Repression of glucocorticoid receptor gene transcription by c-Jun

Molecular and Cellular Endocrinology 175 (2001) 67 – 79 www.elsevier.com/locate/mce

Repression of glucocorticoid receptor gene transcription by c-Jun Ana L.B. Cabral a,b, Angela N. Hays c, Paul R. Housley c, Maria M. Brentani d, Vilma R. Martins e,* a Ludwig Institute for Cancer Research, Sa˜o Paulo 01509 -900, Brazil Departamento de Bioquı´mica, Instituto de Quı´mica da Uni6ersidade de Sa˜o Paulo, Sa˜o Paulo, SP, Brazil c Department of Pharmacology and Physiology, Uni6ersity of South Carolina School of Medicine, Columbia, SC 29208, USA d Laborato´rio de Oncologia Experimental, Disciplina de Radiologia, Faculdade de Medicina da Uni6ersidade de Sa˜o Paulo, Sa˜o Paulo, Brazil e Centro de Tratamento e Pesquisa Hospital do Caˆncer, Rua Prof. Antoˆnio Prudente 109 /4A, 01509 -010, Sa˜o Paulo, Brazil b

Received 11 October 2000; accepted 9 January 2001

Abstract The regulation of glucocorticoid receptor gene expression by members of the AP-1 family was examined in glucocorticoid-free NIH3T3 cells transfected with the human glucocorticoid receptor gene promoter driving expression of a CAT reporter gene. c-Jun inhibited the promoter activity by 80% and JunB by 30%, whereas c-Fos and JunD had no inhibitory effect. Electrophoretic mobility shift assays showed that c-Jun is unable to efficiently interact with the AP-1-like site present in the human glucocorticoid receptor promoter. Moreover, c-Jun was still able to repress promoter mutants in which the region containing the AP-1-like site was deleted. NIH3T3 cell clones overexpressing c-Jun exhibited lower glucocorticoid receptor mRNA levels, which suggests that the murine glucocorticoid receptor gene can also be regulated by AP-1. These results provide a new mechanism for cross-talk between the glucocorticoid receptor and the AP-1 family of transcription factors in the absence of glucocorticoid ligands. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Glucocorticoid receptor; AP-1; Glucocorticoid receptor gene promoter; Gene regulation; NIH3T3

1. Introduction The glucocorticoid hormones play important roles in development, differentiation and homeostasis (Evans, 1988). Glucocorticoids are widely used for their ability to inhibit inflammation and to limit proliferative responses in wound healing, and in chronic destructive diseases (Brinckerhoff et al., 1986; Allison, 1988). Additionally, glucocorticoids are able to block tumor promotion in the mouse skin carcinogenesis system (Belman and Troll, 1972; Scribner and Slaga, 1973) and to reverse the transformed phenotype of glioma cells (Valentini and Armelin, 1996). The cellular response to glucocorticoids is mediated by the glucocorticoid receptor (GR) which modulates gene transcription by induction or repression (Evans, * Corresponding author. Present address: Fundacao Antonio Prudente, Rua Antonio Prudente 109 4A, 01509-010 Liberdade, S.P., Brazil. Tel.: + 55-11-2704922; fax: + 55-11-2707001. E-mail address: [email protected] (V.R. Martins).

1988; Beato, 1989; Parker, 1991). Hormone binding induces GR activation and translocation to the nucleus where the hormone-receptor complex binds to cis-acting DNA sequences known as glucocorticoid responsive elements (Evans, 1988; Beato, 1989). The cellular concentration of GR is the most important determinant of cellular sensitivity to hormone (Vanderbilt et al., 1987). Therefore, the regulation of GR gene expression is critical to the cellular response to glucocorticoids. The human GR promoter has been cloned and sequenced (Zong et al., 1990; Encio and DeteraWadleigh, 1991; Leclerc et al., 1991). It contains putative elements for specific transcription factors such as AP-1 and CREB (Zong et al., 1990), which suggests that members of the AP-1 family and cAMP might regulate GR gene expression. The observation that GR levels are regulated during the cell cycle (Cidlowski and Michaels, 1977; Cidlowski and Cidlowski, 1982) supports a model where protein factors regulated during the cell cycle could modulate GR expression. Vig and Vedeckis (1992) observed that cell treatment with a

0303-7207/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S0303-7207(01)00396-3

68

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

phorbol ester, which activates AP-1 family members (Angel et al., 1987; Scho¨ ntal et al., 1988), and with 8Br-cAMP, which activates CREB, decreased GR mRNA levels. Moreover, c-Jun and GR mRNAs are coordinately regulated in triamcinolone acetonidetreated cells (Vig et al., 1994; Barrett et al., 1996) and recent results show that ligand-bound GR can reduce expression of cellular c-Jun through transrepression at the AP-1 sites in the c-Jun promoter (Wei et al., 1998). The AP-1 family of proteins belongs to a class of transcription factors encoded by protooncogenes which are involved in cell proliferation (Angel and Karin, 1991). The proteins c-Jun, JunD and JunB form homodimers or heterodimers with c-Fos, FosB, Fra-1 and Fra-2 (Ryseck and Bravo, 1991). It is likely that a preferential composition of homodimers and heterodimers during the cell cycle determines the regulation of specific gene expression (Kovary and Bravo, 1992). Many results indicate multiple regulatory interactions between GR and AP-1 family members in different cell types (Miner and Yamamoto, 1991; Schu¨ le and Evans, 1991; Pfahl, 1993), and if c-Jun regulates GR gene expression, the potential complexity of these interactions would be increased. Previous work from Wei and Vedeckis (1997) showed that NIH3T3 cells overexpressing c-Fos exhibit enhanced GR promoter activity, and that the promoter AP-1-like site was required for c-Fos stimulation. The GR promoter AP-1-like site is bound by nuclear proteins in complexes containing Fos family members and JunD or JunB, but not c-Jun (Breslin and Vedeckis, 1996b). The present study was undertaken to explore the effects of c-Jun on the expression of the GR gene. Using transient transfection we observed that the human GR promoter is inhibited by c-Jun in a dose-dependent manner. JunB is less effective in promoter repression, and JunD and c-Fos do not repress promoter activity. In contrast to results demonstrating that heterodimers are most effective for induction of AP-1 target genes (Ryseck and Bravo, 1991), c-Jun homodimers appear to be the form of AP-1 which inhibits the GR promoter. This observation was supported by experiments with 15 NIH3T3 clones that overexpress c-Jun protein, in which the endogenous GR steady state mRNA levels are reduced. Moreover, the activity of a transiently transfected human GR promoter was strongly inhibited in these clones compared to parent NIH3T3 cells. It is unlikely that c-Jun acts by interaction with the AP-1-like site present in the promoter, as electrophoretic mobility shift assays showed that c-Jun binds this element with very low affinity. Additionally, deletion mutants in which the AP-1-like element was removed from the GR promoter are still inhibited by c-Jun. Therefore, the repression mediated by c-Jun might be due to its interaction with and inactivation of

other transcription factors involved in GR promoter gene activation. These results represent a new example of cross-talk between the GR and AP-1 signaling pathways. In addition to the particular promoter and cell type differences in gene expression exhibited by interactions between the GR and AP-1 proteins (Miner and Yamamoto, 1991; Schu¨ le and Evans, 1991), c-Jun is specifically able to inhibit the promoter for the GR gene.

2. Results

2.1. Human GR gene promoter regulation by AP-1 NIH3T3 cells were transiently transfected with phGRpromCAT, which contains the bacterial gene for chloramphenicol acetyltransferase (CAT) under control of the human glucocorticoid receptor promoter. Cotransfection of the constitutive RSVc-Jun expression vector inhibits GR promoter activity in a dose-dependent manner, down to approximately 20% of control values (Fig. 1). The effects of other AP-1 family members were analyzed, as shown in Fig. 2. Twenty micrograms of constitutive expression vectors for v-Jun, c-Jun, c-Fos, JunD, JunB, or the mutant c-JunD132 – 220 were co-transfected into NIH3T3 cells. This amount of DNA exceeds the amount necessary for maximum inhibition by pRSVc-Jun (Fig. 1). The GR promoter activity decreased to 239 6% of the control value (P= 0.001) with c-Jun; the same effect was mediated by v-Jun (2295%). These results suggest that the conserved regions in the viral c-Jun protein are responsible for the attenuation of the GR promoter activity. The c-Jun mutant deleted for amino acids 132–220 was

Fig. 1. Repression of human GR promoter activity by c-Jun. NIH3T3 cells were cotransfected with phGRpromCAT, CMVbgal, and RSVcJun at the amount indicated. Cells were harvested 60 h after transfection and cell extracts containing equal amounts of b-galactosidase were assayed for CAT activity, which is expressed as activity relative to the basal promoter activity. Values are the mean of at least four independent experiments 9 SEM.

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

Fig. 2. c-Jun is the form of AP-1 most active in GR promoter repression. NIH3T3 cells were cotransfected with the phGRpromCAT reporter, CMVbgal, and RSVv-jun, RSVc-Jun, RSVc-fos, RSVjunD, RSVjunB or RSVc-JunD132–220. Extracts were assayed as described in the legend to Fig. 1. Values are the mean of at least three independent experiments 9 SEM and (*) indicates values significantly different from control (PB 0.05).

still able to repress (489 5% of control). This inhibition is less (P= 0.002) than that mediated by the wild-type c-Jun protein, suggesting that the deleted region contributes to the effect of c-Jun on the GR promoter. Transfection of RSVc-fos did not attenuate the activity of the GR promoter. Indeed, c-Fos seems to stimulate promoter activity slightly (121929%, P =0.09), consistent with the observations of Wei and Vedeckis (1997). The GR promoter activity decreased to 70921% (P= 0.009) of control values after transfection with JunB, while JunD did not show any significant effect (919 19%). Compared to c-Jun, JunB caused less repression (P= 0.002) and the degree of repression did not increase when larger amounts (50 and 80 mg) of DNA were co-transfected (data not shown). These results demonstrate that AP-1 family members differentially regulate GR promoter activity.

69

It is possible that endogenous c-Fos levels in these cells are in vast excess over c-Jun levels, and that transfection of exogenous c-Jun actually favors formation of more c-Jun/c-Fos heterodimers, resulting in repression of the GR promoter. In order to address this possibility, mRNA was analyzed from NIH3T3 cells as shown in Fig. 4. In asynchronous cells (Asyn) there is a modest amount of c-Jun mRNA and no detectable c-Fos mRNA. Serum-starvation induces quiescence (Q) and greatly reduces the level of c-Jun mRNA, and subsequent serum stimulation for 30 min (FBS 30%) induces the mRNA for both c-Jun and c-Fos. The level of c-Jun mRNA is even higher after 60 min of serum stimulation (FBS 60%), but the level of c-Fos mRNA has already declined appreciably. To compare protein levels we performed immunoblot experiments on nuclear extracts of NIH3T3 cells with multiple antibodies. Nuclear extracts from asynchronous cells, from serum-starved cells, from serum-stimulated cells, and from cells transfected with RSVc-Jun and RSVc-fos were analyzed with antibodies pan-reactive with c-Jun, JunB, and JunD as well as antibody specific for c-Jun. In all cases the amount of c-Jun protein paralleled the amount of c-Jun mRNA (data not shown). Im-

2.2. c-Jun homodimers repress the GR gene promoter In order to analyze the activity of c-Jun/c-Fos heterodimers on the GR promoter we transfected cells with different concentrations of the RSVc-Jun vector alone or with equal amounts of RSVc-fos. Fig. 3A shows that repression increases with increased amounts of RSVc-Jun, as observed in the experiments of Fig. 1. However, co-transfection of equal amounts of RSVcfos did not alter the pattern of inhibition. In order to favor heterodimer formation we co-transfected cells with 0.5 mg RSVc-Jun plus 3, 5 and 10-fold excess RSVc-fos vector (Fig. 3B). This amount of RSVc-Jun alone results in less than maximal repression and any c-Fos effect can be detected. Even under conditions which should favor heterodimer formation between cJun and c-Fos, attenuation of the GR promoter was equivalent to that mediated by transfection with RSVcJun alone.

Fig. 3. Homodimers of c-Jun repress the GR promoter. NIH3T3 cells were cotransfected with the phGRpromCAT reporter, CMVbgal, and RSVc-Jun plus RSVc-fos (ratio 1:1) at the amount indicated (A), or with 0.5 mg RSVc-Jun plus 1.5, 2.5 or 5.0 mg RSVc-fos (B). Extracts were assayed for CAT activity as described in the legend to Fig. 1. Values are the mean from at least three independent experiments 9SEM.

70

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

Fig. 4. NIH3T3 cell mRNA levels for c-Jun and c-Fos. RNA was isolated from NIH3T3 cells cultured under the indicated conditions and probed for c-Jun, c-Fos, and GAPDH mRNA. Control cells were growing in 10% serum in asynchronous condition (Asyn). To achieve quiescence, cells were serum-starved in medium containing 0.5% serum for 48 h (Q). Quiescent cells were stimulated with medium containing 10% serum (FBS) for 30 min (FBS 30%) or 60 min (FBS 60%).

munoblots using antibodies specific to c-Fos or pan-reactive to c-Fos, FosB, Fra-1, and Fra-2 exhibited no reaction for c-Fos in any sample except that from serum-stimulated cells. Endogenous levels of FosB, Fra-1, and Fra-2 were barely detectable in these experiments, with the strongest reaction from serum-stimulated cells (data not shown). These results suggest that, to a first approximation, the level of mRNA is similar to the level of protein for c-Jun and c-Fos in NIH3T3 cells. Thus, c-Jun is not limiting in NIH3T3 cells compared to c-Fos, and transfection with RSVc-Jun should favor c-Jun/c-Jun homodimer formation.

2.3. AP-1 does not bind efficiently to the AP-1 -like element present in the human GR promoter We performed electrophoretic mobility shift assays to test protein binding to the AP-1-like element present in the GR promoter (Fig. 5A). Using nuclear extracts from NIH3T3 cells, nuclear protein binding to the − 905 to −885 region of the human GR promoter (AP-1-like: TCGAAGTGACACACTTCACGC) is less than that observed for the consensus AP-1 sequence (AP-1: CGCTTGATGACTCAGCCGGAA), and similar to that observed for a mutant AP-1 element (AP-1 mut: CGCTGATGACTTGGACTCAGCCGGAA) which is known for its inability to regulate gene expression mediated by AP-1 factors (Lee et al., 1987). The

reaction is specific for c-Jun since extracts immunodepleted with c-Jun antibodies did not show any binding. The specificity of the interaction is also demonstrated when the reaction (using radiolabeled consensus AP-1) is inhibited by competition with 100-fold excess unlabeled AP-1. Similar results were obtained with nuclear extracts from NIH3T3c-Jun clones (vide infra) containing approximately twice the amount of c-Jun protein found in parent NIH3T3 cells. Although these extracts contain more protein able to participate in complex formation with both the consensus AP-1 sequence and the promoter AP-1-like sequence, protein binding to the promoter AP-1-like sequence is much weaker than to the AP-1 consensus sequence (data not shown). The affinity of c-Jun for the AP-1-like site present in the human GR promoter is illustrated in Fig. 5B. NIH3T3 nuclear extract protein binding to the labeled consensus AP-1 oligonucleotide was inhibited by 80% when the reaction included 2.5-fold excess (1.75 ng) unlabeled AP-1 oligonucleotide, while only 10% inhibition was obtained when the GR promoter AP-1-like oligonucleotide was used as the unlabeled competitor. Total inhibition of c-Jun binding was obtained with 20-fold excess (14 ng) consensus AP-1 oligonucleotide; this amount of GR promoter AP-1-like oligonucleotide attenuated binding by only 50%. These data confirm that c-Jun has a lower affinity for the human GR promoter AP-1-like element than for the consensus AP-1 sequence. It is interesting to note that the degree of competition with the AP-1-like oligonucleotide reaches a plateau (Fig. 5B), suggesting that there are at least two populations of factors that bind the consensus AP-1 sequence and only one that binds the AP-1-like sequence.

2.4. GR promoter inhibition by c-Jun is not abolished by deleting the AP-1 -like element Various deletions of human GR promoter sequences were constructed to localize the region that mediates c-Jun repression. Fig. 6 shows a schematic representation of the mutants, their respective basal activity and their activity after cotransfection with RSVc-Jun. Deletion of the region − 1043 to − 469 (mutant 1) or a more restricted region from −899 through −897 at the AP-1-like site (mutant 5) did not change either the promoter basal activity or the repression by c-Jun (P\ 0.05). These results confirm our interpretation of the band shift assays and demonstrate that repression by c-Jun is not due to its interaction with the AP-1-like element. Mutant 2, deleted for nucleotides − 2738 to − 1043, has the same basal activity and degree of c-Jun repression as the intact promoter (P\ 0.05), suggesting that the proximal  1 kb of the promoter contains all the elements for c-Jun repression. Mutant 3, containing only the proximal promoter sequences −469 to +24,

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

exhibited twice the basal activity of the intact promoter (P B 0.05). This deletion apparently removed a silencer element. Mutant 3 was still repressed by c-Jun, although the magnitude of repression is reduced to about 50%. In the absence of this silencer element, promoter

71

activity may be high enough that c-Jun cannot inhibit by more than 50%. Alternatively, it is possible that an element contributing to c-Jun repression has been removed in mutant 3, although the results with mutants 1 and 2 argue against this possibility. Mutant 4, deleted

Fig. 5. The AP-1-like element in the hGR promoter does not bind c-Jun efficiently. (A): 32P-labeled double-stranded probes containing the AP-1 consensus element (CGCTTGATGACTCAGCCGGAA), the mutant AP-1 element (CGCTGATGACT TGGACTCAGCCGGAA), or the AP-1-like sequence corresponding to nucleotides − 905 to − 885 of the human GR promoter (TCGAAGTGACACACTTCACGC) were assayed alone, with 5 mg of NIH3T3 nuclear extract (NE) protein, or with 5 mg of protein from nuclear extract immunodepleted with anti-c-Jun antibody (NE, c-Jun depleted). The reaction in lane 10 also contained 100-fold excess unlabeled consensus AP-1 oligonucleotide (unlabeled AP-1). (B): Binding reactions containing 0.7 ng of 32P-labeled AP-1 consensus probe were incubated with 5 mg of NIH3T3 cell nuclear extract protein in the presence of 1.8, 3.5, 5.3, 7, and 14 ng (2.5, 5, 7.5, 10, and 20-fold excess) of the unlabeled AP-1 consensus, the AP-1 mutant, or the AP-1-like oligonucleotides. Values are expressed relative to the value for the AP-1 consensus oligonucleotide.

72

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

2.5. Endogenous GR gene expression of NIH3T3 cells is attenuated by c-Jun

Fig. 6. Functional activity of hGR promoter deletion mutants. The indicated phGRpromCAT constructs and CMVbgal were cotransfected into NIH3T3 cells in the absence (basal) or presence ( +c-Jun) of 20 mg of RSVc-Jun. Extracts were assayed for CAT activity as described in the legend to Fig. 1. Values are the mean of at least three independent experiments 9 SEM.

for nucleotides − 755 to −242, exhibited only onethird the basal activity of the intact promoter (PB 0.05), suggesting that this region contains induction elements. Mutant 4 was repressed about 40% by c-Jun, although the low level of basal activity complicates an accurate determination of the degree of inhibition. Deletion of the proximal nucleotides − 469 to + 24 abolished all promoter activity (data not shown). Therefore, the region between nucleotides − 469 and + 24 of the human GR promoter most likely contains the sequences involved in c-Jun repression. The results do not exclude the possibility of more than one repression element, with the − 469 to − 242 segment and the − 242 to + 24 segment each containing a site. The presence of cryptic promoters and AP-1-like elements in plasmids derived from the pUC series has been reported (Kushner et al., 1994). Mutant 5 was constructed to remove both the vector AP-1-like element present in the pCAT plasmid backbone and the GR promoter AP-1-like element. As mutant 5 is still repressed by c-Jun, it seems unlikely that the AP-1-like elements contribute to promoter repression by c-Jun. However, to ensure that similar sequences outside the promoter are not augmenting the effect of c-Jun, we performed additional transfections with the hGR promoter inserted into the pXP2D2 plasmid (Grimm and Nordeen, 1999). This promoterless vector has been deleted for all promoter and AP-1-like sequences. Compared to the reporter luciferase activity from cells transfected with hGRprom/pXP2D2 alone, cotransfection of RSVc-Jun and hGRprom/pXP2D2 resulted in 809 2% repression of luciferase activity. As this result is comparable with the results of transfections with the GR promoter in the pCAT vector, it seems highly likely that c-Jun repression of the GR promoter does not involve sequences outside of the cloned promoter insert.

In order to determine if the endogenous mouse GR gene promoter in NIH3T3 cells could be regulated by c-Jun in the same manner as the human promoter, we generated stable RSVc-Jun transfectants from NIH3T3 cells. The transfection yielded 15 clones that exhibited elevated c-Jun mRNA and protein levels compared to parent NIH3T3 cells (Table 1). The steady-state level of GR mRNA is lower in all clones with a minimum level around 35% of that observed for NIH3T3 cells. Two drug-resistant clones which did not exhibit elevated c-Jun protein contained control levels of GR mRNA (100 and 106%). These results demonstrate that c-Jun is able to repress endogenous mouse GR gene expression by attenuating promoter activity, similar to the results obtained with transient co-transfections with the human promoter. Although the attenuation observed for GR mRNA levels is not directly proportional to c-Jun protein levels, constitutive expression of c-Jun may be sufficient to mediate repression of the GR gene. To test whether constitutive expression of c-Jun mediates repression of the human GR promoter, clonal NIH3T3 c-Jun cell lines were transfected with phGRpromCAT. Compared to parent NIH3T3 cells, human GR promoter activity was reduced to 13 and 29% in clone 25 and clone 38 cells, respectively (Table 2). This result demonstrates that repression of both the human and mouse GR promoters occurs in cells with constitutive expression of c-Jun. Table 1 NIH3T3c-Jun clones were obtained by NIH3T3 transfection with RSVc-Jun plus SV2neo and selected for resistance to G418a NIH3T3c-Jun clone c

c-Jun mRNA

NIH3T3 1 6 8 11 14 15 18 21 23 25 26 31 32 37 38

1.00 9.26 2.64 4.96 1.16 3.04 1.41 12.00 13.31 10.55 25.72 1.66 5.02 1.85 7.41 10.36

a

b

c-Jun protein

1.00 1.35 2.10 2.86 2.16 2.32 1.55 1.32 1.60 1.47 3.99 2.46 1.41 1.88 2.59 5.70

b

GR mRNA

b

1.00 0.58 0.73 0.64 0.76 0.53 0.47 0.48 0.49 0.54 0.39 0.84 0.47 0.61 0.35 0.41

c-Jun and GR mRNA levels were determined from total RNA by Northern blot analysis. c-Jun protein concentrations were measured by Western blot using a specific polyclonal antibody. Values are expressed as relative levels compared to the parent NIH3T3 cells. b Relative levels compared to parent NIH3T3 cells.

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79 Table 2 The indicated cell lines were cotransfected with phGRpromCAT and CMVbgal and assayed for CAT activity as described in Section 4a Cell line

Relative CAT activity

NIH3T3 NIH3T3c-Jun c 25 NIH3T3c-Jun c38

1.00 0.139 0.03 0.29 9 0.06

a

CAT activity is listed relative to the values obtained for parent NIH3T3 cells after normalizing to b-galactosidase activity. Values are the mean of three independent experiments 9 SEM.

3. Discussion Glucocorticoid hormones mediate several vital functions in cell physiology through the specific intracellular receptor. The GR can interact with other transcription factors and regulate gene expression in a manner which is specific for cell type, co-factors, and target genes (Miner and Yamamoto, 1991; Schu¨ le and Evans, 1991). Several investigations have shown that there are direct protein– protein interactions between the GR and AP-1 proteins (Diamond et al., 1990; Jonat et al., 1990; Lucibello et al., 1990; Schu¨ le and Evans, 1991; Pearce and Yamamoto, 1993; Beato et al., 1996). These protein– protein interactions mediate hormone-dependent repression or activation of gene transcription (reviewed by Gottlicher et al., 1998). The GR regulates cell differentiation programs in response to cognate ligands, while AP-1 proteins induce expression of cell proliferation genes in response to extracellular stimuli (Angel and Karin, 1991). Cross-talk between these two signaling pathways at the level of protein– protein interactions allows fine tuning of the regulation of important biological processes (Miner and Yamamoto, 1991; Schu¨ le and Evans, 1991). In glucocorticoid-treated cells, there is an oscillatory pattern of GR and c-Jun mRNA expression, due to differences at the level of transcription (Vig et al., 1994). This coordinate regulation in hormone-treated cells is cell type-specific and dose-dependent, and likely is a result of transcriptional interference from GR binding to CREB-binding protein (Barrett et al., 1996). An additional mechanism whereby the level of GR is controlled by the activity of c-Jun in hormone-free cells would allow an even higher degree of control. In this work we have characterized another example of cross-talk between the GR and AP-1 family members. The human GR gene promoter activity can be appreciably repressed by c-Jun in cultured cells in the absence of glucocorticoids (Figs. 1 and 2). These results are consistent with observations that cells treated with the phorbol ester TPA, which stimulates AP-1 activity (Angel et al., 1987; Scho¨ ntal et al., 1988), decreased GR mRNA levels (Vig and Vedeckis, 1992) and inhibited GR-dependent transcription (Vacca et al., 1989). Varia-

73

tions in the level of GR during different phases of the cell cycle (Cidlowski and Michaels, 1977; Cidlowski and Cidlowski, 1982) might be due in part to the particular composition and amounts of AP-1 proteins, which also vary during the cell cycle (Kovary and Bravo, 1992). The AP-1 family members JunD and c-Fos do not repress GR promoter activity. In fact, Wei and Vedeckis (1997) observed that c-Fos stimulates GR gene expression in cells stably overexpressing c-Fos. Consistent with this observation, transient overexpression of c-Fos slightly stimulated the GR promoter (Fig. 2). In the present work, we found that c-Fos is not limiting for c-Jun-mediated promoter repression in transfected NIH3T3 cells. Wei and Vedeckis (1997) also found that c-Fos protein was virtually undetectable in NIH3T3 cells under basal conditions, and that serumstimulation of quiescent cells was necessary for a transient increase of c-Fos levels. Together with the observation that proteins binding to the promoter AP1-like site do not include c-Jun (Breslin and Vedeckis, 1996b), our finding that deletion of this site does not affect c-Jun-mediated repression demonstrates that the promoter AP-1-like site is not required for c-Jun repression. The observation that this site is required for c-Fos stimulation of the promoter (Wei and Vedeckis, 1997) reinforces the complex nature of GR promoter regulation, even in the absence of hormone. Recent results further demonstrate the complexity of GR promoter regulation, as evidenced by the heterogeneity of GR transcripts. In rat tissues, alternative splicing can result in as many as 11 different exon 1 sequences (McCormick et al., 2000). In murine S49 cells, four exon one variants have been described (Chen et al., 1999). For the human promoter, one of three variants of exon 1A may be diagnostic for human leukemia (Breslin and Vedeckis, 1999). Jun/Jun homodimers have been implicated in the repression of other genes. For example, the major histocompatibility complex class I genes are negatively regulated by c-Jun through a response element in the promoter (Howcroft et al., 1993). However, DNA binding activity and transcriptional activation by Jun proteins are dramatically increased upon dimerization with Fos proteins (Halazonetis et al., 1988; Rauscher et al., 1988). The results of crystallography studies (Glover and Harrison, 1995) showed that heterodimers are favored over homodimers. We observed that the GR promoter repression under conditions favoring cJun homodimer formation is similar to that produced by conditions favoring c-Jun/c-Fos heterodimer formation (Fig. 3). Therefore, the mechanism previously described for induction of gene expression where Jun/Fos heterodimers (Ryseck and Bravo, 1991) are more active than homodimers does not seem to apply to repression of GR gene expression. We worked with c-Jun concentrations where the GR promoter inhibition was not

74

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

maximal, in order to detect any additional effect of c-Fos heterodimers. Even when heterodimer formation is favored by transfecting 10-fold more c-Fos vector than c-Jun (Fig. 3B), the repression did not change. These results provide further evidence that different mechanisms are involved in induction and repression of gene expression by AP-1 (Pfahl, 1993). An AP-1-like site has been sequenced in the human GR promoter (Zong et al., 1990). However, electrophoretic mobility shift assays demonstrated that cJun is unable to bind this element efficiently (Fig. 5). In part, this may be due to intrastrand base pair formation under in vitro conditions. In a more extensive study, Breslin and Vedeckis (1996b) designed oligonucleotides with sequences altered on the 5% or 3% side of the GR promoter AP-1-like sequence to prevent such secondary structures from forming. Using these oligonucleotides, they found that complexes formed with nuclear proteins from AtT-20 cells contained JunB and JunD, but not c-Jun. Fra-2, FosB, and c-Fos were also detected in these complexes (Breslin and Vedeckis, 1996b). Additional studies by Barrett and Vedeckis (1996) demonstrated that nuclear extracts from L929, NIH3T3, and AtT-20 cells contained proteins that shifted a band containing a cAMP responsive element (CRE) sequence. Supershift analysis showed that Jun and Fos family members were present in these complexes from NIH3T3 cells, which suggests that Fos proteins, JunD, or JunB may be able to bind this element and affect gene regulation when associated with CREB and CREM. Heterodimers containing Fos and Jun family members and Jun/Jun homodimers bind the AP-1 consensus sequence (Ryseck and Bravo, 1991), while heterodimers containing Fos family members bind the AP-1-like sequence (Breslin and Vedeckis, 1996b). Since AP-1-like competition with consensus AP-1 for binding factors reaches a plateau (Fig. 5B), we speculate that the AP-1-like sequence competes with the AP-1 consensus sequence for heterodimers but not for Jun homodimers. A possible mechanism to account for c-Jun repression of the GR promoter is that high c-Jun levels sequester Fos family proteins, preventing them from interactions with CREB/CREM or co-activators such as CREB binding protein CBP (Arias et al. 1994; Kwok et al., 1994). This notion is supported by the fact that GR promoter activity is repressed by Jun/Jun or Jun/Fos (Fig. 3). Deletion mutants of the hGR promoter permitted us to map a potential site of action for c-Jun. Repression mediated by c-Jun was not affected when the AP-1-like site in the promoter was removed. The hGR promoter mutant which contains the proximal promoter region − 242 to +24 (Fig. 6, mutant 4) was still repressed by c-Jun, which suggests the possibility that c-Jun could interact with CRE elements (Ryseck and Bravo, 1991; Galien et al., 1994) or with transcription factors acting

on specific binding sites present in that region (Johnson, 1995). Footprinting studies show seven protein binding sites in the proximal 400 nucleotides of the human promoter including an AP2 site and a YY1 site (Nobukuni et al., 1995; Breslin and Vedeckis, 1998). The complexity of GR promoter regulation is illustrated by the effect of deleting nucleotides between − 2738 and − 469 (mutant 3), which generated an increase in activity compared to the intact promoter. This suggests that there is a silencer element in that region. Indeed, a silencer element mapping between nucleotides − 2025 and − 1841 has been previously described (Breslin and Vedeckis, 1996a, 1998). Additionally, deletion of nucleotides − 755 to − 242 (mutant 4) decreased promoter activity indicating that a positive element had been removed. This element appears to be located between nucleotides − 469 and − 150 (Breslin and Vedeckis, 1998). Endogenous GR regulation was also examined in NIH3T3 cells clones stably expressing c-Jun (Table 1). The level of GR mRNA is reduced in these cells, demonstrating that overexpression of c-Jun is correlated with attenuation of murine GR gene expression. As expected, there is a variation of c-Jun mRNA expression in these cells that could be due to the copy number or to the insertion sites of the transgene. Genes regulated during the cell cycle have a very efficient synthesis/degradation program. Therefore, variation of c-Jun mRNA degradation, a process which is controlled by specific RNases (Greenberg et al., 1986), could account for these results. No direct correlation between c-Jun mRNA and c-Jun protein levels was observed in stably transfected NIH3T3 clones. It is known that c-Jun can be selectively degraded by ubiquitinylation (Treier et al., 1994; Hermida-Matsumoto et al., 1996) or by the ATP-dependent 26S proteasome (Jariel-Encontre et al., 1995). These mechanisms could be involved in differential post-translational regulation of c-Jun levels in the NIH3T3 clones. However, in all clones where c-Jun protein was elevated, GR mRNA expression was attenuated, consistent with the results from transient transfection experiments. It is interesting to note that even when c-Jun levels were marginally higher than in the parent cells, GR expression was attenuated. This suggests that in stably transfected cells a low constitutive expression of c-Jun could be enough to produce maximum attenuation of GR gene expression. The lack of a strict correlation between c-Jun expression and the degree of GR mRNA attenuation might be due to asynchronous growth conditions and the rapid turnover of c-Jun mRNA and protein compared to the GR mRNA (Dong et al., 1988; Ryseck et al., 1988). Additionally, c-Jun phosphorylation could contribute to these results since this post-translational modification is crucial for c-Jun biological activity (Binetruy et al., 1991; Boyle et al., 1991).

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

Experiments with the transiently transfected human GR promoter or with the endogenous mouse GR promoter in NIH3T3 cells yielded the same results, indicating that both promoters are regulated in a similar manner. Strahle et al. (1992) demonstrated that mouse GR gene expression is controlled by three distinct promoters. The proximal P1C murine promoter, which is the strongest one in fibroblasts, has high homology with the human GR promoter (Zong et al., 1990). The human promoter sequence from −469 to −10 is nearly identical to the mouse promoter, differing by more than two nucleotides at only five sites (Nobukuni et al., 1995). This supports the proposal that this homologous region should be a target for c-Jun regulation both in the human and mouse GR gene promoter. We have previously reported that H-ras transformed NIH3T3 cells exhibit an attenuated response to glucocorticoids which is due to reduced GR levels (Martins and Brentani, 1990; Martins et al., 1995). This reduction is due, at least in part, to an attenuated transcription rate of the GR gene from its proximal promoter (Martins et al., 1995). It has been established that Ras oncoprotein mediates c-Jun phosphorylation, enhancing c-Jun activity (Binetruy et al., 1991; Westwick et al., 1994). In this work we have shown that c-Jun is able to repress the GR gene promoter, which suggests that activated Ras repression of the GR promoter is mediated, at least in part, by c-Jun. Regulation of the cellular response to glucocorticoids is essential for life, since these hormones are involved in embryonic development and cell differentiation (Evans, 1988). Moreover, glucocorticoids are increasingly important in different therapeutic regimens (Brinckerhoff et al., 1986; Allison, 1988). An understanding of GR regulation and GR interaction with other transcription factors is important to elucidate not only the mechanism of action of these hormones but the role of GR in a complex network of transcription factors. The present report adds a new dimension to the cross-talk between the GR and AP-1 family members.

4. Materials and methods

4.1. Materials Deoxycytidine 5%-[a-32P]triphosphate (3000 Ci/ mmole) and adenosine 5%-[g32P] triphosphate (3000 Ci/ mmole) were from New England Nuclear. D-Threo-[dichloroacetyl-1-14C]chloramphenicol and the ECL chemiluminescence kit were from Amersham. Acetyl coenzyme A and o-nitrophenyl phosphate were from Sigma Chemical Co. Antibodies specific for c-Jun (sc-45), pan-reactive for c-Jun, JunB, JunD (sc-44), specific for c-Fos (sc-8047), pan-reactive for c-Fos, FosB, Fra-1, Fra-2 (sc-447), AP-1 consensus double-

75

stranded oligonucleotide (sc-2501) and AP-1 mutant double-stranded oligonucleotide (sc-2514) were from Santa Cruz Biotechnology. The double-stranded oligonucleotide containing the AP-1-like element of the GR promoter was synthesized at the Ludwig Institute for Cancer Research, Sa˜ o Paulo Branch. CMVbgal containing the E. coli b-galactosidase gene under control of the immediate early cytomegalovirus promoter (MacGregor and Caskey, 1989) was provided by Dr Grant R. MacGregor. The phGRpromCAT plasmid, provided by Dr E. Brad Thompson and Dr Jian Zong, contains 2.7 kb of the human GR promoter sequence (Zong et al., 1990) cloned into the Sal I site of the pCATenhancer plasmid (Promega). Expression plasmids for c-Jun, c-Fos, JunB, JunD, and c-JunD132 –220 (Doucas et al., 1991) under control of the Rous sarcoma virus (RSV) promoter were provided by Dr Moshe Yaniv. The RSVv-jun plasmid (Angel et al., 1988) was provided by Dr Michael Karin. The pGPDN5 plasmid containing the rat glyceraldehyde-3phosphate dehydrogenase cDNA (Fort et al., 1985) was provided by Dr P. Fort. The pXP2D2 promoterless vector was provided by Dr Steven K. Nordeen.

4.2. Cell culture NIH3T3 cells and NIH3T3 cells stably transfected with RSVc-Jun were grown in Dulbecco’s modified Eagle’s medium (GIBCO, Grand Island, NY) supplemented with 10% fetal bovine serum (CULTILAB, Campinas, SP).

4.3. Cell transfection and enzyme assays Cells growing in 162 cm2 flasks were transiently transfected using the calcium phosphate technique with 20 mg of the reporter vector phGRpromCAT, 20 mg of CMVbgal, and the indicated amount of the particular AP-1 protein expression vector. Irrelevant DNA (salmon testis DNA) was used to adjust DNA to the same concentration in all transfections. Cells were harvested 60 h after transfection and extracts prepared by freeze/thaw lysis in 250 mM Tris, pH 7.6. A portion of each extract was saved for b-galactosidase determination (Sambrook et al., 1989) and 5mM EDTA was added to the remainder before heating at 65°C (Crabb and Dixon, 1987). Heated extracts were assayed for CAT activity as described by Gorman et al. (1982). CAT activity was normalized to b-galactosidase activity to correct for any differences in transfection efficiency among experiments. The results show the relative values compared to the hGR promoter basal activity, and represent the average and standard error of at least three independent transfections. Statistical analyses were done using the Student’s t-test for paired samples. The values were considered significantly different when

76

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

P B0.05. To assay for reporter activity in cells transfected with hGRprom/pXP2D2, the luciferase assay system from Promega was used. Transfected cells were lysed in reporter lysis buffer and assayed for luciferase and cotransfected b-galactosidase as suggested by the manufacturer. NIH3T3 cells were stably transfected using the calcium phosphate technique with the RSVcJun and SV2neo plasmids in a ratio 10:1. After selection in medium containing 0.5 mg/ml active G418, clones were isolated and termed NIH3T3c-Jun.

4.4. RNA purification and northern blot analysis Total cellular RNA from parent NIH3T3 cells and NIH3T3c-Jun cloned cells was isolated as described by Puissant and Houdebine (1990). Samples containing 20 mg of formamide-denatured RNA were fractionated on formaldehyde–agarose gels, transferred to a Hybond nylon membrane (Amersham) and hybridized to radiolabeled cDNA probes as previously described (Housley and Forsthoefel, 1989). Fragments of pSV2Wrec, RSVc-Jun, RSVc-fos, and pGPDN5 were labeled with [a-32P]dCTP by random priming and used as probes for GR, c-Jun, c-Fos, and GAPDH mRNA, respectively. The signal intensity of each autoradiographic band was quantified using scanning laser densitometry. Exposures were obtained within the linear response range of the film and bands exhibited a similar range of intensity. GR, c-Fos, and c-Jun mRNA signals for each clone were normalized to GAPDH mRNA signals in the same samples.

4.5. Cell extracts, gel electrophoresis and immunoblotting For analysis of c-Jun levels in stable transfected NIH3T3cJun cells, cells were disrupted in RIPA buffer (10mM Tris pH 7.5, 150 mM NaCl, 1% NP-40, 1% sodium deoxycholate) plus 0.2 mM PMSF, 1 mM leupeptin, 1 mM aprotinin, and 1 mM benzamidine using a Dounce homogenizer. After 30 min extracts were clarified by centrifugation at 22000× g for 15 min. Supernatants were recovered and the protein quantified as described by Bradford (1976). Equal amounts (100 mg) of total protein from each extract were resolved by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes. c-Jun was detected using a rabbit anti-c-Jun antibody followed by anti-rabbit-peroxidase. Reactions were developed using the ECL system and quantified using scanning laser densitometry. For comparison of c-Jun and c-Fos levels, cell nuclei were extracted with RIPA buffer for 30 min on ice and centrifuged at 27000× g for 30 min. Equal amounts of protein from the supernatants were analyzed by immunoblotting with antibodies to c-Fos and c-Jun.

4.6. Nuclear extracts, immunodepletion and electrophoretic mobility shift assays Nuclear extracts were prepared as described by Ausubel et al. (1994). Cells were lysed in hypotonic buffer (10 mM Hepes pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.2 m PMSF, 0.5 mM DTT) and Dounce homogenized. Nuclei were recovered by centrifugation at 1500× g and resuspended in low salt buffer (20 mM KCl, 20 mM Hepes pH 7.9, 25% glycerol, 1.5 mM MgCl2, 0.02 mM EDTA, 0.2 mM PMSF, 0.5 mM DTT) followed by dropwise addition of high salt buffer (1 M KCl, 20 mM Hepes pH 7.9, 25% glycerol, 1.5 mM MgCl2, 0.02 mM EDTA, 0.2 mM PMSF, 0.5 mM DTT). After 30 min at 4°C and centrifugation at 1500× g, supernatants were dialyzed to remove KCl. Where indicated, nuclear extracts were immunodepleted by incubation with rabbit anti c-Jun antibody followed by protein A-Sepharose treatment. Five micrograms of nuclear extract protein were incubated with 3 mg poly dI–dC plus 0.7 ng of a 32P-labeled oligonucleotide: either the consensus AP-1 (CGCTTGATGACTCAGCCGGAA), mutant AP-1 (CGCTGATGACTTGGACTCAGCCGGAA), or the oligonucleotide corresponding to the −905 to −885 region of the human GR promoter (TCGAAGTGACACACTTCACGC) in HEGK buffer (67 mM Tris, pH 8, 33 mM sodium acetate, 10 mM EDTA) for 20 min at room temperature. Competition experiments were done in the presence of 2.5–20-fold excess (1.75– 14 ng) unlabeled oligonucleotides. Samples were resolved by electrophoresis on a 4% polyacrylamide gel and autoradiographed. The signals obtained were quantified using scanning laser densitometry and the maximum binding to the consensus AP-1 oligonucleotide was set at 100%.

4.7. hGR promoter mutants Restriction sites in the phGRpromCAT plasmid were used to generate promoter deletions. Pst I sites are located immediately upstream of the promoter sequence and at position − 469 in the promoter. A Sma I site is located at − 1043 and two Sac I sites are located at − 755 and − 242. Digestion with Pst I, followed by fragment purification and re-ligation yielded mutant 3. The Pst I fragment containing nucleotides −2738 to − 469 was digested with Sma I to yield two fragments: − 2738 to − 1043 and − 1042 to − 469. These fragments were each blunt-ended and ligated separately to the proximal 469 nucleotide promoter fragment to yield mutants 1 and 2, respectively. Mutant 4 was obtained by digestion with Sac I, which cuts at − 755 and −242, followed by fragment purification and re-ligation. Mutant 5 was generated using oligonucleotide-directed mutagenesis as described previously (Mason and

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

Housley, 1993) to mutate the AP-1-like sequences in the pCAT vector and in the GR promoter. Five nucleotides were deleted in the pCAT vector backbone, converting TAACTGACACACATT to TAAACACATT. Three nucleotides were deleted in the GR promoter AP-1-like sequence, converting AAGTGACACACTCT to AAGCACACTCT. Each mutant was verified by DNA sequencing.

Acknowledgements The authors wish to thank the following individuals for generously providing plasmids used in this work: Drs Grant R. MacGregor, E. Brad Thompson, Jian Zong, Moshe Yaniv, Michael Karin, P. Fort, and Steven K. Nordeen. We would also to acknowledge Dr Luisa L. Villa from Ludwig Institute for Cancer Research for the synthesis of the AP-1-like oligonucleotide. This work was supported by FAPESP grant 93/3711-0 (V.R.M.), a grant-in-aid from the South Carolina Cancer Center (P.R.H.), and fellowship funds from FAPESP 94/6248-1 (A.L.B.C.).

References Allison, A.C., 1988. Immunopathogenetic mechanisms of arthritis and modes of action of antirheumatic therapies. In: Goodacre, J.A., Dick, W.C. (Eds.), Immunopathogenetic Mechanisms of Arthritis. Klewer Academic Publishers, Boston, pp. 211 –246. Angel, P., Karin, M., 1991. The role of Jun, Fos and the AP-1 complex in cell proliferation and transformation. Biochem. Biophys. Acta 1072, 129 –157. Angel, P., Alegretto, E., Okini, S., Hattori, K., Boyle, W., Hunter, T., Karin, M., 1988. Oncogene jun encodes a sequence-specific transactivator similar to AP-1. Nature 332, 166 –171. Angel, P., Imagawa, M., Chiu, R., Stein, B., Imbra, R.J., Rahmsdorf, H.J., Jonat, C., Herrlich, P., Karin, M., 1987. Phorbol ester-inducible genes contain a common cis element recognized by a TPA-modulated trans-acting factor. Cell 49, 729 –739. Arias, J., Alberts, A.S., Brindle, P., Claret, F.X., Smeal, T., Karin, M., Feramisco, J., Montminy, M., 1994. Activation of cAMP and mitogen responsive genes relies on a common nuclear factor. Nature 370, 226 – 229. Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D., Seidman, J.G., Smith, J.A., Struhl, K., 1994. Current Protocols in Molecular Biology. Greene Publishing Associates, New York. Barrett, T.J., Vedeckis, W.V., 1996. Occupancy and composition of proteins bound to AP-1 sites in the glucocorticoid receptor and c-Jun promoters after glucocorticoid treatment and in different cell types. Receptors and Signal Transduction 6, 179 – 192. Barrett, T.J., Vig, E., Vedeckis, W.V., 1996. Coordinate regulation of glucocorticoid receptor and c-jun gene expression is cell type-specific and exhibits differential hormonal sensitivity for down-and up-regulation. Biochemistry 35, 9746 –9753. Beato, M., 1989. Gene regulation by steroid hormones. Cell 56, 335 – 344. Beato, M., Candau, R., Chavez, S., Mo¨ ws, C., Truss, M., 1996. Interaction of steroid hormone receptors with transcription factors involves chromatin remodeling. J. Steroid Biochem. Mol. Biol. 56, 47 – 59.

77

Belman, S.E., Troll, W., 1972. The inhibition of croton oil-promoted mouse skin tumorigenesis by steroid hormones. Cancer Res. 32, 450 – 454. Binetruy, B., Smeal, T., Karin, M., 1991. H-ras augments c-jun activity and stimulates phosphorylation of its activation domain. Nature 351, 122 – 127. Boyle, W.J., Smeal, T., Defize, L.H.K., Angel, P., Woodgett, J.R., Karin, M., Hunter, T., 1991. Activation of protein kinase C decreases phosphorylation of c-Jun at sites that negatively regulate its DNA-binding activity. Cell 64, 573 – 584. Bradford, M.M., 1976. A rapid and sensitive method for the quantification of microgram quantities of protein utilizing the principle of protein dye binding. Anal. Biochem. 72, 248 – 254. Breslin, M.B., Vedeckis, W.V., 1996a. Characterization of the human glucocorticoid receptor promoter. 10th International Congress of Endocrinology, pp. 3 – 177. Breslin, M.B., Vedeckis, W.V., 1996b. The glucocorticoid receptor and c-jun promoters contain AP-1 sites that bind different AP-1 transcriptional factors. Endocrine 5, 15 – 22. Breslin, M.B., Vedeckis, W.V., 1998. The human glucocorticoid receptor promoter upstream sequences contain binding sites for the ubiquitous transcription factor, Yin Yang 1. J. Steroid Biochem. Mol. Biol. 67, 369 – 381. Breslin, M.B., Vedeckis, W.V., 1999. Untranslated human glucocorticoid receptor exon 1A: differential expression in blood cells and potential role in diagnosing leukemia. Endocrine Society 81st Annual Meeting Abstracts, p. 498. Brinckerhoff, C.E., Plucinska, I.M., Sheldon, A., O’Connor, G.T., 1986. Half life of synovial cell collagenase mRNA is modulated by phorbol myristate acetate but not by all trans-retinoic acid or dexamethasone. Biochemistry 25, 6378 – 6384. Chen, F., Watson, C.S., Gametchu, B., 1999. Multiple glucocorticoid receptor transcripts in membrane glucocorticoid receptor-enriched S-49 mouse leukemic cells. J. Cell. Biochem. 74, 418 – 429. Cidlowski, J.A., Cidlowski, N.B., 1982. Glucocorticoid receptors and the cell cycle: evidence that accumulation of glucocorticoid receptors during the S phase of the cell cycle is dependent on ribonucleic acid and protein synthesis. Endocrinology 110, 1653 –1662. Cidlowski, J.A., Michaels, G.A., 1977. Alteration in glucocorticoid binding site during the cell cycle in HeLa cells. Nature 266, 643 – 645. Crabb, D.W., Dixon, J.E., 1987. Method for increasing the sensitivity of chloramphenicol acetyltransferase assay in extracts of transfected cultured cells. Anal. Biochem. 163, 88 – 92. Diamond, M.C., Miner, J.N., Yoshinaga, S.K., Yamamoto, K.R., 1990. Transcription factor interactions: selectors of positive or negative regulation from a single DNA element. Science 249, 1266 – 1272. Dong, Y., Poellinger, L., Gustafsson, J.A., Okret, S., 1988. Regulation of glucocorticoid receptor expression: evidence for transcriptional and post-translational mechanisms. Mol. Endocrinol. 2, 1256 – 1264. Doucas, V., Spyrou, G., Yaniv, M., 1991. Unregulated expression of c-Jun or c-Fos proteins but not JunD inhibits oestrogen receptor activity in human breast cancer derived cells. EMBO J. 10, 2237 – 2245. Encio, I.J., Detera-Wadleigh, S.D., 1991. The genomic structure of the human glucocorticoid receptor. J. Biol. Chem. 266, 7182 – 7188. Evans, R.M., 1988. The steroid and thyroid hormone receptor superfamily. Science 240, 889 – 895. Fort, P., Piechaczyk, M., Sabrouty, S.E., Dani, C., Jeanteur, P., Blanchard, H., 1985. Various rat adult tissues express only one major mRNA species from the glyceraldeyde-3-phosphate-dehydrogenase multigenic family. Nucl. Acid Res. 13, 1431 –1442. Galien, R., Emanoil-Ravier, R., Mercier, G., 1994. Differential effects of c-jun and CREB on cAMP response element activation by H-ras. Oncogene 9, 1101 – 1108.

78

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79

Glover, J.M.N., Harrison, S.C., 1995. Crystal structure of the heterodimeric bZIP transcription factor c-Fos-c-Jun bound to DNA. Nature 373, 257 – 261. Gorman, C.M., Merlino, G.T., Willingham, M.C., Pastan, I., Howard, B.H., 1982. The Rous sarcoma virus long terminal repeat is a strong promoter when introduced into a variety of eukaryotic cells by DNA-mediated transfection. Proc. Natl. Acad. Sci. USA 79, 6777 –6781. Gottlicher, M., Heck, S., Herrlich, P., 1998. Transcriptional crosstalk, the second mode of steroid hormone receptor action. J. Mol. Med. 76, 480 – 489. Greenberg, M.E., Hermanowski, A.L., Ziff, E.B., 1986. Effect of protein synthesis inhibitors of growth factor activation of c-fos, c-myc and actin gene transcription. Mol. Cell. Biol. 6, 1050 – 1057. Grimm, S.L., Nordeen, S.K., 1999. Luciferase reporter gene vectors that lack potential AP-1 sites. Biotechniques 27, 220 –222. Halazonetis, T.D., Georgopoulos, K., Greenberg, M.E., Leder, P., 1988. c-jun dimerizes with itself and with c-fos, forming complexes of different DNA binding affinities. Cell 55, 917 – 924. Hermida-Matsumoto, M.L., Chock, P.B., Curran, T., Yang, D.C., 1996. Ubiquitinylation of transcription factors c-Jun and c-Fos using reconstituted ubiquitinylating enzymes. J. Biol. Chem. 271, 4830 – 4936. Housley, P.R., Forsthoefel, A.M., 1989. Isolation and characterization of a mouse L cell variant deficient in glucocorticoid receptors. Biochem. Biophys. Res. Comm. 164, 480 – 487. Howcroft, T.K., Richardson, J.C., Singer, D.S., 1993. MHC class I gene expression is negatively regulated by the proto-oncogene c-jun. EMBO J. 12, 3163 –3169. Jariel-Encontre, I., Pariat, M., Martin, F., Carillo, S., Salvat, C., Piechaczyk, M., 1995. Ubiquitinylation is not an absolute requirement for degradation of c-Jun protein by the 26S proteasome. J. Biol. Chem. 270, 11623 –11627. Jonat, C., Rahmsdorf, H.J., Park, K.K., Cato, A.B., Gebel, S., Ponta, H., Herrlich, P., 1990. Antitumor promotion and antiinflamation: down-modulation of AP-1 ( fos/jun) activity by glucocorticoid hormone. Cell 62, 1189 –1204. Johnson, A.D., 1995. The price of repression. Cell 81, 655 –658. Kovary, K., Bravo, R., 1992. Existence of different Fos/Jun complexes during the G0-to-G1 transition and during exponential growth in mouse fibroblasts: differential role of Fos proteins. Mol. Cell. Biol. 12, 5015 –5023. Kushner, P.J., Baxter, J.D., Duncan, K.G., Lopez, G.N., Schaufele, F., Uht, R.M., Webb, P., West, B.L., 1994. Eukaryotic regulatory elements lurking inplasmid DNA: the activator protein-1 site in pUC. Mol. Endocrinol. 8, 405 –407. Kwok, R.P., Lundblad, J.R., Chrivia, J.C., Richards, J.P., Ba¨ chinger, H.P., Brennan, R.G., Roberts, S.G., Green, M.R., Goodman, R.H., 1994. Nuclear protein CBP is a coactivator for the transcriptional factor CREB. Nature 370, 223 – 226. Leclerc, S., Xie, B., Roy, R., Govindan, M.V., 1991. Purification of a human glucocorticoid receptor gene promoter-binding protein. J. Biol. Chem. 266, 8711 –8719. Lee, W., Mitchell, P., Tjian, R., 1987. Purified transcription factor AP-1 interacts with TPA-inducible enhancer elements. Cell 49, 741– 752. Lucibello, F.C., Slater, E.P., Joss, K.U., Beato, M., Mu¨ ller, R., 1990. Mutual transrepression of Fos and glucocorticoid receptor: involvement of a functional domain in Fos which is absent in FosB. EMBO J. 9, 2827 – 2834. MacGregor, G.R., Caskey, C.T., 1989. Construction of plasmids that express E.coli ×-galactosidase in mammalian cells. Nucl. Acid Res. 17, 2365. Mason, S.A., Housley, P.R., 1993. Site-directed mutagenesis of the phosphorylation sites in the mouse glucocorticoid receptor. J. Biol. Chem. 268, 21501 –21504.

Martins, V.R., Brentani, M.M., 1990. The effects of ras gene expression on glucocorticoid receptors in mouse fibroblasts. J. Steroid Biochem. Mol. Biol. 37, 183 – 193. Martins, V.R., Brentani, M.M., Housley, P.R., 1995. Attenuation of glucocorticoid receptor levels by H-ras oncogene. Endocrine 3, 305 – 312. McCormick, J.A., Lyons, V., Jacobson, M.D., Noble, J., Diorio, J., Nyirenda, M., Weaver, S., Ester, W., Yau, J.L.W., Meaney, M.J., Seckl, J.R., Chapman, K.E., 2000. 5%-heterogeneity of glucocorticoid receptor messenger RNA is tissue specific: differential regulation of variant transcripts by early-life events. Mol. Endocrinol. 14, 506 – 517. Miner, J.N., Yamamoto, K.R., 1991. Regulatory crosstalk at composite response elements. Trends Biochem. Sci. 16, 423 –426. Nobukuni, Y., Smith, C.L., Hager, G.L., Detera-Wadleigh, S.D., 1995. Characterization of the human glucocorticoid receptor promoter. Biochemistry 34, 8207 – 8214. Parker, M.G., 1991. Nuclear Hormone Receptors: Molecular Mechanisms, Cellular Functions, Clinical Abnormalities. Academic Press, New York. Pearce, D., Yamamoto, K.R., 1993. Mineralocorticoid and glucocorticoid receptor activates distinguished by nonreceptor factors at a composite response element. Science 259, 1161 – 1165. Pfahl, M., 1993. Nuclear receptor/AP-1 interaction. Endocr. Rev. 14, 651 – 658. Puissant, C., Houdebine, L.M., 1990. Improvement of the single-step method of RNA isolation by acid guanidinium thiocyanate –phenol– chloroform extraction. Biotechniques 8, 148 – 149. Rauscher, F.J., Voulalas, P.J., Franza, B.R., Curran, T., 1988. Fos and Jun bind cooperatively to the AP-1 site: reconstitution in vitro. Genes Dev. 2, 1687 – 1699. Ryseck, R.P., Bravo, R., 1991. c-Jun, JunB and JunD differ in their binding affinities to AP-1 and CRE consensus sequences: effect of Fos proteins. Oncogene 6, 533 – 542. Ryseck, R.P., Hirai, S.H., Yaniv, M., Bravo, R., 1988. Transcriptional activation of c-jun, during the G0/G1 transition in mouse fibroblasts. Nature 334, 535 – 537. Sambrook, J., Fritsch, E.F., Maniatis, T., 1989. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, New York. Scho¨ ntal, A., Herrlich, P., Rahmsdorf, H., Ponta, H., 1988. Requirement for fos gene expression in the transcriptional activation of collagenase by other oncogenes and phorbol esters. Cell 54, 325 – 334. Schu¨ le, R., Evans, R.M., 1991. Cross-coupling of signal transduction pathways: zinc finger meets leucine zipper. Trends Genet. 7, 377 – 381. Scribner, J.D., Slaga, T.J., 1973. Multiple effects of dexamethasone on protein synthesis and hyperplasia caused by a tumor promoter. Cancer Res. 33, 542 – 546. Strahle, U., Schmidt, A., Kesley, G., Stewart, A.F., Cole, T.J., Schmidt, W., Schultz, G., 1992. At least three promoters direct expression of the mouse glucocorticoid receptor gene. Proc. Natl. Acad. Sci. USA 89, 6731 – 6735. Treier, M., Staszewski, L.M., Bohman, D., 1994. Ubiquitin-dependent c-Jun degradation in vivo is mediated by the d domain. Cell 78, 787 – 789. Vacca, A., Screpanti, I., Maroder, M., Petrangeli, E., Frati, L., Gulino, A., 1989. Tumor-promoting phorbol ester and ras oncogene expression inhibit the glucocorticoid-dependent transcription from the mouse mammary tumor virus long terminal repeat. Mol. Endocrinol. 3, 1659 – 1665. Valentini, S.R., Armelin, M.C.S., 1996. Cloning of glucocorticoidregulated genes in C6/St1 rat glioma phenotypic reversion. J. Endocrinol. 148, 11 – 17. Vanderbilt, J.N., Miesfeld, R., Maler, B.A., Yamamoto, K.R., 1987. Intracellular receptor concentration limits glucocorticoid-dependent enhancer activity. Mol. Endocrinol. 1, 68 – 74.

A.L.B. Cabral et al. / Molecular and Cellular Endocrinology 175 (2001) 67–79 Vig, E., Vedeckis, W.V., 1992. Regulatory mechanisms in glucocortcoid receptor gene expression. Proceedings of the 74th Annual Meeting of the Endocrine Society 1239, 361 Vig, E., Barrett, T.J., Vedeckis, W.V., 1994. Coordinate regulation of glucocorticoid receptor and c-jun mRNA levels: evidence for cross-talk between two signaling pathways at the transcriptional level. Mol. Endocrinol. 8, 1336 –1346. Wei, P., Vedeckis, W.V., 1997. Regulation of the glucocorticoid receptor gene by the AP-1 transcription factor. Endocrine 7, 303 – 310. Wei, P., Inamdar, N., Vedeckis, W.V., 1998. Transrepression of c-jun

.

79

gene expression by the glucocorticoid receptor requires both AP-1 sites in the c-jun promoter. Mol. Endocrinol. 12, 1322 – 1333. Westwick, J.K., Cox, A.D., Der, C.J., Cobb, M.H., Hibi, M., Karin, M., Brenner, D.A., 1994. Oncogenic Ras activates c-Jun via a separate pathway from the activation of extracellular signal-regulated kinases. Proc. Natl. Acad. Sci. USA 91, 6030 – 6034. Zong, J., Ashraf, J., Thompson, E.B., 1990. The promoter and first untranslated exon of the human GR gene are GC rich but lack consensus glucocorticoid receptor element sites. Mol. Cell. Biol. 10, 5580 – 5585.