependent Kinu#!S L CdksiDyeur

The eyelin dependent kinases are a highly conserved fantily of serinelthreoninc Icinaseswhose activity is critical for progression through the cell cycle.These proteins of roughly331cDawere initially characterizedinyeast as the gene product cdc2(£ell givision 0'C1e)inSchizosaochoromycespombf:and cdc2SinSaccharomyces cerevisae (Reed.1991). Through the use oftempernture sensitive mutantsofcdc2 and cdc28. itwas revealedthat actively dividing cells became cell cycle arrested at either Glor G2when grown at the nonpennissive temperature (Rowley etai.,1993). This indicated that the

(17)

4

activityofcdc2andcdc28was essemiaIfoc continuedprogressionthrough the01and01 phasesoftbe ceUcycle(WittenbergeJ

m.,

1991).Intheabsenceofcdc2 or cdc18 activity the ceUs could no longer progress through the cell cycle andbecomearrested.

Theactivityof cdc2 andcdc28is dependent00the bindingof regulatory proteins namedcyclinproteins. Theeyelinproteinsbrieflyappearinthe ceU cycle and bind and activate cdc2or cdc28 in a ceil cycJe-specific manner (Had....igereloJ.,1989). Due to obligatory binding ofeyelin proteins to achieve activity,cdc2 and cdc28~renamed cyclin.

dependent kinases (Cdks).

Theobservation that temperature sensitive mutants of the Cdks encoded by cdc2and cdc28 caused arrest atboththeOdS and OM boundaries, indicated that in yeast cells a single Cdk molecule isrequired for progression through both the G lIS and G2IM boundaries. This dual activity is achieved by means oftwotempor.illy expressed subsets of eyelins which bind and activate thek:ina.se.the G1cyclins and the<hcyclins.Inprq:mation for S phase. the Gl cyclins areexpressc:dand activatecdc2 or cdc:28inO. anddirectthese kinasestotargetS thatmediatepreparationof DNAsynthesis(Reed,1991). Up:m completion of Sphasethe level of O. cyclinsdropsignificantlyandthe level of G1cyclins rise. HenceatG~the 01 eyelins bindthe Cdk to activate the oompoP.Cnts of the cell responsiblefoctheprepan1tiODof mitosis. Thus. in yeastcells. a singleCdk:has differential activityattwodifferentphasesofthecell cycle mediated by binding different eyelin panne<S.

(18)

"

b.CdkiallUllMDlliiaacdls

Inmammaliancells,ceU cycleregulat::ionis more complex asthereare at least seven identifiedCdksnwnberedCdkltoCdk7.These proteinssharesimilar-sizeandbomology and are stably expressed throughoutthe cycle.Inaddition there are at leasteight types of mammalian eyelin proteins known(AtoH)which Conn complexeswithone or more ofthe Cdks. In mammalian cells thetimingof Cdk activation isdefinedbythe cyetin partner 10 which they bind (HuntereJaJ., 1994).

iv)MtIMINIllUmCYCh"IU

1bc: eyeLinproteinsare so namedbeca.a!seof their cyclicappearancea!spcclfictimes inthe<:eU cycle.WhIleCdIcsare stablyexpressedthroughout thecycle.theeyeliDproteins have a short duration ofcxpl"CSSionmediatedby both a bricf period of transcription andbya rapid turnover rate oftbe protein (Sherr, 1993). For example, in GI ,the Ot cyetins(Dand E)are brieflyexpressed and the proteins utKIergo rapid degradation due to a sequence00 theirC·termioal which confers instability (Munay el01.,1989b).Incontrastthe OJ. cyetins, A and B. becomee.xpressc:dinlate S orGzph&se(Pineset0/., 1989) and are relatively stable throughoutCh and MThesecyclinsbecome rapK1.Iy degradedatthe conclusion of mitosis by ubiquitin-depcndcnt degradation (GlotteretaI.,1991)througha conserved~onin eyelinAandBknown as the destruction boxwtUchpromotes ubiquitin conjugationand subsequentdegradation (Minshull et al., 198:9). Thus by these mechanisms mammalian eyelins are presentat high levels only at specificstagesof the cycle. Consequently, they are presentto activate theirCdk: partners onlyat~defined times.

(19)

6

.LCydia0

Thelimeyelin tobeexpressedinthe cycleiscyclin D. D-typecyclins are most closely linked withsignalttansduction astheirexpression is dependent on the presence of growthfactors (Sherr. 1993). Stimulation of ceUs by growth factors induces the expression of eyelin0which signals that the cell cycle has exited Go to enter 01phase(Baldin et01., 1993). Inmammalian ceUs, O.type cyetins bind and activate Cdk2 (Xiong et01., 1992), Cdk4 (MatsushimeeJai.,1992) and Cdk 6 (Meyerson etaI.,1994)inQI. Wilhdrawal of growth factorsduring G1stops transcription of eyelin 0 anddue to the rapid degradation of the protein its levels rapidlydeclineand the cycle halts (WoneJaL,1992). This indicated thatthe continued expression of eyelin0inO.is necessary for the cycle toprogresstoS phase. Infact. eyelin0basbeer1identifiedas an oncogene as there are twnour ceU lines withconstitutively elevated levels of eyeliD D which continuously promote progression through01(Quelleetal., 1993).

b.CydinE

[nlate GI ,cyclinE is expressed and reaches peak levels at the OIlS transition and sharply declines asceJls progrc:sstbrough Sphase(KoffeJaL. 1992). Cyclin E binds to Cdk2 andthiscomplexisIeSpOr1$I.bIefor the initiation of DNA synthesis signaling entry of thecycle into S phase (ObtsuboeJaL,1995). Inactivation of cyelin E by microinjcetion of antibodies to cyclin E block DNA synthesis and cause Glarrestoftbe cycle.

c.CycliGA

FoUowingthe initiation ofS phase, eyelin E is degraded and eyelin A levels increase and bind to Cdk2 (Rosenblattet al.,1992). Evidence suggeststhateyelin AlCdk:2 activityis

(20)

7

necessary forthe completion ofDNA replicatioo yet isnotDCCCSS3IY for the OIlS transition.

IntheabsenceofeyelinA,DNA synthesiscanbeinitiatedbutdocsnotcontinueto completion. Hence, progress through S phase is regulated bythe eyelin EICdk2 complex which is required for the initiation of DNA synthesis and eyelin NCdk2 activity wl1ich is requiredforDNAsynthesisto reach completion.

d.CydinB

In01 and M phase. cyclin A and B are present and are boWld to Cdkl. Cdkl activity is required (oc the breakdown of nuclear laminainG} and signals for preparation of cytokincsis (HeichmaneJaJ.,1994). At thec:ndofM pbasetenninatiOllofCdkI function is requiredfoc complete division to occur. hs function isterminatedby ubiquitin.mediated degradation ofcyctinsA and B. With inactivation ofCdkl, mitosisiscompleted.

v) Regullll;on ofCdIrs by phosp#rorylat;on

Inaddition to the binding of a eyetin partner, cyclin-dependent kioases are also regulated by phosphorylation.AllCdksrequire phosphorylation of a conserved threonine- 161 (Thr.I60inCdk2)to becomeactive.Pbospbofylation of this threonine residue causes a conformational shift in thecyclin-Cdk:complex nocessaty for substrate binding (Morgan.

'9OS).

Cyclin dependent kinases are also negatively regulated by phosphorylation at residues tyrosine.15 and threoninc·14 (McGowan and Russell, 1993). Thereare two separatekinasesthatare responsible forthe initial phosphorylation of residues tyr-15 and thr-14.Thesekinaseshavebeencharacterizedinyeastasweeland mild (Lee etal., 1994 ;

(21)

McGowaneJaL,1993). Itisthought that pbospborylation ofCdksbythesekinascs coostitute thefirstpost-translational modification of the Cdk (pooneJoJ., 1995).

Theenzymes that catalyze the removal ofthese inhibitOly phosphates are the cdc25 family ofphosphatases (cdc25a, cdc25b,and cdc25c) Each enzyme of this family acts at a different phase of the cell cycle andisresponsible for activating the corresponding Cdk complexes fonned at those times.

vi) CIt«kpo;1IlColllro/ ofthecell cyde

Iti5 well ac:eepted that it is in Gl, JX10rto DNA synthesis.thatthe cdl maintains the majority of control on theCdksandconsequently00.the cell cycle(HaItWe1iet aL,1989).

Inmammalian cells this checkpointisreferred to as the restriction point, oc the R point, which occurs in late G1priOTto initiation ofS phase (fsaiet

m.,

1993). At this point in the cycle, dividing cells maintain a mechanism to monitor the state of the cell before proceeding to S phase. ShouJd cell division be undesirable the ceU remains at G1and the cell cycle is saidtohaveanested(Sherr,I996).

Theloss of these critical checkpoints can lead to an increase in mutation and eventual transformation of the cell(HanwelletaL,1994).Inthecontext oftbe cell cycle.

tumourigenesis~tswhen cellsDOlonger-respond toinhibitory signals and continuously proceedthrough checkpoints in the cycle.Inthe event ofdamagetothe DNA. the loss of checkpoints allow mutations to accumulate in the genome ultimately leading tofurther deregulation of growth and the development of aggressively growing tumour cells (Coxet aI., 1995). With the awareocss that cancer cells evolve by escaping growth controls, it is

(22)

9

obvious that agooduodcTsrandingofboth thegrowth-promotingand growth-inhibitory signalswhicbcootroI.the function ofCdks at these checkpoints isnecessary.

vii)Cdl-<ycle COil/rot by Rb

The chief meansbywhich the OJ checkpoint is maintained is through the regulation of the retinoblastoma protein (Rb)byCdks (Sherr, 1994). Rb and its two homologues. pI07 (Ewenetai.,1992), and pl30 (HannonetaL,1993).arephospboproteins which act as rcpre:ssor'$of the cell cyclein0.. Rb and its bomologues. bind and inactivate the E2F familyoftransaiptionfactorswbichare n:spc:lllSlble fortransactivating aQLDnberofgenes involvedin DNAsynthesissuchas DNA polymense-a. (DcGregori etaI.,199:5), dihydrofolate reductase (SlanskyeIaI.,1993), thymidylate synthase (Johnsonetof., 1994) andcyclins o(HerbereraJ.,1994) and E(Obtaniela/.,1995). However, when bound to Rb theE2Fproteins are prevented from nausactivating theseSphase genes and the cell cycle remainsillOJ. Inaddition, Rb-E2F complexes act as repressors of transcription at the promoters of these gcnes(Slanslcyelal.,1996).

However, the Rbproteincan only bind E2F proteins whenina bypophospbo()'lated state(Nevins.,1992).Inearly O. Rb protein isinthislowphosphorylatedstate and thusis found in complexeswithE1F proteins. Upon stimuJation of cdk4 and cdk6bycyetin 0and, laterinGl, Cdk2 bycyclinE, the activatedCdkspbospbcxylate the Rb protein converting it toa hyperphosphorylated state. Asa result of this phosphorylation by theCdks.Rb is released from the E2F proteins which are then free to transaetivate the genesrequiredfor the cycle to enter S phase. Hence the ability ofCdksto phosphorylate Rb is an imponant

(23)

10

function rocCdIcs tosignalprogression through the cell cycle.Forcells tomaintaina Gl checkpoint. phosph(Jlylation ofRb byCdksneeds tobekeptin check..

viii)The tumour su.ppressor ge.ne p53

Anothersignificant regulator ofGlCdks in tenns ofcheckpoint control is the tumour supp~pS3.Thisproteinof53 kDa basa~U-documentedrole in twnour devdopment.

Todate.roughly70%ofall primarytumoursindicate a loss of p53 function due to either a point mutation or deletion onone:orboth of the chromosomes (GT"oenbtan etai.,1994)and its absenceintwnours ofall tissue types suggests a universal role for p53in the management of normal cdlgrowth.

Currentlyit is believed that p53 regulates normaJ cellgrowthby way of monitoring DNA integrity. When cellseocouoterDNAdamage..especiallyinthefonnof ONAbreaks.

the levels ofp53 protein~and induce Glanest oftbe cell cycle(Kastan.1993).

Arrestof the cell cycle by p53 is thought to servet'oYOcritical functionsin the Jlftvention of tumourigenesis. Firstly. byhaltingthe cell cycle prior to S phase, the cell is prevented from propagating errors in the DNA onto progeny cells. The arrested cell is glven time to activate therepairmachineryandremove mutationsintheDNAtoensure genomic integrityismaintained (SmithetoJ., 1995.). Secondly,incases~DNA damage: is too severe.the cell cycle arrestalso allows thecelltoaaivateprogrammed cell death (apoptosis) (LoweetaI.,1993). Activation of apolXosis in the face of severe DNAdamageis an evolved safety me:e.'wtism whereby it is more favorable to trigger the death of the cell rather thanaUow propagation of error.;inthe ceU's DNA(Kastanelal.•199.5). Obvlously. cells

(24)

11

lackingfunctionalpSJlosethiscriti<:aJ. respoosc toDNAdamageand permit the accumulation of mutationsinthe genome ultimately contnbuting to nunourigenesis(Smith elo1.,I99Sb).

be)The p21O,t tumour suppressor gene

Withthe successfulcloning of a full length p53 eDNA, functional analysis revealed thatthis twnourst1pp[eSSOr"functions as a transcription factor (Zambettiet ai.,1992). The search for genest:ransactivatod by pS3 discoveiM GADD45 (Chindai.,1997a). mdm·2 (PerT)'eJaJ., 1993) and cyclin G (Smitheral.,1998). Tbemostsigni6c:antgenemcdiatedby pS3in the regulation ofCdks following DNA damagewasa 21kDa proteinwhichwould increaseuponp53activation.

Investigators focusing on pS3 as a tzanscription factor, namedthep21 protein WAF·

1 @ldtypep53 Activated tTagment) or p21,..n (el Deityetai.,1993). Sequence analysis showed that itwasidentical to a 21·kda protem simul!aneouslybeingstUdied for its interaction and intubition of the eyeLin dependent kinases. Investigators focusing on this aspect of p21. termed the protein

elf

I (eyelin dependent kinase inhibitory protein-I)Of p21 ....¹(Harpel"et01.,1993). p2lciplhas beenshownto bind and inhibit all G. cyclinlCdk combinations. UJRgUlation ofp2I<ip1 by p53 halts the function ofall cyclinlCdk complexes eausinganest oftbecell cycle (Xiooge/oJ., 1993).

Tbediscoveryofp2I~1providod direct evidencefor-the role ofpS3 in the inlubition of Cdk activity in maintaining a GlcheckpointInnonna! cells, stimulation of pS3in vivo by DNA damaging agents causes a dramatic increase in the levels ofp21^oipl(el·Deityet al..

(25)

12

1994).Thep2tciplprotein bindsand inactivates Cdk-cyclin activity. With the inactivation of CdksinGl •Rbremainsin abypopbosphorylated state andremainsboundtoE2Fproteins with theconscqueDOCof ceU cyclearrestinG] (DulictUm.. 1994). Inaddition, OverexpressiODof p21Q>1inthe absence of activation by pS3 could causegrowtharrest whentmnsfected in established tumourceUlioe:s(CheneJaJ.,199:S ; MichielietaL.1996).

Incells Lacking pS3function.thebasalexpression ofp21cipl is dramaticallylower and upregulation ofp21c:ip1byU.V.• gamma-irradiation or other DNA damaging agents is not observed (Coxel al.,1995). Functional Cdks accumulalc, phosphorylate Rb, and signal progression throughthe cell cycle. With the loss of this checkpoint, mutationswithinthe DNA remain unrepaired and are duplicated andpropagatedinS phase ultimately contnbuting 10 a malignant phenotype (CoxetaI.,1995). This led to the hypothesis that Cdkscould escape nonnal growth controls due to a lack of lnbJbition by what are now called Cdk-inhtbitoryproteins.

x) CdJv.itUriblJory prouilU

Following the discovery ofp21CipI as an inbibitorofCdk: activity, omer proteins with similar function have been identified(BledgeeJaL.1994).These regulatory proteins bind either directly totheCdk molecule ortoacyclin-Cdk complex to inlubitkinaseactivity.

OnceboWKl,the Cdkactivity is inactivated and progressionthroughthe:cell cycle is halted.

These proteins. generally referred to as Cdk inhibitory proleins. are divided into two families, the CipllKipl familyandInk:4 family.

(26)

13

..pZ~"'~

Theproteinsp2~1(PolyaketaL.1994) andpS~(LeeuaJ.,1995)have ahigh degree of homology withp2l~1and.like p2lcipl. repress Cdk activity by binding cyclin- Cdk complexes.Inparticulartheyinactivate eyelin 0 and cyetin E complexeswithCdk2.

Cdk4, and Cdk6 in 01.Thus, transfcction of these Cdk inhibitory into established cell lines induces01arrest. WhilepS7JCipJ.istissuerestricted(LeeetaI.,(995). p27Kiplappears10 havea universal role in managementafGo.Thelevelp2jCiPl are highest throughoutGoand early G.(PaganoeJai.,1995).The current model rOf" the action ofp2Ti"1 is that it provides athreshold of Cdkinhibition whichmust be ove:rcome by mitogen activation. Infact.

eliminationofp27~lwithantisense removes the requirement ofgrowthfactors for cells to divide (CoatsetoJ.,1996). Although p2,G'1 andpSjGP1are related top21cipl they are not regulated bypS3DorOO they participate in aOt checkpoint inresponsetoDNAdamage.

b.Thelnk4f.mily

The lnk4 family ofCdks lnhibitors differ from the p21^coplfamily in that they intubit Cdk function by binding din:cdy to a single Cdk molecule to form an inactivehc:terodimer (SettanoeJat,1993). These CdkinhJ.bitors prevent eyelinbindingand retainthe Cdk inan inactive state. Members of this family include pI6W:· (Serrano etoJ•• 1993). plS- (Hannonetai.,1994). p18^1Jl4c(GuanetaL.1994), andpl~(ChanetoJ., 1995)andbind toCdk4aDd Cdk6 in Gt•TheywerenamedInk4 for their observed ability to inlubit Cdk....,!

(Sberrelo/.• I99S).

Each of the Ink4 inhibitors can cause Gtarrestin tumour cells and their loss is implicated in twnourigenesis. Their roleintheinhibition of Cdk4 and Cdk6 appears to

(27)

14

specifically modulate pbospborylation of Rh. Upon tnmsfcctionirttoestablishedoelIlincs.

the 1NK4proteinsbind to Cdk4 and Cdk6.whicbpreomrtsphosphorylation of Rh. thereby causing Glarrest(SherretaI., 1995).

However, unlike the p2lCipI.family

oredle:

inhibitors, the INK4 proteins are unable to induce01arrestinceU lines that lack Rb despite fonning complexeswithCdX.4 or Cdk6.

This provided evidencethai: Rb is the principle substrale for Cdk4 and Cdk6 and in ceUs lackingRb, Cdk4 andCdk6 activitymayberedundant (Lukas

elm.,

1995).

With the awareness that cancercelIsevolvebyescapinggrowthcontrol it is obvious thatagood understandingofthe mechanismofgrowthanest:meritsinvestigation. Inthe studypr-cscntcdherethe role ofp21CipI in Glarrest is examined using the cell line MDA·

468.

xl) The MDA-468 bre4Jt cancer ceU Une

Thebreast cancer cell line MDA468 is a unique cell line characterizedbyan overexpression of the epidermalgrowthfactor (EGF) receptor (Filmuset oJ.,1985). While otbercells ofmammary origin express IO't-IOS EGFreceptOr$,MDA-468 cells expressinthe on:Ier of

uf

recc:plOrSpercell, resulting from a 20..fold duplicatioo.(i.e. tvoenty copies) oCtbe geDefor thereceplof"(Kawamoto etal..1984). Furthermore, MDA-468 is unusual in that uponexposure to EGF at pharmacological concentrations(ic.10~.cells in vitro respond byundergoing reversiblegrowtharrest within 24bows (Filmus etoJ.,1985). Cell cycle analysis by FACS showed that thearrestoccurs in the Glphaseof the cell cycle and upon removal ofEGF from the media the cells re-enter the cell cycle (Chun:helai., 1992).

(28)

.5

TheabilityofEOF 10inducegrowtharrestisadep3JtUrefromitsweU-documenled role as a potentmitogen(Carpenter,1984). rn the case of"MDA-468 cells, it y,uuld seem thatgrowtharrestbyEGFislikelythecoasequeoc:e of signalingthroughtheoverc:xpresscd receptors onitsswface.IndcedanodIcr- wetlstudiedcelllioeA431,anepidermoidcancer cellline.alsoovere:xpr-cssestheEGF~rto thesamedegreeas MDA-468(Merlinoet oJ., 1984)andundergoesEOF-mediatcd Glarrestin thesame manner(Barnes,1982).

However, there are ceillincsthatoverexpress the EGF receptOr whichdoDOtundergo EGF- mediatedgrowtharrest but adopt an proliferative pbenotype (Pandiellael of..1988). Hence EOFreceptor overexpression alone is not a prerequisite :forgrowtharrestand arrest is therefore most likely due to an altered signal transduction pathway within these cells.The exact mechanism by which EGF transduoes a signal for cell cyclearrestin MDA-468 cells remainsunknown.

xii) Possible rokfo, p21O,l in EGF-1Mdilued growtlt arrest

Since targetingCllIlOercells to cease prolifc:rn.tion is agoalofcancc:rresearch.the mechanism by which EGFcan cause a tnmsformedcell line 10undergocell cycle arrest may beof clinical significance. lnitial studies sought toexamine a posstble role for pS3 in the EGF·induced O.arrestof this cell line (Prasadel a/., 1997a). Given its welldefinedrolein growtharrest,itwaspostulated that signal transduction viathe activatedEGFreceptor could increase pS3 function thereby inducing cell cycle arrest

ftwaspostulated thatEGF signalinginMDA-468 migbt cause an induction of pS3 activity by post.translational modification. most likely by phosphorylation, such that this

(29)

16

induction of activity resulted inGtarrest.Thec:oofocmarionalflexibility of p53froman inactive to active state lends well tothis idea(HaJazonetiseJaL. 1993).Indeed.EGF causes changesin pS3 pbosphorylation. protein confonnation and subceI..Iular localization in these cells (prasadelal.,1997.).ThepS3 proteinexpressedinMDA-468 cells is from a single allele with a mutation at codon 273 where an asparagine residue is substituted by histidine (Nigro et aJ., L989). Althoughthisis a mutant p53 molecule, this particular mutation, pS3^l7lbio•hasa curiousphenotype.

The p53 mutantatcodon273(acgtohis)basbeen dubbeda-pseudo-wildtype"

mutant TransfectiOll ofthi$ moleculeintovarious cellsbas shown the capacity to mediate bothoncogenicand,in somecases.growthsuppressive activity (GoUaboo el aI., 1996).

Whilemost pS3 mutants lose their ability to bind DNA {KerneIaI.• 1991}, pS3^m.Hisclearly is able to bind10the pS3 consensussequence(pS3CON) and can transactivate reporter constructswithp53CON while other mutants ofpS3donot (parket af.,1996). Most significantly thep53²ⁿ.Hiomutantiscapable of transaetivating a CAT reporter construct containing the p53CON element when transfected into MDA-468 cells. Inaddition expression of this reporter constructwasincreasedwhcuthe ceUs were treatedwithEGF (Prasadelai.•1997b).

pS3V'l.Hi1also has theuniquecharacteristic ofbeingmcognizedby bothwildtype spocific anti-p,53 annbodies as well as mutant specific antibodies.This observation indicates tba1this peculiar mutant mayretaina certain degree ofwildtypeconformation which may explain its ability toperfonnwildtypefunctions. Inthe context ofMDA-468ceUs, treaonent withEGF causes a shiftinthe proportion of p53~Ieculesdetectedbythese

(30)

17

antibodies such that theproteinbecomespn:dominantly~bythe wild type specific antibody(PrasadeJaL,19978).Coocuneut to theobservedshift to wild type conformation, theproteinmigratesalmost exclusively to the nucleus prior tothe onset ofgrowtharrestof MDA-468 by EGF. Theseobservations may indicate a gain of function and local.ization to the nucleus where it canact as a transcription factor of the genes wbich mediate the growth arrest. Presumably, p21C4>' might be one ofthem.

Thestudy presented here focuses ona possible role for p2leipl in this EGf-mediated growtharrest. Given its potent ability to inactivate the 0,Cdks.itwaslogical to pursue a role for p2ICipI inmediating thegrowthinhibitory response oftbese cells to EGF. Thus it wasproposed thatEGF-mediatedupn:gularionofp21"",1 would lead 10 a saturating amount ofp21.,;p1 in G. Cdkcomplexesto causegrowtharrest.

1bespecific goal ofthisstudy is to use antisense to p2loipl to remove it from a putative role in EGF-mediatedgrowth arrest. Theoretically, antisense technology "knocks out" the expression of an endogenous gene by presenting the complementary RNAstrand whichbtnds 10 the endogenous message.Theresulting double stranded complex becomes a substrate for degradation by the nbonuclease RNAse H (Bonham.eJaJ.,1995). By this manner the levels of an endogenousmaNAcanbe reduced orabolished,consequently eliminating orreducingthe AmOWl ofmessage which can be translated intoprcxcin.

Shouldremovalofp21ciplfromthesecells eliminate or diminish the ability ofEGF to inducearrestthen it could be suggested thatp210iplplays a roleinthe signal transduction pathway by which EGF causesgrowtharrest in these cells.

(31)

18

2. MATERIALS AND METHODS

i)T1I~MDA-46lJ cell line

The cell lineusedinthisstudywasthe breastC<lI1Cttcell line MDA-468obtained fromDr.Ron Buick of the Ontario Cancer Institute. This immortalized cell Line of breast epithelial origin is maintained in Leibovitz medium (L 15) at 37"CinForma Scientific tissue culture incubators. Mediawasprepared by filter Slerilization through 0.221J. filter (Nalgene) 10 whichwasaddedfeta.IbovineselUlD (Gibco BRL) to loo/a, penicillin to 10 mgtml and strepl:omycinto 50 mgfml (Gibco BRL). Mediawasstoredat 4"C andbeatedto 37"C prior tocellcultureU5C. Cellsweremaintained in 75 an²(Costar)flasksuntilreaching80".4 confluence whereupontheyweretrypslnizedandreplatodat a concentration of lxia' cells.

ii)BQcterial cells and plasmid gruwth .. Truasfor'DUIlionofbac:reria

BacterialceUs (JMI09 or MCI06IIPJ) were inoculatedin5mIofLS media and grown overnightat31"Cwithconstantshaking. The oextdaythe cell suspensionwas diluted 1:200 and placedina

sao

mlflask.. Over Jto 4 bows 1mlsamples ofthe suspensionweretaken todeterminethe coooeotnltion of cellsbymeasuring oprical density using aBeckmanDU~spectrophotometer at a wavelength of600 run. Upon reaching an absorbance of 0.400 the cell suspensionwaspouredinto SO ml Oakridge rubes and placed on ice for 10 minutes. then centrifugedat 4000 rpm for 10 minutes at 4°C.1becell pelletwas rinsedin an ice-<:old solution oCO.IM CaChand spunagain at 4000rpmfortominutesat

(32)

19

4"C.The finalpellet: ofcells wasn:suspcndedin 2mIof iee-cold O.IM eaChand left at 4"C for24bows and used for transformatioo with purifiedDNA01"wasaddedglycerol 10 a final concentration of15% to sUn as a stock of transformatiooaJly competent cells at-7rJ'C.

b.TradonurioBornceerialcdls

Inallcasesp1asmidswere b'anSfected into bacterial cellsmadetransfonnationally competentby the calciumchloridemethoddc:scnbedabove. Competent cells were thawed on ice and25·50 ng of plasmid DNA (max.. of 10J.LI)'MiSaddedto the cell whichwerethen left on ice for an additional 30minutes. Cells were beat shocked 81

4rc

for 90 seconds and cbiUed on ice for 2 minutes. 0.8mlof LB minimal brothwasadded10the cells which were then incubated at 37"C for 45 minutes. Each sample was plated in triplicate on LB agar plates containing ampicillin (SOlJ.gtml)and tetracycline (ljJ8fml) and incubated at 37"C overnighl

CoLa..,escaAePI"q)

or

purifiedplasmid

Bacterialcultwes.fromfrozen stocks or from coloniesgrowing00agar plates, containing plasmids ofintcresl weregrownindividually overnight(18hr)at37"Cin200mI of LBwithselectable annbiotic. Recovery and purification of plasmid DNAwasperformed usingthealkalinelysis method(Sambrook::etaI., 1990}. Briefly, overnight cultures~ centrifuged in fourSO ml centrifuge bottles at 5000 rpm for 1S minutes at 4"C. The pellet wasrinsed in 25mloeSTE buffer (0. 1M NaO. 10mM Tris (pH 8.0) and ImM EDTA) and centrifuged again at 7000rpmfor 15 minutes at 4°C.1be pellet was resuspended in 6.5 ml ofGTE buffer (50mM glucose. 25mMTris(pH 8.0), 10mM EDTA) on ice, after whichwas

(33)

20

added10mlofalkalinelysis buffer (02M NaOl:Ul.O%50S)andleft onice for 5 minutes.

Thesolutionwasthenocutnilizedby mixingwith7mlof ice-c::old 5M potassium acetate for 10minutes on ice. Cellular debris and chromosomalDNAwasremoved by centrifugation at 12,000 rpmfor15minutesat4°C.lbe supernatantwasremoved and precipitated at-2O"C with0.5 volumes of isopropanol for 30 minutes. Nucleic acid precipitatewascollected by centrifugation at 8000 rpm for 10 minutes at 4°C. thenwashedwith700/0 ethanoland left to aicdry.Thepellet wasresuspendedin4.2mlof50mM Tris (pH 8.0), lmM EDTA. To generate a cesium chloridegradient.4.7g ofcesium chloride and O.5ml of ethidium bromide (10mgfml)wasadded to the DNA suspension.Thismixturewasspunat8000rpm for\0 minutes at4"Cand the supernatantwastransferredinTOBedcman#342412QuickSeal ultracentrifuge tubes and sealed. Tubeswerecentrifuged at 45.000rpmina Vti 65.2 rotor overnight (-18hr) at ZOGe. At this point the plasmidbandwasvisible by exposme to long wavelength ultraviolet lightand wasextractedbya syringe insencd into the tube where the plasmid band had settled. Plasmid DNAwasextractedwithanequaJvolume of butanol and then diluted to 3 volumeswithSOmMTris (pH 8.0). ImM EDTA and then precipitated in 95% ethanol at-2O"Covernight.. ThepR:Cipiwewascentrifuged at 8000 rpm for 30 minutes,washedYlith 70%etbaool and~in.steriledHzO. Estimation ofDNA cooccntratioowasdonebyspectropbowmeter readings at260nm. PIasmidswerethen storedat4"C.

d. SmaUseale prep ofbacterialptuaJid

Five ml or LBwithantibiotics were inocuJaled under sterile conditionswithceUs pickedfromacolonyon an LB agar plateandgrown overnightwithshaking at 3PC.The

(34)

21

nextday. a l.SmIaliquotwastakenaDd placed in a sterile I.Sml Eppendorftube. The remainder of the cell suspension'MlSplaced on ice and kept ifneeded or stored at·7rI'C as a glycerol stock. TheI.Smlaliquotwas centrifuged atSOOOrpmfor 2 minutes at 4°C. The supernatantwas removed and cells were resuspended in 100J.1\of GTE buffer. To lhiswas addc:d200 ml ofalk:aIinelysissolution. The suspension was then mixed for 5 minutes at room temperature.oeuttalizr.dwith1.50J.llofSM potassium acetateand placed on ice for 10 minutes. This mixturewascentrifuged at14,OOOrpm for5minutes at4"C to remove chromosomalDNA and cellular debris. TIle supernatantwastIansferredto a new Eppendorf microcentrifuge tube and extractedwithone volwne of phenol:chloroform:isoamyl alcohol (25:24:1). Theaqueousphasewasthen precipitatedwith2 volwnes of 95% ethanol and storedat ·20OC foral least an hour.The tubes were centrifugedat14.000rpmCor 20 minutes at 4"C yielding a pelletwhichwaswashc:d with Im1of7O'A. ethanol, centrifuged again and left to air dIy. TheplasmidwasresuspendedinISOJ.1\of sterile dHJ0 for analysisby restriction enzyme digestion.

Iii)ColIStnu:t/on. ofapression vectors .. VectorplaJmid(pcDNA-aeo)

Thevectorcbose:nto deliverp21""1 anti--sense expression into MDA-468oeUsis the eukaryotic expr-ession vector pcDNAI-oeo (lnv;trogen). ThiswidelyusedexpressiOll system makes use of a cytomegalovirus(CMV) promoter sequence, an SV40 poly- adenylationtail.and the gene expressing resistance to neomycin toactas a selectable marker for transfected cells.Thisvector also contains a multiple cloning site adjacent tothe CMV

(35)

22

promoterto allowfat-efficient andspecific ligation of inserts (Fig.SA). This plasmidwas dooak:dbyDr. LauraGillespieand was transfccted and stored in the bacterial. cellline MCI0611P3by the method described above.

b. cDNAinsenofp21O,I

The antisense expression vector was constructed using a eDNA copy of the p21^Cipi gene. Thiswasobtained bygenerous donation from Dr. David Harper who lrindlyoffered thep2IClpI eDNAcloned into the plasmid pa5-Cipl(I7) (Fig.SB) whichwastransfected into thebactcria.I celIlineJMI09forstorage.

TheeDNA insert encoding p21CSp' was cutfrompBS-Cipl(f7) with. XhoIand purifiedfromanaga.rosegelfollowingelectropbore:s:is.Inlhis procedure the entire digestion reactionwasloaded onto a I%agarosc: gelinTAEcontaining Ip.gfmJofethidium bromide andrun at80voltsin IxTAE buffer. lbe excised fragmentwasvisualizedby exposwe to low wavelengthU.V. and cut from the gel. The gel slicewasthen placedindialysis tubing containing0.4 ml of h:TAE. The ends of the tubing were sealed and immersed in amini- gel electrophoresisapparatus..TheDNAfragmentwasthen electroeluted at SOV for 1hour fromthegelsliceintothebuffer withinlitedialysis tube.Thebufferwithinthetubingwas removed andpurifiedby phenol:ch10r0rorm:isoamyl alc:obol (25:24: I) extractionand precipitation intMJvolumes of95%ethanol, 0.1 volumes of 3M sodium acetate and 2J.dof glycogen (Boehringer). Precipitationwas done at-2O"Covernight. FoUowing resuspcnsion insterile dH20theconcentration ofthcbandwas estimated byrunningvarious amounts on a

(36)

23

1% agarose gelwithIJ.lw'mletb.idiwubromideand comparing the intensity oCtile staining with thatofa llcb DNA ladder-marker.

CoLiplioaofpZlOfoiilttoVC'ClOr

For ligation oftheinsert,the\'tlCtOCplasmid pcDNA·Neo-Iwaslinearizedby digestedwith XhoIovernight andpuifiedbypbcooI:chlorofonn extraction and ethanol precipitation as describod above. 100 ng oflinearizcd plasmidwasligared with 500 og of p21Cipl eDNA(isolated as above) with the T4 ligase enzyme (Gibc:o BRL)in a total volume of500J,J..l at 37"C far2.Sbrs.

10IJ.,Iofeach ligase reactionwastransfectedinto competent bacterial cells MCI0611P3 by the method descnbed above. Transfected cells were plated on agar conlaining ampicillin and tetracycline andgrownovernight. Single colonies of transfected bacterial ceUswereselected to identify colOllies containingthe expression vector with ligated p2ICipI eDNA Thiswasdoneby mini-prep ofbactcriaJ plasmid This isolation of pJasmidbymini-prepissonamedasitinvolvesthe same method of bacterial plasmid isolation asthe maxi prep describedabo..-ebut isquicktt and deals withmuch smaller quantities of plasmid. Thisissuitablefor scrceniog a largeDumberof positive bacterial colonies on aplate.

d..Idmtific:atioa ofdoDesbeariDa:"amiinKrt

Identification ofbactcrial clonesbearingpcDNA·neowiththe p21~1nagmentwas done by digestion ofmini·prep plasmidwithXhoI.The plasmid samples from each colony of cellsweredigested and the products of digestion were run on a 0.7% agarose gel

(37)

24

containingI~gImlofethidiwnbromide.ThebandsYoUevisualizedby exposure oCtile gel to ultraviolet light. P!asmids containing the insert will release a 2.0kb insert as a result of thisdigestion. Plasmids withouttheinsert will appear as a single band -6.75 kb.

eo Idmlificadoo of orinuatioo of._kdiD..5erb

To identitY the orientation of the insert within the multiple cloning site of pcDNA-neo, bacterial plasmid was digestedwith bothBgJ II and Hind Ill. The products of the digestionwererun on a 0.7% agarose gel and visualized as described above.The antisense expressionconstruct was labeled p68CiPl-AS (Fig.7A). and the sense cxpn:ssiogconstructwaslabeled p74CiPl-5 (Fig.78). Wherepositive resultswere confirmed the resulting plasmid constructs weretransfectcdinto MCI061/P3and propagated to yield high quantities by maxi· prep isolation descnbed above.

iv) Stable transjection of MDA-468 Cells a. Caldwn pbospbak traDsfecfioo

lvIDA-468cellsweregrown to 8O"/a confluency in10em dishes (Fisher) and b'aDSfectedbythe calcium pbospbate method (Chen, 1988). Afineprecipitatewasfonned inlxHBS using 100 ng of vector by slowty adding (dropwise) a solutionof2MCaa~to a finalcoocentration oro. 125M This mixture was added to the cells in SmILIS media and leftat 37"C

rOT

3 hours. The media was removed and 15% glycerol in IdlBSwasadded to thecellsandleft for I minute. Cells werewashedtwice wiL"- PBS and givenfreshmedia.

1be following day. the cells were trypsinized and replated in media containing 0.6 mglml of G4 I8 (Gibco BRL) to select transfected cloDes. (From this point on all transfected cells are

(38)

25

growninmediacontaining 0.6 mgfml of gcncticin to maintain selection.) CeUswen:

~andfreshmediawithgencticin was addedevery twodays to removedeadcellsand keepselectionstringent Aftereightdaysof selection numerous surviving colonies were viSible. These cellswerepooled together 10 fonn a mixed population of positive transfectants,portions of whichwerefrozen under liquid nitrogen.

b. Frcedot:cdbUDderliquid Ditroana

Incryogenicvials (NaJgeoe)toImIsuspeosion of cells was added 0.1mlFBSand 0.1mlofOMSO. VJalswereplaced at-7O"Cinan isopropanol cootaineT for2hJS and then immersed inliquidnitrogen.

Co

ee-...

Iioaof~1ecto.ecd __

Fromthemixedpopulation oftJansfectedceUs, single cloneswereobtained using 96 microwellplates(Fisher). Cell suspensionswerediluted to a concentration of 0.9 cells per200j.Ualiquot toincreasethe probabilityof generating a colony of cells from a single cell origin. After a week ofgrowthonly those wells containing colonies of a singleorigin weremaintained and developed into a single clonal cell line. Upon growing 10 adequate numbers (-lxHf'ce11s)stocksof each single clone cell linewerefrozenin liquid nitrogen.

Theremainderwerepropagatedand used to study the effectofEGF-mediatedarrestonthe transfcctodceUs.

(39)

26

v) Nortlu!nt Blot A-qsU .. IsoIatiooofRNA

Total RNAwasisolalcdfrom ocllsusingTrizol solution (Gibco BRL). This solution containing pbenol and guanidium thiocyanate:wasaddeddirectly tothe ceUs to lyse andisolate nucleic acids. CeUs were plated atUxlo'cellsin10 em dishes (Fis!ler) and growninthe presence or absence of EGF(lO~for specifiedtime periods. Following incubationthe media was removed. andLO mlofTrizolsolutionwas added directly to the dishat room tcmpetature. After five minutes the cells were scraped using a cell scraper and the resulting lysatewastransferred to 1.5mIEppendorf tubes (Fisher). The lysatewas pipettedup anddownfor 30 seconds to aid homogenization and left to stand at room temp for5minutes.200JJlofchloroformwasaddedand vortexcd for IS secondsandlefttostand for 5 minutesatroom temperature to promote phase scpar.uion. Samples were then centrifuged at10.000rpnforIS minutesat4"C. Theclear upper phasewas carefully removed avoiding theproteininterface.and placedinoew eppendorftubes. Anequal volume of pbeool:chI0r0fonn:isoamyl alcohol (25:24;1)wasadded.briefly vortexed and centrifugedat12,000rpmfor5 minutes at 4"C.The upper phasewasthen precipitatedwith 0.5mIofisopropanoland stored a1 ·WOC. TheRNA precipitatewas collectedby centrifugation at 14,000rpmat 4"C for 25 minutes. Thepelletwas washedwith70%

ethanol, centrifuged again and then left to air dry for 15 minutes. TheRNA was resuspendedin100JJloCDEPe-treated water containing 0.5% 5DS and heated to 5S'C fot ten minutes to aid resuspension. 1be RNA concentration was detennined

(40)

27

spectropbotometcally (BeckmanDU~SpectrOpbotomeler) at wavelengths 260nm and 280tun..Sampleswere then sooted at-2O"C.

b.Prq»aratiollofNOf'1bt:mblof

To measure cxpn::ssion of p21CipIinaUceU lines,total RNAwasisolated and transferredtoGeneScreenmembranes (DuPont) for Northern bioi: analysis. Fifteen 10 twentymicrograms of totalRNA fromspecific timepolnts(+/. EOFtreatment)weremixed ina loading buffer containing 50'%(vlv) fonnamide. O.25M formaldehyde.,lxMOPS (3-{N- mOrpholino]propanesulfonic acid: O.2M MOPS pH 7.0, O.SM sodiumaeewe, O.DIM EDTA) and 4 J,1g1mlethidiwnbromide. Samples were runina 1.2%agarose gel prepared in DEPC- treateddHzO containingL2SM formaldehyde and IX MOPS. Samples were run for four hours at80 volts in I X Mops running buffer. After running the gel the largebandsof nbosomal RNAwere visualized by brief exposure to low wa..-elength U. V. and photographedas a means 10 assess equivalent loading of RNAThegelwas immersed in RNAse·fn:e water for 10 minutes to remove excess formaldehyde while the Gene Screen membranewasimmersedin RNAse-IRe dHz{) for 1 minute and then in 10xSSC (1.5M NaCl D.ISMrrisodiumcirnde211:0)for IS minutes. Acapillaly transfer apparatuswas set up to facilitateovernight transfer of the RNA tothe membraneusing1000SCas buffer. Upon disassembly of the transfer apparn.tus., the gelwas exposed 10 U. V. once again to assess transfer nftbeRNA from the gel to the membrane. A pholograph of this exposure was taken as well.Themembranewaswashedin 2XSSC solution and then placed on Whatman paper

(41)

28

toair dry for10minutes..OncedrytheRNAwasfixedto themembr.mc bybakingin an oven. at8O"Cfor2 hrsundervacuum..

Coee-nltionor probes For Nortbem bIoraoaJyJis

AU Northern blots wereprobedwith2.0kb)(hoUHindlll fragment of p21Cipl.Five micrograms of the plasmid were digested overnight at 37"C. The reactionwasrun on a txTBE gel conta.i.ning I llglml ethidium bromideand the 2.0kb fragmentwasgel purifiedby gel electro-elution(as descnbed above).

d.~primerlabelliD&

Usingarandompriming labelling kit (Gibco BRL),the 2.Okb fragmentwas labelled.withcr.·e:!p]<ICTP(DuPoo.t). Twenty-five tofifty nanograms of the fragment diluted to 20J,l1wasboiledfor5 minutes and placedonice. 15JLI of random primebuffer (GIbcoBRL) and 2!J.1 (O.OSOmM) each ofdGlP. dATP, anddlTPwereadded.5J.lIof a- (l2pJ-dCTP(SOuCi)wasaddedandthen1 lotiofK1enow enzyme (3.0 UnitslJ.L1)beganthe labelling reaction which proceeded at room temperature for 1.5brs. To terminate the reactionand Inseparatethe labeUedprobefromunincorporated nucleotidc:s,SOJ.1Iof lxTE wasadded and the entirem¥::tioamixturewasapplied toaIm1column ofG-SOSepharosc.

1bc labelled fragmentwascollectedbybricfcc:ntrifugation oftbe columnand atJJlsample wasadded to 3mlof Amersham BOS scintillantto mea$lR incorporated radioactivity by scintillation counting(BeckmanLS·380I). Inallcases only probes of specific activity greater than Ixlrl cpm!jJ.g were used for hybridization to the Northern blots.

(42)

29

e.HybridiulioaofprobelID RNAbIof

Priorto bybridizatiorl RNA blocswereplaced inscalable plastic bags and incuba1cd inhybridization solution for 3hours a142°C.The solutioD contained 100 ).lgIml sonicated salmon spennDNA, SXDenhardts(0.1%Fiooll 400, 0.1% polyvinylpyrrolidone, lmgfml BSA).

10% dextran sulphate., 1% 50S. IOxSSC. and SOOA, formamide. Following this incubation the solutionwasremoved and fresh hybridization solution containing O.SxI0⁶cpmlml of boLled radiolabelled fragmentwasadded andincubatedovernightat 42"C.

Blotswerewasbcd with2xSSCJO.S'YoSDSat42"Cand heated to6O"C. Afinal washwasdone aJ: O.sxSSC..(U%SDSat6O"CforISminutes. Following this\WSh the blots wererescaJedinplastic andexposedto XAR film (Kodak) for one tofive:days at

-nrc.

Densitometric scan of thefilmsweredone on an LKB Bromma., Enhanced Laser Densitometerto quantify the intensity ofthe resulting bands.

vi) Western blotllnalysis

For proteinanalysiscellswerelysed bytwomethods:

L W1IoR cdIlyuirt(~IU.riD&)

IxHf cells(+1-EGF treatment)werelysedin1.0mIof denaturing bufferheatedto 80-90"c. This buffer containedO.SMTris (pH 6.8). 2mM EDTA, 10% glycerol, 5%p...

mert:aptoethanol.and 10% 50S. Cellswerescrapedand placedin1.5mIEppendOlftubes andincubated at8Q^G

e

for another5minutes. Samples were then briefly sonicated to break up genomic DNA and any insolublematerialwasremoved by centrifugation at 12,000 rpm forISminutes at4°C.Theprotein concentration ofthesesampleswereassayed by a Bio-

(43)

30

Raddetectioo.kit. 200J.1Iof a diluted sample (-1:10)wasadded to O.8mI of coooentrated BiG-Raddetectionmlgent.Thesamplewasmixedand then read by spectropbotometerat wavelength A-595nm.. CoooentJationsweredeterminedby acomparisontoastandard curve using a10mw'mlstockofBSA(bovineserumalbumen) protem.Samples wcrethen storedat·70"C orpreparedinSD8-sample buffer (SSB)(lS%(v/v)glycerol, O.I25M Tris- C!,. pH6.8. 5mM Nal EDTA., 2%(w/v) 5DS. 0.1%(wlv)bromphenol blue. 1%(vlv) 2- mercaptoethanol)and loaded onto a 12% SDS separating gel using 151-18 of total proteinper sample.

b.ImmWlOpl'edpitlition. (non deutunnc)

1.5x10⁶cells (+/-EGF) were lysed in 1mlnon-denaturing lysis buffer containing 50mM Tris.HCI (pH 7.4), 150mM NaCI. 1% Nonidet.P40. SmM EDTA., 1 J,1g1ml PMSF (Pbeoyimethyl sulfonylOouridc). 2

1JWm1

aprotininand 1 J.l&'ml1eupeptin Cellswereleftat 4"<:for30minutes and then scraped intoEppe:ndorftubes. Sampleswereclarifiedby centrifugationat 10,000rpmfor IS minutcsat 4"C.Thesupc:matantwastransferred to clean tubes towhichwasadded2J.lgofanti-p2.ICipI anbDody(Oncogene Science) and leftrotating overnightat4"C.Thefollowingday. to theimmWlCcomplexwasadded 50JJ1of a 50%

solution protein G-sepharosebeads(Pharmacia) and rotated for Ihr. This complexwas broughtdownbybriefcentrifugatiOll.(10seconds).Thepelletwaswashedthree times with lysis buffer (as above) and thenwashedtwice with ISOmM NaC!. Theproteinwas resuspendedin3SJ.Ll of2xSSBandstoredat-70OC.

(44)

31

CoSDS-PAGEaDdeiectrobloc1iltofproteiDs

Fifteenmicrograms ofproteinwas~in S05-sample bufferandrun on a 12% $OS-PAGEinGibco BRL mini-gelapparatus and then clectroblotted to nitrocellulose.

Proteinsampleswen:boiledr~5 minutes and then loaded omo the gel using a Hamilton syringe. Samples containing immlJllClP«cipitatcs were resuspendedin35J.t!of 2xSSB buffer and the entire samplewas loaded onto the gel. Samples from whole cell lysate were R:SUSpendcd asIS 1J.8 ofprotcin inODevolumeof2xSSB.lnctudedwith the protein samples were low-rangepre-stainedmolecularweightmarkers(BiOo-Rad) and biotinylated markers (Bio-Rad). Gels were run at IS rnA for 30 minutes and then 30rnAfor 60 minutes in an 50S-Page running buffer consisting oro. 125M Tris. O.3M glycine, and 0.1% 5DS. At the endofthc run, the gelwassoaked in transfer buffer conlaining O.025M Tris (pH 7.4), O.20M glycine, and2()o~methanol forIS minutes. Nitrocellulose (Amersham, Hybood-ECL)was also soakedintransfer buffer. Gel and membnmcwm: assembledina sandwich and placed inagelapparatustotransferthe proreinsfromthe:geltothe nitrocelluloseat100 V forI.S hourat4"C.The transfer bufferwaschangedat4S minutes toIceepthebuffer cold.The membrane was then placedina solutionofTBST (1M Tris, SM NaCl. 0.1%T~)until needed.To assaypropertransfertbc:gelwasstainedwithCoomasie Blue and the blotswere reversibly stainedwithPonceau-S dye.

d.lmmuaodeta:tioa ofprolftD blots

Protein blots were blockedfOfDOn~spccificbindingwithTBST containingS%milk:

powder forIhr. Theprimaryantibody, anti_p2lCip1(Oncogene Science), diluted to21-lWmi

(45)

32

inTBST containing 5% milkpowderand 0.02% azidewasincubatedwiththe blot with gentle rocking overnight at room temperature:.lbc blotswerethenwasbedinTBST for one hour with vigorous shaking. ThesecondaryantI.body, goat anti-mouse whole antibody conjugated 10 biotin (Amersham). [and in one case itwassheepanti-mouse whole anobody conjugated to horseradish peroxidase(HRP)], was diluted to LlOOO in mST and added to the blot and rocked forthr. Blots were washed as above. Where the secondary antibody was conjugated tobioti:n.strqXavidin·horscmlishperoxidase(Amersbam) wasadded at a dilution of (:soo in TBST and rocked for Ih!. Blotswerewashedagain and then 1.0m1of detection reagent wasadded (Amersbam, ECL-dettctionkit). AftttIminute blots were placed~two sheets ofpta.sricpageprotectorsandrapeddown in a film cassette (Fisher). Chemiluminescent-sensitive film (Amersham; Hyper·film-ECL) wasexposedto the blot togetsufficient detection of proteinbands(S minutes to an hour). When required.

blots werestrippedinlOOmM 13-meTCaptoethanol. 2% 5DS. 62.SmM Tris-HCl (pH 6.7) for 30 minutes in shaking 70°C walt:r bath and rinsed extensivelyinTBST before storing at 4°C.

Effectiveness ofstrippingwasassayed by adding detection reagents and exposing to film.

vii)EGF~dgrrnvtJlarrest

MDA-468cellsor transfectod single cloneswen:plated intriplicatein60mm dishes(Nunclon)ata cooccntration of2xlO· ceUs. EpidennaIgrowthfactor (CoUaborative Research; natural mouse, culture gmde) was mixedwithLISmedia to a concentration of 10·

~andaddedto cells10Wldergo EGF-treatment; LIS alonewasaddedto control cells.

After EGF treatment for a givenperiod(i.e. four or eightdays) ceUsweretrypSinized in l.0

(46)

33 mI ofS%trypsin inPBS and resuspended in 0.5mlofmediatobecountedby haemocytometc:r.CeUstreatedfMIoogez-thanfour dayshad theirmediachangedon day four(washedoncewithPBSthenfreshmediawithorwithoutEGF).

(47)

34

3. RESULTS

i)EGF iruluces "Prep/4non ofp2lCipl

To elucidate a role for p2l^Cipiinthe EGF-roediated growtharrest of MDA-468, it was of interest toexamineifEGFtreatment alone affected the expressionorp21~1during thecourse ofGIarrestShould EOF treatment foster a change in the expression of p21Clp\

coocurrentwith growth MInt.then a rolef(Xp21Q>1 itIlhatarrest could be suggested.

To rnea.stR the effect orEGF on p21Q>1 expession. MDA468 cellsWCfetreated with EGF andanalysedby Northern blot analysis using probesto detect p21Q>'mRNA.

Cells were serum starved for 48 hours to removegrowthfactorsfromthemedia and then incubatedin serumfreemedia supplementedwithEGF at concentrations ranging from 10·

12Mto IO.3M for12 hows. Control cells were also serum starved for 48 hours but with subsequent addition of media containing100/0 FBS to give an indication of the expression of p21Ciplduringnormal cell culture conditions.

Figun:: IA shows that EGF can induce the expression ofp2lapl mRNA and does so inadosedependeot manner. Oearly the expression ofp21apl is dramatically inducedwhen cells are exposed toEafata concentration ofIO~the coocentrationwhich induces Or IlIIestin these cells (Prasadand ChwdJ. 1992). At lowerconccntIaIionswhichdonor:induce Orarrest.(lO·I<M or 1O"'M). the level ofp21CipI mRNA isnotsignificantlyincreasedabove control levels. At an EOF concentration of10'1~noexpressionofp21~1is seen. Since

(48)

35

A. EGF:

B. EGF:

Figure I: Northern blot analysis of the dose response and long term effect of EGF on p21Cipl expression in MDA-468 cells

(A) Northern blot analysis of the dose re5pGnse off2t^Clt'mRNA to EGF. MDA-468 plated ath:I06 andexposed toEGFatconcentrations ranging from10'2M ,10,10M. 10--9 M, and 10" M for 12 hours. After exposure tbe cells were lysed for RNA isolation using Triml solution (Gibco BRL). Twenty micrograms of totalRNA was run on a formaldehyde gel. transferred to GeneScreen membranes, and probed with a radiolablelled fragment ofp21Cipl eDNA. Labelled blot was washed and then exposed to Kodak X-OMAT filmovemight

(B) Northern blot analysis on the long-term elTert of EGF on p21^0tluprtssion. MDA-468 cells wert grown in the presence or absence of EGF al 10" M for 4, 8, and 12 days. After exposure the cells were lysed for RNA isolation using TriZo! solution (Gibco BRL). Twenty micrograms of I()(a! RNA was run on a formaJdehydege!and detected as described above.

(49)

36

the levels of p21CiplmRNA significantly increase when the cells are exposed to the growth inhibitory concentration of EGF (i.e, IO-8M), further references to EGF treatment in this text are assumed tobeat10-llM.

Induction of p21^CiplmRNA in these cells can also be demonstrated upon long term exposure to EGF. Figure 18 shows a Northern blot of cells treated with EGF (atIO~)for 4.8 and 12days. Clearly, strong upregulation of p21^Ciplis maintained in the continued presence ofEGF while levels remain low in its absence.

Inthe presence ofEGF at 1O.,'IM, MDA-468 cells begin to accumulate in 01within 48-72 hours (Prasad etaJ.,1991). Figure 2A demonstrates that induction of p21^Ciplalso occurs within this time frame. In cells treatedwithEGF from 12,24,48, and 72 hours.

expression ofp21CipI increaseswiththeduration of exposme to EGF, reaching a maximum

at72hours.

To examine the immediate effect ofEGF on the expression of p2Ic 'Pl.MDA-468 cells were exposed to EGF from 2 to 15 hours (Fig. 3A). Surprisingly, the early effect of EGF on p21^crplis quite dramatic. Within two hours of treatment there is a rapid increase in the level of mRNA (Fig. 3A). This initial response remains high for the first six hours then begins to decline, yet remains above basal levels, Previous studies have revealed that the G1arrest initiated by EGF in these cells is reversible UJXln removal of EGF from the media. Upon removal of EGF from the media, the cells were able to return to their proliferative growth rate (Prasadet al.,1991). The Northern blot analysis of Figure 38 demonstrates that induction ofp21^Cip1message by EGF is also reversible upon removal of

(50)

37

A.

control +14brs +48 hn +72hn

B.

• p21O,l exprc:ssion

EGYtreatment

Figure 2: Northern blot analysi of the effect of EGF on p210pI expression during onset ofG. arrest

(A) Northern bioi analysll of pIll"'l mRNA prior to onset of EGF·induee(i growth IrTtsL MDA-468 cells weregrowninthe presence or absence ofEGF III Icr' M for 24, 48,and72 houts. After exposurethe cells were lysed for RNA isolation using TriZolsolution(Gibco BRL) Twenty miaugrulsofIota! RNA wasrun on • formaldehydegetanddetca:cdasdescribed aboYe

(8)Hisl....m of intensitia of bud inrllUrt2A.Thefilm oftheI*ltio Figure 2Awasscannedby laserdensitometryandthevalues pre:seoted as fold inductionOYerthesianaJfromuntreatedMDA-468 cells.

(51)

38

EGFfromthe media.Ineach lane of Figure 38 all ceUs wen: grown in theprcscnceofEGF

ro..

72 hours to irritialcastrong inductionofp2I~'_1ben the cells~washedwithPBS and incubatedin media lacking EGF for anadditiooaJ2, 4, 6, 8, or 12 hours before lysis for RNA isolation. Clearly it can be seen that upon removal ofEGF, induction ofp21Cip\ ceases andits expression decreases within 12 hours. Lane n of Figure 38 depicts cells which were notwashedwithPBS but remainedinthecontinuous presence ofEGF for another 12 hours.

Inthiscase significant p21c",l expression is maintaioed.

Thusit can beSCICDthatthe level ofp2ICipI mRNAis drnmatically upregWaled by EGF.OCCW'SpriortoonsetofG\arrest,and thathighexpression ofp2IC1pI is maintained onlyinthecontinuedpre:seooe of EGF. Thehistogramof Figure 3C is generatedfroma scanaCmecontinuously.treated samplesfromfigure 3A,B (ie.Lanesa-g,and h and n). This histogramshows that EGF induction of p21c",1is somewhat bi.phasicinnalUR:. Initially, thereis a rapid increase in p2ICipI mRNA in thefirst.four hours, which quickly drops off.

However a second phase of induction occurs between 15and 72 hours.

;i)AlUl1ysis ofpZ/Cirl protd";/1MDA468 cdJs

Havingdemonstratedan induction ofp2ICipl mRNAbyEOFit was of interest to examinethe effect ofEGF 00the le..-el ofp21^CillI protein. Western blot analysiswas done using a mouse:monoclonalannbody top21~1(Oncogene: Science) as aprimaryannbody andchemilwnenescent detectionwasdone using either abiotin~njugatedsecondary annbody (FigAA) or an HRP-conjugated secondary antibody (FigAB). The isolation of proteinfromMDA-468wasby using either a whole ceU lysate or using protein samples

TRANSFECfION OF THE BREAST CANCER CELL LINE MDA-468 WITH ANTISENSE RNA TO P21

1.1

Canada

TRANSFECfION OF THE BREAST CANCER CELL LINE MDA-468 WITH ANTISENSE RNA TO P21

IN

ORDER TO INVESTIGATE THE MECHANISM OF EGF-MEDIATED G, ARREST IN THESE CELLS

ACKNOWLEDGEMENTS:

TABLE OF CONTENTS

~ ~E~~:~~===~:::=:=:~:=~::::=~==::=::=:==:=~~:=:=:~~:~

~ =\~::=~~:~;~i:&~:~o,:.;;:-ei~·:

LIST OF ABREVIATIONS

sse

LIST OF TABLES

owr'

LIST OF FIGURES

m.,

"

m.,

elf

oredle:

elm.,

uf

.5

2. MATERIALS AND METHODS

sao

4rc

or

rOT

ee-...

-nrc.

e

1JWm1

3. RESULTS

A.

EGF:

B.

EGF:

A.

B.

ro..

~ ~E:===~:::=:=:~:=~::::=~==::=::=:==:=:=:=::~