HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Le Menuet, D. Articles by Lombès, M. Search for Related Content PubMed PubMed Citation Articles by Le Menuet, D. Articles by Lombès, M. Social Bookmarking                      What's this?

J Biol Chem, Vol. 275, Issue 11, 7878-7886, March 17, 2000

Targeted Oncogenesis Reveals a Distinct Tissue-specific Utilization of Alternative Promoters of the Human Mineralocorticoid Receptor Gene in Transgenic Mice^*

Damien Le Menuet^§, Say Viengchareun, Patrice Penfornis, Francine Walker^¶, Maria-Christina Zennaro, and Marc Lombès

From  INSERM U478, Institut Fédératif de Recherche Cellules Epithéliales, Faculté de Médecine Xavier Bichat and ^¶ Service d'Anatomo-pathologie, Hôpital Bichat, 75018 Paris, France

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The human mineralocorticoid receptor (hMR) is a nuclear receptor mediating aldosterone action, whose expression is driven by two alternative promoters, P1 and P2, flanking the two first 5'-untranslated exons. In vivo characterization of hMR regulatory regions was performed by targeted oncogenesis in mice using P1 or P2 directing expression of the large T antigen of SV40 (TAg). While transgenic P1.TAg founders rapidly developed lethal hibernomas from brown fat, cerebral primitive neuroectodermal tumors and facial leiomyosarcomas occurred in P2.TAg mice. Quantitative analyses of mouse MR (mMR) and transgene expression indicate that P1 promoter was transcriptionally active in all MR-expressing tissues, directing strong TAg expression in testis and salivary glands, moderate in lung, brain, uterus, liver, and heart but, unlike mMR, rather low in colon and kidney. Importantly, the renal transgene expression colocalized with mMR in the distal nephron. In contrast, P2 promoter was approximately 10 times less potent than P1, with no activity in the brain and colon. Several immortalized cell lines were established from both neoplastic and normal tissues of transgenic mice. These cells exhibited differentiated characteristics and maintained MR expression, thus providing useful models for further studies exploring the widespread expression and functions of MR. Our results demonstrate that hMR gene expression in vivo is controlled by complex regulatory mechanisms involving distinct tissue-specific utilization of alternative promoters.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Aldosterone effects are mediated by the mineralocorticoid receptor (MR),¹ a ligand-dependent transcription factor belonging to the nuclear receptor superfamily. MR is closely related to the glucocorticoid receptor, with which it shares a highly homologous primary structure organized in distinct functional domains (1). MR binds with the same affinity aldosterone and glucocorticoid hormones; however, in epithelial tissues, mineralocorticoid selectivity of aldosterone action is ensured by the presence of the enzyme 11-hydroxysteroid dehydrogenase 2 (2), which converts active glucocorticoids into non-MR-binding metabolites (3, 4). In addition, other mineralocorticoid selectivity-conferring mechanisms seem to be necessary to fully account for receptor specificity. At the receptor level, MR is able to discriminate between aldosterone and glucocorticoids in ligand-dependent conformational changes and stabilization of the ligand binding domain (5, 6), as well as in transcriptional activation (7, 8). Activated MR regulates transcription of target genes by binding to hormone response elements (9), where its direct or indirect interaction with coactivators or corepressors modulates the activity of the general transcriptional machinery (10).

MR is expressed in a wide variety of tissues, including polarized epithelial cells of the distal nephron, colon, and sweat glands, which all co-express 11-hydroxysteroid dehydrogenase 2, and where MR is known to regulate transepithelial ionic transports, mostly sodium reabsorption (11). MR is also found in non-epithelial target cells of the central nervous (12) and cardiovascular systems (13, 14); however, in these tissues, the exact nature of MR actions remains largely unknown.

The regulatory mechanisms controlling the tissue-specific expression of MR is still poorly understood. We have recently identified two functional promoter regions in the human MR (hMR) gene (15), which have been shown to direct expression of two hMR mRNA isoforms, hMR and hMR, that differ by their 5'-untranslated sequences (16). Using a series of hMR promoter constructs in transient transfection assays, we have characterized the 5'-flanking regions of the two first untranslated exons, 1 and 1, and defined two regulatory sequences corresponding to a proximal P1 and a distal P2 promoter respectively. The P1 promoter (1 kb of exon 1 5'-flanking region) possesses a stronger transcriptional activity than the P2 promoter (1.7-kb fragment of exon 1 5'-flanking region). In addition, although both promoters were inducible by glucocorticoids, only the distal P2 promoter appears to be sensitive to mineralocorticoids (15). The fact that alternative promoter utilization controls in a extremely fine-tuned manner the expression of hMR isoforms may represent a key step in regulating multiple aldosterone actions in various target tissues.

In the present study, transgenic mouse models were created to better understand the in vivo function of hMR alternative promoters as well as to study mechanisms of aldosterone action by means of establishment of novel mineralocorticoid-sensitive cell lines, which are not yet available. For this purpose, we have developed a targeted oncogenesis strategy in which the P1 and P2 promoter regions were fused to the simian virus 40 large T antigen (TAg) to generate two transgenes, called P1.TAg and P2.TAg, in order to drive oncogene expression in normally MR-expressing cells. TAg was used as a reporter gene to analyze cis-acting elements of each promoter sequence as well as for its properties that facilitate immortalization of cells in a variety of lineages by inactivating the retinoblastoma gene product and p53 (17).

This study describes the phenotypic characterization of P1.TAg and P2.TAg transgenic mice. All P1.TAg animals developed tumors of the brown adipose tissue called hibernomas, whereas P2.TAg mice exhibited primitive neuroectodermal tumors (PNET) and leiomyosarcomas. More importantly, as compared with the endogenous mMR expression in wild type animals, analyses of transgene expression patterns revealed a distinct tissue-specific utilization of hMR alternative promoters. Finally, we have succeeded in generating various novel cell lines originating from neoplastic as well as normal tissues of transgenic mice, which constitute original cellular models to further identify the molecular and cellular mechanisms of mineralocorticoid signaling.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Construction of P1.TAg and P2.TAg Transgenes and Generation of Transgenic Animals-- The P1.TAg and P2.TAg were constructed using the optimized transgenic vector H31 kindly provided by Dr. L. M. Houdebine (INRA, Jouy en Josas, France). It essentially contains an untranslated region for stabilizing mRNA transcripts and the terminal block of the human -globin including the last intron and polyadenylation site. As presented in Fig. 1, a HindIII-AvaII fragment (HA, 965, +216) containing approximately 1 kb of hMR P1 proximal promoter and the beginning of exon 1 blunt-ended with Klenow DNA polymerase (Life Technologies, Inc.) and a SspI-SspI fragment (SS, 1673, +123 bp) including 1.7 kb of the P2 distal promoter and the beginning of exon 1 were ligated into the unique SmaI site of H31 vector to generate the P1-H31 and P2-H31 plasmids, respectively. Subsequently for each construct, the 2.2-kb StuI-BamHI fragment, which contains the entire coding sequence of the SV40 TAg, was inserted into the unique BstXI site of P1-H31 and P2-H31 plasmid after filling in all recessive ends with Klenow treatment. P1.TAg and P2.TAg transgenes were separated from plasmid vector sequences by NotI digestion and purified after 0.7% low melting gel agarose electrophoresis with Elutip-d columns (Schleicher & Schuell, Dassel, Germany) and ethanol precipitation. Transgene DNA resuspended in 10 mM Tris-HCl, pH 7.4, 0.1 mM EDTA, were microinjected into fertilized oocytes obtained from B6D2 mice at the Service d'Expérimentation Animale et de Transgenèse (SEAT, CNRS, Villejuif, France).

View larger version (17K): [in this window] [in a new window]   Fig. 1.   Schematic representation of the 5'-flanking region of hMR gene and construction of the two transgenes P1.TAg and P2.TAg. The 5'-flanking region of hMR gene contains the two first untranslated exon 1 and 1 (solid boxes). The HindIII-AvaII (965, +216) P1 fragment and SspI-SspI (1673, +123) P2 fragment that each includes part of exon 1 or 1 were used to construct P1.TAg and P2.TAg, respectively. As described under "Materials and Methods," the transgenesis H31 vector contains an untranslated region (UTR) that stabilizes RNA messengers. The 2.2-kb coding sequence of TAg was inserted in the terminal human -globin block, including the last intronic 5'-sequence (glob5') and the polyadenylation site (glob3').

Founders harboring the transgene were first identified from polymerase chain reaction analyses of tail DNA from 2-week-old mice with oligonucleotides specific to the SV40 TAg, using forward and reverse primers (S5108, 5'-TTGAAAGGAGTGCCTGGGGGAAT-3'; A4920, 5'-CAGTTGCATCCCAGAAGCCTCCA-3') with an annealing temperature of 54 °C for a 189-bp amplified fragment. To ascertain the presence of the integrated transgene, PCR analyses were also performed with antisense primer A4920 and specific forward primers in each of the two hMR promoters, S180 (5'-TGCAACAGGTAGACGCGAGAGA-3') and SC8 (5'-CCGCTGCCTCGCCGCCTCTTGTA-3') for P1 and P2 promoters, respectively. Thirty cycles (95 °C for 45 s, 55 °C for 45 s, 72 °C for 45 s) were used to amplify the ~1.3-kb fragments. Number of integrated copies of transgenes was determined by Southern blot analysis of 10 µg of genomic DNA digested by SalI using standard protocols (18) with P1.TAg or P2.TAg ³²P-labeled probes (Rediprime II random prime labeling system; Amersham Pharmacia Biotech, Buckinghamshire, United Kingdom).

Ribonuclease Protection Assays-- Total RNA was extracted from various tissues of wild type and transgenic mice. Tissues were disaggregated with a Polytron homogenizer in Trizol reagent (Life Technologies, Inc.) and total RNA isolated following manufacturer's instructions. The TAg riboprobe synthesis was achieved using a 1665-bp PstI-PstI fragment of TAg coding region inserted in pBluescriptII KS+ plasmid (Stratagene, La Jolla, CA). This plasmid linearized by PvuII was used to synthesize a 377-base ³²P-labeled riboprobe with T3 RNA polymerase (riboprobe kit; Promega Madison, WI). The expected protected fragment is 302 bases long. The rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) probe was synthesized using T7 RNA polymerase after digestion of a pBluescript SK-GAPDH plasmid (kindly provided by Dr. Escoubet, Paris, France) by PvuII and StyI. Unprotected riboprobe was 184 bases long with a 164-base protected fragment. The mMR riboprobe was synthesized with a PstI-PstI fragment of mMR exon 2 inserted into pBlueScript KS (kindly provided by Drs. S. Berger and G. Schütz, German Cancer Research Center, Heidelberg, Germany), linearized by StuI, using T7 polymerase. This generated a 452-base probe and a protected fragment of 380 bases.

Ribonuclease protection assays (RPA) were performed as described previously (19). Briefly, 50 µg of total RNA are hybridized overnight at 50 °C in a formamide Pipes hybridization buffer with 4 × 10⁵ cpm TAg or mMR probe and 5 × 10⁴ cpm GAPDH probe as an internal control. Non-hybridized RNA was digested at 30 °C for 1 h with a ribonuclease A and T1 mixture followed by a 30-min proteinase K and SDS treatment. After phenol-chloroform extraction and ethanol precipitation, protected fragments were electrophoresed on a 6% polyacrylamide/urea gel. Gels were dried and fixed in 10% acetic acid. Radioactivity was counted generally overnight with an InstantImager (Packard, Meriden, CT), followed by an autoradiography. Results are expressed in arbitrary units and correspond to the ratio between TAg- or mMR-specific counts versus GAPDH signal.

Histology, Immunohistochemical, and Immunocytochemical Analyses-- Tissues and organs were collected at the time of necropsy and fixed in 4% paraformaldehyde in PBS, dehydrated, and embedded in paraffin. Sections (7 µm) were cut and stained with hematoxylin-eosin. Immunohistochemical studies were performed on some tissue sections of different transgenic animals. To enhance signal, sections were immersed in a 10 mM sodium citrate buffer, pH 6, and passed through microwave heating for 5 min, three times at 500 watts. To reduce background staining due to endogenous peroxidase activities, some tissue sections were preincubated with 3% H₂O₂. The presence of SV40 TAg protein was detected using a 1:20 to 1:50 dilution of the monoclonal anti-SV40 TAg antibody (Ab-2; Calbiochem, La Jolla, CA). Specific antibodies against vimentin (Dako, Copenhagen, Denmark), desmin (Dako), glial fibrillary acidic protein (ICN, Costa Mesa, CA), neurofilament (Novocastra, Le Perret en Yvelines, France), and vascular smooth muscle actin (Sigma) were also used. Sections were then incubated with a secondary biotinylated horse anti-mouse IgG (Vector Laboratories, Burlingame, CA) and the avidin-biotin peroxidase complex (Vectastain Elite ABC kit, Vector Laboratories). Tissues were counterstained with Mayer's hemalum. Specificity of the reaction was tested by omitting primary antibodies. For immunocytochemistry, cells were cultured on glass coverslides, fixed with 4% paraformaldehyde in PBS, rinsed for 5 min with PBS, incubated with 100% ethanol for 5 min, and stored at 20 °C before processing the immunocytochemical experiments.

In Situ Hybridization-- To prepare TAg sense and antisense riboprobes, a 885-bp HindIII-HindIII fragment of TAg coding sequence subcloned in pBlueScriptII KS+ (Stratagene) was used. Uridine ³⁵S-labeled probes (1000 Ci/mmol; Amersham Pharmacia Biotech) were synthesized using T3 RNA polymerase and plasmid linearized by BamHI for sense probe while antisense probe was generated using KpnI-linearized plasmid and T7 RNA polymerase. Other reagents are from Promega Corp. (Madison, WI).

In situ hybridization was performed on 7-µm sections of paraffin-embedded organs. All steps were carried out as described previously (20). After rehydration of sections in graded ethanol solutions, sections were postfixed in 4% paraformaldehyde (in PBS); proteinase K (0.8 µg/ml) treatment and a 10-min acetylation step were then carried out (0.1 M triethanolamine with 0.25 M acetic anhydride, and successively 5 min with PBS, 5 min with 145 mM NaCl). Sections were covered with 100 µl of the hybridization mixture (50% formamide, 1 mM DTT, 2× SSC, 10% dextran sulfate, 100 µg/ml salmon sperm DNA, and 4 × 10⁶ cpm of ³⁵S-labeled probe), covered with a parafilm (American National Can, Neenah, WI), and incubated overnight at 50 °C. Posthybridization treatment consists of an initial wash in 5× SSC, 10 mM DTT, at 50 °C for 30 min, followed by high stringency wash in 50% formamide, 2× SSC, 0.1 M DTT at 65 °C for 20 min and several washes in NaCl-Tris-EDTA (0.5 M NaCl, 10 mM Tris-HCl, 5 mM EDTA) at 37 °C. Sections were incubated in RNase A (20 µg/ml) at 37 °C for 20 min. After rinsing with 0.1× SSC for 15 min, sections were dehydrated and dried. Slides were dipped in Kodak NTB2 (melted at 42 °C), dried, and exposed at 20 °C for 2-4 weeks. Slides were developed and counterstained with toluidine.

Establishment of Cell Lines-- Different P1.TAg and P2.TAg animals were used in an attempt to derive various cell lines originating from normal as well as neoplastic tissues. Briefly, at the time of surgery of anesthetized animals with 0.5 ml of Avertin (Fluka, L'Isle d'Abeau Chesne, France), (150 µg/ml) intraperitoneally per animal, organs were excised under sterile conditions, cut in small pieces, rinsed twice with culture medium, and teased with two forceps in a Petri dish containing 10 ml of defined medium containing fungizone (1 µg/ml). The medium was composed of Dulbecco's modified Eagle's medium/Ham's F-12 medium (Life Technologies, Inc.) supplemented with 10% fetal calf serum, 2 mM glutamine, 100 IU/ml penicillin, 100 µg/ml streptomycin, and 20 mM Hepes. After centrifugation, pelleted cells were rinsed twice with medium and grown at 37 °C in a humidified atmosphere with 5% CO₂. For some tissues, special protocols were used to isolate specific cells. Distal nephron fragments were microdissected, colonic mucosa was scrapped, and small ventricular heart samples were treated with collagenase.

The cell culture medium was generally replaced every 2 days. All cell lines were cultured for at least 20 passages for approximately 6 months, stored under liquid nitrogen, and successfully recultured after thawing. For some cell lines, cells have been subcloned by limiting dilution or after microscopic isolation of a particular clone.

RT-PCR Analyses-- Total RNA was extracted from cells using the trizol reagent (Life Technologies, Inc.) and subjected to DNase I treatment in order to avoid any genomic DNA contamination. Briefly, 2 µg of total RNA were reverse-transcribed with 200 units of reverse transcriptase using the Superscript^TM II kit (Life Technologies, Inc.) according to the manufacturer's recommendations. cDNA was then amplified for 30-35 cycles in a total volume of 25 µl containing 1× PCR buffer (50 mM KCl, 20 mM Tris-HCl, pH 8.4), 200 µM dNTPs, 1.5 mM MgCl₂, 10 pmol of sense and antisense primers, and 0.25 units of Taq polymerase (Life Technologies, Inc.).

Specific gene expressions were detected using specific oligonucleotides to involucrin (sense primer mInv 359, 5'-AACAGCAGCAAGAGCCACAAATG-3'; antisense primer mInv 612, 5'GCTTTTCTCCTGGAATCAGTTTC-3'), MUC1 (sense MUC1 3103, 5'-TAACGGAGATTTTCTGGGGATCT-3'; antisense MUC1 3653, 5'-AGGAAATAGACGATAGCCAAAGC-3'), nestin (sense nestSC129, 5'-GGAGGGTTGCGTCGGGGAAG-3'; antisense nestBR568, 5'-CTCCTCCTCGTGCGCGGCTCG-3'), HNF1 (sense HNFSC154, 5'-AGCGGGGCGGATCTCGACACCAA-3'; and antisense HNFBR465, 5'-GGGTCTTCATGGGGGTGCCCTTG-3'), smooth muscle -actin (sense ACTSC61, 5'-GGAAGACAGCACAGCCCTGGTGT-3'; antisense ACTBR391, 5'-TCATTTTCTCCCGGTTGGCCTTA-3') according to the mouse vascular smooth muscle -actin cDNA (GenBank^TM accession nos. X13297 and X07935).

The thermal cycling program was 95 °C for 45 s, following an hybridization step at 56 °C (involucrin), 57 °C (MUC1), 60 °C (HNF1), 59 °C (-actin) or 62 °C (nestin), each for 45 s, and an elongation step at 72 °C for 45 s. Involucrin (276 bp), MUC1 (289 bp), nestin (460 bp), HNF1 (334 bp), and -actin (353 bp) PCR products were run on a 2% agarose gel.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Expression of Endogenous Mouse MR mRNA-- In an attempt to determine the tissue specific pattern of endogenous mMR expression, we first analyzed mMR mRNA distribution in various tissues of B6D2 wild type mice by ribonuclease protection assays. As shown in Fig. 2, a reproducible expression pattern was obtained with five different animals, although with some individual differences. The colon was always the site of the highest level of mMR mRNA expression. The receptor was also strongly expressed in the lung, salivary glands, and brain. Interestingly, mMR transcripts were clearly detected in the testis and uterus, a finding that has not been reported to date. As expected, mMR mRNA was found in the kidney and heart, but a detectable signal could also be evidenced in the liver, spleen, muscle, and skin. Quantitative analyses were performed to precisely define the relative abundance of mMR transcripts among mouse tissues. Fig. 2B presents the results of mRNA levels expressed as a percentage of the colonic expression. Note that renal mMR level is relatively low, due to the fact that, in kidney, MR expression is restricted to a minority of cells of the distal nephron. This pattern of tissue-specific mMR expression was subsequently used as a basis to compare expression of the transgene driven by each of the hMR promoters.

View larger version (46K): [in this window] [in a new window]   Fig. 2.   Expression of mMR mRNA in various tissues of wild type mice. A, ribonuclease protection assay was used to detect mMR messengers in various organs of wild type B6D2 mice. A rat GAPDH probe was used as loading control. Co, colon; Lu, lung; SG, salivary glands; Br, brain; Te, testis; Ut, uterus; K, kidney; Li, liver; Sp, spleen; H, heart; Mu, muscle; Sk, skin; t, tRNA. B, the relative abundance of mMR mRNA was expressed as percentage of colonic expression (arbitrarily assigned to a 100% value). The mMR/GAPDH ratio was calculated from specific quantification by InstantImager; results are mean ± S.E. of four to five independent determinations from at least five different animals.

Analysis of P1.TAg Transgenic Mice-- The P1.TAg transgene consists of 1 kb of hMR P1 proximal promoter and 0.2 kb of exon 1 inserted upstream of the TAg coding sequence (see Fig. 1). Table I summarizes the main characteristics of transgenic animals obtained after two series of injections (Ref. 21 and present work). As already described for the first transgenic animals (21), all founders presented with malignant liposarcomas originating from brown adipose tissue (hibernomas) before 4 months of age. Hibernomas mostly developed in the dorsal region and were generally extremely voluminous (up to few grams), resulting in a buffalo-like appearance of the transgenic animal. Unfortunately, the moribund condition and/or the premature death of the transgenic P1.TAg animals prevented derivation of mice lines. Most transgenic founders were sacrificed before their condition deteriorated except mice 18 and 22, which appeared normal at sacrifice at 2 weeks of age. Complete autopsy was performed for each founder animal. Tissues samples were collected for histological evaluation, RNA extraction, and/or tissue culture attempts.

SV40 TAg mRNA expression was analyzed in various tissues of several transgenic founders by RPA. TAg signal was quantified and normalized by the GAPDH signal. There was no hybridization of TAg riboprobe with total RNA samples of wild type mouse organs (data not shown). Fig. 3 illustrates the tissue-specific pattern of transgene expression observed in mouse 22, a 2-week-old female. Although no hibernoma was observed in this animal, its interscapular brown adipose tissue already expressed high levels of TAg, suggesting that there might already be a transformation process ongoing, and that P1 promoter is probably transcriptionally active during early life in this tissue. Moreover, RPA revealed transgene expression in almost every tissue examined. Indeed, TAg was detected in the salivary glands, lung, brain, and liver and, at lower levels, in the skin. The level of transgene transcripts was relatively low in the kidney and colon of this animal.

View larger version (37K): [in this window] [in a new window]   Fig. 3.   Expression of TAg mRNA in various tissues of P1.TAg founder 22. TAg expression in different organs was analyzed by RPA. Fifty µg of total RNA were hybridized with TAg and GAPDH antisense riboprobes. B, brain; Li, liver; Sk, skin; M, muscle; SG, salivary glands; BAT, brown adipose tissue; Lu, lung; K, kidney; Th, thymus; R, rachis; Sp, spleen; Co, colon; t, tRNA.

As the integration site is known to influence the expression of the transgene, we compared the tissue-specific pattern of TAg expression in three other P1.TAg founder animals, 36, 40, and 44, in order to analyze the P1 promoter utilization (data not shown). Although absolute TAg expression levels vary among animals, relative tissue-specific expression was comparable and exhibited some common features. Most importantly, in all founders highest levels of transgene expression were detected in the testis and the salivary glands, while the heart, kidney, and colon expressed relatively low transgene mRNA amounts. In other tissues, intermediary levels of TAg messengers were detected with some gender differences in the liver where expression was weaker in males.

To compare the relative strength of hMR P1 promoter in vivo with that of the endogenous mMR regulatory sequences, we measured TAg and mMR expression in the same tissue samples. As shown in Fig. 4 for P1.TAg mouse 40, the tissue-specific pattern of mMR expression was similar to that observed in the wild type animals (see Fig. 2), indicating that the presence of TAg does not modify mMR transcription. The TAg/mMR ratio was then calculated for each tissue. In the salivary glands and testis, transgene expression far exceeds that of mMR; in contrast, TAg and mMR were almost equally expressed in the lung, liver, and brain. Although TAg expression was low in the heart, it was in the same range than that of mMR in this organ. Finally, in the kidney and colon, TAg mRNA levels were clearly lower than those of mMR transcripts.

View larger version (38K): [in this window] [in a new window]   Fig. 4.   TAg and endogenous mMR expression in P1.TAg founder 40. The levels of TAg, endogenous mMR, and GAPDH mRNA were determined in the same samples by RPA experiments on total RNA of different organs. Co, colon; Li, liver; Br, brain; H, heart; K, kidney; Te, testis; Lu, lung; SG, salivary glands. Specific signals were quantified by InstantImager and normalized to the amount of GAPDH transcripts. For each tissue, the calculated TAg/mMR ratio is indicated on the bottom of the figure.

Immunohistochemical studies with an anti-TAg antibody were used to determine which type of cells expressed the oncogene (Fig. 5). A strong nuclear staining was observed in the distal nephron, most notably in the distal convoluted tubule (Fig. 5A) and the cortical collecting ducts (Fig. 5B), whereas the proximal tubules and glomeruli were negative. This localization was identical to that of renal MR. In the liver (Fig. 5C), some hepatocytes presented a nuclear labeling, while some others were clearly negative, indicating that the transgene was not evenly expressed in hepatocytes. Similarly, TAg expression was clearly evidenced in some cells of the central nervous system (Fig. 5D), most likely neurons, while smaller nuclei of other cells were devoid of immunoreactivity. In the heart, some nuclei of cardiomyocytes exhibited a positive labeling, most of them presented nuclear abnormalities (data not shown).

View larger version (186K): [in this window] [in a new window]   Fig. 5.   Immunolocalization of transgene expression in kidney, liver and brain of P1.TAg transgenic mice. The presence of TAg was detected by immunohistochemistry using monoclonal anti-TAg antibody as described under "Materials and Methods." A, photograph of mouse 37 kidney section shows a strong nuclear labeling over the cells of the distal convoluted tubule (DCT) while the glomeruli (Gl) and proximal tubules (Pr) were devoid of immunostaining (magnification, ×350). B, photograph of renal section of mouse 40 illustrates the positive staining of the cortical collecting duct (CCD) (magnification, ×280). C, liver section of mouse 34 illustrates a positive although heterogeneous nuclear staining over hepatocytes (magnification, ×250). D, on a brain section of mouse 37, some cells nuclei present a clear immunostaining while other cells were negative (magnification, ×280).

Altogether, our results demonstrate that 1 kb of P1 promoter region is sufficient to direct transgene expression in all MR-expressing tissues, although the strength of its transcriptional activity varies among organs. In addition, the P1-driven TAg is expressed in an appropriate localization especially in the kidney.

Analysis of Transgenic P2.TAg Mice-- The P2.TAg transgene was constructed with 1.6 kb of the distal hMR P2 promoter and 0.1 kb of exon 1 driving expression of TAg cDNA (see Fig. 1).Table II presents results concerning the three founder animals obtained (mice 45, 54, and 55) and their offspring. Lines could only be developed for mice 45 and 54, as mouse 55 died before breeding. The three transgenic founders and all F1 animals examined presented with neurological abnormalities such as major motor disorders related to the formation of brain tumors. Importantly, none of the P2.TAg animals developed hibernomas. Transgenic mice died at approximately 6 months of age. At necropsy, macroscopic examination of all founders and some of their offspring revealed the presence of cerebral tumors associated with thymus hyperplasia. Histological studies disclosed neoplastic tissue in the brain. As presented in Fig. 6A, a typical field shows a highly cellular primitive neuroectodermal tumor (PNET) (22) composed of undifferentiated cells, most of them have a small round basophilic nucleus while other cells have larger, more pleiomorphic nuclei. Immunohistochemistry demonstrated a positive although heterogeneous staining with anti-vimentin antibodies (Fig. 6B), whereas neoplastic cells did not seem to express neurofilament, S100 protein, or glial fibrillary acidic protein (data not shown), markers of the neuronal and glial lineage, confirming the highly undifferentiated nature of the tumor. Moreover, PNET was shown to express substantial amounts of TAg mRNA as demonstrated by the very strong signal obtained with the antisense riboprobe in in situ hybridization experiments (Fig. 6, C1 and C2), as well as at the protein level since a clear nuclear staining was observed with the anti-TAg antibody (Fig. 6D).

View larger version (124K): [in this window] [in a new window]   Fig. 6.   Morphological, histological, and immunohistological analyses of P2.TAg transgenic mice. A, histological field of mouse 54 brain section, stained with hematoxylin-eosin, illustrates a highly cellular primitive neuroectodermal tumor (PNET) in the lower part of the picture. PNET was composed of undifferentiated cells with basophilic nuclei. Note the normal aspect of the surrounding brain tissue, which contains numerous neuron cell bodies. B, immunohistochemistry of mouse 54 PNET demonstrates a positive although heterogeneous staining with anti-vimentin antibodies. C, photographs of in situ hybridization of mouse 54 PNET with a 35S-TAg-labeled antisense riboprobe. C1 is a brightfield view, whereas C2 illustrates darkfield view of the same zone. A high specific signal was observed on neoplastic cells of PNET but no specific signal was found in the brain. D, immunohistochemical experiment with anti-TAg antibodies of mouse 55 cerebral PNET. TAg was also detected at the protein level with most of the neoplastic cells presenting a clear nuclear staining. A-D, magnification, ×290. E, the P2.TAg mouse 54.1.10, a 5-month-old female, exhibits a voluminous leiomyosarcoma above its left eye (white arrow) and some small tumors developed over its inferior right eyelid and its lips. F, hematoxylin-eosin stained section of this tumor shows a well differentiated leiomyosarcoma. G and H, immunohistochemistry discloses a cytoplasmic staining of the tumoral cells with anti-smooth muscle actin antibodies and a nuclear staining with anti-TAg antibodies. F-H photographs, magnification ×180.

All the eight F1 littermates of mouse 45 died before breeding. Some animals of line 54 and founder 55 also developed thymus hyperplasia with no sign of tumorigenesis. The structural organization was conserved, and immunohistochemical studies demonstrated that some cells, most probably the epithelial reticular cells of the thymus, were clearly positively stained with anti-TAg antibodies (data not shown). This finding is consistent with the presence of MR in some cells of the thymus detected by immunohistochemical techniques (23) and in situ hybridization in the mouse fetus (24). However, no immortalized cell line could be established from this tissue.

Most F2 and F3 animals of line 54 did not develop PNET, and no TAg expression could be detected in their brain. These results were confirmed by in situ hybridization experiments that failed to detect any specific signal in the normal part of the brain of mouse 54. In contrast to founders and F1 transgenic animals, they developed facial tumors preferentially located on their lips and eyelids (Fig. 6E), which generally appeared before the first year of age. One animal (54.1.10) presented with a metastatic localization in the axillary region, while another transgenic female (54.1.2.11) developed a large tumor in one uterine horn. Histopathological analyses of these tumors revealed well differentiated leiomyosarcomas. As shown in Fig. 6F, there was an increased cellularity and a nuclear pleiomorphism with numerous mitotic figures. Immunohistochemical studies showed a cytoplasmic staining with anti-smooth muscle actin (Fig. 6G) and anti-desmin antibodies (data not shown), while a nuclear staining was observed with anti-TAg antibodies (Fig. 6H). All these results confirm the malignant nature and the smooth muscle origin of the tumors.

We next examined the tissue distribution and levels of TAg mRNA and endogenous mMR in the P2.TAg transgenic mice. RPA experiments performed on different organs of founders 55, 54, and 45 are presented in Fig. 7. As expected, a very strong expression of TAg was observed in all transgenic mice brains in which PNET had developed, whereas no signal was detected in the brain of a F2 mouse of line 54, which did not present cerebral PNET. Beside neoplastic brain and thymus (data not shown), all other tissues examined, such as uterus, testis, and lung, expressed very low levels of TAg compared with endogenous mMR messengers. Some differences were observed between transgenic lines. For instance, renal transgene expression was detected only in line 45 animals, while TAg messengers were present in salivary glands of line 54 animals and the heart of founder 55 (Fig. 7). No transgene expression was found in colon, skin, or spleen of any line (data not shown).

View larger version (89K): [in this window] [in a new window]   Fig. 7.   Expression of TAg and mMR mRNA in various tissues of P2.TAg mice. TAg and mMR mRNA expression in some organs of founder 55, line 54 animals, and founder 45 were analyzed in the same sample by RPA. Fifty µg of total RNA were hybridized with TAg or mMR antisense riboprobe and with a GAPDH antisense riboprobe as loading control. Br, brain; H, heart; Lu, lung; K, kidney; Co, colon; SG, salivary glands; Te, testis; Ut, uterus; F2, F2 generation mice; t, tRNA. The ratio TAg/mMR is indicated on the bottom of the figure for each organ. * indicates neoplastic brains, where a shorter exposition for autoradiography was performed.

Establishment of New Cell Lines by Targeted Oncogenesis-- As we were interested in developing cellular models that remain sensitive to aldosterone action, numerous attempts were performed to derive cell lines originating from different neoplastic as well as normal tissues of P1.TAg and P2.TAg transgenic mice. Table III summarizes the main characteristics of different cell lines that were successfully established using our targeted oncogenesis strategy. Apart from the T37i cell line, which was derived from an hibernoma and has been previously described (21), P1.TAg mice allowed us to generate cell lines originating from the whole brain (F6), hippocampus (BZ), lung (PP), salivary glands (SAL), glabrous skin of palms (PAL), and liver (LUCA). Unfortunately, despite several attempts and an appropriate oncogene expression, no cellular model could be obtained from isolated cardiomyocytes, colonic enterocytes, or various tubular segments of the distal nephron such as cortical collecting ducts or the medullary and papillary ascending limb. In contrast to P1.TAg-derived cell lines, no cell lineage could be generated from normal organs of P2.TAg transgenic mice. This is perhaps related to the very low expression level of the oncogene in non-neoplastic tissues of these animals. However, we could establish several cell lines (JO, LIPT, and AXI) from leiomyosarcomas, but we failed to obtain cell lines from the primitive neuroectodermal tumor.

View this table: [in this window] [in a new window]   Table III Cell lines established by targeted oncogenesis TAg and mMR mRNA were analyzed by RPA. UCP1 (uncoupling protein 1) RNA were detected by Northern blot. MUC1, INVO (involucrin), NESTIN, ACTIN, and HNF1 were detected by RT-PCR. TAg protein was detected by immunocytochemical experiments with an anti-TAg antibody in all these cell lines.

As expected, all immortalized cells expressed high levels of TAg at both mRNA level and protein level, as revealed by RPA and immunocytochemistry, respectively (data not shown). More importantly, RPA experiments also indicate that mMR mRNA expression was maintained in most cell lines as presented in Fig. 8A. All these cell lines have been propagated continuously for 6 months and can be stored in liquid nitrogen.

View larger version (56K): [in this window] [in a new window]   Fig. 8.   Characterization of cell lines derived from P1.TAg and P2.TAg animals. A, detection of mMR mRNA by RPA with a 32P-specific labeled antisense probe in the following cell lines: BZ, F6, LUCA, JO, PP, PAL, AXI, SAL; t = tRNA. RT-PCR analyses of marker gene expression in various cell lines: MUC1 in PP cells (B), involucrin in PAL (C), nestin in BZ and F6 (D), HNF1 in LUCA (E), and -actin in AXI and LIPT (F) by RT-PCR. Lengths of specific PCR product amplified were 289 bp (MUC1), 276 bp (involucrin), 460 bp (nestin), 334 bp (HNF1), 353 bp (-actin). Reactions were separated on a 2% agarose gel with ethidium bromide, visualized under UV light, and photographed. , molecular weight (X174RF/DNA/Hae fragments); Lu, wild type mouse lung; Sk, wild type mouse skin; INVO, involucrin; NES, nestin; ACT, smooth muscle -actin; Br, wild type mouse brain; Li, wild type mouse liver; T, mouse 54.1.4 facial tumor (leiomyosarcoma); RT, negative control (no RT template).

To determine precisely the origin of these cell lines, expression of some specific marker genes was investigated by RT-PCR. As shown in Fig. 8B, MUC1 transcripts, specifically expressed in glandular epithelial cells notably in bronchi (25, 26), were detected in PP cells as well as in wild type mouse lung. We also demonstrated the presence of involucrin RNA messengers, an epidermal marker gene expressed in the granular and spinous layers of the skin (27), in two samples of PAL cells as well as wild type mouse skin (Fig. 8C), confirming that these cells were derived from epidermis. Nestin, an intermediate filament marker of neuronal as well as glial stem cells (28), was also expressed in both F6 and BZ cells and in wild type brain (Fig. 8D). In LUCA cells and wild type mouse liver, the liver-enriched transcriptional factor HNF1 was detected, confirming the hepatic origin of this cell line (Fig. 8E) (29). Finally, LIPT and AXI leiomyosarcoma-derived cell lines as well as a facial tumor both exhibited smooth muscle -actin, which is known to be a specific marker of smooth muscle cells (Fig. 8F). Altogether, these results demonstrate the relatively well differentiated features of all mMR-expressing cell lines obtained.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

The hMR expression is driven by two alternative promoters, the proximal P1 and distal P2 promoters, located upstream of the two first untranslated exons 1 and 1, respectively (16). The relative contribution of P1 and P2 promoter sequences in directing tissue-specific expression of hMR  and  transcripts is still unclear. Functional characteristics of P1 and P2 regulatory regions have previously been investigated in two cellular models, i.e. undifferentiated fibroblasts and differentiated epithelial renal cells (15). Transient transfection assays demonstrated that P1 drives a stronger expression of a luciferase reporter gene than P2 promoter, but that only P2 is responsive to mineralocorticoid induction. In an attempt to study P1 and P2 promoter functions in vivo, fragments of each promoter (HA, 1.2 kb of P1; SS, 1.7 kb of P2) were used to direct TAg expression in transgenic mice, TAg being used both as reporter gene and oncogenic protein.

In the first part of this study, the endogenous mMR expression pattern was analyzed by ribonuclease protection assays in B6D2 wild type mice, as no such extensive study in mouse has been reported as yet. In accordance with results obtained in human and rat (20, 30, 31), mMR was strongly expressed in classical mineralocorticoid target organs such as the colon, salivary glands, and also in the brain and lung. Lower levels of mMR were found in the kidney (keeping in mind that MR expression is restricted to the distal parts of the nephron) and heart. Highest levels of mMR transcripts were always detected in the colon, a finding that has not been observed in other species. Importantly, we provide evidence that mMR is also present in the testis and uterus, suggesting that MR might play a role in the physiology of reproduction. Finally, low levels of mMR have also been detected in the skeletal muscle. It is not known whether this is related to the expression of MR in blood vessels and hematopoietic cells (32, 33). Altogether, this tissue-specific expression pattern raises the question of the exact role of MR protein and aldosterone in non-epithelial tissues and how MR could discriminate mineralocorticoid and glucocorticoid hormones in these different cell types.

The nature of tumors developed in P1.TAg and P2.TAg transgenic animals was totally unexpected. The peculiar and unique phenotype of P1.TAg mice was the formation of lethal hibernomas that arose 2 weeks to 4 months after birth, indicating early activation of P1 sequences in brown adipose tissue, most probably during fetal life. On the other hand, two types of tumors were observed in P2.TAg animals, cerebral PNET, which always occurred before 6 months of age, and leiomyosarcomas, which generally appeared in older animals, approximately at 1 year of age. Leiomyosarcomas had essentially a facial tropism and usually arose at the junction of the keratinized and non-keratinized epithelia. The reason of such localization remains unknown. However, similar phenotypes have been described in transgenic mice harboring a hsp70 promoter fused to TAg (34). It is noteworthy that leiomyosarcoma-carrying animals were all offspring of the same 54 F1 transgenic mouse. The formation of these low proliferating neoplasms appeared to be an independent clonal expansion rather than emergence from a precise organ. In addition, the primitive and/or undifferentiated features of these tumors suggest that P2 promoter might preferentially be utilized during fetal life and could be activated by transcription factors expressed at a specific developmental stage. This is consistent with the early expression of MR detected in mouse and human fetus (24, 35). The lack of neoplastic formation in other tissues and organs despite the presence of TAg expression, suggests that additional factors cooperating in a tissue-specific manner might confer distinct oncogenic susceptibilities.

Our study also demonstrates important differences in P1 and P2 hMR promoter characteristics not only in their basal transcriptional activities but also in their tissue-specific utilization. Table IV summarizes transgene expression patterns observed in several transgenic mice of different lines for each construct. It presents the relative strength of P1 and P2 promoters compared with the mMR expression pattern already established (see Fig. 2). The comparison of the ratio TAg/mMR in P1.TAg and P2.TAg mice (Figs. 4 and 7) allowed us to define the tissue-specific utilization of P1 and P2 and their relative strength in vivo compared with that of endogenous mMR regulatory sequences. Thus, P1 appears to be a relatively strong and widespread active promoter in all MR-expressing tissues. A strong activity was detected in the salivary glands and testis, indicating that 1 kb of P1 promoter is sufficient to drive a high level of transgene expression in these organs. It is worth noting that, in P1.TAg mice, renal expression of the transgene colocalizes with that of MR, which is only found in the distal nephron (36), indicating that the P1 promoter contains all elements necessary to drive cell-specific expression in the kidney. The proximal P1 promoter could be therefore useful to specifically target a gene of interest in the mineralocorticoid-sensitive parts of the distal nephron. However, due to the relatively low level of TAg expression, it is likely that other enhancer elements outside the 1-kb HindIII-AvaII (HA) fragment of P1 are involved in renal MR expression. In contrast to P1.TAg animals, a weak and very restricted activity driven by 1.6 kb of exon 1 5'-flanking region was detected in P2.TAg mice. P2 strength was generally 10 times weaker than that of P1, confirming our previous results obtained by transient transfection assays (15). The transcription factor Sp1, which recognizes GC-rich sequences, has been shown to interact with several TATA-binding protein-associated factors and thereby to be implicated in transcription initiation of TATA-less promoters (37). In this context, it is interesting to note that within the first 400 bp upstream of the transcription start site, seven consensus Sp1 binding sites are present in the P1 promoter while only three Sp1 sites are found in the P2 region (15). This could account for differences in the basal strength of the two hMR regulatory regions in vivo. Altogether, we hypothesize that P1 might govern the tissue-specificity and the level of hMR expression, while P2 could mainly participate to the fine-tuning of MR gene expression under different physiological and pathological states. Unfortunately, due to the high lethality of the phenotypes, it was not possible to examine modification of transgene expression under various experimental conditions (adrenalectomy, hormonal treatment, alterations of water and salt balance). The use of inert reporter genes such as -galactosidase or luciferase instead of TAg could have been suitable for analyzing the developmental activity of hMR alternative promoters but obviously precludes establishment of immortalized cell lines.

Indeed, targeted oncogenesis programs using TAg in transgenic mice have been largely employed (17, 38). Using this strategy, we succeeded in generating many different cell lines. Numerous tissues and tumors of P1.TAg and P2.TAg transgenic mice were cultured. The T37i cells derived from a dorsal hibernoma of P1.TAg founder 37 have been shown to undergo terminal differentiation into mature brown adipocytes, which express UCP1, a specific marker of brown fat (21, 39). In addition, we could establish other cell lines from normal tissues of P1.TAg animals. They originated from the liver, brain, palmary skin, lung, and salivary glands. These cellular models offer interesting opportunities to better analyze independently P1 and P2 promoter activities in a specific cellular context and in an appropriate chromatin structure and to study their regulatory factors. It is of particular interest that mMR mRNA expression is maintained in most of these cells. Since TAg expression is under the control of regulatory sequences at least in part similar to those controlling endogenous mMR expression, this appears to be a decisive factor for the maintenance of some differentiated features of these cells, most notably various elements of the mineralocorticoid signaling pathway. Indeed, we showed that all cell lines in which mMR mRNA expression is maintained expressed a specific marker from their tissue of origin under standard culture conditions. If the presence of functional mMR protein could be demonstrated, these models would obviously be of great interest with respect to the study of MR expression and function in various cellular contexts. Such MR-expressing cells may constitute powerful systems to identify target genes that are transcriptionally regulated by aldosterone.

In summary, targeted oncogenesis in transgenic mice demonstrates a distinct tissue-specific utilization of P1 and P2 hMR promoters. These promoters clearly differ by their relative strength in a given tissue and govern differential expression pattern. Our results indicate that in vivo hMR gene expression is under the control of complex regulatory mechanisms including utilization of alternative promoters and generation of different mRNA isoforms. Elucidation of the molecular events involved in tissue-specific expression of hMR constitutes an important step toward better knowledge of widespread and pleiotropic actions of aldosterone.

	ACKNOWLEDGEMENTS

We are indebted to Dr. L.-M. Houdebine for providing us with the transgenic vector H31 and to Dr. I. Cerutti and the Service d'Expérimentation Animale et de Transgenèse (SEAT) for the generation of transgenic animals. We also thank Drs. G. Schütz and S. Berger for their generous supply of the mMR probe, Drs. N. Farman and J.-P. Bonvalet for helpful discussions, and J. Grellier and S. Roger for pictures and illustrations.

	FOOTNOTES

^* This work was supported in part by INSERM and by a grant from the Ligue Nationale Française contre le Cancer (to M. L.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^§ Recipient of a doctoral fellowship from the Ministère de l'Education Nationale et de la Recherche.

To whom correspondence should be addressed: INSERM U478, Institut Fédératif de Recherche Cellules Epithéliales, Faculté de Médecine Xavier Bichat, 16, rue Henri Huchard, BP416, 75870 Paris Cedex 18, France. Tel.: 33-1-44-85-63-19; Fax: 33-1-42-29-16-44; E-mail: mlombes@bichat.inserm.fr.

	ABBREVIATIONS

The abbreviations used are: MR, mineralocorticoid receptor; hMR, human mineralocorticoid receptor; mMR, mouse mineralocorticoid receptor; TAg, large T antigen; bp, base pair(s); kb, kilobase pair(s); RT, reverse transcription; PCR, polymerase chain reaction; RPA, ribonuclease protection assay; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; PNET, primitive neuroectodermal tumor; Pipes, 1,4-piperazinediethanesulfonic acid; DTT, dithiothreitol; PBS, phosphate-buffered saline.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES