Mammary-specific inactivation of E-cadherin and p53 ...initiating and causal event in the development of mouse ILC (mILC) (Derksen et al., 2006), is a common feature of LCIS as well

INTRODUCTIONBreast cancer affects a large number of females in the Westernworld, accounting for half a million deaths worldwide on an annualbasis. Carcinoma of the breast is a heterogeneous disease based onpathological criteria, which is probably due to the multiplicity ofgenetic lesions that have accumulated during tumor development,resulting in distinct tumor types. The most frequently observedsubtypes, invasive ductal carcinoma (IDC) and invasive lobularcarcinoma (ILC), are very distinct phenotypically as well asbiochemically (Coradini et al., 2002; Korkola et al., 2003; Mathieuet al., 2004; Zhao et al., 2004; Stange et al., 2006). ILC is a subtypeof breast cancer that accounts for 10-15% of all cases and has a

greater tendency for multifocal development and bilateralpresentation than other primary breast tumors (Newstead et al.,1992; Krecke and Gisvold, 1993; Helvie et al., 1993). Classical ILCis characterized by non-cohesive invasive cells that are arranged intrabecules without mass formation and calcification, a feature thathinders diagnosis using physical examination or mammography(Simpson et al., 2003). The vast majority of ILCs are mostlyestrogen receptor (ER) positive and therefore responsive toendocrine therapy; however, they are mostly refractory to standardchemotherapy treatment once hormone receptor expression is lost(Gonzalez-Angulo et al., 2007), resulting in comparable survivalrates for ILC and IDC (Molland et al., 2004; Tubiana-Hulin et al.,2006). Controversy still exists concerning the etiology of invasivebreast cancer (IBC). Surprisingly, most clinicians do not regardlobular carcinoma in situ (LCIS) a precursor lesion for ILC, eventhough LCIS is regarded as a marker for progression to ipsilateralIBC, and 90% of LCIS-containing IBCs show a lobular phenotype(Wheeler et al., 1974; Rosen et al., 1978; Frykberg et al., 1987; Gump,1993). Also, loss of E-cadherin, which we have postulated as theinitiating and causal event in the development of mouse ILC (mILC)(Derksen et al., 2006), is a common feature of LCIS as well of theadjacent ILC cells (Vos et al., 1997).

E-cadherin is a key component of adherens junctions, structuresthat control the maintenance of epithelial integrity (Perez-Morenoet al., 2003). E-cadherin is a cell-cell adhesion molecule thatfunctions as a scaffold in the formation of catenin-containingcomplexes that link E-cadherin to the actin and microtubulecytoskeleton (Hulsken et al., 1994; Takeichi, 1995; Perez-Morenoet al., 2003). In multiple types of cancer, loss of E-cadherin functionthrough genetic or epigenetic mechanisms has been implicated in

Disease Models & Mechanisms 347

Disease Models & Mechanisms 4, 347-358 (2011) doi:10.1242/dmm.006395

1Division of Molecular Biology, Netherlands Cancer Institute, Plesmanlaan 121,1066 CX Amsterdam, The Netherlands2Department of Medical Oncology, University Medical Center Utrecht,3508 AB Utrecht, The Netherlands3Department of Pathology, University Medical Center Utrecht, Heidelberglaan 100,3584 CX Utrecht, The Netherlands4Department of Pathology, Netherlands Cancer Institute, Plesmanlaan 121,1066 CX Amsterdam, The Netherlands5Division of Molecular Genetics, Netherlands Cancer Institute, Plesmanlaan 121,1066 CX Amsterdam, The Netherlands*Present address: Department of Pathology, UMC Utrecht, Heidelberglaan 100,3584 CX Utrecht, The Netherlands‡Authors for correspondence ([email protected]; [email protected])

Received 23 July 2010; Accepted 16 December 2010

© 2011. Published by The Company of Biologists LtdThis is an Open Access article distributed under the terms of the Creative Commons AttributionNon-Commercial Share Alike License (http://creativecommons.org/licenses/by-nc-sa/3.0), whichpermits unrestricted non-commercial use, distribution and reproduction in any medium providedthat the original work is properly cited and all further distributions of the work or adaptation aresubject to the same Creative Commons License terms.

SUMMARY

Breast cancer is the most common malignancy in women of the Western world. Even though a large percentage of breast cancer patients showpathological complete remission after standard treatment regimes, approximately 30-40% are non-responsive and ultimately develop metastaticdisease. To generate a good preclinical model of invasive breast cancer, we have taken a tissue-specific approach to somatically inactivate p53 andE-cadherin, the cardinal cell-cell adhesion receptor that is strongly associated with tumor invasiveness. In breast cancer, E-cadherin is found mutatedor otherwise functionally silenced in invasive lobular carcinoma (ILC), which accounts for 10-15% of all breast cancers. We show that mammary-specific stochastic inactivation of conditional E-cadherin and p53 results in impaired mammary gland function during pregnancy through the inductionof anoikis resistance of mammary epithelium, resulting in loss of epithelial organization and a dysfunctional mammary gland. Moreover, combinedinactivation of E-cadherin and p53 induced lactation-independent development of invasive and metastatic mammary carcinomas, which showedstrong resemblance to human pleomorphic ILC. Dissemination patterns of mouse ILC mimic the human malignancy, showing metastasis to thegastrointestinal tract, peritoneum, lung, lymph nodes and bone. Our results confirm that loss of E-cadherin contributes to both mammary tumorinitiation and metastasis, and establish a preclinical mouse model of human ILC that can be used for the development of novel intervention strategiesto treat invasive breast cancer.

Mammary-specific inactivation of E-cadherin and p53impairs functional gland development and leads topleomorphic invasive lobular carcinoma in micePatrick W. B. Derksen1,2,3,*,‡, Tanya M. Braumuller1, Eline van der Burg1, Marten Hornsveld1, Elly Mesman4, Jelle Wesseling4,Paul Krimpenfort5 and Jos Jonkers1,‡

RESEARCH ARTICLED

iseas

e M

odel

s & M

echa

nism

s

DM

M

progression and metastasis (Frixen et al., 1991; Vleminckx et al.,1991; Cleton-Jansen et al., 1994; Hulsken et al., 1994; Oda et al.,1994; Berx et al., 1995; Graff et al., 1995; Takeichi, 1995; Yoshiuraet al., 1995; Savagner et al., 1997; Perl et al., 1998; Batlle et al., 2000;Cano et al., 2000; Comijn et al., 2001; Fujita et al., 2003; Yang etal., 2004; Moody et al., 2005). Nonetheless, the molecularmechanisms that drive tumor development and progression uponloss of E-cadherin in breast cancer remain ill-defined.

To study all aspects of tumor pathology, mouse models areneeded that mimic not only tumor phenotype, but also the initiatingsteps of human tumor development. In addition, it is importantthat mouse models reflect the events that are common in humanILC pathology such as lymphatic dissemination and subsequentdistant metastasis, especially to bone. We have previously shownthat tissue-specific inactivation of E-cadherin and p53 leads to thedevelopment of mILC (Derksen et al., 2006). While this mousemodel mimics many aspects of human pathology, includingmetastatic dissemination, it is not attractive as a preclinical modelbecause of the severity of skin-related problems due to cytokeratin14 (K14) promoter driven Cre recombinase (K14cre) expression inmultiple epithelial tissues. In addition, the stochastic activity ofK14cre in mammary gland epithelium precludes in vivo analysis ofthe direct consequences of E-cadherin loss, alone or in combinationwith loss of p53. In this study, we have developed a new mousemodel for human ILC, based on mouse whey acidic protein (Wap)gene promoter driven Cre recombinase (Wcre)-mediatedconditional inactivation of E-cadherin and p53. Whereasmammary-specific loss of E-cadherin alone results in mammaryepithelial cell death by apoptosis (Boussadia et al., 2002), combinedloss of E-cadherin and p53 confers anchorage-independent survivalof mammary cells, resulting in a nonfunctional mammary glandowing to aberrant lobulo-alveolar development and a severelydisrupted ductal architecture. Wcre-mediated loss of E-cadherinand p53 resulted in a similar tumor spectrum and latency comparedwith the K14cre;Cdh1F/F;Trp53F/F model of human ILC. However,Wcre;Cdh1F/F;Trp53F/F mice showed increased rates of tumor cellinvasion and metastasis compared with the previous model butwithout the development of skin-related problems, yielding asuperior model of human invasive lobular breast cancer.

RESULTSWe have previously shown that somatic inactivation of E-cadherinand p53 in K14-expressing cells induces the formation of mILC(Derksen et al., 2006). Although this model recapitulates severalaspects of human ILC, it is not very attractive as a preclinical modelbecause of the induction of multiple skin lesions and tumors dueto robust K14 expression in the epidermis. To circumvent thisproblem, we generated Wcre transgenic mice, in which Creexpression is controlled by the mouse Wap gene promoter(Hennighausen and Sippel, 1982), which is known to be expressedin the lumenal compartment of mammary epithelium (Fig. 1A).We examined the efficiency and tissue-specificity of Cre-mediatedrecombination by crossing Wcre founders with mice carrying theRosa26-lacZ reporter (R26R mice) (Soriano, 1999). Bi-transgenicvirgin and uniparous female littermates were analyzed for lacZ (-galactosidase) activity using X-gal-stained tissue sections andwhole mount preparations. We started by analyzing the Wcre-mediated recombinase activity upon lactation. Examination of

whole mount preparations from uniparous female Wcre;R26R miceshowed robust -galactosidase activity throughout the mammarygland (Fig. 1B, right panel). Histological analyses suggested lacZexpression in both the myoepithelial and lumenal compartmentsof the uniparous mammary gland. In contrast to previouslypublished results (Wagner et al., 1997), we could readily detect cellsthat had expressed Cre recombinase in mammary glands fromvirgin Wcre mice (Fig. 1B, left panel, and Fig. 1C). Examination ofsectioned whole mounts revealed a weak and patchy stainingpattern in 4-week-old virgin mammary glands, which wasmaintained throughout mammary gland development (Fig. 1C).The -galactosidase staining patterns again suggested that Crerecombinase activity was not confined to lumenal cells of themammary gland but was also present in myoepithelial cells (Fig.1C, bottom right panel). To substantiate this, we performedimmunohistochemistry for cytokeratin (CK)8, CK14 and smoothmuscle actin (SMA) on sections from X-gal-stained Wcre;R26Rvirgin female mice. Unfortunately, CK14 immunohistochemistrydid not yield interpretable results due to the incompatibility of thestaining protocol with antigen retrieval (data not shown). However,although the majority of lacZ-positive (Cre-expressing) cells werecoexpressing CK8, we could also detect lacZ positivity in CK8-negative cells (Fig. 1D, top panel). Moreover, we could detect activityof lacZ in SMA-positive cells (Fig. 1D, bottom panel), suggestingthat Wcre is expressed in both lumenal and myoepithelial cells inWcre;R26R female mice.

Mammary-gland-specific conditional inactivation of E-cadherin isnot toleratedTo study the effects of E-cadherin loss during mammary glanddevelopment and tumor formation, we crossed our Wcre transgenicanimals to conditional E-cadherin knockout (Cdh1F) mice (Derksenet al., 2006). Despite the extensive Cre-mediated recombination inthe mammary gland of Wcre females (Fig. 1), we did not findmorphological abnormalities in virgin, pregnant or parousWcre;Cdh1F/F mice (Fig. 2A, left panels). Although Wcre;Cdh1F/F

females lactated and were able to nurse their litters uponparturition, the produced quantity of milk and litters was markedlylower than in wild-type animals (data not shown). Upon histologicalexamination, we did not observe abnormalities in mammaryarchitecture, nor did we detect E-cadherin-negative ductalstructures, indicating that – in agreement with previous studies(Boussadia et al., 2002; Derksen et al., 2006) – loss of E-cadherinis not tolerated in the mouse mammary gland (supplementarymaterial Fig. S1). Moreover, Wcre-mediated loss of E-cadherin didnot predispose mice to tumor formation, because none of theWcre;Cdh1F/F animals developed mammary carcinomas duringtheir live span, even after the induction of multiple pregnancies,which is consistent with previously published studies usingMMTVcre and K14cre conditional E-cadherin knockout mice(Boussadia et al., 2002; Derksen et al., 2006).

Developmental and lactational defects upon somatic inactivationof E-cadherin and p53Mammary-specific loss of E-cadherin induces cell death byapoptosis (Boussadia et al., 2002). To investigate whether thisapoptotic response to E-cadherin loss could be blocked by p53inactivation, we introduced a conditional Trp53 allele (Jonkers et

dmm.biologists.org348

E-cadherin loss leads to pleomorphic mILCRESEARCH ARTICLED

iseas

e M

odel

s & M

echa

nism

s

DM

M

al., 2001) into the Wcre;Cdh1F/F mouse line. Next, we studiedwhether dual loss of E-cadherin and p53 would influence normalmammary gland development and function. To achieve effectiveCre-mediated deletion of both Cdh1 and Trp53, we induced Wcreexpression by mating and compared mammary gland morphologyin Wcre;Cdh1F/F;Trp53F/F females with that in Wcre;Cdh1F/F andCdh1F/F;Trp53F/F control mice. To this end, we examined carmine-stained whole mount preparations of fourth (inguinal) mammaryglands, harvested at 14 and 17 days of pregnancy and at parturition(day 19.5). Mammary glands from virgin mice were used as astarting point. Whereas mammary glands from virginWcre;Cdh1F/F;Trp53F/F or Wcre;Cdh1F/F females showed no grossmorphological abnormalities as compared with control femalemice, Wcre;Cdh1F/F;Trp53F/F mammary glands harvested at day 14of pregnancy displayed architectural abnormalities, showing severeectasia (dilated ducts) and incomplete lobulo-alveolar development(Fig. 2). These effects became more pronounced as pregnancyprogressed, with a filling of the mammary gland with nonfunctionaltissue, resulting in complete disruption of the ductal structure atparturition (Fig. 2, bottom center panel). Wcre;Cdh1F/F andCdh1F/F;Trp53F/F control animals showed no morphologicalmammary gland abnormalities during pregnancy (Fig. 2, leftand right panels). Although Wcre;Cdh1F/F;Trp53F/F andWcre;Cdh1F/F;Trp53F/+ female mice produced healthy newbornoffspring, all pups fostered by Wcre;Cdh1F/F;Trp53F/F dams diedbefore weaning age due to starvation. Also, pups fromWcre;Cdh1F/F;Trp53F/+ dams showed reduced survival rates due tothe inhibition or absence of lactation. This phenotype could be

rescued by fostering pups from Wcre;Cdh1F/F;Trp53F/+ dams towild-type recipient dams (supplementary material Fig. S2).Furthermore, we did not detect significant differences in thesizes of newborn litters from Wcre;Cdh1F/F, Wcre;Cdh1F/+,Wcre;Cdh1F/F;Trp53F/+, Wcre;Cdh1F/+;Trp53F/F andWcre;Cdh1F/F;Trp53F/F female mice (data not shown).

Next, we analyzed the histology of the mammary gland atdifferent time points during pregnancy and at parturition. Westained for myoepithelial cells using CK14, for lumenal epithelialcells using CK8 and we analyzed Cre-mediated switching throughE-cadherin expression. We also studied proliferation afterincorporation of BrdU, and visualized apoptosis by means ofcleaved caspase-3. During pregnancy, loss of E-cadherin wasapparent multifocally in mammary glands fromWcre;Cdh1F/F;Trp53F/F mice at day-of-pregnancy (DP) 14, 17 and19.5 (Fig. 3; supplementary material Fig. S3). Structures thatshowed an absence of E-cadherin expression had lost normal ductalarchitecture, showed no lobulo-alveolar development, andconsisted of CK8-positive cells, interrupted by cells with afibroblastic and/or mesenchymal appearance (Fig. 3;Wcre;Cdh1F/F;Trp53F/F, DP 14, 17 and 19.5). At parturition,Wcre;Cdh1F/F;Trp53F/F glands showed strong proliferation whencompared with control glands (Fig. 3; DP 19.5), but no signs ofinvolution-induced apoptosis were observed after weening (datanot shown). Surprisingly, we did not detect histologicalabnormalities in mammary glands from Wcre;Cdh1F/F mice(supplementary material Fig. S1 and data not shown). Normalductal architecture was maintained throughout pregnancy, and E-


E-cadherin loss leads to pleomorphic mILC RESEARCH ARTICLE

Fig. 1. Cre expression in virgin and parous Wcrefemale mice. (A)Schematic representation of the Wcre

transgenic construct. The mouse Wap gene promoter, therabbit intron from the gene encoding -globin, thenuclear localization signal (nls)-Cre coding sequences andthe human growth hormone (HGH) poly(A) arerepresented. IVS, intervening sequence/intron. (B)Wcre

activity in the mammary gland during gestation.3-month-old Wcre;R26R mice were analyzed for -galactosidase activity during pregnancy. Blue stainingindicates Cre-expressing cells. Scale bars: 100m. (C)Creexpression in Wcre virgin and parous female mice. Shownis -galactosidase activity of carmine-counterstainedwhole mount preparations (top) and fast-nuclear-red-counterstained sections (bottom) from 4-week- and 12-week-old Wcre;R26R heterozygous virgin mice. Scale bars:10 mm (top panels); 100m (bottom panels). (D)Wcre isexpressed in lumenal and myoepithelial cells. Wholemounts from Wcre;R26R female mice were stained for lacZ

activity, sectioned and analyzed for expression of CK8(top panel) and SMA (bottom panel) usingimmunohistochemistry. Sections were counterstainedwith fast nuclear red. Arrowheads indicate non-lumenallacZ-positive cells, whereas arrows point to CK8-positivelumenal cells (top panel) and SMA-expressingmyoepithelial cells (bottom panel). Scale bars: 50m.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

cadherin-negative ducts were not observed in these glands at eithertime point. These results show that loss of E-cadherin and p53 leadsto aberrant lobulo-alveolar development during gestation. Thedevelopmental abnormalities are probably caused by loss of cellpolarity and acquisition of anoikis resistance, resulting in loss ofmammary gland organization and enhanced proliferative capacityof mammary epithelial cells, leading to a nonfunctional mammarygland.

E-cadherin loss collaborates with p53 loss in mammarytumorigenesisWe have previously shown that K14cre-mediated somaticinactivation of p53 in epithelial tissues results in the developmentof noninvasive adenocarcinoma and carcinosarcoma tumor typeswith a median latency of ~330 days (Liu et al., 2007). To targetconditional loss of p53 to the mammary epithelial compartment,we crossed Trp53F/F conditional mutant mice (Liu et al., 2007) withour Wcre mice. All resulting Wcre;Trp53F/F female mice developedmammary tumors, with a median latency of 290 days (Fig. 4A). Tostudy the tumor suppressor functions of E-cadherin, we crossedour Cdh1F conditional mice (Derksen et al., 2006) with theWcre;Trp53F/F mouse model to produce cohorts ofWcre;Cdh1F/+;Trp53F/F and Wcre;Cdh1F/F;Trp53F/F female mice,which were mated once and monitored for subsequent tumordevelopment. All mice were on a mixed genetic background ofFVB/N and Ola129/sv (Derksen et al., 2006). Wcre;Cdh1F/+;Trp53F/F

animals developed tumors with a similar median latency asWcre;Trp53F/F females (Fig. 4A; P0.4638), showing that E-cadherinis not haploinsufficient for suppression of mammary tumor

formation in these mice. By contrast, Wcre;Cdh1F/F;Trp53F/F micedeveloped mammary tumors with a significantly reduced tumor-free survival age (T50) of 194 days (Fig. 4A; P<0.0001). Tumor onsetand progression in these mice were relatively uniform, with mosttumors arising between 150 and 250 days. These findingsdemonstrate that combined inactivation of E-cadherin and p53contribute synergistically to mammary tumor formation in thesemice. Subsequent genetic analyses revealed uniform loss of bothmutant Cdh1 alleles in tumors derived from Wcre;Cdh1F/F;Trp53F/F

females (data not shown). As in the K14cre mILC model, wealso detected uniform loss of the conditional and wild-type Trp53alleles in mammary tumors from Wcre;Cdh1F/+;Trp53F/F,Wcre;Cdh1F/F;Trp53F/+ and Wcre;Cdh1F/F;Trp53F/F females,indicating that loss of functional p53 is a prerequisite for mammarytumor formation in these mouse models (data not shown).

Lactation does not affect tumor onset, incidence, latency ormetastasis formation in Wcre;Cdh1F;Trp53F miceBecause our analysis of Wcre;R26R mice also revealed Crerecombinase activity in virgin females (Fig. 1), we set outto investigate whether Wcre;Cdh1F/+;Trp53F/F andWcre;Cdh1F/F;Trp53F/F female mice would develop mammarytumors in the absence of lactation. To this end, we produced tenfemale mice of each genotype and monitored tumor developmentin virgin animals. Interestingly, tumors arose inWcre;Cdh1F/+;Trp53F/F virgin females with identical incidence andlatency when compared with uniparous females (Fig. 4B). In linewith this finding, tumor-free latency was also similar in virgin andparous Wcre;Cdh1F/F;Trp53F/F female mice (Fig. 4C). Moreover, the


E-cadherin loss leads to pleomorphic mILCRESEARCH ARTICLE

Fig. 2. Developmental and lactational defectsupon inactivation of E-cadherin and p53. Loss ofE-cadherin and p53 leads to gross abnormalities inthe mammary glands during gestation. Shown arecarmine-stained whole mount preparations ofmouse mammary glands from Wcre;Cdh1F/F mice(left panels), Wcre;Cdh1F/F;Trp53F/F mice (middlepanels) and Cdh1F/F;Trp53F/F control animals (rightpanels). Part., parturition. Original magnification:6.5�.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

tumor spectrum, invasiveness and metastatic dissemination weresimilar in virgin and parous Wcre;Cdh1F/+;Trp53F/F andWcre;Cdh1F/F;Trp53F/F female mice (data not shown), indicatingthat Wcre expression levels in virgin female mice in the Wcreconditional E-cadherin and p53 mouse model (Wcre;Cdh1F;Trp53F)are sufficient to induce stochastic Cre-mediated loss of E-cadherinand p53.

Loss of E-cadherin induces a shift from noninvasiveadenocarcinoma to ILC with pleomorphic featuresMammary tumors of Wcre;Trp53F/F females were categorized intotwo groups and diagnosed as intermediate-grade adenocarcinoma(AC) or solid carcinoma/carcinosarcoma (SC/CS) tumor types,both characterized by expansive growth patterns. Tumors displayedbenign noninvasive features and consisted of large epithelial cellsforming solid nests or irregular glands. ACs showed membranousexpression of E-cadherin and displayed a mixed but exclusiveexpression pattern of CK8 and CK14, but lacked expression ofvimentin and SMA (Fig. 5A; supplementary material Table S1).SC/CS lesions were characterized by a metaplastic and biphasichistology that consisted of epithelial and mesenchymal elements(Fig. 5B-D), showing a heterogeneous and mutually exclusiveexpression pattern for CK8 and CK14, occasionally expressingvimentin and mostly lacking expression of E-cadherin(supplementary material Table S1). Expansive growth patterns wereseen in the vast majority of tumors, which only sporadicallymetastasized to the lung (Table 1; supplementary material TablesS1 and S2).

Because most mammary tumors in Wcre;Trp53F/F animals wereneither invasive nor metastatic, we investigated the phenotypic

consequences of loss of E-cadherin. Mammary-specific somaticloss of E-cadherin and p53 in Wcre;Cdh1F/F;Trp53F/F femalesresulted in a significant shift from expansive to invasive mammarytumors (P<0.0001; Table 1), which showed strong phenotypicsimilarities to human pleomorphic ILC (PILC). These tumors,which we have previously designated mILC (Derksen et al., 2006),developed with high incidence multifocally in several mammaryglands (P<0.0001; Table 1). mILC cells grew in a non-cohesivemanner in a trabecular fashion, were small in size, relativelypleomorphic in appearance and diagnosed as high grade (Fig.5E,F). We could also detect signet ring cells, a typical occasionaltrait of human ILC (Fig. 5I). Like the adenocarcinomas fromWcre;Trp53F/F females, lobular carcinomas showed a mixed CK8and CK14 expression pattern, but did not express vimentin norSMA (supplementary material Table S1), indicating that mILC cellsdisplay epithelial properties. Although most mILCs fromWcre;Cdh1F/F;Trp53F/F and Wcre;Cdh1F/F;Trp53F/+ females wereestrogen receptor (ER) negative, we occasionally found mILClesions that weakly expressed ER. ER was mostly expressed by low-grade elements in mILC lesions, suggesting that ER expression isinversely correlated with tumor grade (supplementary materialTable S1 and Fig. S4).

Wcre;Cdh1F/F;Trp53F/F and Wcre;Cdh1F/F;Trp53F/+ females alsodeveloped SC/CS. Although displaying mILC components, thesetumors predominantly exhibited a mixed epithelial andmesenchymal or spindle-shaped cell morphology, presenting largecells with pleomorphic nuclei, coarsely clumped chromatin andsparse cytoplasm. SC/CS tumor cells showed both expansive andinvasive growth patterns, an exclusive and heterogeneousexpression of CK8 and CK14, and predominantly expressed



Fig. 3. Impaired lobulo-alveolar development upon dual inactivation of E-cadherin and p53. Comparative histochemistry during gestation of mammaryglands from control (Cdh1F/F;Trp53F/F) and Wcre;Cdh1F/F;Trp53F/F animals. Lumenal cells were identified using CK8. Also shown are E-cadherin expression andproliferation by means of staining against incorporated BrdU. HE, hematoxylin and eosin. Original magnification: 100�.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

vimentin, but lacked expression of E-cadherin and SMA (Table 1;supplementary material Table S1).

Loss of E-cadherin induces invasiveness and metastasisTo explore whether the metastatic spread of mILC in theWcre;Cdh1F;Trp53F model mimics the metastatic pattern of humanILC, we performed a detailed histological survey of various organsfrom tumor-bearing mice. Approximately 74% of theWcre;Cdh1F/F;Trp53F/F and 70% of the Wcre;Cdh1F/F;Trp53F/+

females that presented mammary tumors of approximately 1 cmin diameter showed extensive local invasion and metastases todraining and distant lymph nodes (Fig. 6 and Table 1). Discohesiveor loosely clustered mILC cells were detected in organs such asskin, lungs, liver, gastrointestinal tract, pancreas and spleen, or werediffusely disseminated throughout the peritoneal cavity (Fig. 6A-H), indicating that mILC in the Wcre;Cdh1F;Trp53F modelrecapitulates the histopathology and tumor biology of human ILC.In addition, several mice developed bone metastases, a feature thatwe have not observed in the K14cre;Cdh1F;Trp53F mouse model,exemplifying the additional value of the Wcre;Cdh1F;Trp53F model(Fig. 6I). mILC metastases showed a mixed and mutually exclusiveexpression pattern of CK8 and CK14, and displayed a cellularmorphology similar to that of the primary tumor (Fig. 6;supplementary material Fig. S5).

From the primary tumors that developed in our mouse model,we isolated tumor cells, which were then cultured and subsequentlytransduced with luciferase-encoding lentiviruses to enablebioluminescence imaging. To further characterize the metastaticspectrum of mILC, we orthotopically transplanted recipientanimals with approximately 10,000 luciferase-expressingCdh1/;Trp53/ (mILC) tumor cells. Using noninvasivebioluminescence imaging, we could image tumor growth anddistant metastases, which developed approximately 5 weeks posttransplantation (Fig. 7A; P<0.001). Metastases were detected in theperitoneal and thoracic cavity, contralateral mammary glands, lungsand bone (data not shown), illustrating that the full metastaticspectrum of human ILC is recapitulated upon orthotopictransplantation of mILC cell lines derived from theWcre;Cdh1F;Trp53F model. By contrast, orthotopic transplantationof luciferase-marked Trp53/ tumor cells showed a mixed growthpattern, depending on the cell line studied. Most E-cadherin-proficient tumor cells showed tumor growth, but did notmetastasize (Fig. 7A). The difference in metastatic capacity betweenTrp53/ and mILC tumor cells in situ correlated well with theacquisition of anchorage independence upon loss of E-cadherin(Fig. 7B), in analogy to tumor cells derived from theK14cre;Cdh1F;Trp53F mILC model (Derksen et al., 2006). In theabsence of cell-matrix interactions, E-cadherin-expressing Trp53/

tumor cells were not able to survive, whereas mILC cells readilysurvived and proliferated in an anchorage-independent fashion (Fig.7B). Anoikis resistance was observed for cell lines derived fromfour independent mILC primary tumors, whereas cell lines fromthree independent Trp53/ tumors showed anoikis inductionunder non-adherent culture conditions, indicating that loss of E-cadherin mediates survival of mammary tumor cells in the absenceof cell-matrix interactions. In conclusion, in vitro anoikis resistanceis a reliable prognosticator of in vivo metastatic dissemination, usingcell lines derived from the Wcre;Cdh1F;Trp53F mouse model.



Fig. 4. Synergistic tumor suppressor activity of E-cadherin and p53 inmammary tumorigenesis. Tumor incidence in Wcre uniparous femalescarrying conditional Cdh1 and Trp53 alleles. (A)E-cadherin and p53 synergizein mammary tumorigenesis. Shown is a Kaplan-Meier tumor-free survivalcurve for mammary tumors from Wcre;Trp53F/F (red) versusWcre;Cdh1F/+;Trp53F/F (blue), Wcre;Cdh1F/F;Trp53F/F (green) andWcre;Cdh1F/F;Trp53F/+ (brown) females. (B,C)Tumor onset and development inthe Wcre;Cdh1F;Trp53F mouse models is lactation independent. Tumor-freesurvival curves (Kaplan-Meier) of virgin Wcre;Cdh1F/+;Trp53F/F (B; black) andWcre;Cdh1F/F;Trp53F/F (C; black) female mice, versus uniparousWcre;Cdh1F/+;Trp53F/F (B; red) and Wcre;Cdh1F/F;Trp53F/F (C; green) females. Micewere sacrificed when mammary tumors reached an average diameter of 10mm.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

DISCUSSIONResearch on breast cancer metastasis has greatly profited fromrecent advances in modeling metastatic disease in mice, using bothtransplantation techniques and genetic modification (Khanna andHunter, 2005; Ottewell et al., 2006). Owing to the complex natureof the metastatic process, models that mimic both de novo tumordevelopment and spontaneous metastasis formation are scarce, butnevertheless emerging. Here, we have used Cre-loxP-basedconditional mutagenesis to develop a lactation-independent andmammary-gland-specific mouse model of human PILC.

Wcre-mediated inactivation of E-cadherin and p53 in miceinduced simultaneous development of multiple tumors in severalmammary glands. This high penetrance implies that only one or afew additional hits are required to induce tumor formation ofpredisposed E-cadherin- and p53-deficient mammary progenitors.Wcre-mediated loss of E-cadherin and p53 induced numerousbiphasic tumors in several mammary glands that displayed bothepithelial and mesenchymal characteristics, suggesting that

conditional inactivation based on Wap promoter activity occurs ina progenitor cell that is not yet fully committed to a distinctepithelial lineage. Although Wap is known to be transcribed duringgestation and lactation (Pittius et al., 1988; Dale et al., 1992), wefound that a substantial amount of Wcre mammary epithelial cellsshow Cre activity independent of gestation and lactation. Moreover,mammary glands from 4-week-old Wcre;R26R female mice alreadydisplayed -galactosidase activity, indicating that the Wap promoterused in our studies might be activated by prolactin in young virginanimals at to the onset of estrus (Topper and Freeman, 1980;Whittingham and Wood, 1983; Clarke et al., 1993). Nonetheless,the majority of mammary reconstitution assays that we haveperformed with mammary epithelial cells from 3-week-oldWcre;Cdh1F/F;Trp53F/F mice yielded phenotypically wild-typemammary glands (Eva J. Vlug and P.W.B.D., unpublished data),implying that the majority of mammary epithelial stem and/orprogenitor cells in 3-week-old Wcre conditional mice have retaineda non-switched (floxed) configuration and that the majority ofWcre-mediated recombination occurs upon the onset the firstestrous cycle, which commences after approximately 28 days inmice (Caligioni, 2009). Because multiple copies of the Wcretransgene will have concatemerized during genomic integration,we envisage that physiological prolactin levels in the virgin animalsmight be sufficient to drive stochastic Wcre-mediated loss of E-cadherin and p53 in the absence of a lactational pulse.

It is likely that dependency on E-cadherin function changesduring differentiation of mammary stem and/or progenitor cellstowards lumenal and myoepithelial descendants. Cells thatdifferentiate into lumenal-type cells will become dependent on E-cadherin-based adherens junctions, and succumb to apoptosis uponloss of E-cadherin as a result of the absence of a redundant classicalcadherin (Boussadia et al., 2002; Derksen et al., 2006). Following



Fig. 5. Tumor spectrum in the Wcre;Trp53F

and Wcre;Cdh1F;Trp53F mouse models.Comparative histology of mammarytumors. Shown are H&E-stained sections of:(A) adenocarcinoma (AC), (B) solidcarcinoma/carcinosarcoma (SC/CS),epithelial type, (C) SC/CS, mesenchymaltype, (D) SC/CS, mixed epithelial-mesenchymal type, (E) mILC, classical type(note the typical trabecular ‘Indian file’-typeinvasive growth patterns), (F) mILC,pleomorphic type, (G) mILC, solid type, (H)mouse LCIS, (I) metastatic mILC, signet-ringcell type [note the characteristic presence ofintracellular lumina containing mucin(arrow)]. Scale bars: 100m.

Table 1. Somatic inactivation of E-cadherin leads to invasive andmetastatic mILC in the Wcre;Cdh1F;Trp53F mouse model

Condition

Wcre;Trp53F/F +

(n=53)

F/FWcre;Cdh1F/F;Trp53F/+ +

Wcre;Cdh1F/F;Trp53F/F

(n=43) 2 P-valueInvasive growth 6 (11%) 37 (86%) <0.0001

Metastasis 5 (9%) 20 (47%) <0.0001

AC 10 (19%) 1 (2%) <0.0001

SC/CS 51 (96%) 29 (67%) <0.05

mILC 0 (0%) 26 (60%) <0.0001Statistical significance (P-values) was determined using a Pearson exact 2 test (twosided).

Wcre;Cdh1F/+;Trp53

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

this event, cells will be cleared from the fat-pad, resulting in theabsence of E-cadherin-deficient glandular structures inMMTVcre;Cdh1F/F and Wcre;Cdh1F/F mice (Boussadia et al., 2002)(and this study). Myoepithelial cells do not express E-cadherin, andhence will not be influenced by Cdh1 deletion, mainly because P-cadherin is the predominant classical cadherin that will maintainthe junctional integrity in these cells (Radice et al., 1997). Wetherefore hypothesize that mammary stem and/or progenitor cellstolerate the absence of E-cadherin until differentiation into lumenaldescendants is commenced.

Our results show that loss of p53 confers resistance to theproapoptotic signals that are initiated upon E-cadherin inactivation,resulting in the survival of lumenal cells in the absence of afunctional adherens junction. In Wcre;Cdh1F/F;Trp53F/F femalemice, inactivation of E-cadherin and p53 during gestation resultedin the accumulation of mostly CK8-positive epithelial cells andmesenchymal-type cells that fill the periductal environment,thereby inhibiting formation of a functional mammary gland.Interestingly, Wcre-mediated gene switching did not result inaccelerated tumor development in parous females when compared

with the incidence in virgin females or the tumor-free survival inour models, indicating that Cre recombinase activity duringgestation and lactation is mostly restricted to differentiated cellsthat do not contribute to mammary tumor formation. By contrast,lactation-independent inactivation of p53 and E-cadherin inmammary stem and/or progenitor cells will give rise to transformedepithelial cell types that display enhanced survival and growth and,in addition, harbor stem-cell-like characteristics that might facilitateprogression into carcinomas with a non-cohesive growth patterntypical for mouse and human ILC. In the rare event that themyoepithelial compartment is not affected by somatic inactivationof the mutant alleles and the integrity of the basement membraneremains unaffected, this might give rise to LCIS, as in the humansituation. Nonetheless, most early mILC lesions show a disturbedmyoepithelial and/or basement membrane architecture, which ismost probably causal to their highly invasive character. Also, thestromal component that is abundantly present in the periductalregions might contribute to tumor cell invasiveness throughdeposition of extracellular matrix, as has been recently proposed(Levental et al., 2009).



Fig. 6. The metastatic spectrum of mILC. (A)Local invasion ofmILC, showing infiltration of the skin dermis by tumor cells.(B)Infiltration of the striated muscle of the hind limb by mILCcells, which originated from a primary tumor present in theadjacent fifth mammary gland. (C,D)Distant metastasis ofmILC. Carcinoma cells are infiltrating an axillary lymph node (C)and lungs (D). (E-H)Peritoneal involvement in mILC. Sectionsshowing invasion of mILC cells into the muscularis externa ofthe stomach (E), metastasis of mILC to liver (F) and spleen (G),and colonization by mILC cells of a pancreatic islet ofLangerhans (H). (I)Bone metastasis in mILC. Metastatic mILCcells infiltrating the right femur of a Wcre;Cdh1F/F;Trp53F/+

female mouse. Arrows point to metastatic mILC cells. Scalebars: 100m.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

Although human ILCs are mostly ER positive, we only detectedsporadic ER-positive cells in low- or intermediate-grade mILClesions that still display features of glandular and/or ductalarchitecture. High-grade invasive and metastatic mILC nevershowed ER expression, which is in line with previous observationsthat mouse mammary tumors in general show physiologicaldifferences compared with human breast cancer with respect tohormone receptor expression (Yoshidome et al., 2000; Jonkers etal., 2001; Derksen et al., 2006; Alvarez et al., 2006; Liu et al., 2007).We assume that the vast majority of the mammary tumors thatform in Wcre;Cdh1F/F;Trp53F/F female mice have developed froman ER-negative progenitor cell. Occasionally, preneoplastic cellswill express ER, but will either not develop into a high-gradeinvasive tumor, or will transit to an ER-signaling independentmILC.

In contrast to human classical ILC, mILC depends oninactivation of p53, which is thought to occur in a minority (4-25%) of human cases (Marchetti et al., 1993; Rosen et al., 1995;Soslow et al., 2000; Coradini et al., 2002; Arpino et al., 2004). Moststudies have employed immunohistochemical detection of mutantp53 protein, leaving the possibility of functional p53 loss throughalternative mechanisms. Supporting this are recent studiesshowing that 30-40% of human lobular carcinomas have lost theTP53 locus on chromosome 17p13 (Mohsin et al., 2005; Stangeet al., 2006). Most mILC lesions showed atypical nuclearmorphology and would be classified as PILC in human pathology.PILC shows a more aggressive clinical behavior than ILC but isthought to share a common genetic pathway (Buchanan et al., 2008;Orvieto et al., 2008; Simpson et al., 2008; Gudlaugsson et al., 2009).The basis for the poorer prognosis of PILC when compared withILC is thought to emanate from the fact that PILC is characterizedby a higher frequency of ER and/or progesterone receptornegativity (Bentz et al., 1998) and loss of p53 function (Middletonet al., 2000; Kaya et al., 2002; Sneige et al., 2002; Reis-Filho et al.,2005).

In conclusion, we have generated a novel Wcre transgenic mouseline that displays Cre recombinase activity in mammary stemand/or progenitor cells. Consequently, Wcre-driven somaticinactivation of E-cadherin and p53 results in mammary tumorformation with similar incidence and latency in both virgin andparous mice. Combined inactivation of p53 and E-cadherin leadsto mILC, which is highly invasive and shows (lymph)angiogenicand diffuse dissemination with metastasis to the gastrointestinaltract and bone, similar to the human situation. Our mouse modeltherefore represents an excellent preclinical model to test novelintervention strategies for invasive and metastatic breast cancer.

METHODSGeneration of Wcre transgenic miceWe constructed the Wcre transgene from an expression cassettethat includes a 4.5-kb genomic BamHI-SalI mouse Wap genepromoter fragment followed by a 0.65-kb rabbit -globin intronand a 0.63-kb transcription termination/polyadenylation fragmentderived from the human growth hormone gene. Cre codingsequences were inserted between the intron and the poly(A)fragment (Fig. 1A). Next, we separated the 7.8-kb transgenefragment from vector sequences, purified and injected it into thepronuclei of one-cell-stage embryos of FVB/N mice. Microinjectedeggs were transferred at the two-cell stage into the oviducts ofpseudopregnant recipient females. Heterozygous transgenicanimals were identified using PCR analysis.

DNA analysis and genotypingGenomic DNA isolation and Southern blot analysis were performedas described (Jonkers et al., 2001; Derksen et al., 2006). Detectionof the Trp53F, Trp53, Cdh1F and Cdh1 alleles was done by PCRas described (Derksen et al., 2006). Transgenic Wcre animalswere identified by PCR using primers 5�-ACAGCCATCAGT -CACTTGCC-3� and 5�-CATCACTCGTTGCATCGACC-3�,yielding an amplification product of 432 bp.



Fig. 7. mILC anoikis resistance correlates with in vivo metastasis. (A)mILC metastasis imaging in vivo. Bioluminescence imaging of recipient animals, whichwere orthotopically transplanted with luciferase-transduced Trp53/ cells (top panels) or mILC cells (bottom panels). The color bar represents bioluminescenceintensity counts. Transplantations were performed using two different mILC and Trp53/ cell lines, in a minimum of five recipient animals. (B)Loss of E-cadherininduces anoikis resistance in the absence of p53. Tumor cell lines established from Wcre;Trp53F/F (AC) and Wcre;Cdh1F/F;Trp53F/F (mILC) female mice were platedonto non-coated low cluster wells and the percentage of cells expressing phosphatidylserine was determined using binding to FITC-conjugated annexin-V. Deadcells were detected using DNA binding to ToPro-3. In the presence of E-cadherin, tumor cells are not able to survive in the absence of anchorage (white bars),whereas E-cadherin-deficient Cdh1/;Trp53/ (mILC) cells (black bars) show anoikis resistance (P>0.001). Error bars represent the standard deviation of triplicatemeasurements.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

AntibodiesAntibodies used were: mouse anti-E-cadherin (1:300; BDBiosciences), mouse anti -catenin (1:150; BD Biosciences), rat anti-CK8 (Troma-1; 1:125; Developmental Studies Hybridoma Bank),rabbit anti-CK14 (1:10,000; BabCo), guinea pig anti-vimentin(1:400; RDI), rabbit anti-SMA (1:350; Lab Vision), mouse anti-BrdU(1:1000; DAKO), rabbit anti-ER (1:1000; Santa CruzBiotechnology). Secondary antibodies were: biotin-conjugated anti-mouse, anti-rat and anti-rabbit antibodies (DAKO), and biotin-conjugated anti-guinea pig (Jackson ImmunoResearch).

Stainings of whole mount preparationsReconstituted mammary glands were dissected and stretched ona glass slide. The glands were fixed in a mixture of 6:3:1methanol:1,1,1-trichloroethane:acetic acid (all Sigma) for 4 hoursand processed as whole mounts, which were stained overnight withcarmine aluminum staining solution [2 g/l carmine (Sigma), 5 g/laluminum potassium sulphate dissolved in H2O]. After stepwisedehydration in 70%, 95% and 100% ethanol, the glands were clearedin xylene (Sigma) for 10 minutes before taking pictures. For the -galactosidase staining, we fixed glands in 4% paraformaldehyde(Sigma), 2 mM MgCl2 (Merck) and 5 mM EGTA (Sigma) for 4 hoursat 4°C. After washing with 2 mM MgCl2, 0.02% NP-40, 0.01% Na-deoxycholate in PBS, glands were stained overnight with X-Galstaining solution [5 mM ferro-cyanide (K4Fe(CN)6), 5 mM ferri-cyanide (K3Fe(CN)6), 2 mM MgCl2, 0.02% NP-40 and 1 mg/ml X-Gal (Biosolve)] at 37°C in the dark. After dehydration in ethanol,glands were cleared in xylene (Sigma) for 10 minutes. Fordetermination of cellular proliferation, mice were injectedintraperitoneally with 2 mg/ml BrdU (Sigma) 90 minutes prior toharvest of the mammary glands.

Histological analysisTissues were isolated, fixed and processed as described (Derksenet al., 2006).

Cell cultureCells were isolated and cultured as described previously (Derksenet al., 2006)

Lentiviral production and transduction of cellsLentiviral particles were produced by seeding 106 293T cells ontoa 10 cm Petri dish and performing transient transfection after 24hours with third-generation packaging constructs and a luciferase-encoding transfer vector (LV-luc) (Derksen et al., 2006).Supernatant containing lentiviral particles were harvested andconcentrated tenfold by ultra centrifugation at 75,000 g for 2.5hours. Tumor cells were infected for 16 hours in the presence of 4g/ml polybrene.

Orthotopic transplantations and bioluminescence imaging3-week-old Rag2–/–;IL2Rc–/– BALB/c female recipient mice(Gimeno et al., 2004) were anaesthetized by intraperitonealinjection of a mixture containing 25 l fentanyl citrate/fluanisone(hypnorm; Janssen Pharmaceutica), 25 l midazolam (dormicum;Roche) and 50 l water. The fourth mammary gland was exposedand endogenous mammary epithelial tissue was removed. Next,approximately 10,000 luciferase-transduced tumor cells were

injected in the cleared fat-pad using a 10 l Hamilton syringe, afterwhich the animals were sutured. After a recovery period of 2 weeks,mice were anesthetized with isofluorane (Janssen Pharmaceutica),injected intraperitoneally with 225 g/g body weight n-luciferin(potassium salt; Biosynth AG) and imaged on an IVIS-200bioluminescence imager (Xenogen). All animal experiments wereperformed in accordance with institutional guidelines and nationalregulations.

Anoikis assayCells were plated at a density of 75,000 cells per well in a six-wellultra-low cluster polystyrene culture dish (Corning). FITC-conjugated annexin-V (1:20; IQ Products) and ToPro-3 (1:2000;Molecular Probes) were added and annexin-V-positive apoptoticcells were analyzed by FACS as described (Derksen et al., 2003).Statistical significance was calculated using the Student’s t-test.

ACKNOWLEDGEMENTSWe thank Miranda van Amersfoort, Hermien Boerhout and Sabine Vishnudatt fortechnical support. Members of the Jonkers lab are acknowledged for reagents,help and fruitful discussions. We are also indebted to the animal facility and theanimal pathology lab. Joost Vermaat is kindly acknowledged for the statisticalanalyses. This work was supported by grants from the Netherlands Organizationfor Scientific Research (ZonMw VIDI 917.36.347) and the Dutch Cancer Society (NKI2002-2635 and NKI 2006-3486). P.W.B.D. was supported by grants from theNetherlands Organization for Scientific Research (ZonMw VENI 916.56.135 andVIDI 917.96.318).



TRANSLATIONAL IMPACT

Clinical issueMetastatic disease is the major cause of mortality in breast cancer patients.A hallmark of invasive and metastatic cells is inhibition of E-cadherinfunction. Invasive lobular carcinoma (ILC; the second most common type ofprimary breast cancer) is characterized by early loss of the epithelial cell-celladhesion molecule E-cadherin. The classical form of ILC is characterized bynon-cohesive and invasive cells but without mass formation, often resultingin a false-negative diagnosis using physical examination or mammography.Although most cases of ILC are estrogen receptor (ER) positive andtherefore responsive to initial endocrine therapy, the disease is generallyrefractory to standard chemotherapy treatments once ER expression is lostat later stages.

ResultsIn this paper, the authors establish a new mammary-specific conditionalknockout mouse model to show that combined stochastic inactivation of E-cadherin and p53 results in impaired mammary gland function.Phenotypically, mammary tumors in this model exhibit multiple features ofhuman pleomorphic ILC, both in terms of their appearance and theirmetastatic behavior. The data show that somatic inactivation of E-cadherinrenders mammary epithelial cells anoikis resistant in the context of p53deficiency. Importantly, the female mice in this model develop highly invasiveand metastatic mammary tumors for which the onset, incidence andmetastasis are lactation independent.

Implications and future directionsThe lactation-independent and mammary-gland-specific mouse model ofhuman pleomorphic ILC presented in this paper provides a new tool to gaininsight into the role of E-cadherin loss of function in mammary tumorinitiation, progression and metastasis. The study also addresses how E-cadherin and p53 might cooperate in mammary tumor development. Giventhe similarities between mouse and human ILC, this model might ultimatelycontribute to the development of novel clinical intervention strategies for thetreatment of metastatic breast cancer.

Dise

ase

Mod

els &

Mec

hani

sms

D

MM

COMPETING INTERESTSThe authors declare that they do not have any competing or financial interests.

AUTHOR CONTRIBUTIONSP.W.B.D. and J.J. designed the experiments. P.K. generated the Wcre transgenicmice. P.W.B.D., T.M.B., M.H. and E.v.d.B. performed the experiments. E.M. executedthe immune histochemistry on tissue sections and J.W. diagnosed mousepathology. P.W.B.D. wrote the paper with assistance from J.J.

SUPPLEMENTARY MATERIALSupplementary material for this article is available athttp://dmm.biologists.org/lookup/suppl/doi:10.1242/dmm.006395/-/DC1

REFERENCESAlvarez, J. V., Perez, D. and Chodosh, L. A. (2006). MILC-ing the mouse mammary

gland: A model for invasive lobular carcinoma. Cancer Cell 10, 347-349.Arpino, G., Bardou, V. J., Clark, G. M. and Elledge, R. M. (2004). Infiltrating lobular

carcinoma of the breast: tumor characteristics and clinical outcome. Breast CancerRes. 6, R149-R156.

Batlle, E., Sancho, E., Franci, C., Dominguez, D., Monfar, M., Baulida, J. and GarciaDe Herreros, A. (2000). The transcription factor snail is a repressor of E-cadheringene expression in epithelial tumour cells. Nat. Cell Biol. 2, 84-89.

Bentz, J. S., Yassa, N. and Clayton, F. (1998). Pleomorphic lobular carcinoma of thebreast: clinicopathologic features of 12 cases. Mod. Pathol. 11, 814-822.

Berx, G., Cleton-Jansen, A. M., Nollet, F., de Leeuw, W. J., van de Vijver, M.,Cornelisse, C. and van Roy, F. (1995). E-cadherin is a tumour/invasion suppressorgene mutated in human lobular breast cancers. EMBO J. 14, 6107-6115.

Boussadia, O., Kutsch, S., Hierholzer, A., Delmas, V. and Kemler, R. (2002). E-cadherinis a survival factor for the lactating mouse mammary gland. Mech. Dev. 115, 53-62.

Buchanan, C. L., Flynn, L. W., Murray, M. P., Darvishian, F., Cranor, M. L., Fey, J. V.,King, T. A., Tan, L. K. and Sclafani, L. M. (2008). Is pleomorphic lobular carcinomareally a distinct clinical entity? J. Surg. Oncol. 98, 314-317.

Caligioni, C. S. (2009). Assessing reproductive status/stages in mice. Curr. Protoc.Neurosci. Appendix 4, Appendix 4I.

Cano, A., Perez-Moreno, M. A., Rodrigo, I., Locascio, A., Blanco, M. J., del Barrio, M.G., Portillo, F. and Nieto, M. A. (2000). The transcription factor snail controlsepithelial-mesenchymal transitions by repressing E-cadherin expression. Nat. CellBiol. 2, 76-83.

Clarke, D. L., Arey, B. J. and Linzer, D. I. (1993). Prolactin receptor messengerribonucleic acid expression in the ovary during the rat estrous cycle. Endocrinology133, 2594-2603.

Cleton-Jansen, A. M., Moerland, E. W., Kuipers-Dijkshoorn, N. J., Callen, D. F.,Sutherland, G. R., Hansen, B., Devilee, P. and Cornelisse, C. J. (1994). At least twodifferent regions are involved in allelic imbalance on chromosome arm 16q in breastcancer. Genes Chromosomes Cancer 9, 101-107.

Comijn, J., Berx, G., Vermassen, P., Verschueren, K., van Grunsven, L., Bruyneel, E.,Mareel, M., Huylebroeck, D. and van Roy, F. (2001). The two-handed E box bindingzinc finger protein SIP1 downregulates E-cadherin and induces invasion. Mol Cell. 7,1267-1278.

Coradini, D., Pellizzaro, C., Veneroni, S., Ventura, L. and Daidone, M. G. (2002).Infiltrating ductal and lobular breast carcinomas are characterised by differentinterrelationships among markers related to angiogenesis and hormonedependence. Br. J. Cancer 87, 1105-1111.

Dale, T. C., Krnacik, M. J., Schmidhauser, C., Yang, C. L., Bissell, M. J. and Rosen, J.M. (1992). High-level expression of the rat whey acidic protein gene is mediated byelements in the promoter and 3� untranslated region. Mol. Cell Biol. 12, 905-914.

Derksen, P. W., de Gorter, D. J., Meijer, H. P., Bende, R. J., van Dijk, M., Lokhorst, H.M., Bloem, A. C., Spaargaren, M. and Pals, S. T. (2003). The hepatocyte growthfactor/Met pathway controls proliferation and apoptosis in multiple myeloma.Leukemia 17, 764-774.

Derksen, P. W., Liu, X., Saridin, F., Van Der Gulden, H., Zevenhoven, J., Evers, B.,van Beijnum, J. R., Griffioen, A. W., Vink, J., Krimpenfort, P. et al. (2006). Somaticinactivation of E-cadherin and p53 in mice leads to metastatic lobular mammarycarcinoma through induction of anoikis resistance and angiogenesis. Cancer Cell 10,437-449.

Frixen, U. H., Behrens, J., Sachs, M., Eberle, G., Voss, B., Warda, A., Lochner, D. andBirchmeier, W. (1991). E-cadherin-mediated cell-cell adhesion prevents invasivenessof human carcinoma cells. J. Cell Biol. 113, 173-185.

Frykberg, E. R., Santiago, F., Betsill, W. L., Jr and O’Brien, P. H. (1987). Lobularcarcinoma in situ of the breast. Surg. Gynecol. Obstet. 164, 285-301.

Fujita, N., Jaye, D. L., Kajita, M., Geigerman, C., Moreno, C. S. and Wade, P. A.(2003). MTA3, a Mi-2/NuRD complex subunit, regulates an invasive growth pathwayin breast cancer. Cell 113, 207-219.

Gimeno, R., Weijer, K., Voordouw, A., Uittenbogaart, C. H., Legrand, N., Alves, N.L., Wijnands, E., Blom, B. and Spits, H. (2004). Monitoring the effect of gene

silencing by RNA interference in human CD34+ cells injected into newborn RAG2-/-gammac-/- mice: functional inactivation of p53 in developing T cells. Blood 104,3886-3893.

Gonzalez-Angulo, A. M., Morales-Vasquez, F. and Hortobagyi, G. N. (2007).Overview of resistance to systemic therapy in patients with breast cancer. Adv. Exp.Med. Biol. 608, 1-22.

Graff, J. R., Herman, J. G., Lapidus, R. G., Chopra, H., Xu, R., Jarrard, D. F., Isaacs, W.B., Pitha, P. M., Davidson, N. E. and Baylin, S. B. (1995). E-cadherin expression issilenced by DNA hypermethylation in human breast and prostate carcinomas.Cancer Res. 55, 5195-5199.

Gudlaugsson, E., Skaland, I., Janssen, E. A., van Diest, P. J., Voorhorst, F. J.,Kjellevold, K., Hausen, A. Z. and Baak, J. P. (2009). Prospective multicentercomparison of proliferation and other prognostic factors in lymph node negativelobular invasive breast cancer. Breast Cancer Res. Treat. 121, 35-40.

Gump, F. E. (1993). Lobular carcinoma in situ (LCIS): pathology and treatment. J. Cell.Biochem. Suppl. 17, 53-58.

Helvie, M. A., Paramagul, C., Oberman, H. A. and Adler, D. D. (1993). Invasive lobularcarcinoma. Imaging features and clinical detection. Invest. Radiol. 28, 202-207.

Hennighausen, L. G. and Sippel, A. E. (1982). Mouse whey acidic protein is a novelmember of the family of ‘four-disulfide core’ proteins. Nucleic Acids Res. 10, 2677-2684.

Hulsken, J., Birchmeier, W. and Behrens, J. (1994). E-cadherin and APC compete forthe interaction with beta-catenin and the cytoskeleton. J. Cell Biol. 127, 2061-2069.

Jonkers, J., Meuwissen, R., Van Der Gulden, H., Peterse, H., van der Valk, M. andBerns, A. (2001). Synergistic tumor suppressor activity of BRCA2 and p53 in aconditional mouse model for breast cancer. Nat. Genet. 29, 418-425.

Kaya, H., Aribal, E. and Yegen, C. (2002). Apocrine differentiation in invasivepleomorphic lobular carcinoma with in situ ductal and lobular apocrine carcinoma:case report. Pathol. Oncol. Res. 8, 151-152.

Khanna, C. and Hunter, K. (2005). Modeling metastasis in vivo. Carcinogenesis 26, 513-523.

Korkola, J. E., DeVries, S., Fridlyand, J., Hwang, E. S., Estep, A. L., Chen, Y. Y., Chew,K. L., Dairkee, S. H., Jensen, R. M. and Waldman, F. M. (2003). Differentiation oflobular versus ductal breast carcinomas by expression microarray analysis. CancerRes. 63, 7167-7175.

Krecke, K. N. and Gisvold, J. J. (1993). Invasive lobular carcinoma of the breast:mammographic findings and extent of disease at diagnosis in 184 patients. AJR Am.J. Roentgenol. 161, 957-960.

Levental, K. R., Yu, H., Kass, L., Lakins, J. N., Egeblad, M., Erler, J. T., Fong, S. F.,Csiszar, K., Giaccia, A., Weninger, W. et al. (2009). Matrix crosslinking forces tumorprogression by enhancing integrin signaling. Cell 139, 891-906.

Liu, X., Holstege, H., van der Gulden, H., Treur-Mulder, M., Zevenhoven, J., Velds,A., Kerkhoven, R. M., van Vliet, M. H., Wessels, L. F., Peterse, J. L. et al. (2007).Somatic loss of BRCA1 and p53 in mice induces mammary tumors with features ofhuman BRCA1-mutated basal-like breast cancer. Proc. Natl. Acad. Sci. USA 104, 12111-12116.

Marchetti, A., Buttitta, F., Pellegrini, S., Campani, D., Diella, F., Cecchetti, D.,Callahan, R. and Bistocchi, M. (1993). P53 mutations and histological type ofinvasive breast carcinoma. Cancer Res. 53, 4665-4669.

Mathieu, M. C., Rouzier, R., Llombart-Cussac, A., Sideris, L., Koscielny, S., Travagli,J. P., Contesso, G., Delaloge, S. and Spielmann, M. (2004). The poorresponsiveness of infiltrating lobular breast carcinomas to neoadjuvantchemotherapy can be explained by their biological profile. Eur. J. Cancer 40, 342-351.

Middleton, L. P., Palacios, D. M., Bryant, B. R., Krebs, P., Otis, C. N. and Merino, M. J.(2000). Pleomorphic lobular carcinoma: morphology, immunohistochemistry, andmolecular analysis. Am. J. Surg. Pathol. 24, 1650-1656.

Mohsin, S. K., O’Connell, P., Allred, D. C. and Libby, A. L. (2005). Biomarker profileand genetic abnormalities in lobular carcinoma in situ. Breast Cancer Res. Treat. 90,249-256.

Molland, J. G., Donnellan, M., Janu, N. C., Carmalt, H. L., Kennedy, C. W. andGillett, D. J. (2004). Infiltrating lobular carcinoma-a comparison of diagnosis,management and outcome with infiltrating duct carcinoma. Breast 13, 389-396.

Moody, S. E., Perez, D., Pan, T. C., Sarkisian, C. J., Portocarrero, C. P., Sterner, C. J.,Notorfrancesco, K. L., Cardiff, R. D. and Chodosh, L. A. (2005). The transcriptionalrepressor Snail promotes mammary tumor recurrence. Cancer Cell 8, 197-209.

Newstead, G. M., Baute, P. B. and Toth, H. K. (1992). Invasive lobular and ductalcarcinoma: mammographic findings and stage at diagnosis. Radiology 184, 623-627.

Oda, T., Kanai, Y., Oyama, T., Yoshiura, K., Shimoyama, Y., Birchmeier, W.,Sugimura, T. and Hirohashi, S. (1994). E-cadherin gene mutations in human gastriccarcinoma cell lines. Proc. Natl. Acad. Sci. USA 91, 1858-1862.

Orvieto, E., Maiorano, E., Bottiglieri, L., Maisonneuve, P., Rotmensz, N., Galimberti,V., Luini, A., Brenelli, F., Gatti, G. and Viale, G. (2008). Clinicopathologiccharacteristics of invasive lobular carcinoma of the breast: results of an analysis of530 cases from a single institution. Cancer 113, 1511-1520.


E-cadherin loss leads to pleomorphic mILC RESEARCH ARTICLED

iseas

e M

odel

s & M

echa

nism

s

DM

M

Ottewell, P. D., Coleman, R. E. and Holen, I. (2006). From genetic abnormality tometastases: murine models of breast cancer and their use in the development ofanticancer therapies. Breast Cancer Res. Treat. 96, 101-113.

Perez-Moreno, M., Jamora, C. and Fuchs, E. (2003). Sticky business: orchestratingcellular signals at adherens junctions. Cell 112, 535-548.

Perl, A. K., Wilgenbus, P., Dahl, U., Semb, H. and Christofori, G. (1998). A causal rolefor E-cadherin in the transition from adenoma to carcinoma. Nature 392, 190-193.

Pittius, C. W., Sankaran, L., Topper, Y. J. and Hennighausen, L. (1988). Comparisonof the regulation of the whey acidic protein gene with that of a hybrid genecontaining the whey acidic protein gene promoter in transgenic mice. Mol.Endocrinol. 2, 1027-1032.

Radice, G. L., Ferreira-Cornwell, M. C., Robinson, S. D., Rayburn, H., Chodosh, L. A.,Takeichi, M. and Hynes, R. O. (1997). Precocious mammary gland development inP-cadherin-deficient mice. J. Cell Biol. 139, 1025-1032.

Reis-Filho, J. S., Simpson, P. T., Jones, C., Steele, D., Mackay, A., Iravani, M.,Fenwick, K., Valgeirsson, H., Lambros, M., Ashworth, A. et al. (2005). Pleomorphiclobular carcinoma of the breast: role of comprehensive molecular pathology incharacterization of an entity. J. Pathol. 207, 1-13.

Rosen, P. P., Kosloff, C., Lieberman, P. H., Adair, F. and Braun, D. W., Jr (1978).Lobular carcinoma in situ of the breast. Detailed analysis of 99 patients with averagefollow-up of 24 years. Am. J. Surg. Pathol. 2, 225-251.

Rosen, P. P., Lesser, M. L., Arroyo, C. D., Cranor, M., Borgen, P. and Norton, L. (1995).P53 in node-negative breast carcinoma: an immunohistochemical study ofepidemiologic risk factors, histologic features, and prognosis. J. Clin. Oncol. 13, 821-830.

Savagner, P., Yamada, K. M. and Thiery, J. P. (1997). The zinc-finger protein slugcauses desmosome dissociation, an initial and necessary step for growth factor-induced epithelial-mesenchymal transition. J. Cell Biol. 137, 1403-1419.

Simpson, P. T., Gale, T., Fulford, L. G., Reis-Filho, J. S. and Lakhani, S. R. (2003). Thediagnosis and management of pre-invasive breast disease: pathology of atypicallobular hyperplasia and lobular carcinoma in situ. Breast Cancer Res. 5, 258-262.

Simpson, P. T., Reis-Filho, J. S., Lambros, M. B., Jones, C., Steele, D., Mackay, A.,Iravani, M., Fenwick, K., Dexter, T., Jones, A. et al. (2008). Molecular profilingpleomorphic lobular carcinomas of the breast: evidence for a common moleculargenetic pathway with classic lobular carcinomas. J. Pathol. 215, 231-244.

Sneige, N., Wang, J., Baker, B. A., Krishnamurthy, S. and Middleton, L. P. (2002).Clinical, histopathologic, and biologic features of pleomorphic lobular (ductal-lobular)carcinoma in situ of the breast: a report of 24 cases. Mod. Pathol. 15, 1044-1050.

Soriano, P. (1999). Generalized lacZ expression with the ROSA26 Cre reporter strain.Nat. Genet. 21, 70-71.

Soslow, R. A., Carlson, D. L., Horenstein, M. G. and Osborne, M. P. (2000). Acomparison of cell cycle markers in well-differentiated lobular and ductalcarcinomas. Breast Cancer Res. Treat. 61, 161-170.

Stange, D. E., Radlwimmer, B., Schubert, F., Traub, F., Pich, A., Toedt, G., Mendrzyk,F., Lehmann, U., Eils, R., Kreipe, H. et al. (2006). High-resolution genomic profilingreveals association of chromosomal aberrations on 1q and 16p with histologic andgenetic subgroups of invasive breast cancer. Clin. Cancer Res. 12, 345-352.

Takeichi, M. (1995). Morphogenetic roles of classic cadherins. Curr. Opin. Cell Biol. 7,619-627.

Topper, Y. J. and Freeman, C. S. (1980). Multiple hormone interactions in thedevelopmental biology of the mammary gland. Physiol. Rev. 60, 1049-1106.

Tubiana-Hulin, M., Stevens, D., Lasry, S., Guinebretiere, J. M., Bouita, L., Cohen-Solal, C., Cherel, P. and Rouesse, J. (2006). Response to neoadjuvant chemotherapyin lobular and ductal breast carcinomas: a retrospective study on 860 patients fromone institution. Ann. Oncol. 17, 1228-1233.

Vleminckx, K., Vakaet, L., Jr, Mareel, M., Fiers, W. and van Roy, F. (1991). Geneticmanipulation of E-cadherin expression by epithelial tumor cells reveals an invasionsuppressor role. Cell 66, 107-119.

Vos, C. B., Cleton-Jansen, A. M., Berx, G., de Leeuw, W. J., ter Haar, N. T., van Roy,F., Cornelisse, C. J., Peterse, J. L. and van de Vijver, M. J. (1997). E-cadherininactivation in lobular carcinoma in situ of the breast: an early event intumorigenesis. Br. J. Cancer 76, 1131-1133.

Wagner, K. U., Wall, R. J., St Onge, L., Gruss, P., Wynshaw-Boris, A., Garrett, L., Li,M., Furth, P. A. and Hennighausen, L. (1997). Cre-mediated gene deletion in themammary gland. Nucleic Acids Res. 25, 4323-4330.

Wheeler, J. E., Enterline, H. T., Roseman, J. M., Tomasulo, J. P., McIlvaine, C. H.,Fitts, W. T., Jr and Kirshenbaum, J. (1974). Lobular carcinoma in situ of the breast.Long-term followup. Cancer 34, 554-563.

Whittingham, D. G. and Wood, M. J. (1983). The Mouse in Biomedical Research,Reproductive Physiology (Foster, H. L., Small, J. D. and Fox, J. G., eds), pp. 137-164. NewYork: Academic Press.

Yang, J., Mani, S. A., Donaher, J. L., Ramaswamy, S., Itzykson, R. A., Come, C.,Savagner, P., Gitelman, I., Richardson, A. and Weinberg, R. A. (2004). Twist, amaster regulator of morphogenesis, plays an essential role in tumor metastasis. Cell117, 927-939.

Yoshidome, K., Shibata, M. A., Couldrey, C., Korach, K. S. and Green, J. E. (2000).Estrogen promotes mammary tumor development in C3(1)/SV40 large T-antigentransgenic mice: paradoxical loss of estrogen receptoralpha expression during tumorprogression. Cancer Res. 60, 6901-6910.

Yoshiura, K., Kanai, Y., Ochiai, A., Shimoyama, Y., Sugimura, T. and Hirohashi, S.(1995). Silencing of the E-cadherin invasion-suppressor gene by CpG methylation inhuman carcinomas. Proc. Natl. Acad. Sci. USA 92, 7416-7419.

Zhao, H., Langerod, A., Ji, Y., Nowels, K. W., Nesland, J. M., Tibshirani, R., Bukholm,I. K., Karesen, R., Botstein, D., Borresen-Dale, A. L. et al. (2004). Different geneexpression patterns in invasive lobular and ductal carcinomas of the breast. Mol. Biol.Cell 15, 2523-2536.


E-cadherin loss leads to pleomorphic mILCRESEARCH ARTICLED

iseas

e M

odel

s & M

echa

nism

s

DM

M

Documents

Mammary-specific inactivation of E-cadherin and p53 ...initiating and causal event in the development of mouse ILC (mILC) (Derksen et al., 2006), is a common feature of LCIS as well