Mouse models for multistep tumorigenesis

:evlew :evie~ |

82

TRENDS in Cell Biology Vo1.11 No.ll November 2001

Mouse models for multistep tumorigenesis Xuemei Wu and Pier Paolo Pandolfi The mouse is an ideal model system for studying the molecular mechanisms underlying the pathogenesis of human cancer. The generation of transgenic and gene-knockout mice has been instrumental in determining the role of major determinants in this process, such as oncogenes and tumor-suppressor genes. In the past few years, modeling cancer in the mouse has increased in its complexity, allowing in vivo dissection of the fundamental concepts underlying cooperative oncogenesis in various tumor types. In this review, we discuss how this transition has been facilitated, providing relevant examples. We also review how, in the post-genome era, novel methodologies will further accelerate the study of multi-step tumorigenesis in the mouse.

A TRENDS Guide to

Cancer Biology

]

It is widely accepted that tumorigenesis is a multistep

are crucial to the development of distinct forms of cancer.

process that involves a series of genetic and epigenetic

Mutation or loss of these genes has highly penetrant pheuo-

alterations, such as activation of dominantly acting onco-

typic consequences, and germline mutations of these genes

genes and inactivation of tumor-suppressor genes. These

predispose the carriers to tumor development. However,

mutations accumulate in the cells and change their be-

the study of cancer genetics in humans has certain limi-

havior from normal growth to unrestrained growth and

tations. For the sporadic forms of cancer in particular,

eventually lead to invasion into surrounding tissue a n d / o r

tumor susceptibility in each individual is determined not

metastasis. This multiple-stage process is reflected by a

only by these key players but also by several modifier genes

range of observations, including clinical, epidemiological

whose activity might also have a key influence on cancer

and laboratory experiments. The process can be extremely

development. The diversity in genetic background and

complex. For example, during the development of col-

the difficulty in tumor and tissue procurement make it

orectal carcinoma, it has been proposed that at least seven

difficult to study tumorigenesis in humans for the pur-

sequential genetic alterations occur, including mutation of

pose of identifying weak tumor determinants. By con-

the adenomatous polyposis coli (APC) tumor-suppressor

trast, the mouse offers several distinct advantages and has

gene, mutation of Ki-ras and loss of fnnctional p53 gene / .

proved to be a valuable model system for the study of

These sequential mutations of key growth-regulatory genes

multistep tumorigenesis at two distinct levels: (i) identi-

Xuemei Wu Pier Paolo Pandolfi*

in the somatic cells and their progeny are generally re-

fication of novel genes involved in cancer pathogenesis;

garded as 'multiple-hits' in a broader interpretation of the

and (ii) experimental in viv0 assessment of the role of

Molecular Biology Program,

original Knudson 'two-hits' theory, which was based on

genetic hits in multistep tumorigenesis.

Dept of Pathology,

a statistical analysis of the relationship between age and

Memorial Sloan-Kettering

incidence of retinoblastoma 2.

Mice with a controlled genetic background can be raised easily in vast colonies, facilitating the analysis of

Cancer Center, 1275 York

Over the past decade, our genetic understanding of

cancer-relevant genes in tumor development, particularly

Avenue, New York,

tumor development has been greatly enhanced by the

the low-penetrance genes. The large number of inbred

NY 10021, USA.

direct study of h u m a n cancer. Studies on rare forms of

mouse strains that show enormous variation in their sus-

*e-mail: p-pandolfi@

familial inherited cancer syndromes have led to success-

ceptibility to different types of tumors provide a good

ski.mskcc.org

ful identification of many tumor-suppressor genes that

resource for various tumor models. Most importantly, the

http://tcb.trends.com 0962-8924/01/$-seefrontmatter©2001ElsevierScienceLtd.AII rights reserved. PII:S0962-8924(01)02127-4

eviev

Epidermal

I

L

cells

DMBA

TRENDS in Cell Biology Vo1.11 No.ll November 2001

T PA

Initiated cells

~Initiation H-ras mutation

Benign papilloma

J"

Squamous carcinoma

~"

Promotion

Progression

Progression

Cyclin D1 overexpression

Loss of p53

E-cadherin TGF-13

Spindle carcinoma

TRENDS in Ceil Biology

similarity between the mouse and human genomes, and

the regulation of cell growth and differentiation 3. This

particularly between the histological stages and genetic

promotes the selective clonal expansion of initiated epi-

pathways underlying tumor development in h u m a n and

dermal cells and leads to the formation of multiple squa-

mouse, make the mouse an ideal model system for the

mous papillomas. The progression stage is generally a

study of human cancer pathogenesis. Furthermore, it is

spontaneous process. It is characterized by a high level of

now possible to manipulate the mouse genome either by

genetic instability, particularly chromosomal alterations

overexpressing a dominantly acting oncogene in trans-

resulting in elevated expression of genes encoding Ha-Ras

genic mice or by deleting a tumor-suppressor gene from

and cyclin D 1, as well as loss of expression of the tumor-

the germline of knockout mutants in a tissue- and cell-

suppressor p53 (Ref. 7). These changes confer to papilloma

type-specific manner. Many of these transgenic and knock-

cells a further growth advantage, leading to the conversion

out mutants now provide faithful model systems for most

of the papillomatous lesion into malignant squamous car-

human cancers and will also permit analysis of the roles

cinoma. A further advanced stage in mouse skin carcino-

of strong and weak determinants and establish a link

genesis is the progression of squamous carcinoma to

between individual gene mutations and specific cellular

spindle carcinoma. This stage is associated with the changes

changes that lead to tumor development.

$3

Figure 1. Chemicallyinduced multistep skin carcinogenesis in the mouse Epidermal cells are initiated by 7,12-dimethyl-benzanthracene (DMBA) by causing a Ha-rasactivating mutation. Treatment with 12-O-tetradecanoylphorbol-13-acetate (TPA) results in upregulationof oyclin D1 gene expressionand promotes the formation of papilloma, Loss of functional p53 gene acceleratesthe progression of benign papillomato malignant carcinoma. Finally,transforming growth factor beta (TGF-~}) enhances the conversion of squamous carcinomato metastatic spindle cell carcinoma, accompaniedby downregulation and delocalizationof E-cadherin.

of epithelial differentiation markers, such as decreased expression of E cadherin, keratins K1 and K 10, and increased

Molecular carcinogenesis in the mouse:

expression of 0~6[~4 integrin 8.

the skin as a paradigmatic model system for the study of multistep tumorigenesis

lishes a l i n k between genetic pathways and histological

One of the best-established multistage tumorigenesis mod-

stages of tumor development but also serves as a model

The mouse skin carcinogenesis model not only estab-

els is mouse skin carcinogenesis. The sequential application

system to study tile functions of many other genes in the

of a carcinogen, followed by treatment with a non-carcino-

modulation of the tumorigenic process. For instance,

genic promoter can effectively induce skin tumors. Studies

DMBA-TPA treatment of cyclin D1-deficient mice re-

of chemically induced mouse skin tumors established the

suited in a decrease in skin tumor development, indicating

three stages of tumor development: initiation, promotion

that cyclin D1 is an important target of the Ha-Ras path-

and progression 3,4 (Fig. 1). Initiation is generally accom-

way in skin tumorigenesis 9. By contrast, pS3-deficient

plished by topical application of a single subcarcinogenic

mice displayed an accelerated progression of papilloma to

dose of skin carcinogen, such as 7,12-dimethyl-benzan-

carcinoma upon treatment, suggesting that p53 plays an

thracene (DMBA). The treatment causes irreversible DNA

important role in the tumor-progression phase 10. The in-

damage, resulting in mutation of the Ha-ras oncogene in

volvement of the c-los gene in malignant conversion in the

epidermal cells. Subsequent studies demonstrated that the

skin is corroborated by studies on the c-f0s knockout mice 11.

type of mutation of Ha-ras was dependent on the chemical

The promyelocytic leukemia (PML) tumor-suppressor

initiator, suggesting that the initiator has a direct effect on

gene has also been found to antagonize the initiation

the Ha-r0s gene s,6. The promotion is achieved by repeated

phase of skin tumorigenesis, resulting in increased tumor

application of skin promoters, most commonly the phor-

development in PML-null mice lz. This model system was

bo] esters, such as 12-O-tetradecanoyl-phorbol- 13-acetate

also used to elucidate the dual functions exerted by the

(TPA). Although most tumor promoters do not bind co-

gene encoding TGT-~ in skin tumorigenesis 13,14. TGF-~

valently to DNA and are not mutagenic per se, they cause a

transgenic mice were more resiitant to the induction of

range of cellular and biochemical changes, such as elevated

benign papilloma, but the malignant conversion rate was

activation of protein kinase C and increased expression

greatly increased, and most of the carcinomas progressed

of transforming growth factor alpha (TGF-Ct), c-Jun and

further to invasive spindle carcinomas. More recently, the

c-Fos, which have an obvious and direct connection with

epidermal growth factor (EGF) receptor has also been

http://tcb.trends.com

A TRENDSGuide to CancerBiology

$4

I~t~'ll~'A'

TRENDS in Cell Biology Vo1.11 No.ll

November 2001

shown to play an essential role in SOS-dependent skin

to induce leukemia and that additional genetic events are

tumor development. In fact, SOS transgenic mice bearing

probably required. Furthermore, transgenic mice express-

a disrupted gene for the EGF receptor do not develop skin

ing another APL-specific fusion gene, PLZF-RARO~, dis-

tumors ~s. Finally, the mouse skin carcinogenesis model

play a very different phenotype z2. Leukemia that develops

has been of great use in defining the concept of haplo-

in PLZF-RARO~ transgenic mice more closely resembles

insufficiency in tumor suppression. For instance, skin

h u m a n chronic myeloid leukemia (CML). This leukemia

carcinogenesis studies show that p27 ~P~ exerts the tumor-

is characterized by an expansion of myeloid cells that can

suppression function in a dose-dependent manner. The

terminally differentiate, whereas classic APL is character-

nullizygous and p27 kiP1heterozygous mice are both pre-

ized by a distinctive block in myeloid differentiation at

disposed to tumorigenesis, and the heterozygous mutants

the promyelocytic stage. Once again, leukemia in PLZF-

do not lose the wild-type allele 16. Further studies in other

RARer transgenic mice develops after a long latency.Thus,

mutant mouse models suggest that the p27 ~PI gene is

like PML-RARO~, PLZF-RAROt does not seem to be sufficient

not an exception, and that many other tumor suppressors

for APL leukemogenesis, and additional genetic alterations

are haplo-insufficient in their function, such as PML

have to occur in order to cause full-blown leukemia. The

(see below) 17.

phenotypic difference displayed by the two transgenic models indicates that PML-RAR0~ and PLZF-RARCt differ

Interspecific crosses of knockout and transgenic mutants for the study of multistep tumorigenesis: the APL paradigm

in their oncogenic activities and represent distinct RAR~

One of the most effective approaches for the study of

genetic events are required to block the differentiation at

cooperative tumorigenesis in the mouse is to cross a spe-

the promyelocytic stage.

mutants in spite of their identity in the RAR0t portion, suggesting that, in APL harboring PLZF-RAR(I, other

cific tumor-prone transgenic or knockout line with other

As the APL-specific chromosomal translocations are

transgenic or knockout mutants, thus allowing the assess-

reciprocal, RARO~-PML and RARO~-PLZF transgenic mice

ment of whether these additional mutations can acceler-

have also been generated. Interestingly, although RAR(~-

ate tumor development. The selection of'additional' gene

PML transgenic mice do not develop leukemia, RAROt-PML

mutations can be divided into two categories: (i) genes

acts as a tumor modifier, whose presence significantly

that are involved in the molecular biochemical pathway

increases the penetrance of PML-RARO~ and accelerates

that is thought to control the development of a tumor of

leukemia development (Fig. 2). Double transgenic mice

a specific histological origin; and (ii) genes that maintain

coexpressing PML-RARO~ and RAROt-PML reciprocal fusion

genome integrity. The first category includes genes that

genes develop early onset APL-like leukemia z3. The PML

regulate cell proliferation, differentiation or apoptosis.

gene also behaves like a tumor-suppressive modifier of

Genes that fall into the second category are those involved

PML-RARO~-induced leukemogenesis. Inactivation of both,

in DNA repair and the control of genome stability.

or even one, PML allele dramatically increases the inci-

A compelling example of the power of transgenic and

dence of APL-like leukemia in PML-/-/PML-RARot or

knockout methodology in the study of muhistep carcino-

PML+/-/PML-RARo~ double mutants 17.The data also dem-

genesis is provided by the attempt to dissect acute pro-

onstrate that PML is haplo-insufficient in antagonizing the

myelocytic leukemia (APL) pathogenesis in the mouse. APL

function of PML-RAR0~, In this respect, it must be em-

is associated with reciprocal chromosomal translocations,

phasized that, in the APL blast, the dose of the two genes

which always involve the retinoic acid receptor (RAR)O~

involved in the translocation is, in fact, reduced to hetero-

gene variably translocated and fused to five distinct genes:

zygosity, because one allele is involved in the translocation.

PML, promyelocytic leukemia zinc finger (PLZF), nuclear

However, coexpression of PLZF-RARO~and RAR0t-PLZF

mitotic apparatus protein (NuMA), nuclear phosmin (NPM)

in double transgenic mice resulted in a completely different

and STAT5b (Re£ 18). As a consequence of these translo-

outcome. Rather than accelerating leukemia develop-

cations and in view of their reciprocity, two fusion genes

ment, the presence of RARO~-PLZF transformed the bio-

are generated, encoding two distinct fusion proteins, that

logical features of the disease from a CML-like leukemia in

coexist in the APL blast.

single TM to an APL-like leukemia in double TM (Fig. 2).

Transgenic mice expressing PML-RAR~ under the con-

In contrast, leukemia onset was unaffected. Thus, RAR~-

trol of a promyelocytic-specific promoter develop a form

PLZF did not function as a classic minor modifier, but rather

of leukemia that closely resembles h u m a n APL19 ~1. How-

as a tumor 'metamorphoser '2~. Similarly, loss of PLZF func-

ever, the incidence of leukemia development in these mice

tion also metamorphoses, once again in a dose-dependent

is low and the latency is long. These results suggest that

manner, the CML-like leukemia in PLZF-RAROt mice into

PML-RAROt translocation is necessary but not sufficient

APL-like leukemia in PLZF-/-/PLZF-RAROt mice 24,

http://tcb.trends.com

A TRENDSGuide to CancerBiology

]i~'J[~,,,

TRENDS in Cell B i o l o g y

Vo1.11No.ll November2001

S~,

The leukemia incidence and phenotype in PML-RARo~ or PLZF-RARo~ transgenic mice in an RARe-null back-

,a,

Ear,yonse:

ground has not yet been reported. It will be interesting to see whether partial or complete loss of RAR(~ function

APL

also plays a role in leukemogenesis.

X

Taken together, these studies in transgenic and knock-

APL

out models demonstrate that the two translocation prod-

~

Earlyonset > ~

ucts, as well as the inactivation of the genes involved in

APL

these translocations, are essential for the development of a leukemia with APL features. However, given the fact that the reciprocal translocation product is not expressed in 100% of

~

human APL patients, and that the earliest leukemia onset observed in double transgenic mutants is still approxi-

Phenotypicchange (

(b)

mately six months, it is possible that further genetic events are required. Several approaches could be applied to such APL mouse models in the search for these additional

~

~

>

APL x ~

Phenotypicchange ~, ( ~ ~ ~

genetic events, as we will discuss below. These additional hits could participate in the pathogenesis of h u m a n APL

APL

and also in the pathogenesis of other human cancers.

TRENDS in Cell Biology

I d e n t i f i c a t i o n of t u m o r - m o d i f i e r

g e n e s in t h e

mouse

F i g u r e 2. E f f e c t o f a d d i t i o n a l g e n e t i c hits in m o u s e m o d e l s o f A P L

Mouse models of cancer not only enable the determi-

(a) Crossesbetweenpremyelocyticleukemia (PML)-retinoic acid receptor c~(RARe)transgenic mice andRARc(-PML transgenic mice or PML-deficient mice result in acceleration of leukemia onset and alsoa higher incidence of leukemia.Thus,RAR~-PMLandPMLappear to function as tumordevelopmentmodifiers.(b) Bycontrast, crosses between promyelocyticleukemiazincfinger (PLZF)-RAR~transgenic mice and RARcc-Pt.ZFtransgenicmice or PLZF-defieient mice metamorphose the chronic myeloid leukemia(CML)-Iikephenotypeobservedin PLZF-RAR(x transgenicmiceintoacute promyelocyticleukemia(APL)-Iikeleukemia.Thus,RAR~-PI7F andpITF functionin APLastumor phenotypemodifiers or tumormetamorphosers,

nation of the role of specific genetic events in multistep tumorigenesis, as discussed above, but also help in identifying novel genes involved in this process, particularly tumor modifiers, which are notoriously difficult to discover by performing population-based genetic studies. A successful example of the potential of this approach is represented by the identification of Mom- 1. This tumor modifier was identified in an attempt to characterize

The molecular identification of Mom-1 represents a

genetic factors that influence tumor incidence in the Min

paradigm and exemplifies the power of mouse genetics

mouse, a strain that harbours a point mutation in the APe

for the analysis and identification of tumor modifiers.

gene and that is prone to develop multiple adenomatous

So far, the loci of many more tumor modifiers have been

polyposis in the small intestine zs. The incidence of poly-

identified, and their influences on tumor development

posis in the Min mouse varies considerably depending on

extend from affecting the size and number of the

the strain background. In a C57BL/6 background, Min

tumors to the latency and even the morphology of the

mice develop a large n u m b e r of tumors, whereas an AKR

tumors 17,24,28. Moreover, strong genetic interactions

background is resistant to the effects of the APC mutation.

between tumor modifiers have also been implicated in

Senetic mapping linked a locus on chromosome 4 to the

several linkage studies in various mouse models of h u m a n cancers z9,3°.

resistance. As this locus could change tumor susceptibility Lriggered by the Mill allele, it was named 'modifier of Min', )r Morn 1. Subsequent studies led to the identification of

N e w a p p r o a c h e s to i d e n t i f y a n d v a l i d a t e t h e

he gene encoding secreted phospholipase 2a (Pla2g2a)

role o f a d d i t i o n a l g e n e t i c e v e n t s in m o u s e

ts the gene encoding Morn-1 z6. Pla2g2a is highly ex-

models of human cancer

)ressed in the resistant strains, but weakly expressed in

SKY and CGH microarray in the mouse

he sensitive strains. Furthermore, mutation analysis in

Novel methodologies are allowing for more comprehen-

he sensitive and resistant strains identified a thymidine

sive study of the multistep process towards full-blown

nsertion that results in aberrant Pla2g2a splicing in the

transformation and metastasis in animal model systems.

~nsitive strain. Reintroduction of a functional Pla2g2

Additional genetic events occurring throughout cancer

ene into a sensitive strain renders the mice resistant to

evolution can now be studied in mouse models at various

amor development 27.

different levels with greater sensitivity.

http://tcb.trends.com

A TRENDS Guide to Cancer Biology

$6

[leview evlew I

TRENDS in Cell Biology

Vo1.11 No.ll November 2001

• At the chromosomal level by spectral karyotypmg (SKY). SKY has been developed to detect particular

Retroviral insertional mutagenesis has been successfully

chromosome translocations, regional chromosomal

used to identify various cancer-associated genes in the

amplifications and deletions 3~'32. SKY is based on

mouse. Random insertion of proviral sequence into the

hybridization of metaphase chromosomes with chro-

mouse genome can in fact alter gene expression. Some of

mosome-specific probes differentially labeled with a

these insertions can result in overexpression of proto-

combination of fluorophores. Utilizing a combination

oncogenes or silencing of tumor-suppressor genes, thus

of conventional fluorescent light microscopy, Fourier-

leading to acceleration in tumor development. As an

transform spectroscopy and CCD-camera, fluorescence-

example, the proviral insertion resulting in elevated

emission spectrum of each metaphase chromosome is

expression of a truncated N0tcM gene led to accelerated

measured simultaneously and their spectral images are

development of%cell tumors in c-rage transgenic mice *s.

visualized.

More recently, a newly improved method has been devel-

• At the gene locus level. Many methods have been

oped that uses inverse polymerase chain reaction (PCR)

applied, such as standard fluorescent in situ hybridiz

to allow quick identification of the tag sequences of

ation (FISH), comparative genornic hybridization

affected genes .6. The availability of the mouse genome

(CGH) 3a, genetic linkage and loss of heterozygosity

sequence and expressed sequence-tagged databases will

(LOH) analysis, whereas the development of oligo-

further facilitate the detection of cax~cer-associated genes.

nucleotide and eDNA microarrays allow the detection of changes in gene expression throughout tumor progression34 a6.

t h e s t u d y o f c o o p e r a t b J e o n c o g e n e s i s in th~

The combination of all of these approaches makes them extremely powerful for the precise location of

A novel retroviral gene delivery system has been proven

genetic changes during muhistep tumorigenesis. For

to be extremely useful for the study of cooperative on-

example, CGH and FISH can refine the region of trans-

cogenesis in the mouse 47. This system is based on the

location breakpoint after SKY narrows the search to a

utilization of the receptor of subgroup A avian leucosis

particular chromosomal region. Such combination has

virus, TVA. Mice do not harbour a TVA gene homolog and

allowed the identification of a recurrent translocation

are therefore resistant to avian virus infection. However,

on chromosome 14 in lymphomas in the ataxia-telangi-

in transgenic mice the expression of TVA receptor gene

ectasia mutated

under the control of tissue- or cell-specific promoter ren-

(ATM)-deficient m o u s e 37. Similar

analysis in PML-RAR~ and RARot-PML transgenic mice

der them susceptible to avian retroviral infection, allow-

revealed that deletion of chromosome 2 is a frequent

ing genes of interest to be introduced somatically in a

event in leukemia development 38. Furthermore, the

tissue- and cell type-specific manner 48. In practice, the

combination of CGH and microarray technology (CGH

gene of interest is cloned into an avian Rous sarcoma

microarray) makes it possible to analyse the whole

virus-derived replication-competent cloning vector

genome for DNA copy-number variation at higher

(RCAS). The proviral vector is transfected into chicken

resolution 39,*°. Conventional CGH, which was devel-

fibroblast cells to produce a high titer of infectious viral

oped to compare the changes of DNA copy n u m b e r

stock. The virus can then be delivered to TVA-expressing

across the genome, is based on hybridization of tumor

transgenic mice by injecting the virus-producing cells or

genomic DNA to the normal metaphase chromosomes.

by injecting the virus itself (Fig. 3). The TVA-based retro-

Therefore, the resolution is low and the technique can

viral gene delivery system offers several advantages over

only been used to determine regions of chromosomal

current transgenic and knockout mouse models. First, the

amplification and deletion. By contrast, DNA micro-

virus is replication competent and can be consistently

array-based CGH increases the resolution of CGH to

propagated in avian cells to produce a higher titer of viral

such an extent that, not only can it detect changes

stock. Second, infected murine cells do not produce

in known genomic regions, but it can also directly

infectious virus. The viral receptor in infected transgenic

identify candidate genes that are important to tumor

mouse cells remains available for multiple rounds of

development *',*z. Moreover, the combination of array

infection. Therefore, several genes of interest can be intro-

technology and genomic mapping makes it possible to

duced into the same cells simultaneously or sequentially,

study single-nucleotide polymorphism (SNP) in a high-

allowing analysis of the phenotypic consequence of

throughput manner, thus speeding-up enormously the

genetic mutations, both individually and in combination.

mutational analysis and the linkage mapping of cancerassociated genes 43,44.

Furthermore, TVA transgenic mice can be crossbred with

http://tcb.trends.com

other transgenic or knockout mice to study the role of A TRENDS Guide to Cancer Biology

eviev

TRENDS in Cell B i o l o g y Vo1.11 No,ll

November 2001

cooperative genetic mutations in accelerating tumor development and metastasis. A recent study on the mecha-

TVA-transgenic mouse

RCAS-gene X

I Transfection

nisms of gliomagenesis demonstrates the power of the TVA-based approach for dissecting cooperative oncogen-

Tissue-specific promoter (t)

,

DF-1 cell

esis in the mouse 49. To investigate the role of epidermal

$7

TVA

I

growth factor receptor (EGFR) mutation in glioma pathogenesis, transgenic mice were generated that expressed the TVA receptor under the control of the astrocytespecific glial fibrillary acidic protein (GFAP) promoter.

I

X

Propagation

Infection of GFAP-TVA transgenic mice with the virus

t-TVA

encoding a constitutively active mutant EGFR did not result in glioma development. However, coinfection of

Infection >

GFAP-TVA transgenic mice with viruses carrying CDK4 and mutant t~GFR or infection of GFAP-TVA transgenic mice deficient in INK4a-ARF with the mutant EGFR virus triggered gliomagenesis, suggesting that genetic alterations of EGFR and INK4a-ARF are two crucial genetic events in gliomagenesis 49. h~cibie

transgene

and coaditiona~

knockout;

Recent advances in tissue-specific manipulation of tar-

TRENDS in Cell Biology

geted genes achieve a tighter spatial and temporal control of gene expression. These overcome the limitations encountered by generating conventional mouse mutants, such as embryonic lethality in the knockout approach or constitutively high levels of gene expression in transgenic mice. Inducible systems, originally developed for in vitro cellbiology studies, have been applied more recently in vivo in transgenic approaches to regulate gene expression in a controlled manner. One of the most widely used systems is the tetracycline system, which uses a tetracycline-responsive

Figure

3. T V A - b a s e d

gene

delivery

approach

Chicken DF-1 cells are transfected with a plasmid encoding the replication-competent subgroup A avian viral vector, Rous sarcoma virus-derived replication-competent cloning vector (RCAS), and a gene of interest (gene X). The retrovirus is produced and propagated in DF-1 cells to obtain high-titer virus particles. Next, the virus is used to infect transgenic mice expressing the TVA receptor under the control of a tissue- or cell-type-specific promoter (t). Mice do not harbour a TVA gene homolog and thus are resistant to avian viral infection. Only cells engineered to express TVA receptor can be infected. As a result, the retroviral infection enables the introduction of gene X into a specific cell-type or tissue (t-X). Mice carrying multiple genetic alterations can be obtained by crossing t-TVA transgenic mice with knockout or transgenic mutants (Y). The Y mutant harbors one mutation and additional mutations can be delivered by retroviral infection. The cooperative interaction between gene X and Y mutations can result in the development of tumors in the mouse.

promoter to drive the expression of the gene of interest s°. Another widely utilized method takes advantage of the

excision of the region located between loxP sites. Crosses

ability of the estrogen receptor (ER) to shuttle from the

of these two murine lines create a double mutant where

cytosol to the nucleus, upon binding to the ligand or ana-

the gene of interest is now disrupted in a tissue-specific

logues such as tamoxifen. The generation of an in-frame

manner. Several tumor-suppressor gene mutants have been

fusion between the protein of interest (e.g. a transcription

generated using the Cre-lox system in order to study the

factor) and the ER renders the activity of the fusion pro-

consequences of their inactivation in a given tissue (e.g. the

tein tamoxifen dependent sl.

intestinal-specific inactivation of APC or the mammary-

As far as gene knockout is concerned, the Cre-loxP sys-

specific inactivation of BRCAis3's4.

tem has been widely applied to create a conditional disruption of gene expression in a specific tissue sz. In essence, a first mutant mouse is generated where loxP sites flank

of cance~

the target gene, or part of it. These are short sequences

Recombinant congenic (RC) strains provide an additional

that are recognized only by the Cre (cyclization recombi-

useful tool for cancer study ss. RC strains are produced

nation) bacteriophage recombinase. LoxP sites are usually

by limited backcrossing between a common 'donor'

introduced by homologous recombination within introns

inbred strain and a common 'background' inbred strain,

9enes

of the target gene, so that they do not interfere with its

before proceeding with inbreeding of the progeny de-

proper translation. Next, a transgenic line is generated

rived from the initial backcross. In this way, a series of

where the Cre gene is under the control of a tissue-specific

strains is generated, each of which contains a random

promoter. Cre recognizes and catalyzes recombination and

small subset of genes from the donor strain and the

http://tcb.trends.com

A TRENDS Guide to Cancer Biology

S8

[[evie~ evle~ |

TRENDS in Cell Biology Vol.11 No.11 November 2001

m a j o r i t y o f g e n e s f r o m the b a c k g r o u n d strain. Typically, t h e RC strains carry a p p r o x i m a t e l y 1 2 . 5 % o f t h e d o n o r genome

a n d 8 7 . 5 % o f the b a c k g r o u n d g e n o m e . As a

result, t h e i n d i v i d u a l g e n e o f the d o n o r s t r a i n t h a t m i g h t

References 1 Kinzler, K.W and Vogelstein, B. (I 996) Lessons from hereditary colorectal cancer. Cell87, 159 170 2 Knudson, A.G., Jr et al. (1975) Mutation and childhood cancer: a

play a r o l e i n t u m o r i g e n e s i s can b e s e g r e g a t e d a n d its

probabilistic model for the incidence of retinoblastoma. Proc.Natl.

contribution

AcM. Sci. U.S.A. 72,

to t h e

multiple

complex

traits can b e

m a p p e d a n d s t u d i e d separately. This a p p r o a c h has p r o v e d effective i n i d e n t i f y i n g n e w c a n c e r - s u s c e p t i b i l i t y loci, p a r t i c u l a r l y for l u n g a n d c o l o n cancers i n the m o u s e . So far, n i n e n e w loci, S c c l - S c c 9 , have b e e n i d e n t i f i e d t h a t

5116 5120

3 Yuspa, S.H. (1998) The pathogenesis of squamous cell cancer: lessons learned from studies of skin carcinogenesis. J. Dermatoi. Sci. 17, 1-7 4 Akhurst, R.J. and Balmain, A. (1999) Genetic events and the role of

are i n v o l v e d i n t u m o r i g e n e s i s i n the c o l o n s6,sT, w h e r e a s

TGF beta in epithelial turnout progression. J. Path01. 187, 82-90

f o u r n e w loci, Sluc 1 - S l u c 4 , i n f l u e n c e the s u s c e p t i b i l i t y to

Quintanilla, M. et al. (1986) Carcinogen-specific mutation and

l u n g c a n c e r i n the m o u s e zg. This s y s t e m d o e s have a f e w

amplification of Ha-ras during mouse skin carcinogenesis. Nature

d i s a d v a n t a g e s , however. T h e s e i n c l u d e the l e n g t h o f t i m e

322, 78-80

r e q u i r e d to g e n e r a t e e a c h series o f RC strains, a n d l i m i

6 Bizub, D. et al. (t986) Mutagenesis of the Ha-ras oncogene in mouse

ted f a m i l y m e m b e r s available for e a c h RC series. I n a d d i

skin tumors induced by polycyclic aromatic hydrocarbons. Proc.Natl.

t i o u , the g e n e t i c m a p p i n g r e s o l u t i o n c a n b e p o o r i n RC

Acad. Sd. U.S.A.

strains i n l i m i t e d cases i n w h i c h t h e r e are a l a r g e n u m b e r o f s e g r e g a t e d loci. D e s p i t e these l i m i t a t i o n s , i n the f u t u r e this a p p r o a c h m i g h t p r o v i d e a s u b s t a n t i a l c o n t r i b u t i o n to the i d e n t i f i c a t i o n o f c a n c e r - s u s c e p t i b i l i t y g e n e s , t u m o r modifiers and metamorphosers.

83, 6048-6052

7 Kemp, CJ. et al. (1993) A/lelotype analysis of mouse skin tumors using polymorphic microsatellites: sequential gertetic alterations on chromosomes 6, 7, and 11. CancerRes. 53, 6022-6027 8 Cano, A. et al. (1996) Expression pattern of the cell adhesion molecules. E-cadherin, P-cadherin arid {z6~4 integrin is altered in pre-malignant skin rumors of p53 deficient mice. Iat.J. Cancer 68,

Concluding remarks In t h e p a s t f e w years, m o d e l i n g h u m a n

254-262 c a n c e r i n the

m o u s e h a s already p r o v e n i n c r e d i b l y i n s t r u c t i v e for s t u d y ing the mechanisms underlying cancer pathogenesis, the function of cancer genes and their intricate interactions

9 Robles, A.I. et al. (! 998) Rednced skin tumor development in cydin D1 deficient mice highlights the oncogenic ras pathway in viva.Genes Dev.

12, 2469-2474

10 Kemp, C.J. et al. (1993) Reduction of p53 gene dosage does not

i n m u h i s t e p t u m o r i g e n e s i s . The f u t u r e c o m p l e t i o n o f a

increase initiation or promotion but enhances malignax,t progression

p r o p e r a n n o t a t i o n o f the h u m a n a n d m o u s e g e n o m e , as

of chemically induced skin tumors. Cell 74, 813-822

w e l l as t h e a d v e n t o f n o v e l t e c h n o l o g i e s for the s t u d y o f c o o p e r a t i v e o n c o g e n e s i s i n the m o u s e a n d for the i d e n t i f i c a t i o n o f g e n e t i c events t h r o u g h o u t t u m o r p r o g r e s s i o n , w i l l c e r t a i n l y m a r k e d l y s p e e d u p this l e a r n i n g process. This a c c e l e r a t i o n is v e r y m u c h n e e d e d i n o r d e r to dise n t a n g l e t h e o v e r w h e l m i n g c o m p l e x i t y that s u c h a f l u r r y of genetic information will impose on our analytical capacity.

11 Saez, E. et a]. (1995) c-los is reqnired for malignant progression of skin tumors. Cell82,721 732 1Z Wang, Z.G. et a]. (1998) PML is essential for multiple apoptotic pathways. Nat Genet. 20,266 272 13 Cui, W et al. (1996) TGF~1 inhibits the formation of benign skin tumors, but enhances progression to invasive spindle carcinomas in transgenic mice. Cdl 86, 531 542 14 Glick, A.B. et aL (1994)Targeted deletion of the TGE-~l geue causes rapid progression to squamous cell carcinoma. Genes Dev.8,

Acknowledgements T h i s w o r k is s u p p o r t e d b y the NCI, the De W i t t W a l l a c e F u n d for M e m o r i a l S l o a n - K e t t e r i n g C a n c e r Center, the Mouse Model of Human Cancer Consortium (MMHCC), a n d N a t i o n a l I n s t i t u t e s o f H e a l t h G r a n t s to Pier Paolo Pandolfl a n d the L y m p h o m a & L e u k e m i a Society, o f w h i c h Pier Paolo Pandolfi is a Scholar. W e t h a n k A l a n B a l m a i n a n d M i n g H u for t h e i r s u g g e s t i o n s a n d critical r e a d i n g o f the m a n u s c r i p t . Finally, w e are i n d e b t e d to all the past a n d p r e s e n t m e m b e r s o f the l a b o r a t o r y o f M o l e c u l a r a n d D e v e l o p m e n t a l B i o l o g y (MADB) at M e m o r i a l Sloan-

2429-2440 15 Sibilia, M. et al. (2000) The EGF receptor provides an essential survival signal for SOS deper~dent skin tumor development. Cell 102, 211-220 16 Fero, M.L. et al. (I998) The murine gene p27Kipl is haploinsufficient for tumom suppression. Nature 396, 177-180 17 Rego, E.M. et al. (2001) Role of promyelocytic leukemia (PML) protein in tumor suppression. J. Exp.Med. 193, 521--529 18 Pandolfi, RR In viv0analysis of the molecular genetics of acute promyelocytic leukemia. Oncottene (in press) 19 He, L.Z. et a]. (1997) Acute leukemia with promyelocytic features in

K e t t e r i n g C a n c e r Center, w o r k i n g o n m o u s e m o d e l s o f

PML RARottransgenic mice. Proe. Natl.Acad. Sci. U. S.A. 94,

c a n c e r a n d r e l a t e d subjects.

5302-5307

http://tcb,trends.com

A TRENDS Guide to CancerBiology

:evie~

TRENDS in Cell Biology Vo1,11 No,ll

20 Brown, D. et al. (1997) A PML~RAR0t transgene initiates mnrine acute promyelocytic leukemia, Proc. Natl.Aend. Sci. U. S.A. 94,

S9

39 Pollack, J.R. et a]. (1999) Genome-wide analysis of DNA copy number changes using cDNA microarrays. Not. Genet. 23, 41 46 40 Pinkel, D. et a]. (1998) High resolution analysis of DNA copy

2551-2856 Zl Grisolano, J.L. et al. (1997) Altered myeloid development and acute leukemia in transgenic mice expressing PML-RAR(z under control of cathepsin G regulatory sequences. Blood 89,376 387 22 He, L.Z. et al. (1998) Distinct interactions of PML-RAR{Z and PLZF-RAR0t with co repressors determine differential responses to

23 Pollock, J.L. et al. (1999) A bcr-3 isoform of R A R O ; - P M L potentiates the development of PML RARc~-driven acute promyekicytic leukemia. Proc. Natl. Aend. Sci. U. S.A. 96, 15103-1 S 108 24 He, L. et al. (2000) Two critical hits for promyelocytic leukemia. M0].

41 Albertson, D.G. et al. (2000) Quantitative mapping of amplicon structure by array CGH identifies CYP24 as a candidate oncogene. Nat. Genet. 25, 144-146

ovarian cancer. Nat. Getter. 21, 99-102 43 Hacia, J.G. et d. (1999) Determination of ancestral alleles for human single nucleotide polymorphisms using high-density oligonucleotide arrays. Nat. Genet. 22, 164-167 44 Hacia, J.G. et al. (1998) Strategies for mutational analysis of the large

1141

25 Dietrich, "vVRet el. (1993) Genetic identification of Morn-l, a major modifier locus affecting Min-induced intestinal neoplasia in the mouse. Cell 75,631

number variation using comparative genomic hybridization to microarrays. Nat. Genet. 20,207-211

42 Shayesteh, L. et a]. (1999) PIK3CA is implicated as an oncogene in

RA inAPL. Nat. Genet. 18, t26-135

Cell6, 1131

November 2001

muhiexon ATM gene using high-density oligonucleotide arrays. Genome Res. 8, 1245-1258

45 Girard, L. et a]. (1996) Frequent provirus insertional mntagenesis of

639

26 MacPhee, M. et al. (1995) The secretory phospholipase A2 gene is a candidate for the Moml locus, a major modifier of ApcMin-induced intestinal neoplasia. Ceil 81, 9 S 7-966 27 Cormier, R.T. et al. (1997) Secretory phospholipase Pla2g2a confers resistance to intestinal tumorigenesis. Nat. Genet. 17, 88-91

Notch1 in thymomas of MMTVD/myc transgenic mice suggests a collaboration of c-myc and Notehl for oncogenesis, Genes Dev. 1 O, 1930 1944 46 Li, J. et al. (1999) Leukaemia disease genes: large-scale cloning and pathway predictions. Nat. Genet. 23, 348-353 47 Fisher, G.H. et al. (1999) Development of a flexible and specific gene

28 Tripodis, N. and Demant, R (2001) Genetic linkage of nuclear morphology of mouse lung tumors to the Kras- 2 locus. Exp. Lunfl Res.

delivery system for production of murine tumor models. Oncogene 18, 5253 5260 48 Eederspiel, M.J. et aI. (1996) Expression of transduced genes in mice

27, 185-196 29 Fijneman, R.J. et 4. (1996) Complex interactions of new quantitative trait loci, Slucl, Sluc2, Sluc3, and Sluc4, that influence the susceptibility to lung cancer in the mouse. Nat. Genet. 14,

generated by infecting blastocysts with avian leukosis virus-based retroviral vectors. Proc. Natl. Aend. Sci. U. S.A. 93, 4931-4936 4-9 Holland, E.C. et al. (1998) A constitutively active epidermal growth factor receptor cooperates with disruption of G1 cell-cycle arrest

465-467 30 Nagase, H. et al. (2001) Epistatic interactions between skin tumor modifier loci in interspecific (spretus/musculus) backcross mice. Cancer Res. 61, 1305-1308

pathways to induce glioma-like lesions in mice. Genes Dev. 12, 3675-3685 50 Gossen, M. et al. (199 S ) Transcriptional activation by tetracyclines in

31 Schrock, E. et al. (1996) Muhicolor spectral karyotyping of human chromosomes. Science 2 7 3 , 4 9 4 - 4 9 7 3"* Liyanage, M. et al. (1996) Muhicolour spectral karyotyping of mouse chromosomes. Nat. Genet. 14, 312-31 S 33 Lichter, R et al. (2000) Comparative genomic hybridization: uses and limitations. Semin.Hematok 37, 348-357 34 DeRisi, J. et al. (1996) Use of a cDNA microarray to analyse gene expression patterns in human cancer. Nat. Genet. 14, 457 460 35 Lipshutz, R.J. et al. (1999) High density synthetic oligonucleotide

mammalian cells. Science 268, 1766-1769 S 1 Pelengaris, S. et al. (1999) Reversible activation of c-Myc in skin: induction of a complex neoplastic phenotype by a single oncogenic lesion. M01.Cell 3,565 577 $Z Kuhn, R. et al. (1995) Inducible gene targeting in mice. Science 269, 1427-1429 53 Shibata, H. et al. (1997) Rapid colorectal adenoma formation initiated by conditional targeting of the Apc gene. Science 2 7 8 , 120-123 54 Xu, X. et al. (1999) Conditional mutation of Brcal in mammary

arrays. Nat. Genet. 2 I, 20-24 36 Duggan, D.J. et al. (1999) Expression profiling using cDNA microarrays. Nat. Genet. 21, 10-14

epithelial cells results in blunted ductal morphogenesis and tumour formation. Nat. Genet. 22, 37-43

37 Liyanage, M. et al. (2000) Abnormal rearrangement within the alpha/delta Tcell receptor locus in lymphomas from Aim-deficient

55 Moen, C.J. et al. (1991) The recombinant congenic strains - a novel genetic tool applied to the study of colon tumor development in the mouse. Marnm. Genome 1, 217-227

mice. Blood 96, 1940-1946 38 Zimonjic, D.B. et al. (2000) Acquired, nonrandom chromosomal abnormalities assodated with the development of acute promyelocytic leukemia in transgenic mice. Proc. Natl. Aend. Sci. U. S,A. 97, 13306-13311

56 van Wezel, T. et a]. (1996) Gene interaction and single gene effects in colon m m o u r susceptibility in mice. Nat. Genet. 14, 468-470 57 van Wezel, T. et al. (1999) Four new colon cancer susceptibility loci, Scc6 to Scc9 in the mouse. Cancer Res. 59, 4216-4218

http://tcb.trends.com

A TRENDS Guide to Cancer Biology