Of mice and models: improved animal models for … › ~ireneyan › CursoPG › PDFs › Bockamp et al...Of mice and models: improved animal models for biomedical research. Physiol

doi: 10.1152/physiolgenomics.00067.200211:115-132, 2002. ;Physiol. Genomics

Fraser, Leonid Eshkind, Franz Oesch and Bernhard ZabelErnesto Bockamp, Marko Maringer, Christian Spangenberg, Stephan Fees, Stuartbiomedical researchOf mice and models: improved animal models for

You might find this additional info useful...

165 articles, 69 of which you can access for free at: This article citeshttp://physiolgenomics.physiology.org/content/11/3/115.full#ref-list-1

19 other HighWire-hosted articles: This article has been cited by http://physiolgenomics.physiology.org/content/11/3/115#cited-by

including high resolution figures, can be found at: Updated information and serviceshttp://physiolgenomics.physiology.org/content/11/3/115.full

can be found at: Physiological Genomics about Additional material and informationhttp://www.the-aps.org/publications/physiolgenomics

This information is current as of August 17, 2012.

http://www.the-aps.org/. 20814-3991. Copyright © 2002 the American Physiological Society. ESSN: 1531-2267. Visit our website at 24 times a year (twice monthly) by the American Physiological Society, 9650 Rockville Pike, Bethesda MDsystems with techniques linking genes and pathways to physiology, from prokaryotes to eukaryotes. It is published

publishes results of a wide variety of studies from human and from informative modelPhysiological Genomics

at CA

PE

S - U

sage on August 17, 2012

http://physiolgenomics.physiology.org/

Dow

nloaded from

http://physiolgenomics.physiology.org/content/11/3/115#cited-by


invited reviewOf mice and models: improved animal modelsfor biomedical research

ERNESTO BOCKAMP,1 MARKO MARINGER,1 CHRISTIAN SPANGENBERG,2

STEPHAN FEES,2 STUART FRASER,1 LEONID ESHKIND,1 FRANZ OESCH,1

AND BERNHARD ZABEL2

1Laboratory of Molecular Mouse Genetics, Institute of Toxicology, Johannes Gutenberg-UniversityMainz, D-55131 Mainz; and 2Laboratory of Molecular Genetics at Children’s Hospital,Johannes Gutenberg-University Mainz, D-55101 Mainz, Germany

Bockamp, Ernesto, Marko Maringer, Christian Spangenberg,Stephan Fees, Stuart Fraser, Leonid Eshkind, Franz Oesch, andBernhard Zabel. Of mice and models: improved animal models forbiomedical research. Physiol Genomics 11: 115–132, 2002; 10.1152/physiolgenomics.00067.2002.—The ability to engineer the mouse ge-nome has profoundly transformed biomedical research. During thelast decade, conventional transgenic and gene knockout technologieshave become invaluable experimental tools for modeling genetic dis-orders, assigning functions to genes, evaluating drugs and toxins, andby and large helping to answer fundamental questions in basic andapplied research. In addition, the growing demand for more sophisti-cated murine models has also become increasingly evident. Goodstate-of-principle knowledge about the enormous potential of second-generation conditional mouse technology will be beneficial for anyresearcher interested in using these experimental tools. In this reviewwe will focus on practice, pivotal principles, and progress in therapidly expanding area of conditional mouse technology. The reviewwill also present an internet compilation of available tetracycline-inducible mouse models as tools for biomedical research (http://www.zmg.uni-mainz.de/tetmouse/).

transgenic mice; knockout mice; conditional mouse models; Cre and Flprecombinase; tetracycline; doxycycline; IPTG; isopropyl-�-D-thiogalacto-sidase; rapamycine; progesterone; RU486; ecdysone

ONE OF THE PRINCIPAL ISSUES facing biomedical researchtoday is to convert analytical data and sequenceinformation into knowledge about function. Withcontinuously expanding databases of sequencedgenes and a seemingly unlimited amount of biologi-cally compelling and clinically significant data, thelaboratory mouse has become an increasingly impor-tant tool for facilitating functional in vivo data ongene function.

During the last two decades fundamental insightinto gene function has been efficiently providedthrough conventional knockout and transgenic experi-ments. However, these experiments frequently re-vealed serious limitations intrinsic to conventionalgene-targeting and transgenic approaches. Introducinggenetic changes to the germ line of a mouse may trackdown the effects of a particular gene but may also havesevere developmental consequences, complicating oreven precluding the desired experimental analysis.Good examples illustrating these restrictions are em-bryonic lethal knockout phenotypes but also includeadaptive gene expression, compensating for the ab-sence or hyperexpression of a specific gene, or cyto-static and cytotoxic effects in response to inappropriategene expression. To overcome these undesired limita-

Article published online before print. See web site for date ofpublication (http://physiolgenomics.physiology.org).

Address for reprint requests and other correspondence: E. Bockamp,Laboratory of Molecular Mouse Genetics, Institute of Toxicology, Jo-hannes Gutenberg-University Mainz, Obere Zahlbacher Str. 67,D-55131 Mainz, Germany (E-mail: [email protected])

Physiol Genomics 11: 115–132, 2002;10.1152/physiolgenomics.00067.2002.

1094-8341/02 $5.00 Copyright © 2002 the American Physiological Society 115

at CA

PE

S - U



Dow

nloaded from


tions and to precisely control gene expression or extinc-tion of gene function in a spatiotemporal fashion, con-ditional mouse models are becoming increasinglypopular. It is therefore expected that the second gen-eration of mouse models will significantly improve ourability to address the myriad of questions about genefunction.

In this review we will examine different conditionaltransgenic and gene-targeting techniques and also pro-vide a brief overview of conventional mouse transgen-esis. In addition, we will present a novel databasecompiling already established tissue-specific effectormice for conditionally expressing recombinant genesusing the “tet on/off” system. It is not intended to becomprehensive, and the reader is encouraged to ex-plore other excellent recent reviews on this topic (50,84, 117, 153).

Basic Constitutive Transgenic Strategies

Since the first report of transgenic mice generated byinjecting DNA into the pronucleus of one-cell mouseembryos, this technique has been immensely useful increating model organisms for research purposes (44).Normally, the transgenic construct consists of a se-lected enhancer and/or promoter, which may also di-rect gene expression to a specific tissue or developmen-tal stage, linked to the sequence to be expressed. Aschematic overview illustrating the basic methodologyfor generating transgenic mice is shown in Fig. 1.Using this approach, one may directly test in themouse the role of selected gene products, dominantnegative mutants, or specifically designed proteins.

Moreover, antisense mRNA expression or expression ofribozymes can diminish a specific endogenous genefunction (30, 32, 66, 79, 87). In addition, overexpressionof toxin-encoding genes might be used to mimic hy-poplasia (82, 126) and expression of a prodrug metab-olizing enzyme like the herpes simplex virus type 1thymidine kinase is useful to conditionally ablate spe-cific cell populations (18, 29, 111, 135, 144). Alterna-tively, transcriptional elements including promoters,enhancers, silencers, or complete locus control regionscan be assayed in transgenic mice (38, 71, 105, 120,131).

In theory, once integrated into the murine genome,the injected DNA can manifest its function. However,as the insertion occurs at random, positional variega-tion effects may be considered, and both the function ofendogenous genes might be affected by the insertion ofa transgene (insertional effects; Ref. 154) as well as theexpression of the transgene itself may also be severelycompromised by surrounding elements (variegation ef-fects; Ref. 110). As a solution to this recurring problemsingle-copy integration of the transgene into a selectedsite or the use of so-called insulator elements has beendescribed (14, 23, 36, 41, 97). Alternatively, transientgeneration of transgenic mice is often used for studyingtranscriptional regulatory elements in mice (131). Thissystem is useful in that no stable transgenic mouselines need to be established; however, a sufficientlylarge number of founder animals must be generatedand examined. A comparison of all founders will ensurethat phenotypes due to variegational position effects ofthe transgene will become obvious. However, such anapproach will be expensive, time consuming, and tech-nically challenging.

Conventional Germ Line Gene-Targeting

With the advent of murine embryonic stem (ES) cellsin 1981 and the insight that mammalian cells have theenzymatic machinery to appropriately recombine ho-mologous DNA sequences exactly with their counter-parts on the chromosome, the necessary experimentaltools for engineering genetic modifications in themouse were in place (34, 90, 134, 155). Indeed, buildingon these scientific milestones, the first knockout micewere almost immediately generated by homologousrecombination in ES cells (28, 137). A general schemeoutlining the basic strategy for generating knockoutmice is depicted in Fig. 2.

A crucial advantage of gene-targeting approachesover transgenic techniques is that homologous recom-bination in ES cells clearly defines the site of integra-tion and allows for precisely designing the geneticchange to be introduced. As shown in Fig. 2A a gene-targeting construct will usually consist of a core regioncontaining the desired genetic change together with apositive selection cassette (conferring resistance to an-tibiotics such as neomycin or hygromycin). This core isflanked by two regions of absolute sequence identitywith the targeted region (homology arms) required forhomologous recombination to take place. Given that

Fig. 1. Schematic representation for producing transgenic mice: atransgenic construct containing a specific promoter (Pspec), the geneof interest, and a polyadenylation signal (pA) are shown. This con-struct is injected into fertilized oocytes to be subsequently trans-ferred into pseudo-pregnant recipient mothers. Pups are thenscreened for integration of the transgene.

116 INVITED REVIEW

Physiol Genomics • VOL 11 • www.physiolgenomics.org

at CA

PE

S - U



Dow

nloaded from


homologous recombination in ES cells is a relativelyrare event compared with random integration of DNA,improved strategies to enrich for recombined ES cellclones are needed. These enrichment strategies usu-ally make use of negative selection markers such as theherpes simplex virus type 1 thymidine kinase gene orthe diphtheria toxin �-chain coding region, which areplaced outside the homology region of the targetingconstruct (10, 20, 161). Using a negative selectionmarker, ES cell clones with correct homologous recom-bination will not contain this marker, as removal ofnonhomologous flanking sequences will take placeprior to homologous integration. By contrast most ran-domly integrated targeting constructs still retain thenegative selection marker, thus permitting counter-selection against randomly integrated clones. In case ofplanning more complicated multistep genome engi-neering strategies, suitable selection markers allowingboth positive and negative selection have also beendeveloped (20).

With the use of conventional gene-targeting strate-gies, hundreds of genes have been successfully engi-neered and the corresponding phenotypes analyzed(see http://tbase.jax.org/ and http://research.bmn.com/mkmd for electronically searchable databases).

In addition, the careful comparison of identical geneticchanges in different mouse strains revealed the impor-tance of genetic background in the function or loss offunction of a particular gene (106, 121). As the geneticbackground is the collection of all different genespresent in the organism it is not surprising that inmany cases the observed phenotype, which was intro-duced by a single genetic modification, might bestrongly influenced by other genes. These modifiergenes are often completely unlinked to the targetedlocus but are closely related to its function. For exam-ple, mice null for �-protein kinase C showed a markeddifference in sensitivity and tolerance to ethanol solelydepending on the genetic background (11). This issue iscompounded by the desired intercrossing of severalmodified lines which may have been generated on ge-netically distinct backgrounds. For this reason it be-comes critical for the interpretation of any knockoutand transgenic genotype to distinguish whether theobserved phenotypic difference is brought about by thetargeted mutation or instead is the result of back-ground genetic variation. Nevertheless, there are anumber of strategies available to the researcher, ap-propriate to minimize any undesired variegational

Fig. 2. Generation of null mutant mice using homolo-gous recombination in embryonic stem (ES) cells. A:strategy for generating a knockout genotype using ho-mologous recombination. The wild-type locus here de-picted with 5 exons (E1–E5) will recombine with thetargeting construct, thus both introducing a selectionmarker and deleting E3. P, promoter driving expressionof the selectable marker; pA, polyadenylation signal. B:homologously recombined ES cells are injected into ablastocyst, which is then transferred into a pseudo-pregnant recipient mother. As the injected ES cellscontribute to all tissues, chimeric mice can be gener-ated. The previously introduced recombination is thentransmitted to the germ line resulting in hemizygousknockout mice.

117INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


phenotype as brought about by genetic backgrounddifferences.

Of course, the “gold standard” for any knockout ex-periment would be to use genetically identical animalsas controls. This can be achieved for example by estab-lishing an 129 isogenic inbred knockout mouse strain(usually 129/Sv, 129/SvEv, or 129/Ola) or by usingmurine ES cell lines derived from additional mousestrains with better reproductive performance than 129derivatives (26, 73, 81). As an additional possibility,the analysis of mice from the same litter but showingwild-type, heterozygous, and homozygous modifiedpups can also help to uncover the influence of geneticbackground on the phenotype observed. Furthermore,analysis of a large number of litters should help toclarify this issue. A very elegant alternative to excludegenetic background effects is to perform all plannedexperiments with the same individual mouse prior andafter the induction of the reversible phenotype (83).Most importantly, reversible gene-targeting strategiesmay demonstrate that the induced phenotype is clearlyassociated with the mutated gene as reversion back toa functional gene should reverse the observed effect.However, the most common practice used to minimizegenetic background effects is the time-consuming gen-eration of congenic knockout strains. Congenic animalsare genetically identical except for a single region con-taining the introduced change. In general, 2–3 years ofbackcrosses to a defined strain are required before theresulting genetic backgrounds are statistically �99%homogeneous (129). The Jackson Laboratory offers apossible shortcut in this time-consuming backcrossingby implementing marker-assisted selection of mice.This technique can reduce the time needed to generatea defined genetic background on backcrossed mice from10 generations to 5 generations, shortening the timerequired by half (http://jaxmice.jax.org/html/services/services-speedcongenic.shtml). Commercial suppli-ers including the Jackson Laboratory may also offer F2hybrid mice that approximate the mixed strain back-ground under analysis. As a last resort to avoid phe-notypic variations, it is possible to examine large num-bers of knockout mice with mixed background, thusensuring that the range of phenotypes due to geneticbackground are detected.

Although it is not always feasible to use geneticallyidentical mice for analyzing the effect of a given knock-out, and the majority of previously published pheno-types are from genetically mixed backgrounds, it al-ways makes sense to use the most genetically similarbackground to the knockout as a control.

Finally, gene trap approaches in ES cells have beenused for generating entire knockout libraries (33, 132,152, 166). With this approach, the unique integrationsite of the transfected vector serves as a signpost forthe subsequent identification of the trapped gene. Inaddition, through the combination of a promoterlesssplice acceptor sequence together with a reporter geneas part of the gene trapping strategy, informationregarding the spatiotemporal expression of the trappedgene can often be obtained in ES cell-derived hemizy-

gous animals. Electronically searchable databases oflarge-scale public domain and commercial gene trapinitiatives can be accessed via http://socrates.berkeley.edu/�skarnes/resource.html, http://tikus.gsf.de/,and http://www.lexgen.com. Taking a closer look atthese sites, especially at the noncommercial ones, priorto initiating a planned knockout project is thereforegood practice and may also save time and money.

Conditional Switchable Transgenic Mice

In experimental settings, use of conventional trans-genic technology control over the onset of transgeneexpression will strictly depend on positional integra-tion effects and on the nature of the chosen regulatoryelements. However, constitutive expression of a trans-gene is often too inflexible to meet the needs of aspecific experimental question. For example, too earlyor to widespread expression of the transgene may leadto phenotypic or physiological aberrations producingsecondary pleiotropic responses as a result of the in-troduced genetic alteration. Distinguishing effects ofthe resulting phenotype might turn out to be extremelydifficult, as cell autonomous versus cell non-autono-mous effects are not clearly divisible and compensatorysystemic changes are often concealed. In addition, formany experimental questions it might be necessary toanalyze the function of a transgene within a specificdevelopmental window or in a particular cell lineage.Unfortunately, adequate temporal or tissue-restrictedpromoters may not always be available. For these rea-sons, the perfect conditional transgenic mouse should,in principle, include the following criteria. First, in-duced overexpression of the transgene should betightly controlled such that no leaky background ex-pression precludes the accurate analysis. Second, theinducing compound should be nontoxic and highly spe-cific for the target gene. Third, induction kineticsshould be fast and expression levels sufficiently high toproduce a rapid and detectable effect. Fourth, the in-duced switch should be reversible so that defined de-velopmental periods or critical stages in disease can beappropriately monitored.

To date, several transgenic mouse systems havebeen successfully used for tightly controlling spatio-temporal reversible transgene expression in mice.Nearly all of them share a common denominator inthat they hinge on the sequence of three principalevents: 1) Ligand-mediated activation of a transcrip-tional transactivator, 2) DNA binding of the transacti-vator, and 3) transactivator-induced transcriptionalactivation (see Fig. 3A). Translating this system into aconditional mouse model usually requires an effectormouse, expressing a ligand-inducible transcriptionaltransactivator, and, in addition, a responder mouseline, which has the capacity to specifically express achosen transgene upon stimulation by the transactiva-tor. Subsequent intercrossing of effector mice with theresponder line should lead to the birth of bi-transgenicoffspring, capable of conditionally expressing the de-

118 INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


sired transgene (see Fig. 3B for a schematic represen-tation of the underlying strategy).

Historically, these binary systems were plagued byleaky transcription of the transgene even in the ab-sence of the inducing ligand (9, 69). Furthermore, asthe initial activators were derived from oncogenic vi-ruses, undesirable neoplastic activation of endogenousmouse genes was observed (99). By now, these initialproblems have been principally solved, presenting theresearcher with several effective options as to how heor she will set up tight “genetic switches” for theprojected conditional mouse.

The Tetracycline System

Over the last decade the tetracycline (tet) regulatorysystem has been extremely useful for generating re-versible transgenic mouse models. This has led to alarge collection of tet-controlled transgenic mice suit-able for studying gene function in many different celltypes and also during selected time points (for recentreviews, see Refs. 3 and 88, as well as http://www.zmg.uni-mainz.de/tetmouse/ for an electronicallysearchable database).

Pioneering work by Gossen and Bujard (45) demon-strated the general usefulness of the Escherichia coliTn10-derived tet resistance operon for tightly regulat-ing conditional transgene expression in mammaliancells. In this study the tet repressor (TetR) was fused tothe VP16 transactivation domain of herpes simplexvirus, resulting in the tetracycline controlled transac-tivator (tTA). In the absence of tetracycline [or thecommercially available low-cost alternative, doxycy-cline (DOX)] tTA binds to its specific tetracycline oper-ator consensus sequence (tetO), being able to also acti-vate transcription of a chosen transgene whencombined with a synthetic minimal promoter se-quence. By contrast, addition of DOX to this systeminduces a conformational change in the tTA effector,preventing DNA binding and leading to a shutdown ofthe target gene. In summary, use of the tet systemrequires two coordinate building blocks: the ligand-dependent transactivator tTA as the effector, and atetO-CMV minimal promoter cassette governing theexpression of the transgene as the responder. Figure4A demonstrates the underlying mechanism of thetet-off system.

Fig. 3. Conditional transgenesis. A: binary control of condi-tional gene expression at the molecular level. 1) Ligand-mediated activation of the effector composed of an activationdomain (yellow) and a ligand-binding domain (light blue).Ligand is the solid brown box. 2) Specific interaction ofligand-bound effector with its cognate binding site (CBS). 3)Transcriptional induction of the gene of interest. B: imple-menting binary switches in mice. In general, for the produc-tion of conditional transgenic mice, two independent genet-ically modified mice strains are required: first, a lineexpressing a ligand-dependent effector, and second, a re-sponder line which conditionally expresses the transgeneunder the control of the effector. Crossing of the effectorwith the responder line results in a bi-transgenic mousecontaining all necessary elements for conditionally regulat-ing the expression of the gene of interest. Pmin, minimalpromoter; pA, polyadenylation signal; Pspec, specificpromoter.

119INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


The well-established pharmacological properties oftetracycline and many of its derivatives made thesecompounds very promising candidate regulators forestablishing conditional regulatory systems in themouse. These properties include predictable pharma-cokinetics, good tissue distribution, known half-lifetimes, lack of toxicity, and the ability to cross mamma-lian cell membranes and also the placenta. In actualfact, almost immediately after the initial publication ofthe tTA tet-off regulatory system in tissue culture cells,the first report of tet-based transgenic mice demon-strated the excellent potential of this system (42). De-spite the impressive induction levels, reaching in sometissues five orders of magnitude, and excellent shut-down kinetics which were reported, certain limitationsof the tet system also became evident. Initial majorproblems included repeatedly detectable residual ac-tivity of the tetO-CMV minimal promoter, cellular tox-icity of the transactivator, low sensitivity to DOX incertain tissue types, internal cryptic splice acceptorswithin the transactivator coding sequence leading toincorrect translation, instability of the transactivatorcoding mRNA, and slow in vivo induction kinetics ofthe transgene. During the last decade these problemshave been effectively addressed, and at present anexquisite assortment of “second-generation” tet-based

regulatory systems for generating conditional mice areavailable to the researcher.

One of the first crucial advances building on theoriginal tTA system was the development of the re-verse tet-on tetracycline-controlled transactivator(rtTA, tet-on shown in Fig. 4B). This rtTA molecule isinduced upon administration of DOX thus comple-menting the original tTA system and also solving theproblem of slow activation kinetics (46). Introducingthe rtTA system to transgenic mice increased the speedof transgene expression substantially, in some casesreaching complete activation in about 1 hour comparedwith the slow activation kinetics of up to 1 week re-ported for the original tTA (51, 72).

A second recurring drawback of the tet system wasthe unwanted residual activity of the tetO-CMV re-sponder even in the absence of an effector (42, 58, 68,72). This problem was tackled by direct transcriptionaltargeting of the leaky tetO-CMV responder. To thisend, tet-dependent transcriptional repressors were de-signed which do not physically interfere with the tet-transactivators but will repress any residual transcrip-tion of the responder during noninduced periods. Thesetetracycline-controlled repressors are fusion proteinscombining the DNA-binding domain for the tetracy-cline operator consensus sequences (tetO) fused to a

Fig. 4. The Tet system. The Tet system can be used to conditionally activate gene expression in the mouse. A: theTet-off system (tTA) will activate expression in the absence of its ligand doxycycline (DOX, shown as brown box).Upon addition of DOX, transcription of the gene of interest is extinguished. B: in contrast, addition of DOX to theTet-on system (rtTA) results in transcriptional induction of the gene of interest. tTA, tetracycline-dependenttransactivator; rtTA, reverse tetracycline-dependent transactivator; DOX, doxycycline (ligand); TRE, Tet-respon-sive element.

120 INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


strong transcriptional silencing element. A very ele-gant strategy to ensure stringent DOX-dependent reg-ulation of the tetO-CMV responder is coexpression ofthe activator together with the repressor using a bi-cistronic approach in genetically modified animals (39,40, 58). Interestingly, first experiments in transgenicmice provided convincing evidence for DOX-mediatedrepression capable of completely suppressing tran-scription of the target gene without compromisingrtTA-mediated activation (167).

In certain cell types and transgenic animals, stableexpression of the original rtTA and tTA appeared to bedifficult, probably due to toxicity resulting from theVP16 transactivation domain (8). To overcome thisproblem a set of tetracycline-responsive transactiva-tors, which are tolerated at higher intracellular con-centrations, have been developed (1, 5, 139). Finally,remodeling of the original tTA and rtTA transactiva-tors has been efficiently used for substantially improv-ing several other important features of the tet system.These tailor-made tet-inducible effectors include mu-tated transactivators with increased DOX sensitivityand improved activation kinetics, engineered transac-tivators with increased mRNA stability through hu-manized codon usage, abrogation of cryptic splice ac-ceptor sites, and also a new set of transactivatorsshowing novel DNA binding and dimerization proper-ties (6, 138, 141, 151).

Moreover, the tetO responder cassette has also beensignificantly improved, allowing the simultaneous ex-pression of two tet-controllable transgenes by virtue ofa bidirectional tetO-CMV minimal promoter unit (4).Combining the expression of the transgene with areporter gene using bidirectional tetO-CMV promotershas already proven to be a formidable experimentaltool allowing direct visualization of concurrent tran-scription of two genes (75). In this respect a recentreport demonstrated that it was possible to noninva-sively monitor the expression status of a transgene inthe living animal (51).

Finally, two groups have reported the successful useof a single construct containing both a tet-responsivetransactivator together with the transgene responderon the same plasmid (123, 140). This strategy circum-vents the need to generate two independent transgeniclines in the first place and greatly facilitates the mousebreeding strategy as effector and responder will notsegregate.

Taken together, during the last years the tet systemhas seen numerous remarkable improvements. Theseimportant advances will help the researcher to betterdesign conditional transgenic models and also to adjustthe experimental setup according to his or her specificneeds. As the overall number of individual tet-respon-sive transgenic mouse strains is abundant, the scope ofthis review does not allow for the detailed descriptionof each and every single study. The reader is, however,encouraged to explore the excellent expert reviews andto also visit the database at http://www.zmg.uni-mainz.de/tetmouse/, which provides direct links tomost of the original papers.

Hormone Receptor-Based Transgenics

Over the past couple of years, several groups havereported progress in developing conditional mousemodels based on nuclear hormone receptor fusionsincluding progesterone-, estrogen-, and ecdysone-basedtransgene switches. In the following sections we willprovide an overview of these three similar systems.

Progesterone and estrogen receptor-based models.Upon stimulation with an adequate ligand nuclearsteroid hormone receptors will form active dimers ca-pable of binding to DNA and activating transcription(19, 35, 109). This property of nuclear receptors makesthem amenable as molecular switches in mammaliancells. Indeed, tissue culture experiments suggestedthat progesterone as well as estrogen nuclear receptorfusions might be suitable effector molecules for regu-lating target gene expression in mammalian cells (12,149).

Progesterone receptor-based models. Interestingly,COOH-terminal truncated versions of the human pro-gesterone nuclear receptor not capable of binding toendogenous progesterone could nonetheless be acti-vated by RU486 or other synthetic progesterone antag-onists (149, 150). As the concentration of syntheticinducers required for the induction of transgene ex-pression turned out to be well below any physiologicalactive threshold, unwanted endogenous anti-glucocor-ticoid and anti-progesterone effects were not observed.

To establish a suitable system for conditionally ex-pressing a chosen transgene in a mouse, a chimericfusion protein containing three active domains wasconstructed. This tripartite polypeptide consisted ofthe COOH-terminal truncated human progesteronenuclear receptor as the ligand-inducible molecularswitch, the GAL4 DNA-binding domain as a specificDNA-anchor [GAL4 binds exclusively to the UAS (“up-stream activator sequence”), of Saccharomyces cerevi-siae but not to any murine genomic sequence], and thestrong transcriptional activation domain of herpes sim-plex virus VP16. By putting these three elements to-gether, a system for the generation of conditionaltransgenic mouse models was successfully established,known as GLVP-fusion (for “GAL4/herpes simplex vi-rus VP16 transcriptional activation domain and hu-man truncated progesterone receptor fusion”). Uponbinding to a suitable ligand, chimeric GLVPs will as-sociate into active homodimers, capable of specificallybinding to GAL4-responsive UAS elements and thusultimately inducing the desired expression of thetransgene. The GLVP system will also allow to delib-erately switch between activated and repressed ex-pression of the transgene by simply using a syntheticprogesterone antagonist to control the GLVP switch.Good examples for the effectiveness of progesterone-based transgenic mouse models include conditionallyswitchable lacZ reporter mice (43), tissue-restrictedexpression of the Cre recombinase (67), the conditionalexpression of TGF� in the epidermis (147), and therescue of the lethal �-inhibin phenotype (108).

121INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


Estrogen receptor-based models. By means of a sim-ilar fusion strategy, the estrogen receptor encodingsequence was fused to a GAL4 DNA binding domainand the VP16 transactivator. In the presence of estro-gen the resulting tripartite fusion protein becomes anactive dimer, binds to DNA, and will lead to transgeneactivation through the GAL4 promoter (12). An obviousdrawback of estrogen-based systems in the mouse isthat estrogen plays an important physiological role inthe regulation of cell proliferation and differentiation.As a consequence, other endogenous estrogen-respon-sive genes will also be induced upon estrogen admin-istration (56, 165). Conversely, endogenous estrogenlevels might also activate transcription of the trans-gene leading to unwanted “leaky” transgene expres-sion. For this reason estrogen receptor fusions to GAL4and VP16 have not been a preferential choice for gen-erating conditional transgenic mice. By contrast, directfusions of the estrogen receptor to a variety of heterol-ogous protein coding sequences including transcriptionfactors, oncogenes, RNA-binding proteins, kinases, andthe Cre recombinase have been very effective for con-ditional expression of biologically active fusion pro-teins (107).

Ecdysone receptor-based models. Ecdysone is a ste-roid hormone necessary during metamorphosis of in-sects. The natural ligand for the ecdysone receptor isthe 20-OH-ecdysone molting hormone, capable of spe-cifically binding to a heterodimer of the Drosophilaecdysone receptor (EcR) and the product of the ultra-spiracle (USP) gene. The activated EcR-USP het-erodimeric complex will bind to an ecdysone-respon-sive element and induce transcription of genesinvolved in the onset of metamorphosis (162). Severalgroups have taken advantage of the ecdysone systemfor conditionally expressing genes in mammalian cells(48, 148, 163). The potential of this system for gener-ating conditional transgenic mice was first shown us-ing a fusion of the VP16 and the glucocorticoid receptortransactivation domains to a truncated version of theecdysone receptor (100). In addition, it could be shownthat the retinoid X receptor, which is the mammalianhomolog of the natural ecdysone heterodimer partnerUSP in invertebrates, will also form functionally activedimers with an engineered ecdysone effector. Thesestudies also demonstrated that transcriptional activa-tion could be selectively targeted to a synthetic ecdys-one-responsive element, thus bypassing any unwantedactivation through endogenous nuclear hormone recep-tors (100, 136). As a still simpler alternative to theobligate EcR-USP heterodimeric system, a chimericBombyx/Drosophila ecdysone receptor has been cre-ated that does not require any retinoid X receptorheterodimeric partners for regulating conditionaltransgene expression (54). Although the ecdysone sys-tem seems very appropriate for the generation of in-ducible transgenic mouse models, to our knowledgeonly a small number of additional ecdysone-basedtransgenic mouse models have been reported in theliterature (2, 119).

Other Inducible Transgenic Models

In addition to the nuclear receptor- and tetracycline-based conditional systems, two different additional in-novative approaches have been examined in transgenicmice. Both regulatory systems hold great promise forgenerating conditional transgenic mouse models butalso provide the opportunity to be employed in concertwith other switchable systems, thus allowing the con-trol of two and more genes in the same mouse.

The cytochrome P-450 induction system. A prerequi-site for most binary switchable systems is the need forestablishing two independent transgenic strains. Bycontrast, the use of ligand inducible endogenous mam-malian promoters/enhancers can overcome this neces-sity, avoiding in some cases the complexity of bi-trans-genic animals. However, it is important to bare inmind, that deciding on a mono-transgenic regulatorysystem might restrict transgene induction to specifictissues and moreover might also preclude the opportu-nity of crossbreeding the generated effector mouse withother available responder lines and vice versa. Thecytochrome P-450 regulatory system is a good exampleof an effective mono-transgenic approach and has beenused to stringently control transgene expression in themouse and also the rat (17, 65).

Cytochrome P-450 monooxygenases comprise a mul-tigene family whose gene products are essential fordetoxifying lipophilic xenobiotics (101, 133). One mem-ber of the monooxygenase family, the cytochromeP-450 1A1 (CYP 1A1), is not constitutively expressed inmouse tissues and only becomes activated upon expo-sure with certain compounds including polycyclic aro-matic hydrocarbons (104). In a proof-of-principle exper-iment, a combination of CYP 1A1 promoter/enhancerelements was successfully utilized to conditionally con-trol the expression of a lacZ reporter gene (17). Thisstudy demonstrated tight and reversible regulation,reaching induction levels over four orders of magnitudein certain tissues. Expression was predominantlyfound in the liver but included several other organs. Asthe molecular pathway leading to the induction of CYP1A1 gene activation depends on the amount and chem-ical nature of the inductor the possibility for dose-dependent fine-tuning of transgene expression wasalso reported (17). A recent publication by Kantachu-vesiri and colleagues (65) exquisitely illustrated theenormous potential of CYP 1A1 promoter/enhancertransgene expression to answer fundamental ques-tions in physiology. Using both dose-dependent andreversible transgene expression of renin-2, the authorscould tightly control hypertension in the rat. Thisswitchable renin-2 rat model represents an invaluabletool for studying malignant hypertension including ac-celerated rise in blood pressure, endothelial injury,activation of the renin-angiotensin system, and in-duced microangiopathy (65).

IPTG-based inducible systems. The recently pub-lished isopropyl-�-D-thiogalactosidase (IPTG)-induc-ible two-compound system for generating conditionallycontrollable transgenic mice relies on the lactose

122 INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


operon switch originally described in a classic paper byJacob and Monod (61). The E. coli lactose operon con-sists of two main parts the lacI repressor and itscognate DNA binding site lacO. In the absence oflactose the lacI repressor will tightly bind to lacOpreventing the transcription of genes important formetabolizing lactose. Only in the presence of lactosedoes lacI dissociate from lacO and transcription ofgenes involved in lactose utilization can begin. Initialattempts to utilize the lactose operon in transgenicmice failed as no functional lacI gene was detected as aresult of transgene inactivation through DNA methyl-ation (125, 158). To adjust the bacterial lactose operonto eukaryotic transcription and translation standards,Cronin and colleagues (22) meticulously modified theoriginal sequence and gene structure of lacI resultingin the synthetic synlacI repressor. To generate trans-genic mice which express the synlacI repressor in aconstitutive fashion, the synlacI cDNA was put underthe transcriptional control of a human �-actin pro-moter element. Using tyrosinase, which is an essentialenzyme for melanin biosynthesis in the mouse, as areporter gene they demonstrated that the synlacI re-pressor system perfectly allowed for conditionally con-trolling tyrosinase expression. In addition, these exper-iments showed complete reversibility of transgeneexpression in the absence of inducer and also, whatmay be important for developmental experiments, goodconditional regulation of tyrosinase in the adult mouse,embryo, and nursing pup. Taken together this firstreport suggests that the newly developed synlacIIPTG-inducible system provides tight and switchablecontrol of transgene expression in the mouse, making itan excellent candidate tool for generating conditionaltransgenic models.

Conditional Gene Targeting

Each gene-targeting experiment normally involvestwo basic steps. First, homologous recombination willbe used to stably introduce a desired genetic changeinto the DNA of pluripotent ES cells. After aggregationwith or injection into the early embryo, ES cells will becapable of colonizing all its tissues leading eventuallyto the birth of chimeric offspring. In case the engi-neered ES cells have contributed to germ cells, thedesired genetic modification will be transmitted to fu-ture generations (for an excellent hands-on gene-tar-geting manual see Refs. 53 and 64). With this tech-nique, it has been possible to delete gene function(knockout), to insert novel selected genes or DNA frag-ments into a given locus (“knock-in”), to introducesubtle changes like point mutations, and to producechromosomal rearrangements in the germ line of themouse.

Conventional gene-targeting experiments can be de-signed to introduce the desired genetic changes intothe germ line, thus affecting all tissues during theentire lifespan of the resulting mouse. Several limita-tions of this germ line gene-targeting approach areobvious. As the mutation will be already present in the

first developing cell, an embryonic lethal phenotypemight be provoked, precluding any further functionalanalysis during adulthood. This is particularly true forclassic gene knockouts of developmentally importantgenes. Nearly as undesirable as embryonic lethal phe-notypes are pleiotropic effects often observed as a com-pensatory reaction to the introduced germ line muta-tion obscuring or preventing a clear-cut analysis.Moreover, the selected gene might have a wide expres-sion pattern, and its general invalidation might thusinduce a highly complicated accumulative pheno-type involving multiple tissues. Finally, it might beimportant to only extinguish gene function at a specificdevelopmental time point or during a particularstage in disease. For all these reasons it can be ofimportance to create mouse models allowing the dele-tion of genetic material/gene function in selected cellsat a specific time. In this manner, one can efficientlyavoid embryonic lethality, prevent unwanted pleiotro-pic side effects and exclude accumulative compensa-tory developmental changes starting from the earliestdevelopmental stage. The use of site-specific recombi-nases of the Cre and Flp �-integrase family and therecent development of the first reversibly switchableknockout models have opened an exciting new avenuefor creating conditional loss or gain of function mousemodels.

Cre and Flp recombinases. The first site-specific re-combination system used for introducing conditionallyinducible genetic changes in the mouse was the Cre-loxsystem (78, 102). Based on the enzymatic activity of theCre (for “causes recombination”) recombinase, origi-nally isolated from the bacteriophage P1, it is possibleto induce site-specific genetic recombination to se-quences flanked by so-called loxP sites (“locus of cross-over P1”) (49, 52). In a similar fashion the S. cerevisiaerecombinase Flp will cause recombination betweenFRT (for “Flp recombination target”) recognition sites(70). LoxP and FRT recognition sequences are 34-bplong DNA motifs consisting of an 8-bp core regionflanked by palindromic sequences of 13 bp. As shown inFig. 5, Cre and Flp recombinases mediate differenteffects on their DNA target sequences depending ex-clusively on the orientation of the specific recognitionsequence and the number of different DNA moleculesinvolved. Cre- and Flp-induced changes include se-quence excision, duplication, integration, inversion,and chromosomal translocation. All of these are ex-tremely useful for tailoring the mouse genome, but asCre and Flp recombinases have been most extensivelyused for small scale DNA rearrangements affectingsingle genes, we will focus in this review on conditionalgene knockout strategies. Proof-of-principle experi-ments have shown that Cre and Flp recombinases willefficiently work in any cellular environment and onany kind of DNA target. In addition, the Cre/loxP andthe Flp/FRT systems do not need any additional cofac-tors or sequence requirements for site-specific recom-bination in the mouse. Further reading about Cre- andFlp-based strategies for generating chromosomal rear-rangements including large deletions and transloca-

123INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


tions is provided by several excellent reviews (77, 80,98, 164).

Conditional Cre and Flp gene knockout. CombiningES cell technology with Cre recombinase action, pio-neering work from the laboratory of Klaus Rajewsky(47) demonstrated the exquisite potential for creatinginducible mouse models. Normally, inducible modifica-tion of a specific gene is achieved by homologous re-combination in ES cells flanking the sequence to beexcised by loxP or FRT sites (such an allele is called“floxed” for “flanked by loxP,” or “flrted” in the case ofusing FRT sites). For creating conditional null mu-tants, it is essential to introduce recombinase recogni-tion sites around an essential part of the gene of inter-est without compromising its functionality. In an idealscenario this is achieved by flanking the entire codingregion by loxP/FRT sites. As this may be difficult, itoften is sufficient to simply position the loxP/FRT se-quences in the intronic sequence around an essentialexon/essential exons or to conditionally target the pro-moter region of the gene. Figure 6A gives an example ofhow such a floxed locus might look. In any case, acareful positioning of the chosen recombinase recogni-tion sites is vital for the proper outcome of the experi-ment. Depending on the site of recombination, it ispossible that insertion of loxP/FRT sites might directlyinterfere with important transcriptional regulatory el-ements leading to improper expression of the targetedgene in the noninduced state. Furthermore, care has tobe taken that recombinase-mediated excision of theallele will not eventually produce a gene product withresidual or altered function. Lastly, to avoid unwantedtranscriptional interference with the natural tran-scriptional program of the targeted locus, it will be

important to remove the selectable marker (for exam-ple, a neomycin resistance expression cassette) afterthe chosen allele has been targeted. This can be ar-chived by the use of three loxP sites in the targetingvector, two of which are flanking the marker and thethird placed in a location that after removal of themarker the two remaining loxP sites are positionedaround the part to be removed (47). Alternatively, acombination of loxP sites and FRT sites can be used toselectively remove the resistance marker from the tar-geted locus (95). Bearing these basic safety measuresin mind when designing the targeting vector will max-imize the chances of creating the desired well-definedgenetic changes in the planned conditional knockoutmouse model.

In many circumstances, Cre or Flp recombinationwill be used for the generation of lineage/cell type-specific ablation of gene function. To this end, micederived from Cre or Flp targeted ES cells can becrossed with other “deleter” mice expressing Cre or Flpunder the control of a tissue- or cell type-specific pro-moter thus allowing the time- or tissue-restricted ex-cision of the floxed or flrted segment. Figure 6B depictsboth ubiquitous as well as tissue-specific gene inacti-vation. With this strategy, the outcome of any recom-binase-mediated conditional knockout experiment willdepend very much on the individual deleter strain

Fig. 6. Ubiquitous or tissue-specific recombinase-mediated gene in-activation. A: LoxP site flanked gene locus exhibiting wild-type genefunction prior to recombinase excision. B: general deletion of exon 1using a ubiquitous promoter to drive Cre expression (top) and tissue-specific exon 1 deletion resulting from directed expression of Cre inliver (bottom). P, locus-specific promoter; LoxP, Cre-recognition sites(solid triangles); Pub, ubiquitous promoter; Pliv, liver-specific pro-moter; Cre, Cre recombinase.

Fig. 5. Possible genetic modifications mediated by Cre or Flp recom-binases. Both Cre and Flp recombinases can be used for introducingdifferent molecular alterations into the mouse genome. A: excision ofgenomic DNA can be mediated by flanking the sequence to be excisedby unidirectional recombination sites. B: inversion requires the se-quence of interest to be flanked by bidirectional recombination sites.C: a single recombination site within the mouse genome can be usedas a molecular acceptor for inserting a DNA sequence containing asecond recombination site. D: interchromosomal translocations canbe engineered by placing single recombination sites on differentchromosomes. Solid triangles indicate recombination sites. Whitearrow indicates orientation of the sequence to be excised.

124 INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


expected to express the recombinase at a high enoughlevel and specificity to introduce the desired alteration.A common problem with many deleter lines is that theywill induce incomplete excision of the targeted alleleleading to mosaicism of the mouse (145). As the ana-lyzed phenotype might directly reflect the ratio of de-leted to wild-type cells, a precise interpretation of theeffects might be difficult or impossible. In addition, theoften observed lack of control over the spatiotemporaltiming of recombination is in many cases due to thenature or leakiness of the promoter and will lead toinappropriate excision of the targeted allele in celltypes other than those desired (27). Conversely, formany applications, including those for studying cancerin the mouse, partial inactivation of gene functionmight be favorable (115, 142).

Overlaying promoter-specific spatiotemporal activitywith separately inducible activation of a recombinasepromises to further enhance the potential of site-spe-cific recombination. Over the past years several labo-ratories have developed systems for exogenously con-trolling Cre activity in mice. These include the controlof Cre recombinase activity itself by a synthetic ligandand ligand-dependent transcriptional activation of Creexpression.

The first inducible Cre mouse made use of the Mx1promoter to drive expression of Cre (76). Since tran-scriptional activation of the Mx1 promoter is directlydependent on interferon-�, interferon-�, or double-stranded polyinosinic-polycytidilic nucleic acids, ex-pression of Cre can be exogenously controlled. How-ever, the use of this Mx1-Cre deleter strain wasessentially limited to excision of alleles in the liver andthe immune system since recombination in other tis-sues was poor (76, 114). As an alternative severalgroups have developed combined tet-Cre mice allowingCre expression to be tightly controlled through the teton/off system. Using DOX-inducible expression of Cre,researchers have reported efficient excision of floxedalleles in the brain and in the intestine (51, 86, 118). Inaddition, Utomo and colleagues (140) demonstratedtight tissue-specific expression of Cre in transgenicmice with an approach that combined tetO-dependentCre transcription with rtTA tetracycline transactivatorexpression on the same construct thus greatly facilitat-ing breeding schemes.

A very different but also effective strategy to exog-enously control Cre activity in mice has recently beendeveloped. This strategy required the fusion of Creproteins to progesterone or ecdysone receptor ligandbinding domains (see section on Hormone Receptor-Based Transgenics). Proof-of-principle experimentshave shown the versatility of these ligand-dependentrecombinase fusions, and several groups have adaptedthis strategy for temporal control of Cre activity in themouse (93, 124, 145). The first brand of inducible Crerecombinases employs a truncated progesterone recep-tor which is directly fused to the Cre recombinase (96,157), whereas the second approach makes use of afusion between the Cre recombinase and an estrogenreceptor (48, 93, 94, 116, 124, 127). Nevertheless, close

inspection of the induced recombination frequenciesclearly demonstrated that recombination mediated bysome first-generation fusion-recombinases might bemosaic depending very much on the tissue type (13, 24,67, 124, 143, 145). It is therefore to be assumed thatbetter inducible recombinase molecules are needed toefficiently regulate Cre or Flp activity in the mouse (57,59, 85, 157).

Protein engineering has also led to the availability ofimproved second-generation recombinases with novelproperties. These changes include codon-improved Creand Flp recombinases with better in vivo efficiencies(15, 74, 113, 127), engineered recombinases with novelrecognition sites (16, 122), and Cre recombinases capa-ble of permeating cell membranes (62, 103).

Despite all existing limitations and the irreversiblenature of recombinase-induced genetic changes, Cre-and Flp-based conditional knockout mice have alreadyproven to be a reliable tool for generating conditionalmouse models. Investigators considering inducibleCre- and Flp-based mouse models should visit thehttp://www.mshri.on.ca/nagy/cre.htm database whichcontains an excellent compilation of general and tis-sue-restricted deleter strains including also a usefullist of different Cre and Flp reporter mice.

Switchable gene knockout. In an ideal mouse modelthe exogenous control over the knockout phenotypeshould be reversible. Such a completely “switchable”knockout mouse has significant advantages over “one-way” inducible systems allowing the in vivo analysis ofgene function (or loss of gene function) during preciselydefined developmental periods and at critical stages ofdisease. So far, two different groups have been success-ful in generating such switchable knockout pheno-types. Both approaches made use of restoring genefunction by virtue of ligand-mediated overexpression ofthe switchable gene in an otherwise null background.

One example of a reversibly switchable knockoutphenotype in the mouse was the rescue of the �-inhibinnull mouse (108). When the inhibin �-subunit is de-leted from the germ line, inhibin-deficient mice developgonadal sex cord-stromal tumors with a penetrance of100% and ultimately die of a cachexia-like syndrome(91, 92). To be able to exogenously control the expres-sion of �-inhibin in transgenic mice, a progesterone-based binary mouse model was established. In thisbi-transgenic mouse, expression of �- and �-inhibin inthe liver was strictly dependent on the presence of theinducing ligand mifepristone. By crossing these bi-transgenic mice into an �-inhibin null background, itwas possible to completely rescue the lethal phenotypeand to also prevent the onset of gonadal sex cord-stromal tumors. Additional experiments with theseconditional �-inhibin knockout mice further demon-strated that the system is reversible, as “switching off”�-inhibin gene function did lead to gonadal tumorigen-esis. Interestingly, the �-inhibin hormone is not nor-mally produced in the liver but transgenic expressionof �-inhibin in hepatic tissues fully rescued the pheno-type. This raises important concerns about limitationsand general feasibility of switchable knockout mouse

125INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


models. Does the induced or rescued phenotype dependon the level of expressed transgene product? How im-portant is the site of transgene expression? How is itpossible to exactly direct transgene transcription to theright cells at the right time, thus recapitulating thenormal spatiotemporal expression pattern of the en-dogenous gene?

To some extent these important questions have beenaddressed using a tet on/off based approach to generatea completely switchable knockout mouse for the G-protein-coupled Endothelin receptor B (Ednrb) (128).To ensure that switchable expression of the Ednrbgene is regulated in a spatiotemporally appropriatemanner, a tet-inducible transactivator has been inte-grated into the endogenous Ednrb locus. With thisknock-in strategy two essential objectives are met.First, expression of the inserted tetracycline-depen-dent transactivator will closely mirror expression ofthe endogenous Ednrb gene; second, the knock-in leadsto complete inactivation of one allele of the Ednrb gene(knock-in effector). A further functional requirementfor this approach is a tetO-CMV responder expressing

Ednrb in a DOX-dependent fashion. Shin and col-leagues (128) introduced this DOX-responsive elementinto the Ednrb locus which will equally lead to amono-allelic inactivation of the endogenous Ednrb lo-cus (compare wild-type locus in Fig. 7A with theswitchable Ednrb locus in Fig. 7B). Crossing of theknock-in effector with the knock-in responder shouldlead to mice harboring both the tet-responder andtet-effector in an otherwise Ednrb null background.Analysis of these mice demonstrated that DOX-in-duced Ednrb expression could fully rescue the nullphenotype and that DOX-mediated repression of Ed-nrb transcription recapitulated the null phenotype (seeFig. 7C for DOX-dependent regulation of Ednrb ex-pression). Moreover, these experiments identified therestricted time periods during which Ednrb gene func-tion is essential for the development of melanoblastsand enteric neurons. This pioneering and elegant ex-perimental approach is not only a good example forworkable conditional mouse models but also illustratesthe enormous potential of binary systems for generat-

Fig. 7. An example of switchable gene inactivation-endothelin receptor B conditional knockout. A: wild-type locus; endothelin receptor B (Ednrb) single exonlocus. B: switchable locus; the engineered Ednrb locusconsists of a knocked-in Tet transactivator (tTA) bothsubstituting and inactivating its first allele (effector). Inaddition, a conditional Ednrb responder knock-in cas-sette replaces the second allele. This “knock-in/knock-out” strategy results in completely switchable and re-versible expression of Ednrb in an otherwise Ednrb nullbackground. C: regulation of Ednrb gene expression isentirely dependent on administration of DOX. Knock-inof the tTA cDNA into the Ednrb locus places the ex-pression of tTA under transcriptional control of theEdnrb locus leading to the precise recapitulation ofEdnrb expression by tTA. In the presence of DOX, noEdnrb is expressed from the responder cassette locatedwithin the second Ednrb allele. However, in the ab-sence of DOX, Ednrb expression will be resumed. DOX,doxycycline (brown box); Pednrb, Ednrb promoter; tTA,tetracycline-dependent transactivator; TRE, Tet-re-sponsive element.

126 INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


ing completely switchable and appropriately spatio-temporally regulated knockout models.

Most of the inducible strategies discussed abovehave been already applied for research in physiology. Ageneral overview as to how inducible mouse modelsserve as experimental tools for answering many of theopen questions in diverse research areas such as can-cer, metabolic/endocrine dysfunction, cardiovasculardisease, and behavioral physiology is outlined in Fig. 8.Applied to experimental physiology, it is clear thatswitchable mouse models will have a huge advantageover conventional strategies, as they allow for the tightexternal and in many cases also reversible control of asingle gene at a certain time in development or in acertain cell type. For this reason the modeled diseasecondition will often mirror the clinical situation moreappropriately. Very good examples illustrating the

power of inducible and reversible mouse models forasking fundamental questions in cancer research havebeen provided by several groups (21, 25, 37, 55). All ofthese murine models provided a means to assess thephysiological impact of oncogenes in tumorigenesis andfor the first time to directly examine the consequencesof their extinction after the onset of tumor formation.These model systems have important consequences forany anti-cancer drug development. In this respect ex-periments performed by Chin and colleagues (21) usinga tet-inducible RAS melanoma mouse model (Fig. 8A)showed clearly that extinction of H-RASVAL12 led toregression of tumors in the skin, thus confirming mu-tant RAS as a principle therapeutic target.

New insights into metabolic/endocrine physiologycan also be gained by using inducible mouse models.One of the many published studies illustrates that

Fig. 8. How can conditional mouse models be applied to research in physiology? Four experimental approachesrepresenting different research areas in physiology are presented to illustrate the power of conditional mousemodels. A: cancer physiology. Conditional expression of the RAS oncogene in skin induces formation of melanomas.After extinction of oncogenic RAS progression of the tumor growth stops and is followed by a complete regression(21). B: metabolic/endocrine dysfunction. Tissue-specific disruption of the nuclear gene encoding the mitochondrialtranscription factor A (mtTFA) sets off diabetes (130). C: cardiovascular disease. Cardiac fibrosis arises afterknocking-down the mineralocorticoid receptor (MR) in cardiomyocytes. Surprisingly, restoring the expression ofMR protein was sufficient to completely reverse cardiopathy (7). D: brain dysfunction and learning/memory. Toremember or not to remember was the question asked in this spatial and visual memory test. Behavioralexperiments carried out in the laboratory of Eric Kandel demonstrated that long-term memory is lost afterexpression of a truncated dominant negative form of calcineurin in the hippocampus of trained mice. However, theability to precisely remember the previously learned tasks can be completely restored after extinction of thedominant negative calcineurin (89). Readers interested in exploring further applications using second-generationmouse models in physiology are referred to several additional but more specialized reviews for cardiovasculardisease (112), for neurobiology (88, 159), for endocrinology/metabolism (117), and for cancer (60, 63).

127INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


tissue-restricted disruption of the nuclear gene encod-ing the mitochondrial transcription factor A (mtTFA)induced diabetes in mice. In this very elegant studySilva and colleagues (130) were able to show thatdepletion of mtTFA in pancreatic �-cells not only reca-pitulated the human mitochondrial disease but pro-vided direct evidence for a critical role of the respira-tory chain in insulin secretion. This tissue-specificmurine diabetes model provides an important tool tofurther dissect the molecular mechanisms responsiblefor human mitochondrial diabetes and opens the pos-sibility for testing novel therapeutic approaches in thecontext of an appropriate animal model (Fig. 8B).

In the area of cardiovascular disease new possibili-ties for asking old questions and developing novel ther-apeutic approaches have also emerged. In a recentexample of how reversible transgene expression can beapplied to gain insight into cardiac disease, Beggahand colleagues (7) developed a reversible mouse modelfor cardiac fibrosis and heart failure. By conditionallyknocking down the expression of the mineralocorticoidreceptor (MR) in cardiomyocytes, they could inducecardiac hypertrophy, ventricular dysfunction, intersti-tial fibrosis, and finally heart failure. Most interest-ingly, when expression of the MR protein was restored,not only was improvement of cardiac function ob-served, but surprisingly the cardiomyopathy re-gressed. This finding identified MR as a candidatetherapeutic target, demonstrating for the first time itsrole in active remodeling of abnormal extracellularmatrix depositions.

Last but not least, the area of behavioral neuro-physiology has also considerably benefited from thedevelopment of second-generation conditional mousemodels. For example, to assess the role calcineurinplays in the transition from short- to long-term mem-ory, Mansuy and colleagues (89) trained mice in amaze to learn specific spatial and visual recognitiontasks. Induced overexpression of a truncated domi-nant negative form of calcineurin in the hippocam-pus of the trained cohort of mice disabled these miceto remember previously learned tasks without affect-ing their short-term memory. Most strikingly, whenexpression of the dominant negative calcineurin wasextinguished in these mice, long-term memory of thepreviously learned spatial and visual tasks was com-pletely restored.

As illustrated by the above examples, the use ofsecond-generation mouse models together with the to-be-expected further improvement of these technologieswill greatly enhance our possibilities to appropriatelyaddress many open but principal questions about thein vivo function of genes in physiology and humandisease.

Concluding Remarks

In this review we have summarized the rapidly ex-panding number of methods to generate spatially andtemporally regulated gene expression within themouse. These massive advances as well as the incre-

mental improvements in established systems presentresearchers with very powerful tools for analyzing genefunction during development, as well as in disease.Today mouse models for a variety of monogenic disor-ders solely including the modification of a single genehave been generated. However, as many developmen-tal and disease models require the coordinate action oftwo or more genes, the combination of different switch-able systems in one mouse will become increasinglypopular. While the technologies presented here arevery powerful, it is to be expected that RNA interfer-ence-based approaches (31) as well as cloning of mice(146) will soon be additional tools to be applied forbuilding mouse models.

With many groups producing a rapidly expandingnumber of tet on/off and also Cre/loxP conditional mice,any improvements in the communication and sharingof available strains will be of great help to the scientificcommunity. Toward this end, we now present a com-prehensive list of currently available general and tis-sue-specific tet on/off effector and reporter strains.This compilation can be accessed via http://www.zmg.uni-mainz.de/tetmouse/. If there are any tet on/offstrains we unintentionally overlooked or which arecurrently being tested, we would be delighted to alsoinclude these upon notification. As mentioned, an ex-cellent list of Cre- and Flp-based conditional strains ispublished elsewhere (http://www.mshri.on.ca/nagy/cre.htm).

One very attractive use for the methodologies de-scribed above is the genetic marking and tracing ofspecific cellular populations for fate mapping analysisduring development. With the limitations presented byin utero mammalian development (compared to usingin ovo developmental models), fate analysis of progen-itors is currently technically challenging. We envisagerecombinase-mediated cell marking of specific popula-tions such as hematopoietic precursors or stem cells,neural progenitors or germ cells as a very powerful toolfor following cell fates during development (156, 160).

In the future to come, it will be extremely exciting tosee how the further development of conditional mousetechnology will transform biomedical research in gen-eral.

We thank members of the Bockamp and Zabel laboratories, in-cluding Rosario Heck, Cerstin Christel, Svetlana Ohngemach, An-gelique Renzaho, Tatjana Trost, Dirk Reutzel, Dirk Prawitt, andEkkehart Lausch, for helpful comments in the preparation of thisreview. We also wish to thank Jens Rudolph and Marcel Engel forimportant help concerning the tet mouse base.

We apologize that many significant contributions to this field mayhave been omitted from this review.

Work in the Bockamp and Zabel laboratories is supported by theDeutsche Forschungsgemeinschaft, the European Union, the Stif-tung Rheinland-Pfalz for Innovation, and the Sander Stiftung forCancer Research.

REFERENCES

1. Akagi K, Kanai M, Saya H, Kozu T, and Berns A. A noveltetracycline-dependent transactivator with E2F4 transcrip-tional activation domain. Nucleic Acids Res 29: E23, 2001.

2. Albanese C, Reutens AT, Bouzahzah B, Fu M, D’Amico M,Link T, Nicholson R, Depinho RA, and Pestell RG. Sus-

128 INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


tained mammary gland-directed, ponasterone A-inducible ex-pression in transgenic mice. FASEB J 14: 877–884, 2000.

3. Baron U and Bujard H. Tet repressor-based system for reg-ulated gene expression in eukaryotic cells: principles and ad-vances. Methods Enzymol 327: 401–421, 2000.

4. Baron U, Freundlieb S, Gossen M, and Bujard H. Co-regulation of two gene activities by tetracycline via a bidirec-tional promoter. Nucleic Acids Res 23: 3605–3606, 1995.

5. Baron U, Gossen M, and Bujard H. Tetracycline-controlledtranscription in eukaryotes: novel transactivators with gradedtransactivation potential. Nucleic Acids Res 25: 2723–2729,1997.

6. Baron U, Schnappinger D, Helbl V, Gossen M, Hillen W,and Bujard H. Generation of conditional mutants in highereukaryotes by switching between the expression of two genes.Proc Natl Acad Sci USA 96: 1013–1018, 1999.

7. Beggah AT, Escoubet B, Puttini S, Cailmail S, Delage V,Ouvrard-Pascaud A, Bocchi B, Peuchmaur M, DelcayreC, Farman N, and Jaisser F. Reversible cardiac fibrosis andheart failure induced by conditional expression of an antisensemRNA of the mineralocorticoid receptor in cardiomyocytes.Proc Natl Acad Sci USA 99: 7160–7165, 2002.

8. Berger SL, Pina B, Silverman N, Marcus GA, Agapite J,Regier JL, Triezenberg SJ, and Guarente L. Genetic isola-tion of ADA2: a potential transcriptional adaptor required forfunction of certain acidic activation domains. Cell 70: 251–265,1992.

9. Bieberich CJ, King CM, Tinkle BT, and Jay G. A trans-genic model of transactivation by the Tax protein of HTLV-I.Virology 196: 309–318, 1993.

10. Borrelli E, Heyman R, Hsi M, and Evans RM. Targeting ofan inducible toxic phenotype in animal cells. Proc Natl Acad SciUSA 85: 7572–7576, 1988.

11. Bowers BJ, Owen EH, Collins AC, Abeliovich A,Tonegawa S, and Wehner JM. Decreased ethanol sensitivityand tolerance development in gamma-protein kinase C nullmutant mice is dependent on genetic background. Alcohol ClinExp Res 23: 387–397, 1999.

12. Braselmann S, Graninger P, and Busslinger M. A selectivetranscriptional induction system for mammalian cells based onGal4-estrogen receptor fusion proteins. Proc Natl Acad Sci USA90: 1657–1661, 1993.

13. Brocard J, Warot X, Wendling O, Messaddeq N, VoneschJL, Chambon P, and Metzger D. Spatio-temporally con-trolled site-specific somatic mutagenesis in the mouse. ProcNatl Acad Sci USA 94: 14559–14563, 1997.

14. Bronson SK, Plaehn EG, Kluckman KD, Hagaman JR,Maeda N, and Smithies O. Single-copy transgenic mice withchosen-site integration. Proc Natl Acad Sci USA 93: 9067–9072, 1996.

15. Buchholz F, Angrand PO, and Stewart AF. Improved prop-erties of FLP recombinase evolved by cycling mutagenesis. NatBiotechnol 16: 657–662, 1998.

16. Buchholz F and Stewart AF. Alteration of Cre recombinasesite specificity by substrate-linked protein evolution. Nat Bio-technol 19: 1047–1052, 2001.

17. Campbell SJ, Carlotti F, Hall PA, Clark AJ, and Wolf CR.Regulation of the CYP1A1 promoter in transgenic mice: anexquisitely sensitive on-off system for cell specific gene regula-tion. J Cell Sci 109: 2619–2625, 1996.

18. Canfield V, West AB, Goldenring JR, and Levenson R.Genetic ablation of parietal cells in transgenic mice: a newmodel for analyzing cell lineage relationships in the gastricmucosa. Proc Natl Acad Sci USA 93: 2431–2435, 1996.

19. Carson-Jurica MA, Schrader WT, and O’Malley BW. Ste-roid receptor family: structure and functions. Endocr Rev 11:201–220, 1990.

20. Chen YT and Bradley A. A new positive/negative selectablemarker, pu�tk, for use in embryonic stem cells. Genesis 28:31–35, 2000.

21. Chin L, Tam A, Pomerantz J, Wong M, Holash J, Bard-eesy N, Shen Q, O’Hagan R, Pantginis J, Zhou H, HornerJW II, Cordon-Cardo C, Yancopoulos GD, and DePinho

RA. Essential role for oncogenic Ras in tumour maintenance.Nature 400: 468–472, 1999.

22. Cronin CA, Gluba W, and Scrable H. The lac operator-repressor system is functional in the mouse. Genes Dev 15:1506–1517, 2001.

23. Cvetkovic B, Yang B, Williamson RA, and Sigmund CD.Appropriate tissue- and cell-specific expression of a single copyhuman angiotensinogen transgene specifically targeted up-stream of the HPRT locus by homologous recombination. J BiolChem 275: 1073–1078, 2000.

24. Danielian PS, Muccino D, Rowitch DH, Michael SK, andMcMahon AP. Modification of gene activity in mouse embryosin utero by a tamoxifen-inducible form of Cre recombinase.Curr Biol 8: 1323–1326, 1998.

25. D’Cruz CM, Gunther EJ, Boxer RB, Hartman JL, Sinta-sath L, Moody SE, Cox JD, Ha SI, Belka GK, Golant A,Cardiff RD, and Chodosh LA. c-MYC induces mammarytumorigenesis by means of a preferred pathway involving spon-taneous Kras2 mutations. Nat Med 7: 235–239, 2001.

26. Dinkel A, Aicher WK, Warnatz K, Burki K, Eibel H, andLedermann B. Efficient generation of transgenic BALB/c miceusing BALB/c embryonic stem cells. J Immunol Methods 223:255–260, 1999.

27. Dobie K, Mehtali M, McClenaghan M, and Lathe R. Var-iegated gene expression in mice. Trends Genet 13: 127–130,1997.

28. Doetschman T, Gregg RG, Maeda N, Hooper ML, MeltonDW, Thompson S, and Smithies O. Targetted correction of amutant HPRT gene in mouse embryonic stem cells. Nature 330:576–578, 1987.

29. Dzierzak E, Daly B, Fraser P, Larsson L, and Muller A.Thy-1 tk transgenic mice with a conditional lymphocyte defi-ciency. Int Immunol 5: 975–984, 1993.

30. Efrat S, Leiser M, Wu YJ, Fusco-DeMane D, Emran OA,Surana M, Jetton TL, Magnuson MA, Weir G, and Fleis-cher N. Ribozyme-mediated attenuation of pancreatic �-cellglucokinase expression in transgenic mice results in impairedglucose-induced insulin secretion. Proc Natl Acad Sci USA 91:2051–2055, 1994.

31. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, WeberK, and Tuschl T. Duplexes of 21-nucleotide RNAs mediateRNA interference in cultured mammalian cells. Nature 411:494–498, 2001.

32. Erickson RP. Antisense transgenics in animals. Methods 18:304–310, 1999.

33. Evans MJ, Carlton MB, and Russ AP. Gene trapping andfunctional genomics. Trends Genet 13: 370–374, 1997.

34. Evans MJ and Kaufman MH. Establishment in culture ofpluripotential cells from mouse embryos. Nature 292: 154–156,1981.

35. Evans RM. The steroid and thyroid hormone receptor super-family. Science 240: 889–895, 1988.

36. Evans V, Hatzopoulos A, Aird WC, Rayburn HB, Rosen-berg RD, and Kuivenhoven JA. Targeting the Hprt locus inmice reveals differential regulation of Tie2 gene expression inthe endothelium. Physiol Genomics 2: 67–75, 2000.

37. Ewald D, Li M, Efrat S, Auer G, Wall RJ, Furth PA, andHennighausen L. Time-sensitive reversal of hyperplasia intransgenic mice expressing SV40 T antigen. Science 273: 1384–1386, 1996.

38. Festenstein R and Kioussis D. Locus control regions andepigenetic chromatin modifiers. Curr Opin Genet Dev 10: 199–203, 2000.

39. Forster K, Helbl V, Lederer T, Urlinger S, Wittenburg N,and Hillen W. Tetracycline-inducible expression systems withreduced basal activity in mammalian cells. Nucleic Acids Res27: 708–710, 1999.

40. Freundlieb S, Schirra-Muller C, and Bujard H. A tetracy-cline controlled activation/repression system with increasedpotential for gene transfer into mammalian cells. J Gene Med 1:4–12, 1999.

41. Fukumura D, Xavier R, Sugiura T, Chen Y, Park EC, LuN, Selig M, Nielsen G, Taksir T, Jain RK, and Seed B.

129INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


Tumor induction of VEGF promoter activity in stromal cells.Cell 94: 715–725, 1998.

42. Furth PA, St Onge L, Boger H, Gruss P, Gossen M, Kist-ner A, Bujard H, and Hennighausen L. Temporal control ofgene expression in transgenic mice by a tetracycline-responsivepromoter. Proc Natl Acad Sci USA 91: 9302–9306, 1994.

43. Gardner DP, Byrne GW, Ruddle FH, and Kappen C. Spa-tial and temporal regulation of a lacZ reporter transgene in abinary transgenic mouse system. Transgenic Res 5: 37–48,1996.

44. Gordon JW, Scangos GA, Plotkin DJ, Barbosa JA, andRuddle FH. Genetic transformation of mouse embryos bymicroinjection of purified DNA. Proc Natl Acad Sci USA 77:7380–7384, 1980.

45. Gossen M and Bujard H. Tight control of gene expression inmammalian cells by tetracycline-responsive promoters. ProcNatl Acad Sci USA 89: 5547–5551, 1992.

46. Gossen M, Freundlieb S, Bender G, Muller G, Hillen W,and Bujard H. Transcriptional activation by tetracyclines inmammalian cells. Science 268: 1766–1769, 1995.

47. Gu H, Zou YR, and Rajewsky K. Independent control ofimmunoglobulin switch recombination at individual switch re-gions evidenced through Cre-loxP-mediated gene targeting.Cell 73: 1155–1164, 1993.

48. Guo C, Yang W, and Lobe CG. A Cre recombinase transgenewith mosaic, widespread tamoxifen-inducible action. Genesis32: 8–18, 2002.

49. Hamilton DL and Abremski K. Site-specific recombinationby the bacteriophage P1 lox-Cre system. Cre-mediated synapsisof two lox sites. J Mol Biol 178: 481–486, 1984.

50. Hardouin SN and Nagy A. Mouse models for human disease.Clin Genet 57: 237–244, 2000.

51. Hasan MT, Schonig K, Berger S, Graewe W, and BujardH. Long-term, noninvasive imaging of regulated gene expres-sion in living mice. Genesis 29: 116–122, 2001.

52. Hoess RH, Ziese M, and Sternberg N. P1 site-specific recom-bination: nucleotide sequence of the recombining sites. ProcNatl Acad Sci USA 79: 3398–3402, 1982.

53. Hogan B, Beddington R, Constantini F, and Lacy E. Ma-nipulating the Mouse Embryo: a Laboratory Manual. ColdSpring Harbor, NY: Cold Spring Harbor Laboratory, 1994.

54. Hoppe UC, Marban E, and Johns DC. Adenovirus-mediatedinducible gene expression in vivo by a hybrid ecdysone receptor.Mol Ther 1: 159–164, 2000.

55. Huettner CS, Zhang P, Van Etten RA, and Tenen DG.Reversibility of acute B-cell leukaemia induced by BCR-ABL1.Nat Genet 24: 57–60, 2000.

56. Hyder SM, Nawaz Z, Chiappetta C, Yokoyama K, andStancel GM. The protooncogene c-jun contains an unusualestrogen-inducible enhancer within the coding sequence. J BiolChem 270: 8506–8513, 1995.

57. Imai T, Jiang M, Chambon P, and Metzger D. Impairedadipogenesis and lipolysis in the mouse upon selective ablationof the retinoid X receptor alpha mediated by a tamoxifen-inducible chimeric Cre recombinase (Cre-ERT2) in adipocytes.Proc Natl Acad Sci USA 98: 224–228, 2001.

58. Imhof MO, Chatellard P, and Mermod N. A regulatorynetwork for the efficient control of transgene expression. J GeneMed 2: 107–116, 2000.

59. Indra AK, Warot X, Brocard J, Bornert JM, Xiao JH,Chambon P, and Metzger D. Temporally-controlled site-specific mutagenesis in the basal layer of the epidermis: com-parison of the recombinase activity of the tamoxifen-inducibleCre-ER(T) and Cre-ER(T2) recombinases. Nucleic Acids Res 27:4324–4327, 1999.

60. Jackson-Grusby L. Modeling cancer in mice. Oncogene 21:5504–5514, 2002.

61. Jacob F and Monod J. Genetic regulatory mechanisms in thesynthesis of proteins. J Mol Biol 3: 318–356, 1961.

62. Jo D, Nashabi A, Doxsee C, Lin Q, Unutmaz D, Chen J,and Ruley HE. Epigenetic regulation of gene structure andfunction with a cell-permeable Cre recombinase. Nat Biotechnol19: 929–933, 2001.

63. Jonkers J and Berns A. Conditional mouse models of spo-radic cancer. Nat Rev Cancer 2: 251–265, 2002.

64. Joyner A. Gene Targeting: a Practical Approach. Oxford: IRLPress, Oxford, 1993.

65. Kantachuvesiri S, Fleming S, Peters J, Peters B, BrookerG, Lammie AG, McGrath I, Kotelevtsev Y, and Mullins JJ.Controlled hypertension, a transgenic toggle switch revealsdifferential mechanisms underlying vascular disease. J BiolChem 276: 36727–36733, 2001.

66. Kashani-Sabet M, Liu Y, Fong S, Desprez PY, Liu S, Tu G,Nosrati M, Handumrongkul C, Liggitt D, Thor AD, andDebs RJ. Identification of gene function and functional path-ways by systemic plasmid-based ribozyme targeting in adultmice. Proc Natl Acad Sci USA 99: 3878–3883, 2002.

67. Kellendonk C, Tronche F, Casanova E, Anlag K, OpherkC, and Schutz G. Inducible site-specific recombination in thebrain. J Mol Biol 285: 175–182, 1999.

68. Keyvani K, Baur I, and Paulus W. Tetracycline-controlledexpression but not toxicity of an attenuated diphtheria toxinmutant. Life Sci 64: 1719–1724, 1999.

69. Khillan JS, Deen KC, Yu SH, Sweet RW, Rosenberg M,and Westphal H. Gene transactivation mediated by the TATgene of human immunodeficiency virus in transgenic mice.Nucleic Acids Res 16: 1423–1430, 1988.

70. Kilby NJ, Snaith MR, and Murray JA. Site-specific recom-binases: tools for genome engineering. Trends Genet 9: 413–421, 1993.

71. Kioussis D and Festenstein R. Locus control regions: over-coming heterochromatin-induced gene inactivation in mam-mals. Curr Opin Genet Dev 7: 614–619, 1997.

72. Kistner A, Gossen M, Zimmermann F, Jerecic J, UllmerC, Lubbert H, and Bujard H. Doxycycline-mediated quanti-tative and tissue-specific control of gene expression in trans-genic mice. Proc Natl Acad Sci USA 93: 10933–10938, 1996.

73. Kontgen F, Suss G, Stewart C, Steinmetz M, and Blueth-mann H. Targeted disruption of the MHC class II Aa gene inC57BL/6 mice. Int Immunol 5: 957–964, 1993.

74. Koresawa Y, Miyagawa S, Ikawa M, Matsunami K,Yamada M, Shirakura R, and Okabe M. Synthesis of a newCre recombinase gene based on optimal codon usage for mam-malian systems. J Biochem (Tokyo) 127: 367–372, 2000.

75. Krestel HE, Mayford M, Seeburg PH, and Sprengel R. AGFP-equipped bidirectional expression module well suited formonitoring tetracycline-regulated gene expression in mouse.Nucleic Acids Res 29: E39, 2001.

76. Kuhn R, Schwenk F, Aguet M, and Rajewsky K. Induciblegene targeting in mice. Science 269: 1427–1429, 1995.

77. Kwan KM. Conditional alleles in mice: practical considerationsfor tissue-specific knockouts. Genesis 32: 49–62, 2002.

78. Lakso M, Sauer B, Mosinger B Jr, Lee EJ, Manning RW,Yu SH, Mulder KL, and Westphal H. Targeted oncogeneactivation by site-specific recombination in transgenic mice.Proc Natl Acad Sci USA 89: 6232–6236, 1992.

79. Larsson S, Hotchkiss G, Andang M, Nyholm T, Inzunza J,Jansson I, and Ahrlund-Richter L. Reduced beta 2-micro-globulin mRNA levels in transgenic mice expressing a designedhammerhead ribozyme. Nucleic Acids Res 22: 2242–2248, 1994.

80. Le Y and Sauer B. Conditional gene knockout using Crerecombinase. Mol Biotechnol 17: 269–275, 2001.

81. Ledermann B and Burki K. Establishment of a germ-linecompetent C57BL/6 embryonic stem cell line. Exp Cell Res 197:254–258, 1991.

82. Lee P, Morley G, Huang Q, Fischer A, Seiler S, HornerJW, Factor S, Vaidya D, Jalife J, and Fishman GI. Condi-tional lineage ablation to model human diseases. Proc NatlAcad Sci USA 95: 11371–11376, 1998.

83. Lee P, Morley G, Huang Q, Fischer A, Seiler S, HornerJW, Factor S, Vaidya D, Jalife J, and Fishman GI. Condi-tional lineage ablation to model human diseases. Proc NatlAcad Sci USA 95: 11371–11376, 1998.

84. Lewandoski M. Conditional control of gene expression in themouse. Nat Rev Genet 2: 743–755, 2001.

85. Li M, Indra AK, Warot X, Brocard J, Messaddeq N, KatoS, Metzger D, and Chambon P. Skin abnormalities gener-

130 INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


ated by temporally controlled RXR� mutations in mouse epi-dermis. Nature 407: 633–636, 2000.

86. Lindeberg J, Mattsson R, and Ebendal T. Timing the doxy-cycline yields different patterns of genomic recombination inbrain neurons with a new inducible Cre transgene. J NeurosciRes 68: 248–253, 2002.

87. Lyngstadaas SP. Synthetic hammerhead ribozymes as toolsin gene expression. Crit Rev Oral Biol Med 12: 469–478, 2001.

88. Mansuy IM and Bujard H. Tetracycline-regulated gene ex-pression in the brain. Curr Opin Neurobiol 10: 593–596, 2000.

89. Mansuy IM, Mayford M, Jacob B, Kandel ER, and BachME. Restricted and regulated overexpression reveals cal-cineurin as a key component in the transition from short-termto long-term memory. Cell 92: 39–49, 1998.

90. Martin GR. Isolation of a pluripotent cell line from earlymouse embryos cultured in medium conditioned by teratocar-cinoma stem cells. Proc Natl Acad Sci USA 78: 7634–7638,1981.

91. Matzuk MM, Finegold MJ, Mather JP, Krummen L, Lu H,and Bradley A. Development of cancer cachexia-like syndromeand adrenal tumors in inhibin-deficient mice. Proc Natl AcadSci USA 91: 8817–8821, 1994.

92. Matzuk MM, Finegold MJ, Su JG, Hsueh AJ, and BradleyA. Alpha-inhibin is a tumour-suppressor gene with gonadalspecificity in mice. Nature 360: 313–319, 1992.

93. Metzger D and Chambon P. Site- and time-specific genetargeting in the mouse. Methods 24: 71–80, 2001.

94. Metzger D, Clifford J, Chiba H, and Chambon P. Condi-tional site-specific recombination in mammalian cells using aligand-dependent chimeric Cre recombinase. Proc Natl AcadSci USA 92: 6991–6995, 1995.

95. Meyers EN, Lewandoski M, and Martin GR. An Fgf8 mu-tant allelic series generated by Cre- and Flp-mediated recom-bination. Nat Genet 18: 136–141, 1998.

96. Minamino T, Gaussin V, DeMayo FJ, and Schneider MD.Inducible gene targeting in postnatal myocardium by cardiac-specific expression of a hormone-activated Cre fusion protein.Circ Res 88: 587–592, 2001.

97. Misra RP, Bronson SK, Xiao Q, Garrison W, Li J, Zhao R,and Duncan SA. Generation of single-copy transgenic mouseembryos directly from ES cells by tetraploid embryo comple-mentation. BMC Biotechnol 1: 12, 2001.

98. Nagy A. Cre recombinase: the universal reagent for genometailoring. Genesis 26: 99–109, 2000.

99. Nerenberg M, Hinrichs SH, Reynolds RK, Khoury G, andJay G. The tat gene of human T-lymphotropic virus type 1induces mesenchymal tumors in transgenic mice. Science 237:1324–1329, 1987.

100. No D, Yao TP, and Evans RM. Ecdysone-inducible geneexpression in mammalian cells and transgenic mice. Proc NatlAcad Sci USA 93: 3346–3351, 1996.

101. Omiecinski CJ, Remmel RP, and Hosagrahara VP. Con-cise review of the cytochrome P450s and their roles in toxicol-ogy. Toxicol Sci 48: 151–156, 1999.

102. Orban PC, Chui D, and Marth JD. Tissue- and site-specificDNA recombination in transgenic mice. Proc Natl Acad SciUSA 89: 6861–6865, 1992.

103. Peitz M, Pfannkuche K, Rajewsky K, and Edenhofer F.Ability of the hydrophobic FGF and basic TAT peptides topromote cellular uptake of recombinant Cre recombinase: a toolfor efficient genetic engineering of mammalian genomes. ProcNatl Acad Sci USA 99: 4489–4494, 2002.

104. Pelkonen O and Nebert DW. Metabolism of polycyclic aro-matic hydrocarbons: etiologic role in carcinogenesis. PharmacolRev 34: 189–222, 1982.

105. Peterson KR, Navas PA, and Stamatoyannopoulos G.�-YAC transgenic mice for studying LCR function. Ann NYAcad Sci 850: 28–37, 1998.

106. Phillips TJ, Hen R, and Crabbe JC. Complications associ-ated with genetic background effects in research using knock-out mice. Psychopharmacology (Berl) 147: 5–7, 1999.

107. Picard D. Regulation of protein function through expression ofchimaeric proteins. Curr Opin Biotechnol 5: 511–515, 1994.

108. Pierson TM, Wang Y, DeMayo FJ, Matzuk MM, Tsai SY,and Omalley BW. Regulable expression of inhibin A in wild-type and inhibin � null mice. Mol Endocrinol 14: 1075–1085,2000.

109. Pratt WB and Toft DO. Steroid receptor interactions withheat shock protein and immunophilin chaperones. Endocr Rev18: 306–360, 1997.

110. Rijkers T, Peetz A, and Ruther U. Insertional mutagenesisin transgenic mice. Transgenic Res 3: 203–215, 1994.

111. Rindi G, Ratineau C, Ronco A, Candusso ME, Tsai M, andLeiter AB. Targeted ablation of secretin-producing cells intransgenic mice reveals a common differentiation pathway withmultiple enteroendocrine cell lineages in the small intestine.Development 126: 4149–4156, 1999.

112. Rockman HA, Koch WJ, and Lefkowitz RJ. Seven-trans-membrane-spanning receptors and heart function. Nature 415:206–212, 2002.

113. Rodriguez CI, Buchholz F, Galloway J, Sequerra R,Kasper J, Ayala R, Stewart AF, and Dymecki SM. High-efficiency deleter mice show that FLPe is an alternative toCre-loxP. Nat Genet 25: 139–140, 2000.

114. Rohlmann A, Gotthardt M, Hammer RE, and Herz J.Inducible inactivation of hepatic LRP gene by cre-mediatedrecombination confirms role of LRP in clearance of chylomicronremnants. J Clin Invest 101: 689–695, 1998.

115. Rossant J and Spence A. Chimeras and mosaics in mousemutant analysis. Trends Genet 14: 358–363, 1998.

116. Rudolph B, Hueber AO, and Evan GI. Reversible activationof c-Myc in thymocytes enhances positive selection and inducesproliferation and apoptosis in vitro. Oncogene 19: 1891–1900,2000.

117. Ryding AD, Sharp MG, and Mullins JJ. Conditional trans-genic technologies. J Endocrinol 171: 1–14, 2001.

118. Saam JR and Gordon JI. Inducible gene knockouts in thesmall intestinal and colonic epithelium. J Biol Chem 274:38071–38082, 1999.

119. Saez E, Nelson MC, Eshelman B, Banayo E, Koder A, ChoGJ, and Evans RM. Identification of ligands and coligands forthe ecdysone-regulated gene switch. Proc Natl Acad Sci USA97: 14512–14517, 2000.

120. Sanchez MJ, Bockamp EO, Miller J, Gambardella L, andGreen AR. Selective rescue of early haematopoietic progeni-tors in Scl(�/�) mice by expressing Scl under the control of astem cell enhancer. Development 128: 4815–4827, 2001.

121. Sanford LP, Kallapur S, Ormsby I, and Doetschman T.Influence of genetic background on knockout mouse pheno-types. Methods Mol Biol 158: 217–225, 2001.

122. Santoro SW and Schultz PG. Directed evolution of the sitespecificity of Cre recombinase. Proc Natl Acad Sci USA 99:4185–4190, 2002.

123. Schultze N, Burki Y, Lang Y, Certa U, and Bluethmann H.Efficient control of gene expression by single step integration ofthe tetracycline system in transgenic mice. Nat Biotechnol 14:499–503, 1996.

124. Schwenk F, Kuhn R, Angrand PO, Rajewsky K, and Stew-art AF. Temporally and spatially regulated somatic mutagen-esis in mice. Nucleic Acids Res 26: 1427–1432, 1998.

125. Scrable H and Stambrook PJ. Activation of the lac repressorin the transgenic mouse. Genetics 147: 297–304, 1997.

126. Seuntjens E, Vankelecom H, Quaegebeur A, Vande VijverV, and Denef C. Targeted ablation of gonadotrophs in trans-genic mice affects embryonic development of lactotrophs. MolCell Endocrinol 150: 129–139, 1999.

127. Shimshek DR, Kim J, Hubner MR, Spergel DJ, BuchholzF, Casanova E, Stewart AF, Seeburg PH, and Sprengel R.Codon-improved Cre recombinase (iCre) expression in themouse. Genesis 32: 19–26, 2002.

128. Shin MK, Levorse JM, Ingram RS, and Tilghman SM. Thetemporal requirement for endothelin receptor-B signalling dur-ing neural crest development. Nature 402: 496–501, 1999.

129. Sigmund CD. Viewpoint: are studies in genetically alteredmice out of control? Arterioscler Thromb Vasc Biol 20: 1425–1429, 2000.

131INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


130. Silva JP, Kohler M, Graff C, Oldfors A, Magnuson MA,Berggren PO, and Larsson NG. Impaired insulin secretionand �-cell loss in tissue-specific knockout mice with mitochon-drial diabetes. Nat Genet 26: 336–340, 2000.

131. Sinclair AM, Gottgens B, Barton LM, Stanley ML, Par-danaud L, Klaine M, Gering M, Bahn S, Sanchez M,Bench AJ, Fordham JL, Bockamp E, and Green AR.Distinct 5� SCL enhancers direct transcription to developingbrain, spinal cord, and endothelium: neural expression is me-diated by GATA factor binding sites. Dev Biol 209: 128–142,1999.

132. Skarnes WC, Moss JE, Hurtley SM, and Beddington RS.Capturing genes encoding membrane and secreted proteinsimportant for mouse development. Proc Natl Acad Sci USA 92:6592–6596, 1995.

133. Sligar SG. Nature’s universal oxygenases: the cytochromesP450. Essays Biochem 34: 71–83, 1999.

134. Smithies O, Gregg RG, Boggs SS, Koralewski MA, andKucherlapati RS. Insertion of DNA sequences into the humanchromosomal beta-globin locus by homologous recombination.Nature 317: 230–234, 1985.

135. Sund NJ, Vatamaniuk MZ, Casey M, Ang SL, MagnusonMA, Stoffers DA, Matschinsky FM, and Kaestner KH.Tissue-specific deletion of Foxa2 in pancreatic � cells results inhyperinsulinemic hypoglycemia. Genes Dev 15: 1706–1715,2001.

136. Thomas HE, Stunnenberg HG, and Stewart AF. Het-erodimerization of the Drosophila ecdysone receptor with reti-noid X receptor and ultraspiracle. Nature 362: 471–475, 1993.

137. Thomas KR and Capecchi MR. Site-directed mutagenesis bygene targeting in mouse embryo-derived stem cells. Cell 51:503–512, 1987.

138. Urlinger S, Baron U, Thellmann M, Hasan MT, Bujard H,and Hillen W. Exploring the sequence space for tetracycline-dependent transcriptional activators: novel mutations yield ex-panded range and sensitivity. Proc Natl Acad Sci USA 97:7963–7968, 2000.

139. Urlinger S, Helbl V, Guthmann J, Pook E, Grimm S, andHillen W. The p65 domain from NF-B is an efficient humanactivator in the tetracycline-regulatable gene expression sys-tem. Gene 247: 103–110, 2000.

140. Utomo AR, Nikitin AY, and Lee WH. Temporal, spatial, andcell type-specific control of Cre-mediated DNA recombination intransgenic mice. Nat Biotechnol 17: 1091–1096, 1999.

141. Valencik ML and McDonald JA. Codon optimization mark-edly improves doxycycline regulated gene expression in themouse heart. Transgenic Res 10: 269–275, 2001.

142. Van Dyke T and Jacks T. Cancer modeling in the modern era:progress and challenges. Cell 108: 135–144, 2002.

143. Vasioukhin V, Degenstein L, Wise B, and Fuchs E. Themagical touch: genome targeting in epidermal stem cells in-duced by tamoxifen application to mouse skin. Proc Natl AcadSci USA 96: 8551–8556, 1999.

144. Visnjic D, Kalajzic I, Gronowicz G, Aguila HL, Clark SH,Lichtler AC, and Rowe DW. Conditional ablation of theosteoblast lineage in Col2.3�tk transgenic mice. J Bone MinerRes 16: 2222–2231, 2001.

145. Vooijs M, Jonkers J, and Berns A. A highly efficient ligand-regulated Cre recombinase mouse line shows that LoxP recom-bination is position dependent. EMBO Rep 2: 292–297, 2001.

146. Wakayama T and Yanagimachi R. Cloning of male micefrom adult tail-tip cells. Nat Genet 22: 127–128, 1999.

147. Wang XJ, Liefer KM, Tsai S, O’Malley BW, and Roop DR.Development of gene-switch transgenic mice that induciblyexpress transforming growth factor �1 in the epidermis. ProcNatl Acad Sci USA 96: 8483–8488, 1999.

148. Wang Y, Blandino G, Oren M, and Givol D. Induced p53expression in lung cancer cell line promotes cell senescence anddifferentially modifies the cytotoxicity of anti-cancer drugs.Oncogene 17: 1923–1930, 1998.

149. Wang Y, O’Malley BW Jr, Tsai SY, and O’Malley BW. Aregulatory system for use in gene transfer. Proc Natl Acad SciUSA 91: 8180–8184, 1994.

150. Wang Y, Tsai SY, and O’Malley BW. Antiprogestin regulablegene switch for induction of gene expression in vivo. MethodsEnzymol 306: 281–294, 1999.

151. Wells KD, Foster JA, Moore K, Pursel VG, and Wall RJ.Codon optimization, genetic insulation, and an rtTA reporterimprove performance of the tetracycline switch. Transgenic Res8: 371–381, 1999.

152. Wiles MV, Vauti F, Otte J, Fuchtbauer EM, Ruiz P, Fucht-bauer A, Arnold HH, Lehrach H, Metz T, von Melchner H,and Wurst W. Establishment of a gene-trap sequence taglibrary to generate mutant mice from embryonic stem cells. NatGenet 24: 13–14, 2000.

153. Williams RS and Wagner PD. Transgenic animals in inte-grative biology: approaches and interpretations of outcome.J Appl Physiol 88: 1119–1126, 2000.

154. Wilson C, Bellen HJ, and Gehring WJ. Position effects oneukaryotic gene expression. Annu Rev Cell Biol 6: 679–714, 1990.

155. Wong EA and Capecchi MR. Analysis of homologous recom-bination in cultured mammalian cells in transient expressionand stable transformation assays. Somat Cell Mol Genet 12:63–72, 1986.

156. Wong MH, Saam JR, Stappenbeck TS, Rexer CH, andGordon JI. Genetic mosaic analysis based on Cre recombinaseand navigated laser capture microdissection. Proc Natl AcadSci USA 97: 12601–12606, 2000.

157. Wunderlich FT, Wildner H, Rajewsky K, and EdenhoferF. New variants of inducible Cre recombinase: a novel mutantof Cre-PR fusion protein exhibits enhanced sensitivity and anexpanded range of inducibility. Nucleic Acids Res 29: E47, 2001.

158. Wyborski DL, DuCoeur LC, and Short JM. Parameters af-fecting the use of the lac repressor system in eukaryotic cells andtransgenic animals. Environ Mol Mutagen 28: 447–458, 1996.

159. Yamamoto A, Hen R, and Dauer WT. The ons and offs ofinducible transgenic technology: a review. Neurobiol Dis 8:923–932, 2001.

160. Yamauchi Y, Abe K, Mantani A, Hitoshi Y, Suzuki M,Osuzu F, Kuratani S, and Yamamura K. A novel transgenictechnique that allows specific marking of the neural crest celllineage in mice. Dev Biol 212: 191–203, 1999.

161. Yanagawa Y, Kobayashi T, Ohnishi M, Tamura S, Tsu-zuki T, Sanbo M, Yagi T, Tashiro F, and Miyazaki J.Enrichment and efficient screening of ES cells containing atargeted mutation: the use of DT-A gene with the polyadenyl-ation signal as a negative selection maker. Transgenic Res 8:215–221, 1999.

162. Yao TP, Forman BM, Jiang Z, Cherbas L, Chen JD, McKe-own M, Cherbas P, and Evans RM. Functional ecdysonereceptor is the product of EcR and Ultraspiracle genes. Nature366: 476–479, 1993.

163. Yu B, Lane ME, Pestell RG, Albanese C, and Wadler S.Downregulation of cyclin D1 alters cdk 4- and cdk 2-specificphosphorylation of retinoblastoma protein. Mol Cell Biol ResCommun 3: 352–359, 2000.

164. Yu Y and Bradley A. Engineering chromosomal rearrange-ments in mice. Nat Rev Genet 2: 780–790, 2001.

165. Zajchowski DA, Kauser K, Zhu D, Webster L, Aberle S,White FA III, Liu HL, Humm R, MacRobbie J, Ponte P,Hegele-Hartung C, Knauthe R, Fritzemeier KH, VergonaR, and Rubanyi GM. Identification of selective estrogen re-ceptor modulators by their gene expression fingerprints. J BiolChem 275: 15885–15894, 2000.

166. Zambrowicz BP, Friedrich GA, Buxton EC, Lilleberg SL,Person C, and Sands AT. Disruption and sequence identifi-cation of 2,000 genes in mouse embryonic stem cells. Nature392: 608–611, 1998.

167. Zhu Z, Ma B, Homer RJ, Zheng T, and Elias JA. Use of thetetracycline-controlled transcriptional silencer (tTS) to elimi-nate transgene leak in inducible overexpression transgenicmice. J Biol Chem 276: 25222–25229, 2001.

132 INVITED REVIEW


at CA

PE

S - U



Dow

nloaded from


Documents

Of mice and models: improved animal models for … › ~ireneyan › CursoPG › PDFs › Bockamp et al...Of mice and models: improved animal models for biomedical research. Physiol