¹Institute for Molecular Biology and Biotechnology, McMaster University, 1280 Main St. W., Hamilton, Ontario, Canada, L8S 4K1

²Department of Biology, McMaster University, 1280 Main St. W., Hamilton, Ontario, Canada, L8S 4K1

³Department of Biochemistry, McMaster University, 1280 Main St. W., Hamilton, Ontario, Canada, L8S 4K1

⁴Department of Pathology and Molecular Medicine, McMaster University, 1280 Main St W., Hamilton, Ontario, Canada, L8S 4K1

Correspondence to: W J Muller, Institute for Molecular Biology and Biotechnology, McMaster University, 1280 Main St. W., Hamilton, Ontario, Canada, L8S 4K1

The pathogenesis of human breast cancer is thought to involve multiple genetic events, the majority of which fall into two categories, gain of function mutations in proto-oncogenes such as c-myc, cyclin D1, ErbB-2 and various growth factors which are involved in supporting cell growth, division and survival, and loss of function mutations in so called 'tumor suppressor' genes, such as p53, which are involved in preventing unrestrained cellular growth. A number of mouse systems exist to address the significance of these mutations in the pathogenesis of breast cancer including transgenic mice expressing high levels of a specific gene in target tissues and knockout mice in which specific genes have been ablated via homologous recombination. More recently, the combination of these techniques to create bigenics as well as the use of 'knockin' and conditional tissue specific gene targeting strategies have allowed the models more reflective of the human disease to be devised. Studies with these models have not only implicated particular genetic events in the progression of the disease but have emphasized the complex, multi-step nature of breast cancer progression. These models also provide the opportunity to study various aspects of the pathogenesis of this disease, from hormonal effects to responses to chemotherapeutic drugs. It is hoped that through the combined use of these models, and the further development of more relevant models, that a deeper understanding of this disease and the generation of new therapeutic agents will result. Oncogene (2000) 19, 6130-6137.

Transgenic mice; knockout mice; mammary gland; cancer; oncogenes; tumor suppressors

Introduction

The pathogenesis of breast cancer is thought to involve multiple genetic events. Karyotypic and epidemiological analyses of mammary tumors at various stages suggest that breast carcinomas become increasingly aggressive through the stepwise accumulation of genetic changes (Dupont and Page, 1985). The majority of genetic changes found in human breast cancer fall into two categories, gain of function mutations in proto-oncogenes, which are involved in supporting cell growth, division and survival, and loss of function mutations in so called 'tumor suppressor' genes, which are involved in preventing unrestrained cellular growth. The majority of gain of function mutations in human primary breast cancers involve amplifications in one of three chromosomal regions, the c-myc and erbB-2 proto-oncogenes or the chromosomal band 11q13 (Lidereau et al., 1988). Loss of function mutations in primary human breast cancers include changes in the known tumor suppressor p53 as well as in the familial cancer markers of the BRCA gene family. Additionally, multiple regions of loss of heterozygosity (LOH) are observed in primary human breast cancers (Bieche and Lidereau, 1995; Callahan et al., 1992; Garcia et al., 1999). It is thought that these regions of LOH affect as yet unidentified putative tumor suppressors. Indeed allelic loss of the PTEN region has been noted in a subset of aggressive breast cancers (Garcia et al., 1999).

A number of mouse systems exist to address the significance of these mutations in the pathogenesis of breast cancer. On the most basic level, the use of transgenic mice expressing high levels of a specific gene in a target tissue allows the involvement of a given gene in the pathogenesis of breast cancer to be addressed. Alternatively, the ablation of specific genes via homologous recombination also allows researchers to determine the role of a gene in breast cancer progression. More recently, the combination of these techniques to create bigenics as well as the use of 'knock-in' and conditional tissue specific gene targeting strategies have allowed the creation of models more reflective of the human disease to be devised.

Transgenic mouse models of gain of function mutations

A number of transgenic promoters have been employed to target transgene expression to the mammary gland. A majority of the transgenics generated have employed either the mouse mammary tumor virus long terminal repeat (MMTV) or the whey acidic protein promoter (WAP). The MMTV-LTR is active throughout mammary development and its transcriptional activity increases during pregnancy (Pattengale et al., 1989). In contrast, the WAP promoter is only active in the mid-pregnant mammary gland. Thus, it is apparent that the phenotypes exhibited by WAP and MMTV transgenics may depend upon the developmental stage of the individual mouse examined. Other less common promoters employed include the 5' flanking region of the C3(1) component of the rat prostate steroid binding protein, beta-lactalbumin, metallothionein and tetracycline responsive promoters.

Models for genetic regions amplified in human breast cancer

c-myc

The c-myc gene encodes for a transcription factor that is frequently amplified in human tumors (Berns et al., 1992; Bieche and Lidereau, 1995; Escot et al., 1986). Multiple transgenic studies in which the myc gene was overexpressed under the control of mammary specific promoters have indicated an important role for myc in the progression of breast cancer (Leder et al., 1986; Schoenenberger et al., 1988; Stewart et al., 1984). The first of these studies used the MMTV promoter to over-express myc in the mammary glands of mice and resulted in spontaneous mammary adenocarcinomas in two distinct lines by 4 to 8 months of age (Stewart et al., 1984). A second transgenic study also using the MMTV promoter to overexpress myc resulted in the formation of locally invasive mammary tumors in four multiparous females by 10 to 19 months of age (Leder et al., 1986). Interestingly, in one of MMTV/myc transgenic strains, c-myc expression was detected in a wide range of tissues. Despite the broad pattern of tissue specific expression, these mice developed a limited subset of tumor types including mammary tumors. Thus elevated expression of c-myc appears capable of inducing tumors in selected tissue sites.

Elevated expression of myc in the mammary gland has also been achieved by placing the c-myc oncogene under the transcriptional control of the WAP promoter (Schoenenberger et al., 1988). In these transgenic strains, 80% of female transgenics develop multiple tumors affecting single or multiple glands after two pregnancies at ages as early as 2 months (Schoenenberger et al., 1988). Together they demonstrate that c-myc can induce mammary tumor formation when overexpressed in the mammary gland. However, the fact that overexpression of c-myc does not result in transformation of the entire mammary gland, as normal mammary epithelium is also present in these strains, reveals that additional genetic events are required for the development of overt mammary carcinomas. In this sense, these models accurately reflect the nature of the progression of human breast cancer.

Cyclin D1

Cyclins regulate the activation of cyclin-dependent kinases (CDK's) allowing cell cycle progression, S phase entry and DNA replication. A variety of lines of evidence have linked cyclins to the progression of breast cancer. Foremost, the cyclin D1 gene is found within the 11q13 region, which is amplified in 15-20% of primary human breast cancers (Bieche and Lidereau, 1995; Brison, 1993). Overexpression of cyclin D1 under control of the MMTV promoter results in proliferative abnormalities in the mammary gland, with significant lobulo-alveolar development shortly after sexual maturity is reached. Significantly, eight of 12 mice from three distinct transgenic lines developed focal mammary tumors with a mean onset of 18 months (Wang et al., 1994). As with the c-myc transgenic models, the long latency and focal nature of these tumors suggests that although cyclin D1 can promote mammary tumorigenesis, additional genetic changes are needed for the development of overt mammary carcinomas. Consistent with this view, mammary epithelial expression of cyclin D1 has been implicated as an important event in mammary tumor induced by activated Src kinases, integrin linked kinase (ILK) and ErbB-2 (Lee et al., 1999, 2000; Radeva et al., 1997). Conversely germline inactivation of cyclin D1 results in impaired mammary epithelial gland development (Fantl et al., 1999). Collectively these observations suggest that cyclin D1 plays a critical role in both normal mammary gland development and mammary tumorigenesis.

ErbB-2

ErbB-2 is a member of the EGFR family of receptor tyrosine kinases (RTKs). This family is comprised of four closely related type 1 RTKs that include the EGFR, ErbB-2 (Neu, HER2), ErbB-3 (HER3), and ErbB-4 (HER4) (Hynes and Stern, 1994; Olayioye et al., 2000). Signaling in these receptors involves the formation of homo and hetero-dimers in response to ligand stimulation. This dimerization results in the phosphorylation of specific tyrosine residues on the receptor. These phosphorylated tyrosines then offer docking sites for the SH2 and SH3 (PTB) domains of various endogenous signaling molecules that are able to interact with the receptor and transduce the signal (Dankort and Muller, 2000; Hynes and Stern, 1994). Originally, erbB-2 was described as the oncogene neu found in chemically induced neuroblastomas in rats (Schechter et al., 1984). Neu possesses a valine-glutamic acid substitution in its transmembrane domain that results in the constitutive aggregation and activation of the receptor in the absence of ligand (Bargmann et al., 1986a,b; Dankort and Muller, 2000; Stern et al., 1986; Xie et al., 1995).

The importance of ErbB-2 in primary human breast cancer is highlighted by the fact that 20-30% of human breast cancers express elevated levels of ErbB-2 due to the genomic amplification of the erbB-2 proto-oncogene (Slamon, 1987, 1989). Furthermore, its amplification and subsequent overexpression strongly correlates with a negative clinical prognosis in both lymph node positive (Hynes and Stern, 1994; Mansour et al., 1994; Ravdin and Chamness, 1995) and node-negative (Andrulis et al., 1998) breast cancer patients. Further evidence that overexpression of ErbB-2 results in an aggressive tumor type stems from studies showing that elevated ErbB-2 expression is observed in many in situ and invasive human ductal carcinomas but is rarely observed in benign breast disorders such as hyperplasias and dysplasias (Allred et al., 1992, Mansour et al., 1994). Significantly, ErbB-2 overexpression may be useful not only as a prognostic marker but as a predictive marker as well as HER-2 overexpression predicts tamoxifen resistance of the primary tumor (reviewed in Pegram et al., 1998).

Multiple transgenic mouse studies have confirmed a direct role for ErbB-2 in mammary tumorigenesis each with their own level of relevance to the human disease. MMTV-driven overexpression of the oncogene neu or an analogous ERbB-2 transgene engineered to possess a similar activating mutation within the transmembrane domain results in the formation of mammary adenocarcinomas that histologically resemble human comedocarcinomas (Muller et al., 1988; Bouchard et al., 1989; Cardiff and Muller, 1993; Guy et al., 1996; Stocklin et al., 1993).

Although these studies suggest a significant role for ErbB-2 in human breast cancer progression, the lack of a comparable mutation in human breast cancers suggests that the primary mechanism operating in human breast cancer is the overexpression of wild-type ErbB-2 and not its mutational activation (Lemoine et al., 1990; Slamon, 1989; Zoll et al., 1992). Consequently, a more relevant model in which a wild-type neu cDNA was expressed under MMTV control was generated to test the oncogenic potential of the wild-type receptor. These mice develop focal mammary tumors of similar comedocarcinoma-type morphology after an average of 7 months which frequently metastasize to the lung (Guy et al., 1992). Further examination of the ErbB-2 status in these tumors revealed that tumors but not the adjacent normal mammary epithelium carried sporadic mutations in neu which resulted in its constitutive activation (Siegel et al., 1994). These mutations were comprised of multiple in frame deletions, insertions or point mutations in the extracellular domain of neu and promoted the transforming ability of neu through the formation of intermolecular di-sulfide bonds (Siegel and Muller, 1996). To directly explore the importance of these activated forms of Neu, transgenic mice carrying altered Neu receptors were derived. Females from these lines develop multiple mammary tumors that frequently metastasize to the lung with a mean onset between that of the normal and point-activated alleles (Siegel et al., 1999). Interestingly, tumor progression in these strains was associated with elevated levels of tyrosine-phosphorylated Neu and ErbB-3 (Siegel et al., 1999). Consistent with these observations, a survey of primary human breast tumors revealed frequent co-expression of both ErbB-2 and ErbB-3 transcripts (Siegel et al., 1999). These results suggest that ErbB-3 may be the critical heterodimerization partner for Neu in breast cancer progression.

Although the transmembrane point mutation has not been detected in primary human breast cancers overexpressing ErbB-2, studies have detected an alternative splice form in human breast cancers and breast cancer-derived cell lines (Kwong and Hung, 1998; Siegel et al., 1999). Significantly, this splice isoform closely resembles the neu deletion mutants observed in the transgenic line overexpressing wildtype neu with a 16 amino acid deletion in the juxta-transmembrane region of ErbB-2. Like the sporadic neu mutants, this splice isoform is oncogenic due to its capacity to form constitutively active dimers. Conceivably the observed high rate of mutations observed in the transgene in the Neu transgenic strain reflects the fact that the transgene was originally derived from a neu cDNA that is incapable of undergoing alternative splicing. Further studies will be required to assess the significance of this splice isoform in the pathogenesis of human breast cancer. Taken together these transgenic studies strongly implicate the activation of ErbB-2 through receptor dimerization as a critical step in mammary tumorigenesis.

Role of growth factors in breast cancer progression

EGFR ligands

As alluded to above, the expression of EGFR family members plays a critical role in the induction of mammary cancers. Another way to activate members of the EGFR receptor family is through elevated expression of their ligands. Indeed, expression of EGFR ligands TGF- and amphiregulin can be detected in erbB-2 induced mammary tumors (Kenney et al., 1996). Transforming factor alpha is a peptide hormone first isolated from retrovirus-transformed cells and subsequently identified in the conditioned media of breast cancer-derived cell lines and and in invasive ductal carcinomas. TGF- possesses strong homology to EGF and like EGF acts as a ligand for the EGFR. TGF and EGFR expression has been found to coincide with normal mammary epithelial proliferation in vivo (reviewed in Humphreys and Hennighausen, 2000). Early studies with transgenic mice expressing TGF- weakly in the mammary gland showed increased cellular proliferation and fat pad developmental defects (Sandgren et al., 1990). These mice displayed mammary epithelial hyperplasias and dysplasias after multiple pregnancies (Sandgren et al., 1990). Transgenic mice expressing TGF- under MMTV control showed developmental defects and hyperplasias in virgin mice (Matsui et al., 1990). These hyperplasias were observed to progress towards dysplasias with multiple pregnancies and increasing age, with 40% of multiparous animals showing tumors at 16 months of age (Halter et al., 1992). More dramatic effects were achieved by targeting TGF- to the mammary gland with the WAP promoter. Mice expressing TGF- under WAP control showed increased incidence and shorter latency of tumor formation as compared to the MMTV models (Sandgren et al., 1995). The relationship between mammary development and tumorigenesis is highlighted by the observation that these mice also displayed significant delays in mammary gland involution (Sandgren et al., 1995). As these mice still required multiple pregnancies for tumors to form, it is possible that this delay in involution may act to provide an expanded population of proliferating epithelial cells predisposed to transformation (Humphreys and Hennighausen, 2000). However, the long latency observed once again indicates that additional genetic events are necessary for tumor progression in this model. In this regard, it is interesting to note that levels of Cyclin D1 were increased in tumors from the WAP-TGF- mice (Sandgren et al., 1995).

Another class of EGFR ligands that has been implicated in the induction of mammary cancers are members of the heregulin family (Chang et al., 1997). Mammary specific expression of one of the neuregulin isoforms in transgenic mice initially resulted in generation of terminal end bud hyperplasias (Krane and Leder, 1996). However these transgenic mice eventually developed focal mammary tumors and co-expressed constitutively tyrosine phosphorylated ErbB-2 and ErbB-3 receptors. These observations reinforce the importance of ErbB-2 and ErbB-3 heterodimers in the induction of mammary cancers.

Hepatocyte growth factor (HGF)

Another growth factor thought to play an important role in modulating the biological behavior of mammary epithelial cells is the hepatocyte growth factor. HGF and its receptor tyrosine kinase Met are involved in the development of the normal mammary gland (Niemann et al., 1998; Yang et al., 1995). Several studies have also shown overexpression of both Met and HGF in human breast cancers (Lamszus et al., 1997; Tuck et al., 1996; Yamashita et al., 1994). Two studies have generated mice which express an activated form of the Met receptor under control of the metallothionein promoter with varying results (Jeffers et al., 1998; Liang et al., 1996). Mice displayed either hyperplastic nodules progressing to tumors between 11 and 14 months of age (Liang et al., 1996) or induction of metastatic mammary tumors (Jeffers et al., 1998). Consistent with these results, generation of mice expressing HGF under metallothionein control developed tumors of various types, the majority mammary tumors. Together, these studies support a role for HGF in mammary tumor progression.

Fibroblast growth factors

Early studies with MMTV insertion sites revealed frequent proviral activation of members of the fibroblast growth factor (FGF) family including Fgf3 (Dickson et al., 1984; Peters et al., 1983), Fgf4 (Peters et al., 1989) and Fgf8 (Kapoun and Shackleford, 1997; MacArthur et al., 1995). Direct evidence supporting a role for these growth factors derives from studies of a number of transgenic models. Mammary epithelial expression of Fgf3 (int2) results in induction of wide spread mammary epithelial hyperplasias that eventually progress towards full malignancy (Muller et al., 1990). In addition to Fgf3, transgenic mice expressing either Fgf8 (Daphna-Iken et al., 1998) or Fgf7 (Kitsberg and Leder, 1996) under MMTV develop pregnancy-dependent mammary hyperplasias that progress to tumors.

The role of tumor suppressors in mammary tumor progression

Recent transgenic studies have also highlighted the role of LOH in breast cancer progression. Studies in multiple transgenic mice lines including MMTV/v-Ha-ras (Radany et al., 1997), MMTV/wild-type neu (Ritland et al., 1997), MMTV/c-myc (Weaver et al., 1999) and MMTV/activated neu (Cool and Jolicoeur, 1999) have demonstrated that tumors from these animals also show LOH. Significantly, amongst the many areas affected by LOH in these tumors, all showed LOH affecting markers in chromosome 4, an area that contains regions syntenic to human chromosomal regions frequently lost in human breast cancers (1p32-36 and 9p21-22). This further validates these transgenics as models of events involved in human breast cancer. Although it is thought that these LOH mutations affect tumor suppressor genes, many of the loci affected have yet to be identified. However, two types of loss of function mutations that frequently occur in primary human breast cancers are those that affect the known tumor suppressor p53 and the BRCA gene family.

BRCA1 and BRCA2 have been strongly implicated in the pathogenesis of familial or heritable breast cancer. In fact, germline mutations in BRCA1 have been detected in 90% of familial breast/ovarian cancers and almost 50% of familial cases involving breast cancer alone (reviwed in Alberg and Helzlsouer, 1997; Paterson, 1998). The p53 tumor suppressor has also been frequently investigated, both in the context of breast cancer and cancer in general. In fact, p53 is the most commonly altered gene by deletion or mutation in human breast cancer (Elledge and Allred, 1994). The advent of gene targeting in embryonic stem cells has enabled researchers to directly assess the importance of both p53 and the BRCA family in mammary tumorigenesis. One problem with this approach is that these mutations either effect viability or life span of the mouse. For instance, mice homozygous for BRCA1 mutations die early during embryongenesis (Liu et al., 1996). Heterozygotes for BRCA1 are no more informative as they are not pre-disposed to develop mammary tumors (Liu et al., 1996). Similarly, although mice homozygous for null p53 do develop a diverse array of tumors, mammary tumors are rarely observed (Donehower et al., 1992). Studies with the p53 knockouts are further complicated by the formation of extensive lymphomas and thymic tumors that result in the death of the animal at an early age (Donehower et al., 1992). To circumvent these limitations, mice carrying a mutant p53 172^Arg-His under WAP control were generated (Li et al., 1998). These mice display low tumor incidence but exhibit increased tumor incidence as compared to controls in response to chemical carcinogens.

Recent technological advances have also allowed the drawbacks of knockouts, such as embryonic lethality, to be circumvented. Using a powerful modification of the original knockout technique, conditional mutants may be generated which excise the gene of interest in a tissue-specific manner via combination of the Cre-Lox recombination system with the knockin approach. The basis for this system is the ability of the Cre recombinase to excise genetic material flanked by loxP sequences from the genome. This can be achieved at the transgenic level through the generation of mice carrying mammary-targeted Cre recombinase under either the MMTV or WAP promoters (for a review of the Cre-Lox system in mice, see Sauer, 1998). These mice are then crossed with mice which have been engineered through homologous recombination techniques to possess loxP sequences flanking a critical region of the gene of interest.

This advanced technique has allowed the question of the role of BRCA in mammary tumorigenesis to be properly addressed. In the case of the BRCA1 conditional knockout, Cre-mediated excision of exon 11 of brca1 in mouse mammary epithelium initially caused increased apoptosis and abnormal ductal development (Xu et al., 1999). Eventually after a long latency, mammary tumors formed which were further associated with genetic instability characterized by aneuploidy, chromosomal rearrangements or alteration of the p53 locus (Xu et al., 1999), supporting the view that BRCA1 is involved in DNA repair and maintenance of genomic integrity.

Transgenic mouse models of multistep carcinogenesis

One of the major lessons the study of these transgenic models has illustrated is that expression of a single activated oncogene or loss of tumor suppressor gene is not sufficient to convert the mammary epithelial cell to the malignant phenotype. To assess the relative contribution of these genetic events to mammary tumorigenesis, investigators have performed genetic crosses between separate transgenic strains harboring these different genetic lesions. One of the first examples of this approach involved the interbreeding of MMTV/v-Ha-ras strains with the MMTV/c-myc mice (Sinn et al., 1987). In contrast to either parental strain, bi-transgenic mice expressing both c-myc and activated ras developed focal mammary tumors with a dramatically shortened latency period. Although these experiments demonstrated that C-myc and activated Ras could cooperate to accelerate mammary tumor formation, both the focal nature of the tumors and the latency period suggested that additional genetic events were required to transform the mammary epithelium to full malignancy.

Another example of this approach is illustrated by crossing the MMTV/c-myc strain to mice expressing TGF under MMTV control (Amundadottir et al., 1995). These bi-transgenics exhibited mammary tumors by a mean time of 66 days as compared to 298 days in MMTV/c-myc mice. TGF- mice developed no observable tumors in this study. Significantly, the entire mammary gland was malignant in the bi-transgenics, demonstrating that these two signals are likely sufficient for mammary tumorigenesis to proceed. Interestingly, it appears that both transgenes contribute to mammary epithelial proliferation but TGF- also acts to prevent c-myc induced apoptosis in these tumors (Amundadottir et al., 1995).

Given the ability of EGFR family members to heterodimerize to transduce signals within the cell and the proven importance of Neu and the EGFR ligand TGF- in mammary tumorigenesis, crosses were generated between transgenics carrying these transgenes under the transcriptional control of MMTV (Muller et al., 1996). Bi-transgenic mice co-expressing TGF and Neu exhibited accelerated tumor kinetics resulting in multifocal tumors involving the entire mammary gland. In contrast to the parental Neu transgenic strain, mutations could not be detected within the Neu transgene. Conceivably, activation of intrinsic Neu tyrosine kinase activity is achieved in the bi-transgenic tumors by trans-phorylation of Neu by EGFR rather through selection of somatic mutations in the transgene. Interestingly, similar results were obtained in crosses between MMTV/Neu transgenic mice and mice carrying a mutant p53 (p53 172 R-H) under WAP control (Li et al., 1997). Again, bi-transgenic mice developed multifocal mammary tumors with a dramatically shorter latency period without evidence of somatic mutations in the transgene. It is possible that inactivation of the p53 tumor suppressor pathway also obviates the selection of activating mutations in the transgene by activating Neu through an independent pathway. In this regard, it is interesting to note that both p53 ablation and TGF- overexpression are potent anti-apopotic signals.

Consistent with these studies, inactivation of p53 appears to be a critical event in other transgenic model systems. For example, mice deficient in p53 have been crossed to a variety of transgenics including MMTV/Wnt1 (Donehower et al., 1995), MMTV/c-myc (Elson and Leder, 1995), MMTV/v-Ha-ras (Hundley et al., 1997), WAP/IGF-1 (Hadsell et al., 2000) transgenic lines as well as BRCA1 heterozygotes (Cressman et al., 1999) and the conditional BRCA1 knockouts (Xu et al., 1999). With the exception of the MMTV/c-myc and MMTV/v-Ha-ras strains which died due to extensive lymphomas, inactivation of p53 resulted in a dramatic acceleration of mammary tumor progression. Mammary tumors found in these bi-transgenics also displayed increased aneuploidy as compared to those found in the mono-transgenic alone (Donehower et al., 1995; Hadsell et al., 2000). These results suggest that an absence of p53 predisposes mammary epithelial cells to genetic instability and tumor formation in the presence of some other initiating event such as a growth signal. This necessity for a proliferative signal balanced with an anti-apoptotic is a common theme in many transgenic models. For instance, overexpression of the cell survival factor, Bcl-2, in the mammary glands of WAP/Tag mice also accelerates tumor formation (Furth et al., 1999). Taken together these studies suggest that suppression of apoptotic cell death is a critical step in mammary tumorigenesis in these transgenic models.

The role of hormones in mammary tumor progression

Many hormones affect the development of the mammary gland and have been tied to breast development and cancer progression. Currently, mouse models exist to address the roles of three of these hormones; prolactin, estrogen, and progesterone. The effects of these hormones on breast cancer have been studied through the use of knockout technologies directed against either their receptors or the hormone itself. Through crosses to established transgenic strains, the contribution of these hormones to the tumorigenesis phenotype is being examined. Together with other lactogenic hormones, prolactin provides signals that drive the development of the mammary gland. Mice lacking the prolactin receptor show defects in mammary gland development with terminal end buds failing to differentiate into proper lobuloalveoli (Brisken et al., 1999). Mice lacking the prolactin hormone itself show arrested mammary gland development. Crosses between the prolactin knockouts and mice expressing the viral oncogene polyomavirus middle T antigen (PyV mT) under MMTV control show slower induced tumor growth than in mice expressing PyV mT alone (Wennbo and Tornell, 2000).

The roles of the steroid hormones estrogen and progesterone in mouse models of breast cancer have both been recently studied. Mice lacking the estrogen receptor alpha, when crossed to MMTV/Wnt1 mice, show greatly decreased tumor kinetics (50%) while having no effect on the formation of early hyperplasias characteristic of the MMTV/Wnt1 strain. Similarly, progesterone receptor function is necessary for mammary gland maturation in normal mammary development (Brisken et al., 1999) as well as for tumorigenesis in a carcinogen-induced mammary tumor model (Lydon et al., 1999). Interestingly, both progesterone and estrogen have been shown to induce the production of cyclin D1 in murine mammary epithelial cells (Said et al., 1997). These results indicate that hormonal effects may play an important role in mammary cancer progression. Further studies using a wider number of established models should greatly increase our knowledge of their precise roles and effects.

One of the primary limitations to many of these transgenic models discussed is their dependency on strong viral, hormonally sensitive promoters such as WAP and MMTV. Consequently, it is difficult to properly address the interactions between the oncogene-coupled signaling pathways and endocrine hormones that affect mammary gland development. This problem is being addressed by the use of knockins in which transgenics are generated which express oncogenes of interest from their endogenous promoters. This is achieved through the use of a modified homologous recombination approach by which oncogenes of interest are introduced into their endogenous loci.

A combination of both tissue specific recombination and knock-in technologies has enabled researchers in this lab to place the activated neu under the endogenous erbB-2 promoter (Andrechek et al., 2000). To prevent the early embryonic lethality that may have resulted from expression of this cDNA, a silencer cassette containing a neo cassette flanked with loxP sites was placed between the erbB-2 promoter and the activated neu allele. This resulted in expression of the endogenous ErbB-2 until the silencer cassette was excised by mammary epithelial specific expression of the Cre recombinase resulting in mammary epithelial specific expression of the activated ErbB-2 allele (Andrechek et al., 2000). Expression of this allele in the mammary gland resulted in accelerated lobuloalveolar development and tumor formation after a long latency period (Andrechek et al., 2000). Significantly, normal levels of expression of the activated allele from the endogenous erbB-2 promoter were not sufficient for tumorigenesis as all tumors showed amplification (2-22 copies) of the activated neu allele relative to normal mammary tissue (Andrechek et al., 2000). Thus like ErbB-2 positive human tumors, mammary tumorigenesis in this mouse model required amplification of the erbB-2 locus. This model thus holds great promise for relevant studies of the pathogenesis of ErbB-2 positive human breast cancer.

Conclusions

It is evident from the models outlined above that it is important to consider many factors when assessing the applicability of a mouse model for breast cancer research to human breast cancer. The nature of the genetic change, the characteristics of the promoter used to target transgene expression, the status of endogenous signaling pathways, the spectrum of additional mutations that may arise during tumor progression in the transgenic, the number of transgenic lines examined and the reliability of the phenotype amongst them, the transgenic's genetic background and the molecular pathology and histology are all important indicators of the relevancy of the model to the human disease.

While no single genetically engineered mouse can offer a complete model of the wide assortment of human neoplasms found in human breast cancer, it is hoped that these multiple approaches will enable us to develop insights into the complex molecular events involved in tumorigenic progression of the breast. One common theme evident from these studies is the involvement of genes necessary for normal mammary gland development in the progression of this disease. Another emergent theme is the complex, multi-step nature of all stages of breast cancer progression from initial tumor formation to final metastasis. Fortunately, researchers now have many models available to them to study these steps in a controlled and rational manner. Furthermore these models provide the opportunity to study many various aspects of the pathogenesis of this disease, from hormonal effects to responses to chemotherapeutic drugs. It is hoped that through the combined use of these models, and the further development of more relevant models that a deeper understanding of this disease and the generation of new therapeutic agents will result.

Alberg AJ and Helzlsouer KJ. (1997). Curr. Opin. Oncol. 9, 505-511. MEDLINE

Allred DC, Clark GM, Molina R, Tandon AK, Schnitt SJ, Gilchrist KW, Osborne CK, Tormey DC and McGuire WL. (1992). Hum. Pathol. 23, 974-979. MEDLINE

Amundadottir LT, Johnson MD, Merlino G, Smith GH and Dickson RB. (1995). Cell Growth Differ. 6, 737-748. MEDLINE

Andrechek ER, Hardy WR, Siegel PM, Rudnicki MA, Cardiff RD and Muller WJ. (2000). Proc. Natl. Acad. Sci. USA 97, 3444-3449. MEDLINE

Andrulis IL, Bull SB, Blackstein ME, Sutherland D, Mak C, Sidlofsky S, Pritzker KP, Hartwick RW, Hanna W, Lickley L, Wilkinson R, Qizilbash A, Ambus U, Lipa M, Weizel H, Katz A, Baida M, Mariz S, Stoik G, Dacamara P, Strongitharm D, Geddie W and McCready D. (1998). J. Clin. Oncol. 16, 1340-1349. MEDLINE

Bargmann CI, Hung MC and Weinberg RA. (1986a). Cell 45, 649-657. MEDLINE

Bargmann CI, Hung MC and Weinberg RA. (1986b). Nature 319, 226-230. MEDLINE

Berns EM, Foekens JA, van Putten WL, van Staveren IL, Portengen H, de Koning WC and Klijn JG. (1992). J. Steroid Biochem. Mol. Biol. 43, 13-19. MEDLINE

Bieche I and Lidereau R. (1995). Genes Chromosom. Cancer 14, 227-251. MEDLINE

Bouchard L, Lamarre L, Tremblay PJ and Jolicoeur P. (1989). Cell 57, 931-936. MEDLINE

Brisken C, Kaur S, Chavarria TE, Binart N, Sutherland RL, Weinberg RA, Kelly PA and Ormandy CJ. (1999). Dev. Biol. 210, 96-106. MEDLINE

Brison O. (1993). Biochim. Biophys. Acta 1155, 25-41. MEDLINE

Callahan R, Cropp CS, Merlo GR, Liscia DS, Cappa AP and Lidereau R. (1992). Cancer 69, 1582-1588. MEDLINE

Cardiff RD and Muller WJ. (1993). Cancer Surv. 16, 97-113. MEDLINE

Chang H, Riese II DJ, Gilbert W, Stern DF and McMahan UJ. (1997). Nature 387, 509-512. MEDLINE

Cool M and Jolicoeur P. (1999). Cancer Res. 59, 2438-2444. MEDLINE

Cressman VL, Backlund DC, Hicks EM, Gowen LC, Godfrey V and Koller BH. (1999). Cell Growth Differ. 10, 1-10. MEDLINE

Dankort DL and Muller WJ. (2000). Oncogene 19, 1038-1044. MEDLINE

Daphna-Iken D, Shanker DB, Lawshe A, Ornitz DM, Shackleford GM and MacArthur CA. (1998). Oncogene 17, 2711-2717. MEDLINE

Dickson C, Smith R, Brookes S and Peters G. (1984). Cell 37, 529-536. MEDLINE

Donehower LA, Godley LA, Aldaz CM, Pyle R, Shi YP, Pinkel D, Gray J, Bradley A, Medina D and Varmus HE. (1995). Genes Dev. 9, 882-895. MEDLINE

Donehower LA, Harvey M, Slagle BL, McArthur MJ, Montgomery Jr CA, Butel JS and Bradley A. (1992). Nature 356, 215-221. MEDLINE

Dupont WS and Page DL. (1985). N. Engl. J. Med. 312, 146-151. MEDLINE

Elledge RM and Allred DC. (1994). Breast Cancer Res. Treat. 32, 39-47. MEDLINE

Elson A and Leder P. (1995). J. Biol. Chem. 270, 26116-26122. MEDLINE

Escot C, Theillet C, Lidereau R, Spyratos F, Champeme MH, Gest J and Callahan R. (1986). Proc. Natl. Acad. Sci. USA 83, 4834-4838. MEDLINE

Fantl V, Edwards PA, Steel JH, Vonderhaar BK and Dickson C. (1999). Dev. Biol. 212, 1-11. MEDLINE

Furth PA, Bar-Peled U, Li M, Lewis A, Laucirica R, Jager R, Weiher H and Russell RG. (1999). Oncogene 18, 6589-6596. MEDLINE

Garcia JM, Silva JM, Dominguez G, Gonzalez R, Navarro A, Carretero L, Provencio M, Espana P and Bonilla F. (1999). Breast Cancer Res. Treat. 57, 237-243. MEDLINE

Guy CT, Cardiff RD and Muller WJ. (1996). J. Biol. Chem. 271, 7673-7678. MEDLINE

Guy CT, Webster MA, Schaller M, Parsons TJ, Cardiff RD and Muller WJ. (1992). Proc. Natl. Acad. Sci. USA 89, 10578-10582. MEDLINE

Hadsell DL, Murphy KL, Bonnette SG, Reece N, Laucirica R and Rosen JM. (2000). Oncogene 19, 889-898. MEDLINE

Halter SA, Dempsey P, Matsui Y, Stokes MK, Graves-Deal R, Hogan BL and Coffey RJ. (1992). Am. J. Pathol. 140, 1131-1146. MEDLINE

Humphreys RC and Hennighausen L. (2000). Oncogene 19, 1085-1091. MEDLINE

Hundley JE, Koester SK, Troyer DA, Hilsenbeck SG, Subler MA and Windle JJ. (1997). Mol. Cell. Biol. 17, 723-731. MEDLINE

Hynes NE and Stern DF. (1994). Biochim. Biophys. Acta 1198, 165-184. MEDLINE

Jeffers M, Fiscella M, Webb CP, Anver M, Koochekpour S and Vande Woude GF. (1998). Proc. Natl. Acad. Sci. USA 95, 14417-14422. MEDLINE

Kapoun AM and Shackleford GM. (1997). Oncogene 14, 2985-2989. MEDLINE

Kenney NJ, Smith GH, Maroulakou IG, Green JH, Muller WJ, Callahan R, Salomon DS and Dickson RB. (1996). Mol. Carcinog. 15, 44-56. MEDLINE

Kitsberg DI and Leder P. (1996). Oncogene 13, 2507-2515. MEDLINE

Krane IM and Leder P. (1996). Oncogene 12, 1781-1788. MEDLINE

Kwong KY and Hung MC. (1998). Mol. Carcinog 23, 62-68. MEDLINE

Lamszus K, Jin L, Fuchs A, Shi E, Chowdhury S, Yao Y, Polverini PJ, Laterra J, Goldberg ID and Rosen EM. (1997). Lab. Invest. 76, 339-353. MEDLINE

Leder A, Pattengale PK, Kuo A, Stewart TA and Leder P. (1986). Cell 45, 485-495. MEDLINE

Lee RJ, Albanese C, Fu M, D'Amico M, Lin B, Watanabe G, Haines III GK, Siegel PM, Hung MC, Yarden Y, Horowitz JM, Muller WJ and Pestell RG. (2000). Mol. Cell. Biol. 20, 672-683. MEDLINE

Lee RJ, Albanese C, Stenger RJ, Watanabe G, Inghirami G, Haines III GK, Webster M, Muller WJ, Brugge JS, Davis RJ and Pestell RG. (1999). J. Biol. Chem. 274, 7341-7350. Article MEDLINE

Lemoine NR, Staddon S, Dickson C, Barnes DM and Gullick WJ. (1990). Oncogene 5, 237-239. MEDLINE

Li B, Murphy KL, Laucirica R, Kittrell F, Medina D and Rosen JM. (1998). Oncogene 16, 997-1007. MEDLINE

Li B, Rosen JM, McMenamin-Balano J, Muller WJ and Perkins AS. (1997). Mol. Cell. Biol. 17, 3155-3163. MEDLINE

Liang TJ, Reid AE, Xavier R, Cardiff RD and Wang TC. (1996). J. Clin. Invest. 97, 2872-2877. MEDLINE

Lidereau R, Mathieu-Mahul D, Escot C, Theillet C, Champeme MH, Cole S, Mauchauffe M, Ali I, Amione J, Callahan R and Larsen C-J. (1988). Biochimie 70, 951-959. MEDLINE

Liu CY, Flesken-Nikitin A, Li S, Zeng Y and Lee WH. (1996). Genes Dev. 10, 1835-1843. MEDLINE

Lydon JP, Ge G, Kittrell FS, Medina D and O'Malley BW. (1999). Cancer Res. 59, 4276-4284. MEDLINE

MacArthur CA, Shankar DB and Shackleford GM. (1995). J. Virol. 69, 2501-2507. MEDLINE

Mansour EG, Ravdin PM and Dressler L. (1994). Cancer 74, 381-400. MEDLINE

Matsui Y, Halter SA, Holt JT, Hogan BL and Coffey RJ. (1990). Cell 61, 1147-1155. MEDLINE

Muller WJ, Arteaga CL, Muthuswamy SK, Siegel PM, Webster MA, Cardiff RD, Meise KS, Li F, Halter SA and Coffey RJ. (1996). Mol. Cell. Biol. 16, 5726-5736. MEDLINE

Muller WJ, Sinn E, Pattengale PK, Wallace R and Leder P. (1988). Cell 54, 105-115. MEDLINE

Muller WJ, Sinn E, Pattengale PK, Wallace R and Leder P. (1990). EMBO J. 9, 907-913. MEDLINE

Niemann C, Brinkmann V, Spitzer E, Hartmann G, Sachs M, Naundorf H and Birchmeier W. (1998). J. Cell. Biol. 143, 533-545. MEDLINE

Olayioye MA, Neve RM, Lane HA and Hynes NE. (2000). EMBO J. 19, 3159-3167. Article MEDLINE

Paterson JW. (1998). Dis. Markers. 13, 261-274. MEDLINE

Pattengale PK, Stewart TA, Leder A, Sinn E, Muller W, Tepler I, Schmidt E and Leder P. (1989). Am. J. Pathol. 135, 39-61. MEDLINE

Pegram MD, Pauletti G and Slamon DJ. (1998). Breast Cancer Res. Treat. 52, 65-77. MEDLINE

Peters G, Brookes S, Smith R and Dickson C. (1983). Cell 33, 369-377. MEDLINE

Peters G, Brookes S, Smith R, Placzek M and Dickson C. (1989). Proc. Natl. Acad. Sci. USA 86, 5678-5682. MEDLINE

Radany EH, Hong K, Kesharvarzi S, Lander ES and Bishop JM. (1997). Proc. Natl. Acad. Sci. USA 94, 8664-8669. MEDLINE

Radeva G, Petrocelli T, Behrend E, Leung-Hagesteijn C, Filmus J, Slingerland J and Dedhar S. (1997). J. Biol. Chem. 272, 13937-13944. MEDLINE

Ravdin PM and Chamness GC. (1995). Gene 159, 19-27. MEDLINE

Ritland SR, Rowse GJ, Chang Y and Gendler SJ. (1997). Cancer Res. 57, 3520-3525. MEDLINE

Said TK, Conneely OM, Medina D, O'Malley BW and Lydon JP. (1997). Endocrinology 138, 3933-3939. MEDLINE

Sandgren EP, Luetteke NC, Palmiter RD, Brinster RL and Lee DC. (1990). Cell 61, 1121-1135. MEDLINE

Sandgren EP, Schroeder JA, Qui TH, Palmiter RD, Brinster RL and Lee DC. (1995). Cancer Res. 55, 3915-3927. MEDLINE

Sauer B. (1998). Methods 14, 381-392. Article MEDLINE

Schechter AL, Stern DF, Vaidyanathan L, Decker SJ, Drebin JA, Greene MI and Weinberg RA. (1984). Nature 312, 513-516. MEDLINE

Schoenenberger CA, Andres AC, Groner B, van der Valk M, LeMeur M and Gerlinger P. (1988). EMBO J. 7, 169-175. MEDLINE

Siegel PM, Dankort DL, Hardy WR and Muller WJ. (1994). Mol. Cell. Biol. 14, 7068-7077. MEDLINE

Siegel PM and Muller WJ. (1996). Proc. Natl. Acad. Sci. USA 93, 8878-8883. MEDLINE

Siegel PM, Ryan ED, Cardiff RD and Muller WJ. (1999). EMBO J. 18, 2149-2164. Article MEDLINE

Sinn E, Muller W, Pattengale P, Tepler I, Wallace R and Leder P. (1987). Cell 49, 465-475. MEDLINE

Slamon DJ, Clark GM, Wong SG, Levin WJ, Ullrich A and McGuire WL. (1987). Science 235, 177-182. MEDLINE

Slamon DJ, Godolphin W, Jones LA, Holt JA, Wong SG, Keith DE, Levin WJ, Stuart SG, Udove J, Ullrich A and Press MF. (1989). Science 244, 707-712. MEDLINE

Stern DF, Heffernan PA and Weinberg RA. (1986). Mol. Cell. Biol. 6, 1729-1740. MEDLINE

Stewart TA, Pattengale PK and Leder P. (1984). Cell 38, 627-637. MEDLINE

Stocklin E, Botteri F and Groner B. (1993). J. Cell. Biol. 122, 199-208. MEDLINE

Tuck AB, Park M, Sterns EE, Boag A and Elliott BE. (1996). Am. J. Pathol. 148, 225-232. MEDLINE

Wang TC, Cardiff RD, Zukerberg L, Lees E, Arnold A and Schmidt EV. (1994). Nature 369, 669-671. MEDLINE

Weaver ZA, McCormack SJ, Liyanage M, du Manoir S, Coleman A, Schrock E, Dickson RB and Ried T. (1999). Genes Chromosom. Cancer 25, 251-260. MEDLINE

Wennbo H and Tornell J. (2000). Oncogene 19, 1072-1076. MEDLINE

Xie Y, Li K and Hung MC. (1995). Oncogene 10, 2409-2413. MEDLINE

Xu X, Wagner KU, Larson D, Weaver Z, Li C, Ried T, Hennighausen L, Wynshaw-Boris A and Deng CX. (1999). Nat. Genet. 22, 37-43. Article MEDLINE

Yamashita J, Ogawa M, Yamashita S, Nomura K, Kuramoto M, Saishoji T and Shin S. (1994). Cancer Res. 54, 1630-1633. MEDLINE

Yang Y, Spitzer E, Meyer D, Sachs M, Niemann C, Hartmann G, Weidner KM, Birchmeier C and Birchmeier W. (1995). J. Cell. Biol. 131, 215-226. MEDLINE

Zoll B, Kynast B, Corell B, Marx D, Fischer G and Schauer A. (1992). J. Cancer Res. Clin. Oncol. 118, 468-473. MEDLINE