cyclin e both regulates and is regulated by calpain 2, a protease … · cyclin e both regulates...

Cyclin E Both Regulates and Is Regulated by Calpain 2, a Protease

Associated with Metastatic Breast Cancer Phenotype

Stephen J. Libertini,1,3Brian S. Robinson,

1Navdeep K. Dhillon,

1Danielle Glick,

1Michael George,

1

Satya Dandekar,1Jeffrey P. Gregg,

2Earl Sawai,

2and Maria Mudryj

1,3

Departments of 1Medical Microbiology and Immunology, and 2Pathology and Laboratory Medicine, University of California, Davis, Davis;and 3Veterans Affairs-Northern California Health Care System, Mather, California

Abstract

Overexpression of cyclin E in breast tumors is associated with apoor response to tamoxifen therapy, greater genomic instabil-ity, more aggressive behavior, and a poor clinical prognosis.These tumors also express low molecular weight isoforms ofcyclin E that are associated with higher kinase activity andincreased metastatic potential. In the current study, we showthat cyclin E overexpression in MCF7 cells transactivates theexpression of calpain 2, leading to proteolysis of cyclin E aswell as several known calpain substrates including focaladhesion kinase (FAK), calpastatin, pp60src, and p53. In vivoinhibition of calpain activity in MCF7-cyclin E cells impedescyclin E proteolysis, whereas in vivo induction of calpainactivity promotes cyclin E proteolysis. An analysis of humanbreast tumors shows that high levels of cyclin E are coincidentwith the expression of the low molecular weight isoforms, highlevels of calpain 2 protein, and proteolysis of FAK. Lastly,studies using a mouse model of metastasis reveal that highlymetastatic tumors express proteolyzed cyclin E and FAK whencompared to tumors with a low metastatic potential. Ourresults suggest that cyclin E–dependent deregulation ofcalpain may be pivotal in modifying multiple cellular processesthat are instrumental in the etiology and progression of breastcancer. (Cancer Res 2005; 65(23): 10700-8)

Introduction

Cyclin E, the regulatory component of the cyclin/cyclin-dependent kinase (CDK) complex, is overexpressed in differenttumor types and is often associated with a poor clinical outcome(1–9). The contribution of cyclin E to tumorigenesis has been beststudied in breast cancers where overexpression is apparent in 10%to 25% of breast cancers, is a strong predictor of endocrine therapyfailure (10), and an accurate predictor of a poor clinical outcome(9, 11). Studies of cyclin E–overexpressing breast tumors andtumor-derived cell lines found that cyclin E is also proteolyticallyprocessed into more active low molecular weight (LMW) isoforms(12) that are resistant to inhibition by the CDK inhibitors p21 andp27, and hence are associated with higher CDK kinase activity (13).Analyses using an in vivo melanoma model suggest that the LMWisoforms are more effective than the full-length cyclin E inenhancing angiogenesis and metastasis (14), thus establishing alink between the expression of cyclin E LMW isoforms andmetastasis.

Although studies substantiate the role of cyclin E LMW isoformsin tumorigenesis, the mechanisms involved in cyclin E proteolysisare unclear. An analysis of LMW isoforms expressed in a cyclinE–overexpressing cell line suggests that elastase is instrumental incyclin E proteolysis (12). Another study indicates that cyclin E canbe cleaved by calpain to generate LMW isoforms that are similar insize to the isoforms present in tumors (15). Therefore, at present,the protease(s) that generate the LMW isoforms in breast tumorsand tumor-derived cell lines are not clearly delineated.Calpains are calcium-dependent thiol proteinases that have been

implicated in many cellular processes. Calpain activity is tightlyregulated by levels of expression, calcium requirement, autopro-teolysis, phosphorylation, intracellular distribution, and theendogenous inhibitor calpastatin (16). In general, calpains cleaveproteins at a limited number of sites to generate large polypeptidefragments (16). Over 100 calpain substrates have been identified,and many are cytoskeleton/membrane attachment proteinsincluding focal adhesion kinase (FAK; ref. 17), a-spectrin (18),talin (19), paxillin (20), E-cadherin (21), and h-catenin (22). Calpainalso cleaves proteins that are involved in signal transduction andtranscriptional regulation such as pp60src (23), p53 (24, 25), and fos(26). Therefore, deregulation of calpain activity would affectproliferation, migration, apoptosis, and metastasis.We previously reported that cyclin E overexpression in MCF7 and

T47D tumor–derived breast cells confers partial tamoxifen resis-tance (27). We also noted that cyclin E–overexpressing cell lines haveincreased levels of the LMW isoforms. The current study shows thatin MCF7 cells, cyclin E transactivates calpain 2 transcription, thus,positively regulating its own proteolysis as well as the proteolysis ofother calpain substrates. An analysis of human mammary tumorsidentified a correlation between cyclin E expression, calpain 2 levels,and calpain-dependent proteolysis. Such modifications are consis-tent with the aggressive behavior and poor clinical prognosis that isevident in cyclin E–overexpressing breast tumors.

Materials and Methods

Cell culture. HepG2 cells, obtained from American Type Culture

Collection (Manassas, VA) and the Hct116 cell line, a gift from Dr. K.

Kaplan, University of California, Davis, Davis, CA, were propagated in

DMEM (Life Technologies, Rockville, MD) supplemented with 10% fetalbovine serum, penicillin/streptomycin at 37jC and 5% CO2. The generation

of the MCF7 and T47D cell lines has been previously described (27, 28).

Microarray analysis. Labeling of samples, hybridization to U133A

GeneChips (Affymetrix, Santa Clara, CA), staining, and scanning were doneas described in the Affymetrix Expression Analysis Technical Manual.

Fluorescence intensity values (.CEL files) generated from hybridized,

stained GeneChips were analyzed with R statistical software (v.2.01, and‘‘affy’’ BioConductor package) and BRB Array Tools to identify genes that

were differentially expressed. The settings used for Robust Multichip

Analysis in R included Microarray Suite 5.0-based background correction,

Requests for reprints: Maria Mudryj, Department of Medical Microbiology andImmunology, University of California, Davis, Tupper Hall, Davis, CA 95616. Phone: 530-754-6090; Fax: 530-753-8692; E-mail: [email protected].

I2005 American Association for Cancer Research.doi:10.1158/0008-5472.CAN-05-1666

Cancer Res 2005; 65: (23). December 1, 2005 10700 www.aacrjournals.org

Research Article

Research. on December 16, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

quantile normalization, and Robust Multichip Analysis–based algorithmsfor calculation of expression values using perfect match only fluorescence

intensities. A detection P V 0.05 and a mean fold change of z1.5-fold (P V0.01) were used as criteria for filtering genes for clustering analyses.

Hierarchical clustering and comparative fold-change analysis were used toidentify and group similar patterns of gene regulation. Assignment of genes

to functional categories was done by annotation of gene lists with the

program, Database for Annotation, Visualization, and Integrated Discovery

(http://apps1.niaid.nih.gov/david) and literature-based classification wasdone by hand. Statistically overrepresented (Fisher exact probability score

<0.05) biological processes within clusters were identified using Expression

Analysis Systematic Explorer v.1.0 analysis software (29).

In vivo calpain induction. MCF7 cells were plated in six-well plates and

cultured overnight. Cells were pretreated with 20 Amol/L of the calpain

inhibitor calpeptin (Calbiochem, La Jolla, CA) or DMSO for 15 minutes and

then treated with 10 Amol/L ionomycin (Calbiochem) for 20 minutes,

harvested, washed with ice-cold PBS, and lysed in radioimmunoprecipita-

tion assay buffer.In vivo calpain inhibition.MCF7 cells were plated in six-well plates and

cultured overnight. Cells were treated with DMSO, 5 or 20 Amol/L of

calpeptin, and 2.3 or 23 Amol/L of N-acetyl-Leu-Leu-methional (ALLM;Calbiochem) for 24 or 48 hours, at which time they were washed with

ice-cold PBS and harvested.

Transfection. Cells were transfected with Superfect reagent (Qiagen,

Valencia, CA) and 5 Ag of pRcCMV-cyclin E or pRcCMV plasmid DNAfollowing the manufacturer’s recommendation. For the promoter studies,

cells were transfected with FuGENE 6 (Roche Applied Science, Indianapolis,

IN) and capnP-luc2, SVh-gal, pRcCMV-cyclin E380, pRcCMV, or p21 plasmids.

Forty-eight hours after transfection, cells were harvested and subjected toanalysis as previously described (28). pRcCMV-cyclin E380 harbors a site-

directed mutation converting the tyrosine 380 residue to alanine, thus

enhancing cyclin E stability. The mutation was generated using the

following primers: 5VCCCAGTGGGCTCGCCCCGCCACAGAGC3V and5VGCTCTGTGGCGGGGCGAGGAGCCCACTGGG3V.

Northern blot analysis. Total RNA was isolated using the Trizol reagent

(Life Technologies) following the manufacturer’s recommendation. TotalRNA (10 Ag) was electrophoresed on a 1% agarose gel and transferred to

Duralon-UV membrane (Stratagene, La Jolla, CA). The following primers

were used to generate calpain 2–specific RT-PCR products: 5VAGGATGAG-GATGAGGACGAGGAGGAT3V and 5VTTTCAGAAAAGACCGGATG3V. TheRT-PCR products were labeled with a 32P-dCTP. The hybridization and wash

conditions have been previously described (27).

Generation of the calpain 2 promoter luciferase plasmid. Sequencescorresponding to �43 to �25 region (5VGCCCTGAACTTTGTGCCAAA 3V)and the �481 to �462 region (5VTTTGGCACAAAGTTCAGGGC3V) of the

calpain 2 promoter were used to PCR-amplify the human calpain 2

promoter using purified human DNA. The PCR product was cloned into the

pCR2.1-TOPO (Invitrogen, Carlsbad, CA) vector to generate pCAPNpro-

TOPO. The promoter sequence was excised by digestion with SacI and XhoI

and cloned into the Sac I and XhoI site of the pGL3 basic (Promega,

Madison, WI) luciferase reporter plasmid to generate capnP-luc. The

promoter sequence was identical to the one previously reported (31)

(National Center for Biotechnology Information accession no. J04700.1).

Western immunoblot analysis. Cells were directly placed in aradioimmunoprecipitation assay lysis buffer that contained E64, leupeptin

and calpeptin, and a protease inhibitor cocktail (Sigma, St. Louis, MO).

Thirty micrograms of protein were separated on 8%, 10%, or 12% SDS-PAGE

gels and transferred to BA-85 membrane (Schleicher & Schuell, Keene, NH).The membrane was blocked with 5% nonfat dry milk in PBS and 0.1%

Tween before the addition of specific antibodies. The following antibodies

were used: cyclin E antibody was a generous gift from Dr. G. Lozano. Actin(sc-8432 horseradish peroxidase), p53 (FL-393; sc-6243), and c-Src (B-12; sc-

8056) antibodies were from Santa Cruz (Santa Cruz, CA). InBa (Ab-2;

PC142), calpain 2-domain III (208755), p53 (Ab-6; clone DO-1; OP43), and

calpastatin, domain IV (208903), antibodies were from Calbiochem.Glyceraldehyde-3-phosphate dehydrogenase (GAPDH; RGM2) antibodies

were from Advanced ImmunoChemical, Inc. (Long Beach, CA), and FAK

(clone 4.47; 05-537) antibodies were from Upstate (Charlottesville, VA). Afterincubation with secondary mouse monoclonal or rabbit polyclonal

horseradish peroxidase–linked NIF 825 antibodies (Amersham Pharmacia,

Piscataway, NJ), proteins were detected using enhanced chemiluminescence

(Amersham Pharmacia) following the manufacturer’s recommendations.Histone HI in vitro kinase assays were done as previously described (28).

In vitro calpain assay. MCF7-RcCMV control cells were resuspended in

calpain assay buffer [50 mmol/L Hepes buffer (pH 7.4), 150 mmol/L NaCl, 1

mmol/L EDTA, and 1% Triton X-100], calpain was activated with differingamounts of CaCl2, and the extracts were incubated at room temperature for

60 minutes. To detect cyclin E proteolysis by purified calpain, MCF7-RcCMV

control cells were resuspended in calpain assay buffer, increasing

concentrations of calpain 2 (Calbiochem) were added to the reactionsand incubated at room temperature for 60 minutes. The reactions were

terminated by boiling.

Preparation of human tumor samples. Tumor tissue was obtainedfrom the Department of Pathology and Laboratory Medicine, University of

California, Davis, Davis, California. The samples were flash-frozen in liquid

nitrogen, and all identifiers were removed. A total of 11 samples (10 tumors

and 1 normal mammary tissue) consisting of a minimum of 100 mg ofbreast cancer tissue were available. This study was approved by the

Institutional Review Board of the School of Medicine, University of

California, Davis, Davis, CA.

Mouse metastasis assay. The metastasis studies were conducted aspreviously described (30). The tumor transplant lines, PD5 and PD6, were

developed from rare pulmonary metastases and were transplanted into the

no. 4 mammary fat pad.

Results

Generation of cyclin E–overexpressing clones. Cyclin E–overexpressing MCF7 mammary cell lines were generated to definecyclin E–dependent alterations in gene expression. Two clones,MCF7-CycE1 and MCF7-CycE3F that expressed intermediate andhigh levels of cyclin E, and had elevated levels of cyclin E–dependentkinase activity were chosen for further studies (Fig. 1A and B).Notably, MCF7-CycE1 and MCF7-CycE3F cells had abundantamounts of the LMW isoforms of cyclin E that have been previouslydetected in cyclin E–overexpressing tumors (12, 15), whereas MCF7control cells expressed only trace amounts of these isoforms.Overexpression of cyclin E in ZR75, LNCaP, PC3, HEK293, and T24cells also resulted in the expression of the same-sized LMW isoforms(data not shown). These observations raised the possibility thatcyclin E, when overexpressed, might promote its own processing.Cyclin E overexpression transactivates genes involved in

catabolism. To determine the effects of cyclin E overexpression inMCF7 cells, microarray analysis was used to analyze changesin gene transcription. The analysis was conducted in triplicate foreach line. Statistical analysis of the microarray data resulted inidentification of 645 genes that were differentially expressed ( foldchange z1.5; P V 0.01) in cyclin E–overexpressing MCF7 cells. Asexpected, MCF7-CycE1 and MCF7-CycE3F expressed increasedlevels of cyclin E, as well as decreased levels of genes that wepreviously found to be repressed in cyclin E–overexpressing cells—Bcl-2 (28) and RB.4 Hierarchical clustering of differentiallyexpressed genes revealed similar patterns of regulation in bothcyclin E–overexpressing cell lines (Fig. 2A). Interestingly, onesubcluster of genes up-regulated by cyclin E expression are genesinvolved in catabolism (Fig. 2B). This subcluster contained calpain2, which was previously implicated in cyclin E proteolysis.

4 B. Robinson and M. Mudryj, unpublished data.

Cyclin E Transactivates Calpain 2 Expression

www.aacrjournals.org 10701 Cancer Res 2005; 65: (23). December 1, 2005



Calpain 2 expression is elevated in cyclin E–overexpressingMCF7 cells. The elevated expression of calpain 2 in cyclinE–overexpressing MCF7 cells was confirmed by Northern blot analysisof mRNA isolated from MCF7 control and cyclin E–overexpressingcells (Fig. 1C). Western immunoblot analysis established thatcalpain 2 protein levels were elevated in cyclin E–overexpressingcells (Fig. 3A). GAPDH and h-actin served as gel loading controls.To determine if cyclin E directly transactivates calpain 2

transcription, we cloned the human calpain 2 promoter into thepGL3 luciferase expression vector. The 456 bp promoter containedsequences from �25 to �481 of the translational start site. Previousanalyses found that this promoter has multiple transcriptional startsites, and harbors Sp1, Ap�1, Ap-2, E-box, and E2F-binding sites (31).The capnP-luc reporter construct was transfected into MCF7 cellsalong with the cyclin E expression plasmid or with a vector controlplasmid (pRcCMV). As shown in Fig. 1D , cyclin E transactivatesluciferase expression. To determine if CDK kinase activity wasinstrumental in promoter transactivation, cells were transfectedwith the capnP-luc plasmid, the cyclin E expression plasmid, and thep21 CDK inhibitor expression plasmid. Inclusion of p21 resulted indecreased promoter activity, arguing that cyclin E–dependent kinaseactivity is required for calpain 2 transactivation.Proteolysis of known calpain substrates in cyclin E–over-

expressing cells. Calpain cleaves a number of cellular substrates togenerate LMW isoforms; therefore, detection of these cleaved formswould be indicative of elevated calpain activity. Previous studieshave shown that FAK, a 120 kDa protein, is cleaved by calpain to a90 kDa form (17). MCF7 control cells predominantly expressed the120 kDa full-length form of the protein (Fig. 3A). MCF7-CycE1 cellsexpressed full-length and cleaved FAK, whereas MCF7-CycE3Fexpressed mostly cleaved FAK. Therefore, progressively higherlevels of cyclin E result in a dose-dependent increase in calpainproteolysis of FAK.

Calpastatin, an endogenous inhibitor of calpain is a verysensitive calpain substrate (32). Whereas calpastatin is encodedby a single gene, multiple calpastatin isoforms were detected inMCF7 control cells (Fig. 3A). In contrast, calpastatin was barelydetected in MCF7-CycE1 cells and was undetectable in MCF7-CycE3F cells. In this system, an increase in cyclin E causes a drasticdecrease in calpastatin expression.Another calpain substrate, pp60c-src, is proteolyzed by calpain

to a 48 kDa form in platelets (23). This truncated protein has adeletion of the NH2-terminal region which harbors the myristoy-lation site as well as a deletion of part of the SH3 domain. As shownin Fig. 3A , the MCF7 control and MCF7-CycE1 cells expressed low,but detectable levels of the 48 kDa isoform. However, MCF7-CycE3F cells expressed equal amounts of the full-length and thetruncated 48 kDa form of pp60c-src. Therefore, whereas pp60c-srcis cleaved by calpain in vivo , it is not as sensitive to calpain-dependent proteolysis as calpastatin or FAK.In vitro studies showed that human and mouse p53 are calpain

substrates (24, 25). Human p53 is cleaved to a 48 kDa NH2-terminally truncated isoform. The full-length, but not the LMWisoform, can be detected by an antibody directed against the NH2-terminal domain of the molecule (DO-1). An antibody generatedagainst the full-length p53 detects the full-length and the calpain-cleaved isoform of p53. As shown in Fig. 3A , the NH2-terminalantibody detected the p53 protein in MCF7 control and MCF7-CycE1 cells. MCF7-CycE3F cells seemed to have very low levels ofp53. To detect the truncated p53, the blot was stripped andreanalyzed with an antibody directed against the full-length p53protein. This analysis revealed that MCF7-CycE3F expressed p53,but most of the protein was cleaved to the NH2-terminallytruncated form. Also, the amount of p53 was reduced in theMCF7-CycE3F cells, suggesting that the truncated form of theprotein may be more labile. This result supports the hypothesis

Figure 1. Calpain transcription is transactivated in cyclinE–overexpressing MCF7 cells. A, expression of cyclin E and actinin MCF7-RcCMV (control, lane 1 ), MCF7-CycE1, andMCF7-CycE3F (cyclin E–overexpressing) cells (lanes 2 and 3).Full-length (FL ; arrows ) and low molecular weight (LMW ;brackets ) forms. B, cyclin E–associated kinase activity inMCF7-RcCMV, MCF7-CycE1, and MCF7-CycE3F cells. C,calpain 2 mRNA expression in MCF7-RcCMV, MCF7-CycE1, andMCF7-CycE3F cells. D, transactivation of the calpain 2 promoter(capnP ) by cyclin E. MCF7 cells were transfected with thecapnP-luc along with RcCMV vector alone, cyclin E expressionplasmid, or an equal amount of cyclin E and p21 expressionplasmids. RcCMV was included to maintain constant levelsof DNA.

Cancer Research




that calpain cleavage of p53 may be an alternative mechanism ofp53 degradation (24).Cyclin E overexpression transactivates calpain 2 in other

cell lines. T47D (breast), HCT116 (colon), and HepG2 (hepatocel-lular carcinoma) cells were used to determine if cyclin Eoverexpression in other cellular backgrounds resulted in a similartransactivation of calpain 2. Cyclin E–overexpressing T47D clones(T47D-CycE4 and T47D-CycE18) expressed higher amounts of thecyclin E LMW isoforms, higher levels of calpain protein, and a dose-dependent decrease in FAK expression (Fig. 3B). Transientexpression of cyclin E in HCT116 resulted in increased expressionof the cleaved isoform of calpain 2 and the complete proteolysis ofFAK to the 90 kDa form (Fig. 3B). HepG2 transiently transfectedwith the cyclin E plasmid expressed the LMW cyclin E isoforms andhad elevated levels of calpain 2. Cyclin E–overexpressing HepG2cells had very low levels of FAK, suggesting that the protein may becompletely proteolyzed (Fig. 3B). Previous studies found that the90 kDa FAK isoform is further degraded to 45 and 40 kDafragments and ultimately is completed degraded (17). This analysissuggests that cyclin E–dependent transactivation of calpain may bea feature of other tumors.Cyclin E is a calpain substrate in vitro . To confirm that cyclin

E can be cleaved by calpain, cellular extracts prepared from MCF7control cells were treated with increasing amounts of CaCl2. Figure4A shows a CaCl2 dose-dependent increase in cyclin E proteolysisto the LMW isoforms in MCF7 cells (lanes 1-4). Additionally, the

cyclin E–overexpressing clones expressed LMW isoforms of similarmass (lanes 5 and 6). To confirm that calpain activity was activatedby the inclusion of CaCl2, the immunoblot was stripped andreprobed with a FAK-specific antibody. Increasing concentrationsof CaCl2 also resulted in the cleavage of FAK. The same analysiswas used to detect the sensitivity of calpastatin to calpain–dependent proteolysis. C-rel, a protein that has not been describedas a calpain substrate, served as a negative control. Increasingconcentrations of CaCl2 had no effect on c-rel, therefore, thedegradation of cyclin E, FAK, and calpastatin are calpain-specificand not due to a general degradation of the cellular extracts.An alternative method was used to confirm that cyclin E is a

calpain 2 target. MCF7 control cellular extracts were treated withincreasing concentrations of purified calpain 2. As shown in Fig.4B , calpain 2 addition resulted in a dose-dependent proteolysis ofcyclin E into the same-size LMW isoforms detected in cyclin E–overexpressing MCF7 cells. The two complementary analyses showthat, in vitro , cyclin E is a calpain substrate.In vivo activation and inhibition of cyclin E proteolysis.

Although previous studies reported calpain-dependent cyclin Ecleavage in vitro (15), it was unknown if calpain-dependentproteolysis occurred in vivo . To address this issue, proliferatingMCF7 control cells that expressed only trace amounts of the LMWisoforms were treated with the calcium ionophore, ionomycin.Previous studies have shown that ionomycin activates endogenouscalpain (21). As shown in Fig. 5A , ionomycin treatment resulted in

Figure 2. Hierarchal cluster analysis of gene transcriptionin MCF7-RcCMV, MCF7-CycE1, and MCF7-CycE3F cells.Hierarchal clustering of the microarray data was used toidentify similar patterns of altered transcription in cyclinE–overexpressing cells. Clusters containing functionallyrelated genes revealed several biological processesincluding catabolism that were deregulated in cyclinE–overexpressing cells. B, a representative list ofup-regulated catabolic genes in cyclin E–overexpressingcells includes calpain 2.





cleavage of cyclin E into LMW isoforms. Inclusion of the cell-permeable calpain inhibitor, calpeptin, prevented cyclin E prote-olysis. To confirm that calpain was activated, the immunoblot wasstripped and reprobed with a FAK-specific antibody. This showsthat cyclin E is a calpain substrate in vivo .If cyclin E cleavage in MCF7-CycE3F cells is indeed due to

calpain, then in vivo inhibition of calpain activity would preventcyclin E proteolysis. To test this hypothesis, MCF7-CycE3F cellswere treated with the vehicle DMSO or with two differentconcentrations of the calpain inhibitors calpeptin or ALLM for 24or 48 hours. As shown in Fig. 5B , calpeptin treatment effectivelyprevented cyclin E cleavage in a dose-dependent manner. ALLMinhibited cyclin E proteolysis, but to a lesser extent than calpeptin.The immunoblot was stripped and reprobed with FAK antibodies.Inhibition of FAK cleavage served as a positive control for theinhibition of calpain activity. Our ability to modulate theexpression of cyclin E LMW isoforms by activation or inhibitionof calpain activity in vivo showed that in MCF7 cells, the LMWisoforms were generated by calpain-dependent proteolysis.Previous reports indicate that the cyclin E LMW forms are

associated with higher cyclin E–dependent kinase activity. To testthis hypothesis in vivo , cellular extracts prepared from MCF7-CycE3F treated with 20 Amol/L calpeptin or DMSO for 24 hourswere used to detect cyclin E–associated kinase activity. CyclinE–dependent kinase activity was reduced 1.8-fold by calpeptin

treatment, arguing that the LMW isoforms are associated withhigher kinase activity (data not shown).Cyclin E overexpression, calpain 2 levels, and focal adhesion

kinase proteolysis in human breast tumors. Cyclin E isoverexpressed in 10% to 25% of human breast tumors and thesetumors also express the LMW isoforms of the protein (9, 11).Therefore, we asked if there is a correlation between cyclin Eoverexpression, calpain 2 levels, and calpain-dependent proteol-ysis of a known calpain substrate. Ten frozen mammary tumorsamples and one control (normal mammary tissue) were used inthe analysis. Equal amounts of protein extracts were analyzed forthe expression of cyclin E. GAPDH was used to confirm equalprotein loading. As shown in Fig. 6A , one tumor, 3440, expressedhigh levels of cyclin E, whereas tumors 571 and 3040 expresseda moderate amount of cyclin E. Tumor 3440 also expressed cyclin ELMW isoforms. Tumor 571 had a small amount of one LMW iso-form. The MCF7-CycE3F extract served as marker for the expres-sion of the LMW cyclin E isoforms. Consistent with the hypothesisthat cyclin E transactivates calpain 2 expression, tumors 3440,3040, and 571 had elevated levels of calpain 2 when compared withnormal mammary tissue (1827). FAK was almost completelyproteolyzed to the 90 kDa form in tumors 3440, 3040, and 571,indicative of elevated calpain activity. In contrast, tumor 2385had low levels of calpain 2 and expressed abundant levels of full-length FAK. This tumor analysis supports the hypothesis that

Figure 3. Cyclin E overexpression in MCF7 cells results in increased calpain 2 protein expression and activity. A, calpain 2 protein levels are increased in cyclinE–overexpressing cells (lanes 2 and 3 ). The FL and LMW forms of cyclin E are marked. Calpain substrates FAK, calpastatin, pp60src, and p53 are proteolyticallyprocessed in cyclin E–overexpressing (lanes 2 and 3 ), but not in control cells (lane 1) indicating an increase in calpain activity. Arrows, different isoforms of calpain 2,FAK, c-src, and p53. B, calpain substrate FAK is proteolytically processed in cyclin E–overexpressing (lanes 2, 3, 5 and 7), but not in control cells (lanes 1, 4 and 6).

Cancer Research




LMW isoforms of cyclin E are associated with increased calpain-dependent proteolysis.Cyclin E and focal adhesion kinase proteolysis in non-

metastatic and metastatic tumors. Many calpain substrates arecytoskeleton/membrane attachment proteins. These proteins areimportant for cellular adhesion, a process that is modified by calpainproteolysis. Alterations in cellular adhesionmay have a role in tumorinvasion and/or metastasis. Therefore, a mouse breast tumormetas-tasis model (30, 33) was used to determine if (a) calpain may have a

role in this process and (b) if cyclin E proteolysis is associated withmetastasis. The study employed four related polyomavirus-middle T(PyV-mT) transgenic mouse mammary carcinoma transplant linesthat display differences in metastatic potential (33). Importantly,PyV-mT has been used as a surrogate for Her2/neu overexpressionand interacts with and activates the 85 kDa regulatory phosphoi-nositide-3-kinase (PI3K) subunit. The Met line, derived from trans-genic mice constructed with the wild-type PyV-mT line, developsrapid mammary carcinoma in all animals with 100% pulmonarymetastasis. The Db7 line is derived from animals with double site-directed mutations of the PyV-mT that decouple it from the PI3Kpathway. The Db7 model has 100% penetrance of mammary tumorbut exhibits significantly fewer pulmonarymetastases (9%). The PD5and PD6 transplantable tumor lines are derived from rare metas-tases of the Db7 line. The order of metastatic potential from leastmetastatic to most metastatic is Db7, PD6, PD5, and Met. Extractsprepared from these tumors were used to detect expression of FAK,calpain 2, and cyclin E. h-Actin and GAPDH served as gel loadingcontrols. As shown in Fig. 6B , FAK is poorly cleaved in the Db7tumor, intermediately cleaved in PD6 and PD5, and greatly cleaved inthe Met tumor. Calpain 2 autolysis results in a more active, but lessstable, form of calpain 2 (34). The Met tumor expressed a smallerform of calpain 2, further suggesting that this tumor had highercalpain activity. Finally, the Met tumor, but not the Db7 tumor,expressed LMW isoforms of cyclin E, demonstrating that cyclin Eproteolysis correlated with calpain activity. These results suggestthat calpain-dependent proteolysis is associated with metastasis.In summary, this study supports the hypothesis that cyclin E

transactivates calpain 2 expression thus regulating its ownproteolysis as well as the proteolysis of additional calpain substrates.This conclusion is substantiated by studies of human mammarytumors. Further analysis using a mouse mammary tumor modelsuggests that calpain-dependent proteolysis may have a role inmammary metastasis.

Discussion

This study identifies a link between cyclin E overexpression andenhanced expression and activity of calpain 2, indicating that cyclinE both regulates and is regulated by calpain 2. Similar autoregulatoryloops have been described for cyclin E and the CDK inhibitor p27, aswell as cyclin E and the E2F/RB transcription factor complex (35–38).The cyclin E–dependent activation of calpain 2 results in increasedcalpain-mediated proteolysis of cyclin E into LMW isoforms that arerefractory to inhibition by p21 and p27, resulting in higher cyclinE–associated kinase activity. Moreover, the increase in calpain 2expression affects proteolysis of the calpain inhibitor and substratecalpastatin. Calpastatin proteolysis further shifts the calpain/calpastatin homeostasis in favor of increased calpain activity, thusleading to the proteolysis of additional calpain substrates.Calpain-dependent proteolysis has been implicated in signal

transduction, adhesion and metastasis. Calpain’s proteolytic activityis distinct from other proteases because calpains have a limited andcomplex site specificity (16) that is determined by both the sequenceand conformation of the polypeptide (39). The sequence specificity isnot dependent on an exact sequence, but rather by preferred aminoacids in distinct sites (39). This rather nebulous definition of calpaincleavage sites makes it difficult to predict substrate specificity in theabsence of experimental data. Because calpain specificity isdependent in part upon conformation, posttranslational modifica-tions such as phosphorylation, significantly affect the rate of

Figure 4. Activation of calpain promotes proteolytic cleavage of cyclin Eand FAK. A, treatment of MCF7-RcCMV control cellular extracts with increasingconcentrations of Ca2+ causes progressively greater cleavage of cyclin E(lanes 1-4). The proteolytic fragments are identical in size to the forms present incyclin E–overexpressing MCF7 clones (lanes 5 and 6 ). Cleavage of calpainsubstrates FAK and calpastatin served as positive controls, whereas c-rel servedas negative control. B, addition of increasing amounts of purified calpain 2 toMCF7-RcCMV cellular extracts (lanes 4-9) promoted proteolysis of cyclin E tothe same-size LMW isoforms that are expressed in cyclin E–overexpressingMCF7 clones (lanes 1 and 2).





proteolysis (40, 41). Therefore, the cleavage of specific calpainsubstrates in a cell may be dependent on posttranslationalmodifications of the substrates aswell as the rates of calpain cleavage.Increased calpain 2 activity promotes proteolysis of FAK, which

is cleaved to a readily detectable 90 kDa form (17). We found thatFAK is a sensitive calpain substrate and the extent of FAKproteolysis is proportional to cyclin E expression. Therefore, FAK

proteolysis is a useful measure of calpain activity. FAK is thought tocoordinate cell adhesion and migration. Previous studies inplatelets reported that proteolysis of FAK resulted in a reductionin its autokinase activity and led to its dissociation from thecytoskeleton. Interestingly, cells devoid of FAK exhibit reducedin vitro mobility and an increase in the number of focal adhesions,suggesting that FAK may have a role in focal adhesion turnover

Figure 6. Cyclin E overexpression correlates with increased calpain 2 expression and proteolysis of FAK. A, Western immunoblot analysis was used to detectcyclin E, calpain 2, and FAK expression in 10 human mammary tumors (lanes 2-11 ) and normal mammary tissue (1827, lane 1). MCF7-CycE3F cellular extracts servedas a marker for the expression of the LMW isoforms (lane 12). Arrows, tumors with elevated cyclin E levels. GAPDH served as a gel loading control. B, analysis ofPyV-mT transgenic mouse mammary carcinomas with differing metastatic potentials. Extracts prepared from the nonmetastatic (Db7, lane 1), weakly metastatic(PD6 and PD5, lanes 2 and 3), and very metastatic (lane 4) tumors were analyzed using Western immunoblots. Highly metastatic (lane 4) mouse tumors have higherlevels of calpain 2, a decrease in FAK, and proteolytically cleaved cyclin E when compared with the nonmetastatic tumor (Db7, lane 1).

Figure 5. In vivo induction of calpain activity promotes cyclin E and FAK cleavage, whereas in vivo inhibition of calpain activity prevents cleavage. A, to inducecalpain activity, MCF7-RcCMV control cells were treated with the calcium ionophore ionomycin (lanes 3 and 4 ) in the presence or absence of the calpain inhibitorcalpeptin for 20 minutes. Control cells were treated with the vehicle DMSO (lane 1) or calpeptin alone (lane 2). Ionomycin treatment of cells resulted in cyclin E and FAKcleavage, whereas inclusion of calpeptin inhibited cleavage. B, calpain activity was inhibited by treating MCF7 cyclin E–overexpressing cells with two differentconcentrations of calpeptin (lanes 2, 3, 7 and 8) or ALLM (lanes 4, 5, 9 and 10) for 24 (lanes 1-5) or 48 hours (lanes 6-10 ). Calpeptin was very effective in preventingcyclin E and FAK cleavage, indicating that cyclin E proteolysis in MCF7 cyclin E–overexpressing cells is dependent on calpain activity.

Cancer Research




(42). A recent study reported a novel role for FAK that isindependent of its kinase activity—the recruitment of calpain intofocal adhesions where it might potentially cleave other focaladhesion proteins (43), modifying cell adhesion and migration.Cells overexpressing high, but not intermediate levels of cyclin E,

exhibited proteolysis of pp60src to a 48 kDa form. The interactionbetween c-src and calpain is complex because studies have shownthat c-src is involved in increasing calpain levels (44). Calpain-dependent proteolysis of pp60src to a 48 kDa isoform has beenpreviously described in platelets and the cleavage site was localizedto the NH2-terminal region of the molecule. We found a consensuscalpain site at amino acids 125 to 135 which would generate a 48kDa polypeptide. This 48 kDa protein would have a deletion of themyristoylation site and part of the SH3 domain, but would retainthe SH2 and catalytic domains, suggesting that the kinase activityof this molecule would not be abolished. However, deletion of themyristoylation site may alter subcellular localization (23), anddeletion of part of the SH3 domain may modify pp60src kinaseactivity as well as interactions with specific proteins includingpaxillin (45).Calpain also cleaves proteins that are instrumental in

transcriptional regulation including p53. In vitro studies foundthat calpain cleaves p53 at the NH2-terminal end to generate a 48kDa protein. Removal of the NH2-terminal region would promotep53 degradation in a ubiquitin-independent manner, thusdiminishing the levels of this tumor suppressor (24). The currentstudy is the first to report the detection of the 48 kDa p53 isoformin vivo . Cells overexpressing high levels of cyclin E expressed onlytrace amounts of the full-length p53, but the 48 kDa form wasreadily detectable. However, the total amount of p53 was reduced,supporting the hypothesis that calpain proteolysis promotes p53degradation. A reduction in the levels of p53 would facilitatetransversal of DNA damage checkpoints, contribute to genomeinstability, modify apoptotic responses, and hence, promotetumorigenesis.Several studies have suggested a link between increased calpain

activity and metastasis (21, 46, 47). The mechanisms by whichcalpain may be involved in this process is poorly explored. Using amouse model of mammary tumor metastasis (30, 33), that in partmimics Her2/neu overexpression, we found that calpain cleavage ofFAK and cyclin E is coincident with increased metastatic potential.The metastatic phenotype of the Met cells is dependent on the PyV-mT protein initiating a signaling cascade through an interactionwith the PI3K pathway (33). How is this pathway linked to calpainactivity? Sonenshein and coworkers found that activation of the

PI3K pathway by overexpression of the HER2/neu leads to a calpain-dependent proteolysis of InB and subsequent activation of NFnB-dependent transcription (48). The exact mechanism involved incalpain activation is unknown. Studies have also shown thatextracellular signal-regulated kinase, a kinase that is instrumentalin growth factor–initiated signal transduction can directly phos-phorylate and activate calpain 2 (49). HER2/neu is overexpressed inf20% to 30% of all breast cancers and, like cyclin E overexpression,is predictive of a poor clinical outcome (50). Therefore, signalingpathways that are perturbed in breast cancer and are associatedwitha poor clinical outcome may share a common feature—anaugmentation of calpain activity.The expression of calpain and calpastatin in human mammary

tumors has been largely unexplored. One study indicates that calpainactivity is higher in human breast tumors than in normal mammarytissue (51). Studies of other tumor types have provided evidence thatincreased calpain expression is associated with tumorigenesis.Calpain 2 expression is elevated in prostate tumors (21) and incolorectal adenocarcinomas (52). Calpain 1 expression is increasedin renal cell carcinomas (53). There are no studies on calpastatinexpression in human breast tumors, but the expression ofcalpastatin is decreased in nasopharyngeal carcinomas and incolorectal adenocarcinomas (52, 54). These studies suggest that aderegulation of the calpain/calpastatin equilibrium may have a rolein tumor progression, but further studies are needed to clearly definethe role of these proteins in mammary tumorigenesis.In conclusion, this study shows that cyclin E–overexpressing

MCF7 cells, human mammary tumors overexpressing cyclin E, andmammary metastatic tumors derived from mouse PyV-mTexpressing cells exhibit an increase in calpain-dependent proteol-ysis. Together, these results suggest that cyclin E–dependentderegulation of calpain 2 may be the pivotal feature that confersa cellular phenotype of aggressive behavior and genomic instabilityassociated with cyclin E–overexpressing tumors.

Acknowledgments

Received 5/13/2005; revised 8/4/2005; accepted 9/16/2005.Grant support: Cancer Research Coordinating Committee (M. Mudryj), the

University of California, Davis, Health Systems grant (M. Mudryj), a VA Merit grant(M. Mudryj), and a President’s Undergraduate Fellowship (B.S. Robinson). Thisinvestigation was conducted in a facility constructed with support from the ResearchFacilities Improvement Program grant no. C06 RR-12088-01 from the National Centerfor Research Resources, NIH.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Drs. Michael Seldin, Gigi Lozano, and Martin Privalsky for reading themanuscript and providing useful comments.

References

1. Keyomarsi K, O’Leary N, Molnar G, Lees E, Fingert HJ,Pardee AB. Cyclin E, a potential prognostic marker forbreast cancer. Cancer Res 1994;54:380–5.

2. Kitahara K, Yasui W, Kuniyasu H, et al. Concurrentamplification of cyclin E and CDK2 genes in colorectalcarcinomas. Int J Cancer 1995;62:25–8.

3. Erlanson M, Portin C, Linderholm B, Lindh J, RoosG, Landberg G. Expression of cyclin E and thecyclin-dependent kinase inhibitor p27 in malignantlymphomas—prognostic implications. Blood 1998;92:770–7.

4. Marone M, Scambia G, Giannitelli C, et al. Analysisof cyclin E and CDK2 in ovarian cancer: geneamplification and RNA overexpression. Int J Cancer1998;75:34–9.

5. Sakaguchi T, Watanabe A, Sawada H, et al. Prognosticvalue of cyclin E and p53 expression in gastriccarcinoma. Cancer 1998;82:1238–43.

6. Fukuse T, Hirata T, Naiki H, et al. Prognosticsignificance of cyclin E overexpression in resectednon-small cell lung cancer. Cancer Res 2000;60:242–4.

7. Xiangming C, Natsugoe S, Takao S, et al. Thecooperative role of p27 with cyclin E in theprognosis of advanced gastric carcinoma. Cancer2000;89:1214–9.

8. Hedberg Y, Davoodi E, Ljungberg B, Roos G, LandbergG. Cyclin E and p27 protein content in human renal cellcarcinoma: clinical outcome and associations withcyclin D. Int J Cancer 2002;102:601–7.

9. Keyomarsi K, Tucker SL, Buchholz TA, et al. Cyclin Eand survival in patients with breast cancer. N Engl J Med2002;347:1566–75.

10. Span PN, Tjan-Heijnen VC, Manders P, Beex LV,Sweep CG. Cyclin-E is a strong predictor of endocrinetherapy failure in human breast cancer. Oncogene2003;22:4898–904.

11. Loden M, Stighall M, Nielsen NH, et al. The cyclin D1high and cyclin E high subgroups of breast cancer:separate pathways in tumorigenesis based on pattern ofgenetic aberrations and inactivation of the pRb node.Oncogene 2002;21:4680–90.

12. Porter DC, Zhang N, Danes C, et al. Tumor-specificproteolytic processing of cyclin E generates hyperactivelower-molecular-weight forms. Mol Cell Biol 2001;21:6254–69.

13. Akli S, Zheng PJ, Multani AS, et al. Tumor-specificlow molecular weight forms of cyclin E induce genomicinstability and resistance to p21, p27, and antiestrogensin breast cancer. Cancer Res 2004;64:3198–208.





Cancer Research


14. Bales E, Mills L, Milam N, et al. The low molecularweight cyclin E isoforms augment angiogenesis andmetastasis of human melanoma cells in vivo . Cancer Res2005;65:692–7.

15. Wang XD, Rosales JL, Magliocco A, Gnanakumar R,Lee KY. Cyclin E in breast tumors is cleaved into its lowmolecular weight forms by calpain. Oncogene 2003;22:769–74.

16. Goll DE, Thompson VF, Li H, Wei W, Cong J. Thecalpain system. Physiol Rev 2003;83:731–801.

17. Cooray P, Yuan Y, Schoenwaelder SM, Mitchell CA,SalemHH, Jackson SP. Focal adhesion kinase (pp125FAK)cleavage and regulation by calpain. Biochem J 1996;318:41–7.

18. Stabach PR, Cianci CD, Glantz SB, Zhang Z, MorrowJS. Site-directed mutagenesis of aII spectrin at codon1175 modulates its mu-calpain susceptibility. Biochem-istry 1997;36:57–65.

19. Franco SJ, Rodgers MA, Perrin BJ, et al. Calpain-mediated proteolysis of talin regulates adhesion dynam-ics. Nat Cell Biol 2004;6:977–83.

20. Yamaguchi R, Maki M, Hatanaka M, Sabe H.Unphosphorylated and tyrosine-phosphorylated formsof a focal adhesion protein, paxillin, are substrates forcalpain II in vitro : implications for the possibleinvolvement of calpain II in mitosis-specific degradationof paxillin. FEBS Lett 1994;356:114–6.

21. Rios-Doria J, Day KC, Kuefer R, et al. The role ofcalpain in the proteolytic cleavage of E-cadherin inprostate and mammary epithelial cells. J Biol Chem2003;278:1372–9.

22. Rios-Doria J, Kuefer R, Ethier SP, Day ML. Cleavage ofh-catenin by calpain in prostate and mammary tumorcells. Cancer Res 2004;64:7237–40.

23. Oda A, Druker BJ, Ariyoshi H, Smith M, Salzman EW.pp60src is an endogenous substrate for calpain inhuman blood platelets. J Biol Chem 1993;268:12603–8.

24. Kubbutat MH, Vousden KH. Proteolytic cleavage ofhuman p53 by calpain: a potential regulator of proteinstability. Mol Cell Biol 1997;17:460–8.

25. Pariat M, Carillo S, Molinari M, et al. Proteolysis bycalpains: a possible contribution to degradation of p53.Mol Cell Biol 1997;17:2806–15.

26. Hirai S, Kawasaki H, Yaniv M, Suzuki K. Degradationof transcription factors, c-Jun and c-Fos, by calpain.FEBS Lett 1991;287:57–61.

27. Dhillon NK, Mudryj M. Ectopic expression of cyclin Ein estrogen responsive cells abrogates antiestrogenmediated growth arrest. Oncogene 2002;21:4626–34.

28. Dhillon NK, Mudryj M. Cyclin E overexpressionenhances cytokine-mediated apoptosis in MCF7 breastcancer cells. Genes Immun 2003;4:336–42.

29. Hosack DA, Dennis G, Jr., Sherman BT, Lane HC,Lempicki RA. Identifying biological themes within listsof genes with EASE. Genome Biol 2003;4:R70.

30. Jessen KA, Liu SY, Tepper CG, et al. Molecularanalysis of metastasis in a polyomavirus middle Tmouse model: the role of osteopontin. Breast Cancer Res2004;6:R157–69.

31. Hata A, Ohno S, Akita Y, Suzuki K. Tandemlyreiterated negative enhancer-like elements regulatetranscription of a human gene for the large subunit ofcalcium-dependent protease. J Biol Chem 1989;264:6404–11.

32. Wang KK, Posmantur R, Nadimpalli R, et al. Caspase-mediated fragmentation of calpain inhibitor proteincalpastatin during apoptosis. Arch Biochem Biophys1998;356:187–96.

33. Webster MA, Hutchinson JN, Rauh MJ, et al.Requirement for both Shc and phosphatidylinositol 3Vkinase signaling pathways in polyomavirus middleT-mediated mammary tumorigenesis. Mol Cell Biol1998;18:2344–59.

34. Elce JS, Hegadorn C, Arthur JS. Autolysis, Ca2+requirement, and heterodimer stability in m-calpain.J Biol Chem 1997;272:11268–75.

35. Swanson C, Ross J, Jackson PK. Nuclear accumulationof cyclin E/Cdk2 triggers a concentration-dependentswitch for the destruction of p27Xic1. Proc Natl Acad SciU S A 2000;97:7796–801.

36. Geng Y, Eaton EN, Picon M, et al. Regulation of cyclinE transcription by E2Fs and retinoblastoma protein.Oncogene 1996;12:1173–80.

37. Toyoshima H, Hunter T. p27, a novel inhibitor of G1cyclin-Cdk protein kinase activity, is related to p21. Cell1994;78:67–74.

38. Lundberg AS, Weinberg RA. Functional inactivationof the retinoblastoma protein requires sequentialmodification by at least two distinct cyclin-cdk com-plexes. Mol Cell Biol 1998;18:753–61.

39. Tompa P, Buzder-Lantos P, Tantos A, et al. On thesequential determinants of calpain cleavage. J BiolChem 2004;279:20775–85.

40. Elvira M, Diez JA, Wang KK, Villalobo A. Phosphor-ylation of connexin-32 by protein kinase C prevents itsproteolysis by mu-calpain and m-calpain. J Biol Chem1993;268:14294–300.

41. Shen J, Channavajhala P, Seldin DC, Sonenshein GE.Phosphorylation by the protein kinase CK2 promotescalpain-mediated degradation of InBa. J Immunol2001;167:4919–25.

42. Ilic D, Furuta Y, Kanazawa S, et al. Reduced cellmotility and enhanced focal adhesion contact formationin cells from FAK-deficient mice. Nature 1995;377:539–44.

43. Carragher NO, Westhoff MA, Fincham VJ, SchallerMD, Frame MC. A novel role for FAK as a protease-targeting adaptor protein: regulation by p42 ERK andSrc. Curr Biol 2003;13:1442–50.

44. Carragher NO, Westhoff MA, Riley D, et al. v-Src-induced modulation of the calpain-calpastatin proteo-lytic system regulates transformation. Mol Cell Biol2002;22:257–69.

45. Weng Z, Taylor JA, Turner CE, Brugge JS, Seidel-Dugan C. Detection of Src homology 3-bindingproteins, including paxillin, in normal and v-Src-transformed Balb/c 3T3 cells. J Biol Chem 1993;268:14956–63.

46. Xu L, Deng X. Tobacco-specific nitrosamine4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone inducesphosphorylation of mu- and m-calpain in associationwith increased secretion, cell migration, and invasion.J Biol Chem 2004;279:53683–90.

47. Mamoune A, Luo JH, Lauffenburger DA, Wells A.Calpain-2 as a target for limiting prostate cancerinvasion. Cancer Res 2003;63:4632–40.

48. Pianetti S, Arsura M, Romieu-Mourez R, Coffey RJ,Sonenshein GE. Her-2/neu overexpression inducesNF-nB via a PI3-kinase/Akt pathway involving cal-pain-mediated degradation of InB-a that can beinhibited by the tumor suppressor PTEN. Oncogene2001;20:1287–99.

49. Glading A, Chang P, Lauffenburger DA, Wells A.Epidermal growth factor receptor activation of calpainis required for fibroblast motility and occurs via anERK/MAP kinase signaling pathway. J Biol Chem2000;275:2390–8.

50. Slamon DJ, Godolphin W, Jones LA, et al. Studies ofthe HER-2/neu proto-oncogene in human breast andovarian cancer. Science 1989;244:707–12.

51. Shiba E, Kambayashi JI, Sakon M, et al. Ca2+-dependent neutral protease (calpain) activity in breastcancer tissue and estrogen receptor status. BreastCancer 1996;3:13–7.

52. Lakshmikuttyamma A, Selvakumar P, Kanthan R,Kanthan SC, Sharma RK. Overexpression of m-calpain inhuman colorectal adenocarcinomas. Cancer EpidemiolBiomarkers Prev 2004;13:1604–9.

53. Braun C, Engel M, Seifert M, et al. Expression ofcalpain I messenger RNA in human renal cell carcino-ma: correlation with lymph node metastasis andhistological type. Int J Cancer 1999;84:6–9.

54. Sriuranpong V, Mutirangura A, Gillespie JW, et al.Global gene expression profile of nasopharyngealcarcinoma by laser capture microdissection and com-plementary DNA microarrays. Clin Cancer Res 2004;10:4944–58.



2005;65:10700-10708. Cancer Res Stephen J. Libertini, Brian S. Robinson, Navdeep K. Dhillon, et al. PhenotypeProtease Associated with Metastatic Breast Cancer Cyclin E Both Regulates and Is Regulated by Calpain 2, a

Updated version

http://cancerres.aacrjournals.org/content/65/23/10700

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/65/23/10700.full#ref-list-1

This article cites 53 articles, 27 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/65/23/10700.full#related-urls

This article has been cited by 11 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

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://cancerres.aacrjournals.org/content/65/23/10700To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/content/65/23/10700.full#ref-list-1

http://cancerres.aacrjournals.org/content/65/23/10700.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



cyclin e both regulates and is regulated by calpain 2, a protease … · cyclin e both regulates...

Documents