Conditional Gene Expression in the Epidermis of Transgenic Mice Using the Tetracycline-Regulated Transactivators tTA and rTA Linked to the Keratin 5 Promoter
2000; Elsevier BV; Volume: 115; Issue: 5 Linguagem: Inglês
10.1046/j.1523-1747.2000.00144.x
ISSN1523-1747
AutoresIlysa Diamond, Timothy Owolabi, Melissa Marco, Christopher Wai Kei Lam, Adam B. Glick,
Tópico(s)Skin and Cellular Biology Research
ResumoTo produce conditional expression of genes in the mouse epidermis we have generated transgenic mouse lines in which the tetracycline-regulated transcriptional transactivators, tTA and rTA, are linked to the bovine keratin 5 promoter. The transactivator lines were crossed with the tetOlacZ indicator line to test for transactivation in vivo. In the absence of doxycycline, the K5/tTA line induced β-galactosidase enzyme activity in the epidermis at a level 500-fold higher than controls, and oral and topical doxycycline caused a dose- and time-dependent suppression of β-galactosidase mRNA levels and enzyme activity. In the K5/rTA lines, doxycycline induced β-galactosidase activity between 3- and 50-fold higher depending on the founder line, and this occurred within 24–48 h after dosing. Histochemical analysis of all lines localized β-galactosidase expression to the basal layer of the epidermis and the outer root sheath of the hair follicle, as well as other keratin 5 positive tissues. In several K5/rTA lines, skin-specific transactivation was restricted to the hair follicle. Treatment of these double transgenic mice with 12-O-tetradecanoyl-phorbol-13-acetate caused rapid migration of β-galactosidase marked cells from the hair follicle through the interfollicular epidermis, demonstrating the usefulness of this specific double transgenic for fate mapping cells in the epidermis. These results show that the tetracycline regulatory system produces effective conditional gene expression in the mouse epidermis, and suggest that it should be amenable to suppression and activation of foreign genes during development and specific pathologic conditions relevant to the epidermis. To produce conditional expression of genes in the mouse epidermis we have generated transgenic mouse lines in which the tetracycline-regulated transcriptional transactivators, tTA and rTA, are linked to the bovine keratin 5 promoter. The transactivator lines were crossed with the tetOlacZ indicator line to test for transactivation in vivo. In the absence of doxycycline, the K5/tTA line induced β-galactosidase enzyme activity in the epidermis at a level 500-fold higher than controls, and oral and topical doxycycline caused a dose- and time-dependent suppression of β-galactosidase mRNA levels and enzyme activity. In the K5/rTA lines, doxycycline induced β-galactosidase activity between 3- and 50-fold higher depending on the founder line, and this occurred within 24–48 h after dosing. Histochemical analysis of all lines localized β-galactosidase expression to the basal layer of the epidermis and the outer root sheath of the hair follicle, as well as other keratin 5 positive tissues. In several K5/rTA lines, skin-specific transactivation was restricted to the hair follicle. Treatment of these double transgenic mice with 12-O-tetradecanoyl-phorbol-13-acetate caused rapid migration of β-galactosidase marked cells from the hair follicle through the interfollicular epidermis, demonstrating the usefulness of this specific double transgenic for fate mapping cells in the epidermis. These results show that the tetracycline regulatory system produces effective conditional gene expression in the mouse epidermis, and suggest that it should be amenable to suppression and activation of foreign genes during development and specific pathologic conditions relevant to the epidermis. cytomegalovirus Targeted overexpression of specific genes to the epidermis of transgenic mice has been widely utilized to study their in vivo role in epidermal growth control, differentiation, and tumorigenesis. These studies have primarily employed promoters of epidermal structural proteins such as keratin 1 (K1), K5, K10, and K14, or loricrin to drive the gene of interest (Bailleul et al., 1990Bailleul B. Surani M.A. White S. et al.Skin hyperkeratosis and papilloma formation in transgenic mice expressing a ras oncogene from a suprabasal keratin promoter.Cell. 1990; 62: 697-708Abstract Full Text PDF PubMed Scopus (214) Google Scholar;Vassar and Fuchs, 1991Vassar R. Fuchs E. Transgenic mice provide new insights into the role of TGF-β during epidermal development and differentiation.Genes Dev. 1991; 5: 714-727Crossref PubMed Scopus (361) Google Scholar;Chung et al., 1994Chung S.Y. Cheng C.K. Rothnagel J.A. et al.Expression of the human keratin gene (K1) in transgenic mice is tissue- and developmental-specific but altered with respect to differentiation state.Mol Cell Diff. 1994; 2: 61-81Google Scholar;Wang et al., 1998Wang X.J. Greenhalgh D.A. Jiang A. et al.Expression of a p53 mutant in the epidermis of transgenic mice accelerates chemical carcinogenesis.Oncogene. 1998; 17: 35-45Crossref PubMed Scopus (76) Google Scholar), resulting in constitutive, high levels of expression from E9.5 (K5, K14) and E13.5 (K1, K10) of gestation (Ouellet et al., 1990Ouellet T. Lussier M. Babai F. Lapointe L. Royal A. Differential expression of the epidermal K1 and K10 keratin genes during mouse embryo development.Biochem Cell Biol. 1990; 68: 448-453Crossref PubMed Scopus (12) Google Scholar;Byrne et al., 1994Byrne C. Tainsky M. Fuchs E. Programming gene expression in developing epidermis.Development. 1994; 120: 2369-2383Crossref PubMed Google Scholar) into adulthood. When growth regulatory genes are used in conjunction with these promoters, the normal homeostasis of the epidermis is altered prior to any experimental manipulation in the adult. The K6 or bovine KIV promoter has also been utilized to regulate expression of target genes, as K6 expression is activated by hyperproliferative stimuli such as treatment with 12-O-tetradecanoyl-phorbol-13-acetate (TPA) (Blessing et al., 1995Blessing M. Nanney L.B. King L.E. Hogan B.L. Chemical skin carcinogenesis is prevented in mice by the induced expression of a TGF-β related transgene.Teratog Carcinog Mutagen. 1995; 15: 11-21Crossref PubMed Scopus (40) Google Scholar;Fowlis et al., 1996Fowlis D.J. Cui W. Johnson S.A. Balmain A. Akhurst R.J. Altered epidermal cell growth control in vivo by inducible expression of transforming growth factor β1 in the skin of transgenic mice.Cell Growth Differ. 1996; 7: 679-687PubMed Google Scholar). This system is confounded, however, by the fact that TPA is itself a potent modulator of epidermal function (Yuspa, 1984Yuspa S.H. Molecular and cellular basis for tumor promotion in mouse skin.in: Fujiki H. Hecker E. Moore T.E. Sugimura T. Weinstein I.B. Cellular Interactions of Environmental Tumor Promoters. Japan Sci. Soc. Press, Tokyo, Japan1984: 315-326Google Scholar;Toftgard et al., 1985Toftgard R. Yuspa S.H. Roop D.R. Keratin gene expression in mouse skin tumors and in mouse skin treated with 12–O-tetradecanoylphorbol-13-acetate.Cancer Res. 1985; 45: 5845-5850PubMed Google Scholar). In addition genes that produce an embryonic or neonatal lethal phenotype require a conditional expression system for meaningful analysis of functional effects in the adult animal. Recently several bi-transgenic systems have been developed that allow tissue-specific conditional expression of a target gene (Saez et al., 1997Saez E. No D. West A. Evans R.M. Inducible gene expression in mammalian cells and transgenic mice.Curr Opin Biotechnol. 1997; 8: 608-616Crossref PubMed Scopus (90) Google Scholar). The tetracycline-regulated system utilizes two tetracycline-dependent transactivator proteins, the tTA and rTA, which are fusion proteins of the tet repressor and the herpes simplex virus VP16 transactivating domain (Gossen and Bujard, 1996Gossen M. Bujard H. Tight control of gene expression in mammalian cells by tetracycline-responsive promoters.Proc Natl Acad Sci USA. 1996; 89: 5547-5551Crossref Scopus (4136) Google Scholar;Kistner et al., 1996Kistner A. Gossen M. Zimmermann F. Jerecic J. Ullmer C. Lubbert H. Bujard H. Doxycycline-mediated quantitative and tissue-specific control of gene expression in transgenic mice.Proc Natl Acad Sci USA. 1996; 93: 10933-10938Crossref PubMed Scopus (647) Google Scholar). These proteins activate transcription of a second transgene linked to a heptameric tetO repeat from the tet operon in a tetracycline-dependent manner (Gossen and Bujard, 1996Gossen M. Bujard H. Tight control of gene expression in mammalian cells by tetracycline-responsive promoters.Proc Natl Acad Sci USA. 1996; 89: 5547-5551Crossref Scopus (4136) Google Scholar). Thus, transactivation by the tTA is suppressed by tetracyclines (Kistner et al., 1996Kistner A. Gossen M. Zimmermann F. Jerecic J. Ullmer C. Lubbert H. Bujard H. Doxycycline-mediated quantitative and tissue-specific control of gene expression in transgenic mice.Proc Natl Acad Sci USA. 1996; 93: 10933-10938Crossref PubMed Scopus (647) Google Scholar), whereas the rTA, a mutant form of tTA, is activated by tetracyclines (Gossen et al., 1995Gossen M. Freundlieb S. Bender G. Muller G. Hillen W. Bujard H. Transcriptional activation by tetracyclines in mammalian cells.Science. 1995; 268: 1766-1769Crossref PubMed Scopus (1953) Google Scholar). Previous studies with transgenic mice expressing tTA linked to a cytomegalovirus (CMV) promoter showed that transactivation of a tetOlacZ or tetOluciferase target gene could be regulated by tetracycline in vivo, although in these mice the expression in the skin was highly variable (Furth et al., 1994Furth P.A. St Onge L. Boger H. et al.Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter.Proc Natl Acad Sci USA. 1994; 91: 9302-9306Crossref PubMed Scopus (653) Google Scholar;Hennighausen et al., 1995Hennighausen L. Wall R.J. Tillmann U. Li M. Furth P.A. Conditional gene expression in secretory tissues and skin of transgenic mice using the MMTV-LTR and the tetracycline responsive system.J Cell Biochem. 1995; 59: 463-472Crossref PubMed Scopus (104) Google Scholar). More recently this system has been used to conditionally overexpress ornithine decarboxylase in the hair follicle after treatment with TPA by linking the tTA to the K6 promoter (Guo et al., 1999Guo Y. Zhao J. Sawicki J. Peralta S.A. O'brien T.G. Conversion of C57Bl/6 mice from a tumor promotion-resistant to a – sensitive phenotype by enhanced ornithine decarboxylase expression.Mol Carcinog. 1999; 26: 32-36Crossref PubMed Scopus (25) Google Scholar). In order to create a conditional expression system in the proliferative compartment of the mouse epidermis, we have generated transgenic mice in which expression of the tTA and rTA genes is driven by the bovine K5 promoter (Ramirez et al., 1994Ramirez A. Bravo A. Jorcano J.L. Vidal M. Sequences 5′ of the bovine keratin 5 gene direct tissue- and cell-type-specific expression of a lacZ gene in the adult and during development.Differentiation. 1994; 58: 53-64PubMed Google Scholar). To characterize the time and dose dependence of target gene regulation by doxycycline, a tetracycline analog, we have crossed the BK5/tTA and rTA transgenic lines with the tetOlacZ transgenic line, which harbors a nuclear targeted β-galactosidase expression cassette linked to the tetO sequences (Furth et al., 1994Furth P.A. St Onge L. Boger H. et al.Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter.Proc Natl Acad Sci USA. 1994; 91: 9302-9306Crossref PubMed Scopus (653) Google Scholar). Here we show that in double transgenic mice β-galactosidase gene expression is specifically and uniformally expressed in the epidermis as well as other K5-expressing tissues. In combination with the appropriate transactivator, systemic and topical doxycycline are able to either suppress or induce β-galactosidase expression. In addition we have used the K5/rTA × tetOlacZ double transgenic mice to map the fate of marked hair follicle cells following treatment of the epidermis with the tumor promoter TPA. These results indicate that this bigenic system will be very useful for inducible expression of genes that perturb epidermal function and influence neoplastic progression. The tTA gene was isolated from the plasmid UHD15-1 (Gossen and Bujard, 1996Gossen M. Bujard H. Tight control of gene expression in mammalian cells by tetracycline-responsive promoters.Proc Natl Acad Sci USA. 1996; 89: 5547-5551Crossref Scopus (4136) Google Scholar) with a BamH1/EcoR1 digest, and the blunt-ended fragment was subcloned into the unique SnaB1 site of pBK5 (R. Jorcano, CIEMAT). The rTA gene was isolated by polymerase chain reaction (PCR) amplification from the pTet-on plasmid (Clontech, Palo Alto, CA) and subcloned into pCRII (Invitrogen, Carlsbad, CA). A Not 1 fragment containing the rTA gene was then subcloned into the Not 1 site of pBK5/97. Transgenic mice were generated in FVB/N mice using standard techniques of pronuclear injection (Hogan et al., 1994Hogan B. Beddington R. Costantini F. Lacy E. Manipulating the Mouse Embryo. Cold Spring Harbor Laboratory Press, New York1994Google Scholar), and identified by southern blot (Hogan et al., 1994Hogan B. Beddington R. Costantini F. Lacy E. Manipulating the Mouse Embryo. Cold Spring Harbor Laboratory Press, New York1994Google Scholar) using the isolated tTA fragment as a hybridization probe with BamH1 restricted genomic DNA. To initially screen tTA founders for transactivation, primary keratinocytes from eight K5/tTA founder lines were transiently transfected with a tetOluciferase reporter plasmid pUHC13-3 (Gossen and Bujard, 1996Gossen M. Bujard H. Tight control of gene expression in mammalian cells by tetracycline-responsive promoters.Proc Natl Acad Sci USA. 1996; 89: 5547-5551Crossref Scopus (4136) Google Scholar) and luciferase activity was measured in the presence and absence of doxycycline. Transactivating lines were tested in vivo by crossing with the tetOlacZ transgenic line (Furth et al., 1994Furth P.A. St Onge L. Boger H. et al.Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter.Proc Natl Acad Sci USA. 1994; 91: 9302-9306Crossref PubMed Scopus (653) Google Scholar), and assaying double transgenic animals for β-galactosidase expression. In these crosses transgenes were identified by PCR with the following primers: tTA and rTA, 5′-CTCGCCCAGAAGCTAGGTGT, 5′-CCATCGCGATGACTTAGTAA; β-gal, 5′-CCAACTTAATCGCC- TTGCAG, 5′-GAGCGAGTAACAACCCGTCG. K5/rTA founders were directly screened for transactivation by crossing to the tetOlacZ line and screening offspring for histochemical expression of β-galactosidase in tail biopsies. Histochemical staining for β-galacto- sidase enzyme activity was done at 37°C overnight on frozen sections postfixed in 0.2% glutaraldehyde, in a phosphate-buffered saline solution containing 0.75 mM X-Gal (Promega, Madison, WI), 1.5 mM MgCl2, 10 mM K3Fe(CN)6, and 10 mM K4Fe(CN)6.H2O. Tissues were counterstained with nuclear Fast Red (Trevigen, Rockville, MD). Images were captured using a Leitz DMRB microscope equipped with a video camera. To quantitate β-galactosidase enzyme activity, tissues were homogenized in a buffer containing 100 mM potassium phosphate pH 7.8, 0.2% Tween 20, 1 mM dithiothreitol, and the cleared supernatant reacted with a luminescent substrate (Galacton Star, Clontech). β-Galactosidase activity was measured using a Turner Designs 20/20 luminometer and expressed as relative light units per microgram of protein. Total RNA was isolated from the skin using Trizol (Gibco/BRL, Rockville, MD), and β-galactosidase RNA levels were detected using semiquantitative reverse transcriptase PCR. Reverse transcriptase reactions were done with 1 μg total DNAse-treated RNA, dNTP, oligo-dT, and 1 unit superscript II reverse transcriptase (BRL) at 42°C for 1 h. cDNAs were amplified with β-galactosidase or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) specific primers for 25 cycles in the presence of 32P-dCTP. Labeled PCR products were resolved on a 10% polyacrylamide-TBE gel, and visualized by exposing the dried gel to autoradiographic film. No PCR products were detected in the absence of reverse transcriptase. tTA and rTA mRNA levels were detected in northern blots of total RNA and a 32P-labeled tTA DNA fragment as a hybridization probe. Filters were rehybridized to a labeled PL7 cDNA to normalize for loading. Recipient cells for transient transfections were plated at a density of 2 × 105 cells per well in 12 well tissue culture trays, or 1.5 × 105 cells per well in 24 well tissue culture trays, and transfected with the transactivator and tetOluciferase reporter plasmids, or tetOluciferase alone using the lipofectamine reagent (Gibco/BRL). Six hours after transfection the medium was changed to complete medium with or without tetracycline/doxycycline, and after a further 42 h cell lysates were prepared and assayed for luciferase activity with 100 μM luciferin (Analytical Luminescence) using a Berthold Lumat luminometer (Berthold, Germany). Mice were dosed with systemic doxycycline by two methods. For dosing in the drinking water, doxycyline (Sigma, St Louis, MO) stock solutions were diluted, to give the appropriate final concentration as indicated in the results, with deionized water containing 5% sucrose and administered using amber water bottles. Alternatively, the mice were fed mouse chow formulated with 200 mg or 1 g per kg doxycycline (Bio-Serve, Laurel, MD). Based on the assumption that a mouse eats 5 g of food per d these formulations result in approximately 1 and 5 mg of doxycycline per d. Doxycycline was applied topically in 25 μl of 40% ethanol vehicle to a 1 cm circular area on the shaved dorsal skin of a mouse. Vehicle-only applications were made to a different region of the same mouse. No histologic changes in the skin were observed in K5/tTA mice that had been dosed with suppressing doses of doxycycline for up to 9 mo. To target expression of the tTA and rTA transactivators to the basal layer of the epidermis DNA fragments containing these fusion genes were subcloned into the pBK5/97 expression plasmid, which contains approximately 5.2 kb of sequences 5′ of the bovine K5 promoter, a β-globin intron, and SV40 poly-A sequence (Figure 1a) (Ramirez et al., 1994Ramirez A. Bravo A. Jorcano J.L. Vidal M. Sequences 5′ of the bovine keratin 5 gene direct tissue- and cell-type-specific expression of a lacZ gene in the adult and during development.Differentiation. 1994; 58: 53-64PubMed Google Scholar). Eight K5/tTA and 21 K5/rTA founders were identified by southern blot analysis with a tTA-specific hybridization probe (Figure 1b). Although four of eight K5/tTA lines were able to transactivate a tetOluciferase construct when it was transfected into primary keratinocytes (data not shown), only one of eight K5/tTA lines (line 1216) was able to transactivate target gene expression in vivo when crossed with the tetOlacZ transgenic line (Furth et al., 1994Furth P.A. St Onge L. Boger H. et al.Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter.Proc Natl Acad Sci USA. 1994; 91: 9302-9306Crossref PubMed Scopus (653) Google Scholar), which contains a nuclear targeted β-galactosidase expression cassette linked to the tetO binding sequence. In contrast, 11 of 21 K5rTA lines were able to transactivate the tetOlacZ transgene expression in vivo, and of these, five were chosen for further characterization. Northern blot analysis of transactivator expression in the different founder lines showed that the tTA or rTA transcript was readily detected in all transactivating lines, and with the exception of the rTA line A10, which was markedly lower, there was little significant variation in levels of expression (Figure 1c). To analyze tissue specificity of target gene activation as well as regulation by doxycycline, β-galactosidase enzyme activity was assayed in different tissues of double and single transgenic mice in the presence or absence of doxycycline. Figure 2(a) shows that, in the absence of doxycycline, K5/tTA × tetOlacZ double transgenic mice expressed β-galactosidase in the skin at levels nearly 500-fold higher than single transgenic mice, or mice in which expression had been suppressed by long-term dosing with 200 g per kg doxycycline chow. β-Galactosidase expression was also elevated 200–300-fold above single transgenic animals in the tongue and forestomach, tissues that normally express K5 in the mouse. In contrast, in the double transgenic mice with or without doxycycline, the level of β-galactosidase expression in the non-K5-expressing tissues, the heart, lung, kidney, or liver, was similar to the background levels found in the single transgenic animals. No changes in epidermal histology, or expression of K1, K10, or K14, detected with immunostaining, were observed in K5/tTA, rTA, or induced double transgenic mice, indicating that nonspecific transactivation of gene expression is unlikely. In the absence of doxycycline, β-galactosidase enzyme activity in all tissues of K5/rTA × tetOlacZ double transgenic lines was near the levels of the single transgenic tetOlacZ mice (Figure 2b). Depending on the founder line, β-galactosidase enzyme activity was increased between 3- and 30-fold in the skin in double transgenic mice kept on 1g per kg doxycycline chow. Thus lines K4, A10, and G5 caused strong transactivation whereas lines F6 and B1 produced only 3-fold transactivation of the tetOlacZ target gene. Transactivation of lacz also occurred in the tongue and forestomach of the double transgenic mice, but again to a lesser extent than in the skin on a microgram basis. These results indicate that in double transgenic animals expressing the tTA and rTA genes from the K5 promoter, target gene expression occurs only in the absence or presence of doxycycline, with negligible background levels of expression. To analyze localization of target gene expression in different tissues from the K5/tTA and K5/rTA × tetOlacZ mice, β-galactosidase activity was determine in situ using a histochemical stain. In K5/tTA and rTA double transgenic mice, β-galactosidase activity was detected in K5-expressing tissues such as the skin and hair follicle (Figure 3a-c), tongue (Figure 3f), forestomach (Figure 3g), and trachea (not shown), but not in non-K5-expressing tissues such as liver, spleen, muscle, heart, or intestine, or in tetOlacZ single transgenics (not shown). β-Galactosidase was not detected in the K5/tTA line in the presence of doxycycline (Figure 3d), or in the K5/rTA lines in the absence of doxycycline (Figure 3e). In the skin of the K5/tTA double transgenic mice, β-galactosidase was induced in the basal layer of the interfolliclular epidermis and in the outer root sheath of the hair follicle (Figure 3a), consistent with previously described expression patterns of the K5 promoter construct used in these studies (Ramirez et al., 1994Ramirez A. Bravo A. Jorcano J.L. Vidal M. Sequences 5′ of the bovine keratin 5 gene direct tissue- and cell-type-specific expression of a lacZ gene in the adult and during development.Differentiation. 1994; 58: 53-64PubMed Google Scholar). The K5/rTA lines differed significantly in the pattern of β-galactosidase transactivation. In the B1 line, β-galactosidase was expressed strongly and uniformly in the outer root sheath of the hair follicles and basal layer of the epidermis (Figure 3b), whereas in the other lines examined such as F6 β-galactosidase expression was restricted to the hair follicles and was expressed rarely in the basal layer of the epidermis (Figure 3c). To test the efficacy of target gene regulation by doxycycline in the K5/tTA and rTA lines we analyzed the dose and time dependence for suppression and induction of β-galactosidase in double transgenic mice following administration of doxycycline. Figure 4(a) shows that after 2 wk of doxycycline administered in the drinking water, there was a dose-dependent reduction in β-galactosidase enzyme activity in the skin of the K5/tTA double transgenic mice. At 0.25, 0.5, and 1 μg per ml doxycycline the level of β-galactosidase activity in the epidermis was 60%, 15%, and 10% of the level in the untreated mice. Above 5 μg per ml β-galactosidase levels were suppressed to near background levels. These results show that different levels of target gene expression can be achieved using a range of doxycycline concentrations. To examine the time course of reinduction after suppression of transactivation, mice were switched to deionized water after 2 wk on different doxycycline doses. Initial studies showed that, whereas 200 μg per ml oral doxycycline caused a rapid suppression of β-galactosidase activity, long-term maintenance of animals on this concentration prevented rapid reinduction after removal of doxycycline from the drinking water (data not shown). Figure 4(b) shows that rapid induction of β-galactosidase only occurred at the lowest doses of doxycycline, whereas higher doses (5–20 μg per ml) required 8 d for significant induction to occur. These results indicate that a dose between 1 and 5 μg per ml is likely to achieve the best balance between suppression and rapid reinduction of target gene expression. To examine the time course of doxycycline-induced transactivation by the rTA, double transgenic mice kept in the absence of doxycyline were placed on diets containing 1 g per kg doxycycline. In the absence of doxycycline, β-galactosidase enzyme activity was similar to that of single transgenic mice (relative light units per μg), but within 24 h after switching to doxycycline chow there was a 20-fold increase in β-galactosidase enzyme activity in the skin, which reached 50-fold by 3 d (Figure 5a). Histochemical analysis showed that the rapid induction of β-galactosidase activity 24 h after doxycycline treatment occurred primarily within the hair follicle (Figure 5b,c). To deter- mine whether target gene expression could be modulated specifically in the epidermis, we treated the skin of K5/tTA × tetOlacZ double transgenic mice that had been kept off doxycycline with topical doxycycline. In situ histochemical analysis showed reduced levels of β-galactosidase activity in the basal layer of the skin of mice 24 h after treatment with doxycycline (Figure 6b). To verify that the changes in β-galactosidase activity reflected altered gene expression we prepared RNA from treated and untreated skin and analyzed β-galactosidase mRNA levels by RT-PCR. Figure 6(c) shows that topical treatment of mice with 1 μg doxycycline reduced levels of β-galactosidase mRNA within 24 h to levels similar to those found in the tetOlacZ single transgenic animals. Densitometric analysis using GAPDH to normalize expression indicated that mRNA levels were suppressed approximately 10-fold. As β-galactosidase was nearly exclusively expressed in the hair follicle of the K5/rTA line F6 we used the double transgenic mice to map the movement of β-galactosidase positive cells in the hair follicle after TPA treatment. Double transgenic mice were put on 1g per kg doxycycline for 1 wk, and then treated once topically with 10 μg TPA. Tissue sections were taken at 24, 36, and 48 h after TPA treatment and analyzed in situ for β-galactosidase expression. Twenty-four hours after TPA treatment β-galactosidase positive cells remained exclusively in the hair follicle in treated and untreated mice (Figure 7a). By 36 h, however, coincident with the onset of hyperplasia, there was a significant increase in blue stained cells in the stratum corneum of the epidermis, as well as within the epidermis. By 48 h, no β-galactosidase positive cells were detected in any layers of the interfollicular epidermis, although the hair follicles remained positive as expected. TPA treatment did not induce rTA expression in the skin of transgenic mice, indicating that the altered localization of the β-galactosidase positive cells was not due to new expression of the transactivator (Figure 7b). These results suggest that within a 24 h period between 24 and 48 h post TPA treatment there is a very rapid movement of cells out of the hair follicle through the interfollicular epidermis and into terminal differentiation. The tetracycline expression system has been widely used to regulate gene expression in cultured cells and transgenic animals (Resnitzky et al., 1994Resnitzky D. Gossen M. Bujard H. Reed S.I. Acceleration of the G1/S phase transition by expression of cyclins D1 and E with an inducible syste.Mol Cell Biol. 1994; 14: 1669-1679Crossref PubMed Scopus (973) Google Scholar;Efrat et al., 1995Efrat S. Fusco-DeMane D. Lemberg H. al Emran O. Wang X. Conditional transformation of a pancreatic β-cell line derived from transgenic mice expressing a tetracycline-regulated oncogene.Proc Natl Acad Sci USA. 1995; 92: 3576-3580Crossref PubMed Scopus (235) Google Scholar;Shockett et al., 1995Shockett P. Difilippantonio M. Hellman N. Schatz D.G. A modified tetracycline-regulated system provides autoregulatory, inducible gene expression in cultured cells and transgenic mice.Proc Natl Acad Sci USA. 1995; 92: 6522-6526Crossref PubMed Scopus (340) Google Scholar;Kistner et al., 1996Kistner A. Gossen M. Zimmermann F. Jerecic J. Ullmer C. Lubbert H. Bujard H. Doxycycline-mediated quantitative and tissue-specific control of gene expression in transgenic mice.Proc Natl Acad Sci USA. 1996; 93: 10933-10938Crossref PubMed Scopus (647) Google Scholar;Yu et al., 1996Yu Z. Redfern C.S. Fishman G.I. Conditional transgene expression in the heart.Circ Res. 1996; 79: 691-697Crossref PubMed Scopus (112) Google Scholar). The advantage of this bigenic system is that, with the tTA and rTA, tetracyclines can be used to both suppress and induce target gene expression. We have described two transgenic lines in which the bovine K5 promoter directs expression of the tTA and rTA to the hair follicle and basal layer of the epidermis. In all lines examined transactivator expression was correctly regulated in a tissue-specific manner as mice containing either the K5/tTA or rTA and the tetOlacZ transgenes expressed β-galactosidase in the epidermis and hair follicles and variably in other K5 positive tissues, but did not express β-galactosidase in non-K5-expressing tissues. Whereas the K5/tTA and one K5/rTA line transactivated strongly in both the follicular and interfollicular keratinocytes, the remainder of the rTA lines tested showed restriction of transactivation to the hair follicle. As most of the founder lines expressed similar levels of the rTA transcript it is unlikely that rTA expression levels account for the different patterns of transactivation. It is possible that different patterns of rTA expression could account for the distinct patterns of transactivation, however. More detailed studies will be required to determine if this restricted pattern reflects a position effect related to transgene integration site, or a common rearrangement of the K5 promoter leading to restricted expression of the transactivator. We also observed that the fold induction of β-galactosidase activity was much greater in the K5/tTA line than in any of the rTA lines treated with doxycycline, even though the level of transactivator mRNA in the K5/tTA line was similar to the majority of rTA lines. Although a previous study comparing both the tTA and rTA showed similar levels of transactivation (Furth et al., 1994Furth P.A. St Onge L. Boger H. et al.Temporal control of gene expression in transgenic mice by a tetracycline-responsive promoter.Proc Natl Acad Sci USA. 1994; 91: 9302-9306Crossref PubMed Scopus (653) Google Scholar), it is possible that this reflects an intrinsic biologic difference in the properties of the specific transactivators. In the majority of double transgenic mice with both transactivator lines, β-galactosidase expression appeared uniform throughout the epidermis, and even in the follicle-restricted lines all hair follicles expressed β-galactosidase. This is a significant improvement from a previously reported CMV-tTA line in which transactivation of the same tetOlacZ indicator line was very patchy in the epidermis (Hennighausen et al., 1995Hennighausen L. Wall R.J. Tillmann U. Li M. Furth P.A. Conditional gene expression in secretory tissues and skin of transgenic mice using the MMTV-LTR and the tetracycline responsive system.J Cell Biochem. 1995; 59: 463-472Crossref PubMed Scopus (104) Google Scholar). It is possible that in keratinocytes the K5 promoter is maintained in an expressed state in more cells than is the CMV promoter. In addition, the suppressed levels of target gene expression in double transgenic animals either in the presence of doxycycline for the tTA or absence for the rTA were close to that of the single transgenic animals. Thus with these K5-targeted transactivator lines there was no significant read-through of target gene expression. Our results clearly show that different levels of target gene expression can be achieved by varying the doxycycline concentration. This will be particularly useful for analysis of the effects of different levels of target gene expression on epidermal biology. It is clear, however, that to achieve rapid induction of target gene expression with the tTA following suppression, relatively low doses of doxycycline must be used, because at doses higher than 5–20 μg per ml more than 2 wk are required to reduce systemic doxycycline levels enough to allow transactivation. It is likely that, at higher doses, sequestration of the antibiotic in tissues prevents rapid removal from the animal (Riond and Riviere, 1988Riond J.L. Riviere J.E. Pharmacology and toxicology of doxycycline.Vet Hum Toxicol. 1988; 30: 431-443PubMed Google Scholar). In contrast to the tTA, significant induction of β-galactosidase occurs within 24 h after addition of doxycycline at 1 g per kg, although slower kinetics of induction were observed at lower doses. In addition, regulation of target gene transactivation by topical doxycycline will be particularly useful for specifically targeting changes in gene expression to the epidermis, as well as producing regional changes in gene expression within the epidermis. Similar topical regulation of target gene expression has been obtained with a skin-targeted bigenic system based on a modified hormone receptor (Wang et al., 1999Wang X.J. Liefer K.M. Tsai S. O'malley B.W. Roo D.R. Development of gene-switch transgenic mice that inducibly express transforming growth factor beta 1 in the epidermis.Proc Natl Acad Sci USA. 1999; 96: 8483-8488Crossref PubMed Scopus (126) Google Scholar). We made use of the unique nature of the follicle-restricted K5/rTA × tetOlacZ lines to test whether they could be used to follow movement of marked hair follicle cells under conditions that perturb normal tissue homeostasis. TPA causes increased proliferation and hyperplasia of hair follicles (Aldaz et al., 1985Aldaz C.M. Conti C.J. Gimenez I.B. Slaga T.J. Klein-Szanto A.J. Cutaneous changes during prolonged application of 12-O-tetradecanoylphorbol-13-acetate on mouse skin and residual effects after cessation of treatment.Cancer Res. 1985; 45: 2753-2759PubMed Google Scholar;Cotsarelis et al., 1990Cotsarelis G. Sun T.T. Lavker R.M. Label-retaining cells reside in the bulge area of pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin carcinogenesis.Cell. 1990; 61: 1329-1337Abstract Full Text PDF PubMed Scopus (1803) Google Scholar), and numerous studies have provided strong evidence for a follicular origin of papillomas induced by two-stage carcinogenesis protocols in mouse skin (Cotsarelis et al., 1990Cotsarelis G. Sun T.T. Lavker R.M. Label-retaining cells reside in the bulge area of pilosebaceous unit: implications for follicular stem cells, hair cycle, and skin carcinogenesis.Cell. 1990; 61: 1329-1337Abstract Full Text PDF PubMed Scopus (1803) Google Scholar;Hansen and Tennant, 1994Hansen L.A. Tennant R.W. Follicular origin of epidermal papillomas in v-Ha-ras transgenic TG.AC mouse skin.Proc Natl Acad Sci USA. 1994; 91: 7822-7826Crossref PubMed Scopus (72) Google Scholar;Binder et al., 1998Binder R.L. Johnson G.R. Gallagher P.M. Stockman S.L. Sundberg J.P. Conti C.J. Squamous cell hyperplastic foci: precursors of cutaneous papillomas induced in SENCAR mice by a two-stage carcinogenesis regimen.Cancer Res. 1998; 58: 4314-4323PubMed Google Scholar). Whereas some evidence suggests that papillomas could arise from promoter-induced migration of initiated hair follicle cells into surrounding normal follicles (Binder et al., 1997bBinder R.L. Gallagher P.M. Johnson G.R. Stockman S.L. Smith B.J. Sundberg J.P. Conti C.J. Evidence that initiated keratinocytes clonally expand into multiple existing hair follicles during papilloma histogenesis in SENCAR mouse skin.Mol Carcinog. 1997; 20: 151-158Crossref PubMed Scopus (22) Google Scholar), little is known about the fate of hair follicle cells after TPA treatment. Our results show that TPA caused rapid movement of β-galactosidase positive cells from the outer root sheath of the hair follicle into the stratum corneum of the interfollicular epidermis, followed by desquamation. As TPA has minimal effect on expression of endogenous K5 gene or the promoter used to generate the rTA lines (Casatorres et al., 1994Casatorres J. Navarro J.M. Blessing M. Jorcano J.L. Analysis of the control of expression and tissue specificity of the keratin 5 gene, characteristic of basal keratinocytes. Fundamental role of an AP-1 element.J Biol Chem. 1994; 269: 20489-20496Abstract Full Text PDF PubMed Google Scholar; Figure 7b), it is unlikely that this represents new β-galactosidase transactivation in the interfollicular keratinocytes due to TPA-induced expression of the rTA. Although further studies are required, these results support the concept of migration of initiated hair follicle cells (Binder et al. 1997a), as initiated cells would be expected to be resistant to terminal differentiation induced by TPA (Hennings and Yuspa, 1985Hennings H. Yuspa S.H. Two-stage tumor promotion in mouse skin: an alternative explanation.J Natl Cancer Inst. 1985; 74: 735-740PubMed Google Scholar;Hennings et al., 1987Hennings H. Michael D. Lichti U. Yuspa S.H. Response of carcinogen-altered mouse epidermal cells to phorbol ester tumor promoters and calcium.J Invest Dermatol. 1987; 88: 60-65Abstract Full Text PDF PubMed Google Scholar). These results also indicate that TPA-induced epidermal hyperplasia is due in part to migration of outer root sheath keratinocytes into the interfollicular epidermis. In summary our results show that the tetracycline-regulated bigenic system can be applied to temporal and quantitative manipulation of foreign gene expression in the mouse epidermis. These tTA and rTA lines should prove useful for regulating expression of genes involved in growth control or differentiation during epidermal development, carcinogenesis, or wound healing where precise temporal control of the gene under study is required. The authors would like to thank Dr. Jose Jorcano for the generous gift of the pBK5 plasmid, Dr. H. Bujard for the tTA and tetOluciferase plasmids, C. Cheng for the PL7 plasmid, Dr. Lothar Hennighausen for the tetOlacZ mice, Katy Bernhard for animal support services, and Dr. Stuart H. Yuspa for critical reading of the manuscript.
Referência(s)