Temporally Regulated and Tissue-Specific Gene Manipulations in the Adult and Embryonic Heart Using a Tamoxifen-Inducible Cre Protein
Abstract—The advent of conditional and tissue-specific recombination systems in gene-targeted or transgenic mice has permitted an assessment of single gene function in a temporally regulated and cell-specific manner. Here we generated transgenic mice expressing a tamoxifen-inducible Cre recombinase protein fused to two mutant estrogen-receptor ligand-binding domains (MerCreMer) under the control of the α-myosin heavy chain promoter. These transgenic mice were crossed with the ROSA26 lacZ-flox–targeted mice to examine Cre recombinase activity and the fidelity of the system. The data demonstrate essentially no Cre-mediated recombination in the embryonic, neonatal, or adult heart in the absence of inducing agent but >80% recombination after only four tamoxifen injections. Expression of the MerCreMer fusion protein within the adult heart did not affect cardiac performance, cellular architecture, or expression of hypertrophic marker genes, demonstrating that the transgene-encoded protein is relatively innocuous. In summary, MerCreMer transgenic mice represent a tool for temporally regulated inactivation of any loxP-targeted gene within the developing and adult heart or for specifically directing recombination and expression of a loxP-inactivated cardiac transgene in the heart.
The use of site-specific recombinase, such as Cre or Flp, has permitted an evaluation of gene function within genetically modified animals in a tissue-specific manner.1 Typically, a gene of interest is flanked with recombinase recognition sequences (loxP sites for Cre), which permit subsequent recombination and gene disruption coincident with expression of the appropriate enzyme.1 Expression of a site-specific recombinase is achieved either through a tissue-restricted transgene or by inserting the recombinase cDNA into a genetic locus with a known tissue expression profile. Tissue-specific gene disruption is often used to circumvent embryonic or fetal lethality associated with complete somatic disruption, thus permitting an examination of gene function in the tissue of interest or at later developmental stages. However, one significant limitation of the present approach is the inability to control the timing of Cre- or Flp-mediated recombination, because gene disruption closely parallels the earliest expression profile of the chosen promoter. Examples have been reported in which Cre-mediated (tissue-specific) gene disruption still results in embryonic, fetal, or neonatal lethality, so that the adult function of a certain gene cannot be examined.2 3 4 5 6 Because the heart is the first functioning organ system within the developing embryo, a traditional tissue-specific recombination strategy could easily result in developmental lethality, excluding an assessment of gene function in the neonatal or adult heart.7 To circumvent this limitation, we have generated transgenic mice expressing a tamoxifen-inducible Cre-fusion protein specifically within the heart.
Materials and Methods
All experiments were conducted in accordance with the Guide for the Use and Care of Laboratory Animals and approved by the Institutional Animal Care and Use Committee.
A cDNA encoding a double-fusion protein between two mutant estrogen receptor domains on either side of Cre recombinase was subcloned as a SalI fragment into the cardiac-specific α-myosin heavy chain (MHC) 5.5-kb promoter construct.8 9 The MerCreMer fusion cDNA (gift of Michael Reth, Max-Plank Institute, Freiburg, Germany) encodes the mutated murine estrogen receptor ligand–binding domain (amino acids 281 to 599, G525R), which is insensitive to estrogen but sensitive to tamoxifen.9 The resulting α-MHC-MerCreMer DNA fragment was injected into pronuclei of freshly fertilized oocytes from FVB mice to produce transgenic animals. FVB founder transgenic mice are presently being crossed into the C57BL/6SV129 genetic background (fourth generation at present) to generate a strain more suitable for traditional gene-targeting experiments. Transgene expression was verified by Western blotting of 100 μg of cardiac protein extract from 3- and 6-week-old FVB transgenic mice using Cre-specific antisera at a dilution of 1:5000 (Covance). Western blotting and the subsequent detection procedure were performed as previously described.10 Similar levels of protein expression were observed between FVB transgenic mice and outbred C57BL/6SV129 transgenic mice (data not shown).
To induce Cre recombination, adult MerCreMer transgenic mice (3 to 12 weeks of age) were treated with tamoxifen (Sigma) by intraperitoneal injection once a day for 4 to 6 days at a dosage of 20 mg/kg per day. To examine recombination in embryos, pregnant mice were treated with tamoxifen by intraperitoneal injection once a day for 4 days at 20 mg/kg day. Tamoxifen was dissolved in 60% ethanol at a concentration of 5 mg/mL. Hearts from tamoxifen- or saline-treated control mice were removed and perfused in a retrograde manner with phosphate-buffered saline through the aorta, left ventricle, and coronary vasculature for 2 minutes followed by overnight perfusion at room temperature with X-gal staining solution to detect lacZ activity throughout the heart.11 RNA dot blotting and transthoracic echocardiography were performed as described previously.12 DNA Southern blotting to detect Cre-mediated recombination used a probe specific to an N-terminal sequence in lacZ and an EcoRI digest of genomic DNA. The lacZ Southern probe (470 base pairs) was generated by polymerase chain reaction using the following primers: lacZ up 5′ GTCACACTACGTCTGAACGT-3′ and lacZ down 5′ CTGCACCATTCGCGTTACG-3′.
The estrogen receptor is a hormone-activated transcription factor that is usually sequestered by heat-shock proteins in the absence of ligand. In the presence of 17β-estradiol, the estrogen receptor is released, allowing participation in transcriptional regulatory complexes. Since its initial description, the ligand-binding domain of the estrogen receptor (amino acids 281 to 599) has been extensively used as a strategy for fusion protein sequestration within mammalian cells.13 More recently, a mutation in the estrogen receptor at amino acid 525 (glycine to arginine) was described that rendered it insensitive to 17β-estradiol but sensitive to the estrogen antagonist tamoxifen.14 Use of this mutant estrogen receptor (Mer) domain in fusion protein strategies permits estrogen-insensitive but tamoxifen-inducible activity within mammalian cells.
A cDNA consisting of Cre recombinase fused to a Mer-encoding sequence at both the N- and C-termini (MerCreMer)9 was subcloned into the cardiac-specific α-MHC 5.5-kb promoter construct to permit generation of transgenic mice (Figure 1A⇓).8 This double-fusion strategy was shown to more effectively reduce promiscuous Cre activity in mammalian cells.9 15 Interestingly, the double-fusion protein also demonstrates substantially greater Cre recombinase activity compared with a single CreMer fusion, suggesting that the double-fusion strategy somehow enhances the intrinsic recombinase activity of Cre or its ability to access DNA sites.15
Two stable transgenic mouse lines were generated expressing the MerCreMer fusion protein in the heart, both of which were viable with phenotypically normal hearts. One of the transgenic lines (line 2) demonstrated only weak MerCreMer protein expression in the neonatal heart, which was downregulated in adulthood (data not shown). However, line 1 transgenic mice demonstrated robust MerCreMer protein expression (107 kDa) in the juvenile and adult heart, as revealed with Cre-specific antisera and Western blotting at both 3 and 6 weeks of age (Figure 1B⇑).
To examine the activity of the MerCreMer protein, line 1 transgenic mice were crossed with the ROSA26 lacZ-loxP reporter strain as an indicator of Cre-mediated recombination in the heart. The ROSA26 lacZ-loxP reporter strain contains a Cre-dependent, loxP-inactivated lacZ cDNA cassette targeted within the ubiquitously expressed ROSA26 locus16 (Figure 2A⇓). Cre-mediated recombination of this allele deletes neomycin and a series of polyadenylation sequences, resulting in the juxtaposition of a splice acceptor site and the lacZ cDNA.5 The MerCerMer transgene (tg/0) was bred into the ROSA26 lacZ-loxP background (lacZ/0), generating double-heterozygous mice (MCM×ROSA) for subsequent analysis.
Hearts from double-heterozygous mice were harvested at embryonic day 17 (E17), 17 days after birth, 6 weeks of age, and 12 weeks of age and subjected to X-gal staining as an indicator of Cre-mediated recombination. Remarkably, untreated double-heterozygous mice (MCM×ROSA) showed no detectable lacZ activity in either the ventricles or atria at E17, 17 days postnatal, or 6 weeks of age (Figures 2B⇑ and 3A⇓). Such a result was unexpected, because even low levels of unregulated Cre activity result in irreversible recombination, which has a cumulative effect on lacZ staining. Whereas the overall lack of lacZ staining up through 6 weeks of age indicated extremely tight regulation, we did occasionally observe a single cell with lacZ staining in histological tissue sections (data not shown). To more carefully examine the extent of background recombination, significantly older (3-month-old) double-heterozygous mice were examined. By 3 months of age, untreated double-heterozygous mice demonstrated approximately 1% spurious recombination within the heart as estimated from whole mount and histological sections (Figure 2B⇑ and data not shown). These results indicate that the MerCreMer transgenic protein is, on the whole, tightly regulated within the heart so that only very low levels of unregulated recombination are observed in the absence of tamoxifen.
Treatment of double-heterozygous mice with tamoxifen for only 5 days (4 separate intraperitoneal injections) was sufficient to induce robust recombination and lacZ activity in a uniform profile throughout the embryonic, neonatal, and adult heart (Figure 2B⇑). To assess tamoxifen-induced recombination in the embryonic heart, pregnant females were treated with drug from E12 to E16, the embryos were harvested at E17, and the hearts were removed for lacZ staining. Although the α-MHC promoter drives high levels of expression in the postnatal and adult heart, significant expression is also observed in the embryonic heart.17 Cre-mediated recombination was observed in utero, indicating that the MerCreMer transgene can be used to manipulate gene expression in the embryonic heart (Figure 2B⇑). Hearts harvested on E17 showed somewhat less-robust lacZ staining when viewed in whole mount, attributable largely to the staining procedure. Given their size, embryonic hearts were simply emersed in staining solution, whereas neonatal and adult hearts were slowly perfused with staining solution through the coronary vasculature, exposing essentially all cardiomyocytes. Despite the differing staining procedures, histological sectioning of E17 hearts revealed a robust and uniform profile of lacZ expression in the outer few cell layers exposed to the staining solution (data not shown). Finally, 3-month-old double-heterozygous mice also demonstrated robust lacZ expression after 5 days of tamoxifen treatment (Figure 2B⇑). Collectively, these data demonstrate efficient Cre-mediated recombination in the embryonic, neonatal, and adult mouse heart.
To more carefully evaluate the characteristics of the MerCreMer transgene and tamoxifen-induced recombination, large cohorts of mice were examined at 6 weeks of age. Six-week-old double-heterozygous mice treated with tamoxifen demonstrated lacZ staining throughout both right and left ventricles and atria, although atrial staining appeared less robust because of its thinness (Figure 3A⇑). Within the ventricles of 6-week-old mice, staining was homogenous throughout the myocardial, endocardial, and epicardial cell layers (Figure 3B⇑). Untreated double-transgenic mice also displayed a homogenous profile of nearly absent lacZ staining in the myocardial, endocardial, and epicardial cell layers at 6 weeks of age (Figures 3A⇑ through 3C). Identical results were observed in 6 independent experiments. Additional controls showed no lacZ staining in single heterozygous ROSA26 lacZ-loxP mice or in single heterozygous MerCreMer transgenic mice treated with tamoxifen (Figure 3A⇑). No difference was observed in levels of unregulated or inducible lacZ staining between male and female mice at 6 weeks of age (data not shown).
At 6 weeks of age, lacZ-stained hearts were histologically sectioned to more carefully evaluate the extent of recombination after 5 days of tamoxifen treatment (Figures 3B⇑ and 3C⇑). Nearly all of the myocytes demonstrated lacZ activity in double-heterozygous mice (MCM×ROSA) treated with tamoxifen, whereas untreated mice showed undetectable or extremely low levels of lacZ staining (Figures 3B⇑ and 3C⇑). Qualitative assessment of histological sections suggested a recombination frequency of >80% after 5 days of tamoxifen administration in 5- to 6-week-old mice. To additionally evaluate the extent of Cre-mediated recombination, Southern blotting was performed with a probe specific to the ROSA 26 locus (neomycin-lacZ cassette). Untreated double-heterozygote mice showed no recombination at baseline at 6 weeks of age (lane 2), whereas 5 days of tamoxifen treatment revealed >70% recombination in the heart (lane 1) (Figure 3D⇑). However, given that the heart contains a fair number of nonmyocytes (fibroblasts, infiltrates, and endothelial cells) that do not express the MerCreMer protein, the myocyte-specific recombination frequency is undoubtedly >70% as measured at 6 weeks of age. No recombination was observed in the brain, kidney, lung, liver, and skeletal muscle of double-heterozygous mice treated with tamoxifen (Figure 3D⇑). Collectively, these data demonstrate efficient tamoxifen-induced Cre-mediated recombination specifically in the heart of MerCreMer transgenic mice. More importantly, the Southern blot data also confirm the histological data, demonstrating nearly undetectable levels of background recombination at 6 weeks of age.
MerCreMer transgenic mice represent a valuable genetic tool for inducing the inactivation of any loxP-targeted gene within the heart. However, for such a reagent to be useful, the MerCreMer protein itself should not adversely affect the heart. Accordingly, echocardiography was performed in 8-week-old MerCreMer transgenic mice (± tamoxifen for 5 days) and compared with nontransgenic littermate controls. The results show no alterations in cardiac chamber dimensions, wall thicknesses, or fractional shortening in MerCreMer transgene either with or without tamoxifen treatment (Table⇓). It should be noted that MerCreMer transgenic mice were allowed to recover from tamoxifen treatment for 6 days before echocardiography to eliminate any potential nonspecific effects associated with momentary tamoxifen administration.
Heart weights of MerCreMer transgenic mice were similar to nontransgenic littermate control hearts at 8 weeks of age, and H&E- and trichrome-stained histological sections failed to identify any pathology (Figure 4A⇓). More importantly, tamoxifen-treated MerCreMer transgenic mice (6 days after treatment) did not show upregulated expression of hypertrophy marker genes compared with wild-type control mice (Figure 4B⇓). In contrast, control MEK1 transgenic mice with demonstrable hypertrophy showed significant upregulated expression of atrial natriuretic factor, brain natriuretic peptide, skeletal α-actin, and β-MHC mRNA (Figure 4B⇓).12 Collectively, these data demonstrate that the MerCreMer transgene is relatively innocuous to the heart.
The ability to control tissue specificity of gene deletions in the mouse using Cre-lox technology has profoundly advanced mouse genetics and the ability to examine single gene function in vivo. However, one significant limitation of the traditional Cre-lox approach is an inability to temporally control genetic recombination. Typically, a loxP-targeted gene undergoes Cre-mediated recombination coincident with Cre expression directed by a promoter region of interest. In this study, we describe transgenic mice that express a modified Cre recombinase fusion protein in the heart under the transcriptional control of the α-MHC promoter. Whereas transgene expression from this promoter is largely constitutive throughout development and in the adult heart, the Cre recombinase fusion protein is inactivated by two mutant estrogen receptor ligand-binding domains. Such an approach only requires a single transgene to achieve cardiac-specific and temporally regulated recombination in vivo. Alternative cardiac-specific gene regulatory strategies, such as the tetracycline-responsive activator and transcriptional regulatory region, have also been recently described.18 However, the tetracycline-inducible system typically requires two independent transgenes (binary), one expressing the tet-activator/repressor in the tissue of interest and one containing a chimeric tetracycline-regulated promoter fused to a cDNA encoding Cre recombinase. Although such a tetracycline-inducible system for regulating Cre-mediated recombination in the heart has yet to be described, it remains entirely feasible, because the proof of concept has already been demonstrated.19 20 21 However, a binary tetracycline-inducible approach requires a more complicated breeding strategy compared with the monogenic MerCreMer approach described herein. In addition, the potential fidelity (leakiness) of a tetracycline-regulated Cre recombinase system is uncertain.
While this manuscript was in review, Minamino et al22 described the generation and characterization of a RU486-regulated Cre protein in the heart. These investigators used a construct in which Cre was fused to the ligand-binding domain of the human progesterone receptor under the control of the α-MHC promoter. Similar to our study, Minamino et al22 demonstrated drug-regulated Cre recombination in a temporally controlled manner in the mouse heart. However, the approach used by Minamino et al resulted in slightly greater background recombination compared with the tamoxifen-inducible MerCreMer approach described here (Figures 2⇑ and 3⇑). It is likely that our approach using both an N- and C-termini Mer domain imparts tighter regulation compared with the single progesterone receptor domain fusion used by Minamino et al.22 Indeed, the double Mer fusion protein (MerCreMer) was previously shown to substantially reduce promiscuous Cre activity in cultured cells and even to enhance Cre-mediated recombination compared with a single CreMer fusion protein.9 15
Another difference between the tamoxifen- and RU486-inducible systems relates to the ability to perform developmental gene manipulations in embryos. RU486, an antiprogesterone, can be more detrimental to early-stage embryos compared with the antiestrogen compound tamoxifen.23 Indeed, a tamoxifen-regulated CreMer transgene expressed in the developing neural tube has already been shown to promote regulated gene recombination in early- and mid-gestation mouse embryos.24 Similarly, we show that tamoxifen administration to pregnant mice between E12 and E16 promoted recombination in the embryonic heart (Figure 2B⇑). Whereas the α-MHC promoter is widely used for its characteristic expression pattern in the neonatal and adult heart, embryonic expression is also significant.17 A similar yet slightly lower dosage of tamoxifen (7.5 mg/kg per day) administered daily between E12 and E18 in pregnant CD-1 mice did not compromise development or induce abortion, although endometrial hyperplasia and polyploid adenomas were identified in 30% to 50% of female offspring at ≈1 year of age.25 Despite these concerns, the results of the present study indicate that MerCreMer transgenic mice can be used to temporally regulate gene manipulations within the developing mouse heart without causing embryonic lethality.
Here we measured Cre-mediated recombination within the ROSA26 gene locus, which is ubiquitously expressed and constitutively active in the mouse. Given its active confirmation with respect to chromatin, it is formally possible that the ROSA26 locus is more permissive to site-specific recombination compared with other genetic loci. A final technical consideration relates to the use of tamoxifen as an inducing agent. Although echocardiographic assessment of tamoxifen-treated MerCreMer transgenic mice did not reveal obvious defects in cardiac chamber dimensions 6 days after drug treatment, nor were hypertrophic marker genes significantly activated, it remains possible that tamoxifen might alter some aspect of cardiac physiology or biochemistry. However, decreasing the dose of tamoxifen (10 mg/kg per day) or the time course of treatment might resolve any potential alterations. Indeed, we have observed that even a single tamoxifen injection in 3-week-old double-heterozygous mice induced significant lacZ expression (data not shown).
The most obvious application of the MerCreMer transgenic line is to promote temporally regulated deletion of loxP-targeted genes in vivo. However, the MerCreMer transgene can also be used to temporally regulate expression of another unrelated transgene within the heart. Such a transgene would be constructed such that a loxP flanked stuffer sequence is placed immediately upstream of a given cDNA, rendering it inactive. Cre-mediated excision of such a stuffer region would result in the juxtaposition of the downstream cDNA and the given promoter, as shown in Figure 2A⇑. In this manner, a transgene could be specifically activated in the adult heart, bypassing developmental affects associated with unregulated transgenesis. In summary, MerCreMer transgenic mice will permit temporally regulated activation or inactivation of a properly designed transgene or a loxP-targeted genetic locus within the heart. Such applications will undoubtedly expand our ability to genetically dissect the function of putative disease-causing genes within either the developing or adult heart.
This work was supported by the National Institutes of Health and a Pew charitable trust scholar award (to J.D.M.).
Original received February 2, 2001; revision received May 8, 2001; accepted May 8, 2001.
- © 2001 American Heart Association, Inc.
Fiering S, Bender MA, Groudine M. Analysis of mammalian cis-regulatory DNA elements by homologous recombination. Methods Enzymol. 1999;306:42–66.
Shafi R, Iyer SP, Ellies LG, O’Donnell N, Marek KW, Chui D, Hart GW, Marth JD. The O-GlcNAc transferase gene resides on the X chromosome and is essential for embryonic stem cell viability and mouse ontogeny. Proc Natl Acad Sci U S A. 2000;97:5735–5739.
Li H, Wang J, Wilhelmsson H, Hansson A, Thoren P, Duffy J, Rustin P, Larsson NG. Genetic modification of survival in tissue-specific knockout mice with mitochondrial cardiomyopathy. Proc Natl Acad Sci U S A. 2000;97:3467–3472.
Haigh JJ, Gerber HP, Ferrara N, Wagner EF. Conditional inactivation of VEGF-A in areas of collagen2a1 expression results in embryonic lethality in the heterozygous state. Development. 2000;127:1445–1453.
Nozaki M, Ohishi K, Yamada N, Kinoshita T, Nagy A, Takeda J. Developmental abnormalities of glycosylphosphatidylinositol-anchor-deficient embryos revealed by Cre/loxP system. Lab Invest. 1999;79:293–299.
Liu JL, Grinberg A, Westphal H, Sauer B, Accili D, Karas M, LeRoith D. Insulin-like growth factor-I affects perinatal lethality and postnatal development in a gene dosage-dependent manner: manipulation using the Cre/loxP system in transgenic mice. Mol Endocrinol. 1998;12:1452–1462.
Copp AJ. Death before birth: clues from gene knockouts and mutations. Trends Genet. 1995;11:87–93.
Palermo J, Gulick J, Colbert M, Fewell J, Robbins J. Transgenic remodeling of the contractile apparatus in the mammalian heart. Circ Res. 1996;78:504–509.
Zhang Y, Riesterer C, Ayrall AM, Sablitzky F, Littlewood TD, Reth M. Inducible site-directed recombination in mouse embryonic stem cells. Nucleic Acids Res. 1996;24:543–548.
De Windt LJ, Lim HW, Haq S, Force T, Molkentin JD. Calcineurin promotes protein kinase C and c-Jun NH2-terminal kinase activation in the heart: cross-talk between cardiac hypertrophic signaling pathways. J Biol Chem. 2000;275:13571–13579.
Cheng TC, Hanley TA, Mudd J, Merlie JP, Olson EN. Mapping of myogenin transcription during embryogenesis using transgenes linked to the myogenin control region. J Cell Biol. 1992;119:1649–1656.
Bueno OF, De Windt LJ, Tymitz KM, Witt SA, Kimball TR, Klevitsky R, Hewett TE, Jones SP, Lefer DJ, Peng CF, Kitsis RN, Molkentin JD. The MEK1-ERK1/2 signaling pathway promotes compensated cardiac hypertrophy in transgenic mice. EMBO J. 2000;19:6341–6350.
Eilers M, Picard D, Yamamoto KR, Bishop JM. Chimaeras of myc oncoprotein and steroid receptors cause hormone-dependent transformation of cells. Nature. 1989;340:66–68.
Littlewood TD, Hancock DC, Danielian PS, Parker MG, Evan GI. A modified oestrogen receptor ligand-binding domain as an improved switch for the regulation of heterologous proteins. Nucleic Acids Res. 1995;23:1686–1690.
Verrou C, Zhang Y, Zurn C, Schamel WW, Reth M. Comparison of the tamoxifen regulated chimeric Cre recombinases MerCreMer and CreMer. Biol Chem. 1999;380:1435–1438.
Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet. 1999;21:70–71.
Luo Y, Ferreira-Cornwell M, Baldwin H, Kostetskii I, Lenox J, Lieberman M, Radice G. Rescuing the N-cadherin knockout by cardiac-specific expression of N- or E-cadherin. Development. 2001;128:459–469.
Fishman GI. Timing is everything in life: conditional transgene expression in the cardiovascular system. Circ Res. 1998;82:837–844.
Yu Z, Redfern CS, Fishman GI. Conditional transgene expression in the heart. Circ Res. 1996;79:691–697.
Bowman JC, Steinberg SF, Jiang T, Geenen DL, Fishman GI, Buttrick PM. Expression of protein kinase C beta in the heart causes hypertrophy in adult mice and sudden death in neonates. J Clin Invest. 1997;100:2189–2195.
Redfern CH, Degtyarev MY, Kwa AT, Salomonis N, Cotte N, Nanevicz T, Fidelman N, Desai K, Vranizan K, Lee EK, Coward P, Shah N, Warrington JA, Fishman GI, Bernstein D, Baker AJ, Conklin BR. Conditional expression of a Gi-coupled receptor causes ventricular conduction delay and a lethal cardiomyopathy. Proc Natl Acad Sci U S A. 2000;97:4826–4831.
Minamino T, Gaussin V, DeMayo FJ, Schneider MD. Inducible gene targeting in postnatal myocardium by cardiac-specific expression of a hormone-activated Cre fusion protein. Circ Res. 2001;88:587–592.
Sadek S, Bell SC. The effects of the antihormones RU486 and tamoxifen on fetoplacental development and placental bed vascularisation in the rat: a model for intrauterine fetal growth retardation. Br J Obstet Gynaecol. 1996;103:630–641.
Danielian PS, Muccino D, Rowitch DH, Michael SK, McMahon AP. Modification of gene activity in mouse embryos in utero by a tamoxifen-inducible form of Cre recombinase. Curr Biol. 1998;8:1323–1326.
Diwan BA, Anderson LM, Ward JM. Proliferative lesions of oviduct and uterus in CD-1 mice exposed prenatally to tamoxifen. Carcinogenesis. 1997;18:2009–2014.