Trans-Retinoic Acid and Glucocorticoids Synergistically ...

Pergamon J. Steroid Biochem. Molec. Biol. Vol. 62, No. 2/3, pp. 129-142, 1997

~%. 1997 Elsevier Science Ltd. All fights reserved Printed in Great Britain

P I I : 8 0 9 6 0 - 0 7 6 0 ( 9 7 ) 0 0 0 3 3 - 2 0960-0760/97 $17.00 + 0.00

T r a n s - R e t i n o i c Acid and Glucocort icoids Synergist ical ly Induce Transcript ion from

the Mouse M a m m a r y T u m o r Virus Promote r in H u m a n Embryonic Kidney Cells

R h e e m D. M e d h and T h o m a s J. S c h m i d t * Department of Physiolo~ and Biophysics, The University of Iowa, College of Medicine, Iowa City, IA 52242, U.S.A.

H u m a n e m b r y o n i c k i d n e y (K293) cei ls t r a n s f e c t e d w i th a m o u s e m a m m a r y t u m o r v i r u s ( M M T V ) p r o m o t e r - l u c i f e r a s e : r epor te r c o n s t r u c t ( p H H - L u c ) w e r e u t i l i zed to i n v e s t i g a t e t he p o t e n t i a l e f fec t s o f trans-retinoic ac id ( tRA) , e i t h e r b y i t s e l f o r in c o m b i n a t i o n w i th g l u c o c o r t i c o i d (GC) hoi+i~iones, on a w e l l - c h a r a c t e r i z e d , G C - s e n s i t i v e t r a n s c r i p t i o n a l r e s p o n s e , t R A o r the s y n t h e t i c G C h o r m o n e d e x a m e t h a s o n e i n d u c e d t r a n s c r i p t i o n f r o m the M M T V p r o m o t e r in a d o s e - d e p e n d e n t m a n n e r , w i th 1 p m o l t R A a n d 1/a~tol d e x a m e t h a s o n e a l o n e c a u s i n g a f o u r - to s ix - fo ld a n d a 40-fold i n d u c t i o n o f b a s a l t r a n s c r i p t i o n , r e s pec t i ve l y . S i m u l t a n e o u s t r e a t m e n t w i th 1 p m o l d e x a m e t h a s o n e a n d 1 p m o l t R A r e s u l t e d in a s y n e r g i s t i c t r a n s c r i p t i o n a l r e s p o n s e t h a t w a s 120-fold h i g h e r t h a n b a s a l level a n d 2.5 t i m e s t he p r e d i c t e d r e s p o n s e , b a s e d on a s i m p l e add i t i ve e f fec t o f b o t h agon i s t s , t R A does no t a p p e a r to m e d i a t e th is s y n e r g i s t i c t r a n s c r i p t i o n a l r e s p o n s e b y e n h a n c i n g G C r e c e p t o r (GR) b i n d i n g c a p a c i t y , a f f in i ty , o r n u c l e a r t r a n s l o c a t i o n , t R A was u n a b l e to p o t e n t i a t e G C - i n d u c e d t r a n s c r i p t i o n a l ac t iv i ty f r o m a m i n i m a l G C r e s p o n s e e l e m e n t ( G R E ) , a n d G C w e r e u n a b l e to p o t e n t i a t e t R A - i n d u c e d t r a n s c r i p t i o n a l ac t i v i t y f r o m a m i n i m a l r e t l n o i c ac id r e s p o n s e e l e m e n t ( R A R E ) . T h e s e d a t a ru le ou t d i r e c t p r o t e ! i n - p r o t e i n i n t e r a c t i o n s b e t w e e n G C a n d r e t i n o i d r e c e p t o r s as a m e c h a n i s m for t he o b s e r v e d s y n e r g i s m , t R A a lso s y n e r g i z e d wi th a l d o s t e r o n e - i n d u c e d , m i n e r a l o c o r t i c o i d r e c e p t o r ( M R ) - m e d i a t e d , t r a n s c r i p t i o n a l a c t i v a t i o n o f t he M M T V p r o m o t e r , r e s u l t i n g in a r e s p o n s e t h a t w a s 1.7 t i m e s the p r e d i c t e d a d d i t i v e r e s p o n s e . T h e M M T V G R E l o c a t e d b e t w e e n -187 a n d -165 was r e q u i r e d fo r G C - i n d u c e d a n d s y n e r g i s t i c a c t i v a t i o n o f the M M T V p r o m o t e r , w h e r e a s s e q u e n c e s l o c a t e d w i t h i n -151 1:o +5 w e r e su f f i c i en t fo r t R A - i n d u c e d t r a n s c r i p t i o n f r o m the M M T V p r o m o t e r . M u t a t i o n o f a c o n s e n s u s R A R E ha l f - s i t e ( C C A A G T ) iden t i f i ed a t p o s i t i o n - 6 5 to - 6 0 w i t h i n the M M T V - L T R d id n o t a f fec t e i t h e r t R A - i n d u c e d t r a n s c r i p t i o n a l a c t i v a t i o n o r s y n e r g i s m wi th G C . We p r o p o s e t h a t t he t R A - i n d u c e d t r a n s c r i p t i o n a l r e s p o n s e f r o m the M M T V p r o m o t e r , as wel l as s y n e r - g i s m wi th G C , m a y b e m e d i a t e d b y the a c t i v a t i o n o r i n d u c t i o n o f a f a c t o r ( s ) t h a t e i t h e r d i r ec t l y b i n d s to t he M M T V p r o m o t e r o r i n d i r e c t l y s t ab i l i zes b i n d i n g o f a n o t h e r t r a n s c r i p t i o n f a c t o r to t h e s e s e q u e n c e s . /D 1997 E l s e v i e r S c i ence Ltd .

J. Steroid Biochem. Molec. Biol., Vol. 62, No. 2/3, pp. 129-142, 1997

I N T R O D U C T I O N

Glucocorticoids (GC) play key roles in mediating diverse physiological processes such as gluconeogen- esis, maintenance of electrolyte balance, cell prolifer-

Present address: DepartmerLt of Human Biological Chemistry and Genetics, University of Texas Medical Branch, Galveston, T X 77555, U.S.A.

*Correspondence to Thomas J. Schmidt. Tel: +1 319 335 7847; Fax: +1 319 335 7330.

Received 14 Oct. 1996; accepted 5 Feb. 1997.

ation, differentiation and anti- inflammatory and immunosuppressive responses [1]. These steroids mediate their actions via binding to specific intracellu- lar receptors (GR) that function as ligand activated nuclear transcription factors [2]. Ligand binding pro- motes activation of the receptor to a D N A binding form and subsequent nuclear translocation. Activated G R bind as homodimers to specific cis-acting D N A sequences (hormone response elements) located within the promoters of target genes [3, 4]. Such G R

129

130 R.D. Medh and T. J. Schmidt

d imer -DNA complexes regulate transcriptional ac- tivity of the promoter via interactions with the basic transcriptional apparatus, although the precise mech- anism(s) underlying this regulation is poorly under- stood [5]. GC-induced transcriptional activity has been shown to be influenced by distinct gene regulat- ory pathways, such as those activated by cAMP [6, 7] and c-jun/c-fos [8-10]. The direction and magnitude of hormone-induced transcriptional responses are thus ultimately determined by the combined effects of several transcriptional modulators and the "cross- talk" between these factors.

The mouse mammary tumor virus (MMTV) pro- moter has been used extensively as a model system to study GC-mediated gene regulatory mechanisms [11, 12]. Steroid hormone-induced transcription from the M M T V - L T R requires sequences between -201 and - 59 , which comprise the hormone response element (HRE) [13, 14]. The M M T V HRE contains four well characterized GR binding sequences (GRE) which cooperatively mediate hormone inducibility. Each binding site contains an imperfect palindromic sequence with a hexanucleotide core motif 5 ' T G T T C T 3 ' [11], and mutational analysis has shown that each GRE is essential for GC responsive- ness. An NF-1 binding site located between - 7 6 and - 6 2 ( 5 ' T T G G A A T C T A T C C A A 3 ' ) is also essential for GC-induced transcriptional activation of the promoter [14, 15]. Similarly, the ubiquitous octamer transcription factor (OTF-1 or Oct- l ) also synergizes with GR to induce transcription from the M M T V promoter [ 12]. OTF- 1 binding sites ( 5 ' A T G T A A A T G C T T A T G T A A A C 3 ' ) have been identified between - 5 6 and - 3 7 bp upstream of the cap site, and have been shown to be required for maximal hormonal induction of the promoter. Interactions between GR and AP1 have also been characterized using the M M T V - L T R , which contains consensus AP1 binding sites at - 1 5 6 to - 1 5 0 and - 1 0 to - 4 [8, 16]. The M M T V promoter is thus an example of the complex interactions that occur among various transcription factors and the effects of these interactions on hormonal regulation of transcription.

Vitamin A and its derivatives (retinoids, e.g. trans- retinoic acid; tRA) have been implicated in a variety of physiological events including embryonic develop- ment and organ formation, the visual cycle, hemato- poiesis, bone formation and the transcriptional regulation of several growth modulatory genes [17, 18]. Like GCs, retinoids modulate their actions via specific intracellular receptors that bind to cis-acting DNA response elements of target genes [19, 20]. Unlike the GR, multiple isoforms of retinoid receptors exist and can be classified into retinoic acid receptors (RAR c~, fl and 7 isoforms) or retinoid X receptors (RXR ~, fl and 7 isoforms) [21, 22]. These bind to specific DNA sequences called retinoic acid response

elements (RARE) to induce (liganded receptors) or suppress (unliganded receptors) target gene tran- scription. Retinoid receptors can bind to DNA as monomers, homodimers or heterodimers [23].

Based on their structure, function and DNA bind- ing properties, the family of nuclear receptors can be divided into two groups: the first group, which includes GC, progesterone, mineralocorticoid and androgen receptors, can bind to the consensus re- sponse element G G T A C A n n n T G T T C T , and the second group, which includes estrogen, thyroid hor- mone (T3), RA and vitamin D3 receptors, can bind to various dimeric arrangements (direct or inverted repeats with distinct spacing requirements) of the basic hexanucleotide consensus motif A/G G TIG T C A [23]. RXR isoforms modulate their actions pri- marily via heterodimerization with several receptors within the second group [24, 25].

Several recent reports have suggested functional in- teractions between GC and retinoids [26]. For example, RA-induced upregulation of G R has been demonstrated in osteosarcoma cells [27], and in rat hepatoma cells GC upregulate RXR expression [28]. RA has also been shown to upregulate GR binding capacity and potentiate agonist-induced G R nuclear translocation in liver cells [29]. Transcription from the phosphoenolpyruvate carboxykinase (PEPCK) and growth hormone gene promoters has been shown to be coordinately regulated by GC and RA [30-32].

These data suggest that interactions between GC- and RA-induced pathways may play an important role in mediating key physiological responses, and may be a more commonly occurring phenomenon than is currently realized.

Here we report that tRA itself induces transcription from the well-characterized M M T V promoter and sig- nificantly potentiates the GC-induced transcriptional response in transfected human embryonic kidney (K293) cells. We have evaluated the possible mechan- ism(s) underlying this synergism between tRA- and GC-induced responses. Our data demonstrate that this synergism is not the result of tRA-induced altera- tions in G R binding capacity, binding affinity, or nuclear translocation, tRA also appears to synergize with mineralocorticoid receptors (MR), which are known to bind to GREs in the M M T V promoter and induce transcription. Using K293 cells in gene trans- fer experiments, we have mapped tRA-responsive DNA sequences between -151 and + 5 b p from the transcription start site within the M M T V - L T R . However, site-directed mutagenesis of a consensus RARE half-site (CCA A G T to CCAtGg) that was identified between - 6 5 and - 6 0 does not appear to alter tRA-induced responses.

tRA-induced Transcription from the MMTV Promoter 131

M A T E R I A L S A N D M E T H O D S

Chemicals and reagents

1,2,4-N-[3H]Triamcinolone acetonide ([3H]TA; 40Ci /mmol) was purchased from N E N - D u p o n t (Boston, MA, U.S.A.). D-threo [dichloroacetyl- 1-14C]chloramphenico]L (55 mCi/mmol) was from Amersham (Arlington Heights, IL, U.S.A.). RU 38486 and RU 28362 were gifts from Roussel- U C L A F (Romainville, France). ZK 98299 was a gift from Schering AG (Berlin, Germany). Dexamethasone, aldo,;terone, tRA, acetyl CoA and other reagent grade chemicals were purchased from Sigma Chemical Co. (St Louis, MO, U.S.A.). D- Luciferin (potassium salt), was purchased from Analytical Luminescence Laboratories (San Diego, CA, U.S.A.). Bradford protein assay reagents were obtained from Bio-Rad Laboratories (Richmond, CA, U.S.A.). Restriction enzymes and molecular biology reagents were from New England Biolabs (Beverly, MA, U.S.A.). Oligonucleotides were synthesized by the Diabetes and Endocrinology Research Center (DERC) DNA core facility at The University of Iowa College of Medicine. Tissue culture media and com- ponents were purchased either from DERC or from The University of Iowa Cancer Center.

Cell culture

The human embryonic kidney cell line K293 was obtained from A T C C (Rockville, MD, U.S.A.). Cells were maintained in HEPES-buffered high glucose Dulbecco's minimum essential medium (DMEM) supplemented with 10% fetal bovine serum, 200 U/ml penicillin, 200 #g/ml streptomycin, 500 ng/ml fungi- zone and 2 mmol L-glutamine (complete DMEM) . The I l d clone that was stably transfected with p H H - Luc and the 5Ik clone that was stably transfected with the GR expression vector p R S h G R , were maintained in the presence of 400 #g/ml Geneticin (Gibco-BRL, Bethesda, MD, U.S.A.). One millimolar stock sol- utions of steroids or retinoids were prepared in etha- nol. Appropriate dilutions were made in culture media immediately prior to treatments, with final concentrations of ethanol never exceeding 0.1%. Ethanol vehicle controls were included in all experiments.

Plasmids

The construct p H H - L u c [33] contains M M T V pro- moter sequences spanning - 2 2 0 to +110 upstream of the luciferase gene, and was a gift from Dr Andrew F. Russo. pRShGR~ and pRShMR are expression vec- tors for the human GR and MR, respectively, and were gifts from Dr Ronald M. Evans [34, 35]. pRARE-TK-Luc , containing two copies of the RAR/~ RARE upstream of ~ minimal thymidine kinase pro- moter and a luciferase reporter was a gift from Dr

Andrew F. Russo. The constructs p G R E 1 - C A T and p G R E 2 - C A T contain one and two copies, respect- ively, of the tyrosine amino transferase GRE upstream of the E l b T A T A box and chloramphenicol acetyl transferase reporter [36] and were gifts from Dr J. Cidlowski. To generate p 19-MTV(225)-Luc and p 19- MTV(156)-Luc, D N A corresponding to - 2 2 0 to +5 and -151 to +5 of the M M T V promoter, respect- ively, was amplified by the polymerase chain reaction (PCR) using appropriate primers incorporating 5'Sal I and 3'Sma I restriction sites. The P CR product was digested with Sal I and Sma I and cloned into the multiple cloning site of the reporter construct p 19Luc that was also digested with Sal I and Sinai . P19- MTV(RAREm)-Luc contains D N A corresponding to - 2 2 0 to +5 of the M M T V - L T R , but with a mutation within the predicted RARE half-site at - 6 5 to - 6 0 that also introduces a N c o I site (CCAAGT to CCAtGg). For site-directed mutagenesis, two separate D N A fragments were amplified by PCR, correspond- ing to - 2 2 0 to - 5 5 and - 6 9 to +5, each incorporat- ing the mutation and Nco I site within the predicted RARE. The two fragments were ligated after digestion with Nco I and then the outside primers were used to amplify the 225 bp fragment incorporating the mu- tation. The mutant fragment was then cloned within the Sal I -Sma I site of p l9Luc .

Transfections

For transient transfections, 15 pg relevant promo- ter-reporter construct and 5 #g of any other ad- ditional plasmids (expression vectors) were transfected using the calcium phospha te -DNA pre- cipitation method [37]. Briefly, D N A was mixed with CaC12 and HEPES-buffered saline and added drop- wise to a subconfluent cell monolayer plated 24 h prior to transfections. Slow precipitation and uptake was allowed to proceed up to 6 h, after which cells were washed twice with D M E M and incubated over- night in fresh medium. Cells were then harvested, washed and plated in six-well dishes, treated with the appropriate agonist for 24 h and harvested. Cell pel- lets were frozen at - 2 0 ° C until used for preparation of cell extracts. For stable transfections, cells were cotransfected with 10#g of either p H H - L u c or pRShGR~ and 5 #g of the selection plasmid pRC/ CMV (Invitrogen, Carlsbad, Calif., U.S.A.) that imparts neomycin resistance. Cells were harvested 48 h after transfection and plated in 24-well dishes at a low cell density in medium containing 400 #g/ml Geneticin (Gibco) to allow selection of stable trans- fectants. Colonies of viable cells were screened for ex- pression of luciferase (pHH-Luc) or G R (pRShGR~) genes by performing luciferase assays or G R binding assays, respectively. Clones I l d and 5Ik expressing the M M T V promoter-driven luciferase gene and the G R gene, respectively, were isolated.

132 R.D. Medh and T. J. Schmidt

Reporter gene assays

The cell pellets were resuspended in 100 ml of pot- assium phosphate buffer p H 7 . 8 containing 0.1 M dithiothreitol ( D T T ) , lysed by three cycles of freezing and thawing, and centrifuged at 13 000g for 20 rain to pellet cell debris. The supernatants were analysed for luciferase or C A T activities and protein content. Luciferase activity was determined in 20 ~tl of cell lysate using a luminometer (monolight 2010) and 1 mmol o-luciferin in a buffer containing 0.1 M pot- assium phosphate p H 7.8, 15 mmol MgSO4, 5 mmol ATP, and 1 mmol D D T [38]. For C A T assays 20/A of cell extracts were incubated at 37°C with 0.05 ktCi of 14C-chloramphenicol and 0.7 mmol acetyl CoA in 0 . 1 M tris-HC1, p H 7 . 8 for l h. Samples were extracted with 1 ml ethyl acetate, the upper organic phase air-dried, resuspended in 10 ~tl ethyl acetate and spotted on a thin layer chromatography (TLC) plate. T L C was run in chloroform: methanol (19:1), and autoradiographed. C A T activity (% acetylated chloramphenicol) was quantitated by liquid scintil- lation counting of cut T L C plates [39]. Protein was estimated by the method of Bradford [40] using the Bio-Rad dye reagent, and reporter gene activity was normalized to protein content to correct for variability in cell numbe r between individual wells.

Binding assays

For G R binding assays, cytosols were prepared by homogenizat ion of ethanol- or 1/~mol tRA-treated cells in buffer A ( 5 0 m m o l potassium phosphate, p H 7.0, 10 mmol Na2MoO4, 10 mmol monothiogly- cerol and 2 mmol D T T , [41]) and centrifugation of homogenate at 105000g for 1 h at 4°C. Specific binding of [3H]TA to G R was determined by cytoso- lic binding assays. One milliliter aliquots of cell extracts were incubated with increasing concen- trations of [3H]TA in the presence (non-specific) and absence (total) of 500-fold molar excess of unlabelled T A for 1 h at 4°C. Triplicate 50 Ftl aliquots of labelled extracts were incubated with 400/A of a hydroxylapa- rite slurry for 30 min at 4°C to facilitate binding of the receptor to the resin. The mixtures were pelleted, washed several times to remove free labelled agonist, and radioactivity was determined by liquid scintil- lation counting. Specific binding was calculated by subtracting non-specific from total binding. T o evalu- ate changes in binding affinity and/or maximal bind- ing capacity, Scatchard analyses were performed [42]. Data were plotted as specific bound/free vs specific bound. Dissociation constants ( K a = - 1 / s l o p e ) for extracts of ethanol and tRA-treated cells were com- pared to determine whether hormone treatment changes binding affinity of GR. The X-intercept indi- cated maximal specific binding capacity.

Nuclear translocation assay

To measure the extent of nuclear translocation of G R in ethanol- and tRA-treated cells, they were har- vested after 24 h of appropriate t reatment and resus- pended in serum-free med ium containing either 30 n M 3H[TA] alone or mixed with a 500-fold molar excess of unlabelled TA. The whole cell suspensions were incubated at 37°C for 1 h. Cells were washed extensively with phosphate-buffered saline (PBS), and an aliquot of the suspension was counted for total (nuclear and cytoplasmic) cellular binding activity. Another aliquot of cells was lysed by freezing and thawing. Cytosolic and nuclear fractions were separ- ated by centrifugation and each was subjected to liquid scintillation counting. Specific binding in cyto- solic and nuclear fractions was calculated as described above, and the percentage of nuclear binding activity was used as an estimation of the percentage of nuclear translocation [41 ].

R E S U L T S

Retinoic acid potentiates basal and GC-induced transcrip- tion from the M M T V promoter

K293 cells (clone I1 d) stably transfected with the p romote r - repor te r construct p H H - L u c , which con- tains 3 5 0 b p of the M M T V - L T R ( - 2 2 0 to +110) ups t ream of the luciferase reporter gene, were treated with increasing concentrations of tRA in the absence or presence of various concentrations of dexametha- sone. tRA caused a dose-dependent induction of basal luciferase activity, with a maximal 5.7-fold induction detected at 10~tmol tRA and a 4.9-fold induction detected at 1/~mol tRA (Fig. 1). In different exper- iments, maximal t R A - i n d u c e d luciferase activity ranged from 2.7- to 10-fold over basal levels, tRA sig- nificantly potentiated 1 0 n M , 1 0 0 n M and 1 ~mol dexamethasone-induced transcriptional activation of p H H - L u c . Trea tmen t with 1 0 n M , 1 0 0 n M and 1/~mol dexamethasone alone resulted in an eight-, 31- and 40-fold induction of luciferase activity, re- spectively, and this induction was potentiated by each concentration of tRA tested. In the presence of 1 #tool tRA, these same concentrations of dexametha- sone caused a 29-, 90- and 94-fold induction over basal levels respectively (Fig. I(B, C, D); solid bars). This response was more than 2.5 times the activity calculated based on a purely additive effect of tRA and dexamethasone. Maximal induction was seen in the presence of 5 ~tmol tRA and 1/~mol dexametha- sone (120-fold, 2.8 times the calculated additive value). Similar results were obtained when K293 cells were transiently transfected with p H H - L u c and the human G R expression vector, pRShGR~ (data not shown). Essentially identical results were obtained when the pure G R agonist RU 28362 was used in place of dexamethasone (data not shown).

tRA-induced Transcription from the MMTV Promoter 133

4000

3000

e" 2000

tRA A 10nMO.x I n --~7-- tRA + 10 nM Oex -<>- tRA + 1 uM Dex ram= tRA 1

~ r ~ l tRA + Dex I

00 25 50 75 100 12.5 • .0 .5 • 100

,4..l J:: 01 . i

._1

3000

2000

1000

0

100 inM Dex

0 01 1.0 2.5 5.0

C

10.0

lp.M Dex - ]

l l o o., 1o 2 5 so 10.0

B 800

. 600

400

2 200 I~.

o

3000 e- D "

J~ O1 .m

,_1 2000

1000

trans.Retinoic a c i d ( g M )

Fig. 1. t-Retinoic acid and dexamethasone synergistically activate transcription from the MMTV promoter: K293-I1 d cells (stably transfected with pHH-Luc) were plated in six-well dishes. Duplicate monolayers were exposed to the indicated agonist(s) for 24 h. Cell extracts were evaluated for luciferase activity and protein content. A. Dose-.dependent induction of luciferase activity by tRA in the absence (o) or presence of 10 nM (V) , 100 nM (HD or 1 / ~ (~)) dexamethasone (dex). Data from A are replotted in B (10 nM dex), C (100 nM dex) and D ( 1 / ~ l dex) to depict the synergism between tRA and dex. Solid bars represent activity induced by tRA alone. Open bars represent activity upon cotreatment with dex and the indicated concentration of tRA.

Data are averages of duplicate treatments from a representative experiment.

Effects of tRA treatment on GR binding affinity, binding capacity and agonist induced nuclear translocation

Ethanol-treated control and 1 pmol tRA-treated K293-I1 d cells or 5Ik cells, another clone that was stably transfected with the human GR expression vec- tor pRShGR~, were used for Scatchard analysis to determine the potential effects of tRA treatment on the affinity or specific binding capacity of GR for the synthetic glucocorticoid, triamcinolone acetonide (TA). Binding capacity of endogenous G R (as measured in cytosolic extracts of I1 d cells) in etha- nol-treated control or 1 #tool tRA-treated cells was 6 x 104 sites/cell and 6.2 x 104 sites/cell, respectively. Similarly, binding affinity of GR in control or tRA- treated cells was similar, with Kd values being 6.1 and 7.6 nM, respectively (Fig. 2(A)). Because endogenous GR expression in these cells is relatively low, sub- sequent experiments were performed using cells stably or transiently transfected with the human G R ex- pression vector pRShGR~. To determine whether tRA altered the binding capacity or affinity of trans- fected GR, Scatchard analysis was also performed in cytosols prepared from 5Ik cells that overexpress transfected GR. The binding capacity of both control and tRA-treated cells was 3 x l0 s sites/cell, whereas

K~ values for TA were 8 and 7.4 nM, respectively (Fig. 2(B)). These data indicate that tRA-mediated potentiation of GC-induced transcriptional activity is not caused by an upregulation of G R binding capacity or increased G R binding affinity. To determine whether tRA can increase the ability of G R to translo- care into the nucleus, thereby potentiating its tran- scriptional activity, the potential effect of tRA on nuclear translocation of G R was determined. Ethanol- treated control and 1 pmol tRA-treated 5Ik cells were subjected to whole-cell binding assays using [3H]TA as the ligand. Specific binding activities in cytosolic and nuclear fractions were determined and the per- centages of nuclear G R binding sites were calculated. In ethanol-treated and tRA-treated cells, 47.3% and 49.7% of the GR binding sites, respectively, were nuclear (Table 1). These data suggest that tRA does not promote GC-induced transcriptional activation by an enhancement of nuclear translocation of the GR.

Effects of GC and tRA on transcriptional responses from pure GRE or pure R A R E sequences

To characterize further the mechanisms of tRA modulation of GC-induced transcriptional activity, the potential effect of tRA on GC-induced transcrip-

134 R.D. Medh and T. J. Schmidt

A l l d c e l l s 0.06 • -tRA

• v +tRA

0.04 v •

0.02

©

0 . 0 0 i i i i

0.~ o B 51k c e i l s

0 . 4

v 0 . 3

0 .2

0 . I

0 . 0 , i

Bound (nM)

Kd: control: 6.1 nM tRA treated: 7.6 nM

Sites/cell: control: 60,000 tRA treated: 62,000

Kd: control: 8.0 nM tRA treated: 7.4 nM

Sites/celh control: 200,000 tRA treated: 200,000

Fig. 2. S c a t c h a r d ana lys i s o f G R b i n d i n g in I l d a n d 5IK cells: I l d (pane l A) or 5IK (pane l B) ce l l s w e r e t rea ted w i t h e i ther e thano l (*) or 1 / tM tRA ( V ) for 24 h, then h a r v e s t e d and r e s u s p e n d e d in s e r u m - f r e e D M E M . O n e n~ill | l (ter a l iquot s w e r e i n c u b a t e d wi th v a r i o u s c o n c e n t r a t i o n s (1.5 to 60 nM) o f [3HITA in the a b s e n c e (total) or p r e s e n c e (non- spec i f i c ) o f 500-fold m o l a r e x c e s s o f u n l a b e l l e d T A for 1 h. Spec i f i c b o u n d a n d free rad ioac t iv i ty w a s d e t e r m i n e d by Hquid s c i n - t i l lat ion count ing . T h e data w e r e subjec t ed to S c a t c h a r d ana lys i s and p lo t t ed to d e t e r m i n e Kd ( s lope = - llKd) a n d total b i n d i n g c a p a c i t y ( X - i n t e r c e p t ) . D a t a r e p r e s e n t s

a v e r a g e s o f tr ip l icate sets .

tional activity of the purely GC responsive minimal promoter-reporter constructs, p G R E 1 - C A T and pGRE2-CAT, was determined. One micromole of the pure GR agonist RU28362 (RU 362) induced CAT activity 6.7- and 18-fold from l x G R E and 2 x GRE, respectively (Fig. 3(A)). In this and subsequent exper- iments we have preferred to use RU 28362 rather than dexamethasone because it is a more specific ligand for GR. tRA at concentrations of 250 nM or 2.5 #tool did not significantly affect basal CAT ac- tivity. Treatment with 250 nM or 2.5 ~tmol tRA in combination with 1 #mol RU 28362 resulted in CAT activity from 1 × GRE and 2 × GRE that was 7.1- or

5.4-fold and 14.3- or 13.2-fold higher than basal levels, respectively. This was similar to the level of in- duction observed with RU 28362 alone, suggesting that tRA does not modulate GC-induced transcrip- tion from a pure GRE. Similarly, using a pure tRA-re- sponsive construct, RARE-TK-Luc , containing two tandemly arranged RARE sequences from the fi-RAR promoter upstream of a minimal thymidine kinase promoter linked to a luciferase reporter gene, we demonstrated that RU 28362 does not affect the dose-dependent tRA-induced transcriptional activity from a pure RARE (Fig. 3(B)). Sequences located within the M M T V promoter thus appear to be directly involved in tRA-induced transcription and synergism between GC and tRA from that promoter.

Effects of tRA on MR-media ted transcriptional activation

of the M M T V promoter

Several other members of the steroid hormone receptor super family, including M R and progesterone receptors are also capable of inducing transcriptional activation of the M M T V promoter by binding to the GRE sequences, and we therefore determined the effect of tRA on MR-induced transcription from pHH-Luc . Because GR and M R have overlapping ligand binding specificities, and the mineralocorticoid hormone aldosterone can bind to and activate the GR, transcriptional activity induced by M R needed to be evaluated under conditions where GR are blocked by a specific antagonist. To determine the efficiency of the GR antagonists in blocking GR-mediated tran- scriptional activation, K293 cells transfected with pRShGRa and p H H - L u c were pretreated with or without each antagonist (RU 38486 or ZK 98299) prior to treatment with aldosterone. Figure 4(A) shows that RU38486 alone stimulated transcription from the M M T V promoter 4.7-fold over basal levels, suggesting that it exhibited partial agonist activity. In contrast, the GR antagonist ZK 98299 did not stimu- late basal level transcription, whereas aldosterone stimulated basal level transcription 25.3-fold. Pretreatment with RU 38486 or ZK 98299 lowered aldosterone-induced activity from 25.3- to 4.5- or 1.8 times basal levels, respectively. The effect of tRA on MR-induced transcriptional activity could thus be stu- died by blocking GR with ZK 98299 and stimulating MR with aldosterone. Figure 4(B) compares the effect of tRA on GR- and MR-mediated transcriptional acti- vation of the M M T V promoter. K293 cells were tran-

Table 1. Effect of rRA on nucleur translocation of GR in 5IK cells

GR (sites per cell)

Total Cytoplasmic Nuclear Nuclear + cytoplasmic % Nuclear translocation

Control 300 000 164 000 142 000 302 000 47. 2 tRA-treated 350 000 159 000 174 000 332 000 51.1

tRA-induced Transcription from the MMTV Promoter 135

12

o 10

o B o

6 v

4 o

F~ 2

tD

1 x GRE [----] 2x GRE

m--]

- I ~ M

I( o ~ RU 3 6 2 - - I ~ M I ~ M

t - R A 2 5 0 n M 2 . 5 ~ M 2 5 0 n M 2 . 5 ~ M

A

120000 EtOH B 10 nM RU 3 6 2

"~ 100 nM RU 362 <× -~ I~M RU 362 \X o \X

O~ 80000 ,,X

~o ,,×

• ,~ \X m \X

,oooo '× '× '× "× '× ,×

0 ~ \ ~

0 2 .5 25 2 5 0 2 5 0 0

tTa~.s--Retinoic Acid (nM)

Fig. 3. RE 28362 and rJRA-mediated act ivat ion of GREt/GI~JE2-CAT and 17J~PJE-TK-Luc, respectively, is not a f f ec t ed b y t h e h e t e r o l o g o u s agon i s t : A. K293 cel ls we re t r a n s i e n t l y t r a n s f e c t e d w i t h t h e G R e x p r e s s i o n v e c t o r p R S h G R ~ a n d a C A T r e p o r t e r c o n s t r u c t s d r i v e n by l x o r 2× G R E s e q u e n c e s . T r a n s f e c t e d cells w e r e p l a t e d in to s ix -wel l d i s h e s a n d t r e a t e d w i t h t h e i n d i c a t e d c o n c e n t r a t i o n s o f R U 28362 a n d / o r t R A for 24 h . C A T a c - t iv i ty d e t e r m i n e d in cell e x t r a c t s w a s n o r m a l i z e d to p r o t e i n c o n t e n t . D a t a a r e p lo t t ed h e r e as a v e r a g e s o f d u p l i c a t e se t s . B. K293 cel ls we re c o t r a n s f e c t e d w i t h p R S h G R ~ a n d t h e r e p o r t e r c o n s t r u c t R A R E - T K - L u c , a n d p l a t e d in to s ix-wel l d i s h e s . I n d i v i d u a l wel ls w e r e t r e a t e d w i t h t h e i n d i c a t e d c o n c e n t r a t i o n o f t R A a n d / o r R U

28362. D a t a r e p r e s e n t a v e r a g e l u c i f e r a s e ac t iv i ty f r o m d u p l i c a t e t r e a t m e n t s o f a r e p r e s e n t a t i v e e x p e r i m e n t .

siently cotransfected with both G R and M R ex- pression vectors (pRShGR~ and pRShMR) , along with p H H - L u c . The pure G R agonist RU28362 was used to evaluate GR--mediated responses. The GR- specific antagonists RU38486 and ZK98299 were used to block G R responses and hence to facilitate evaluation of M R responses induced by aldosterone. GR- (as measured by RU 28362) and M R - (as measured by aldosterone in the presence of Z K 98299) mediated transcription from the M M T V pro- moter was 18.2 and 6.3 times basal levels, respect-

ively, tRA-induced luciferase activity was 2.7 times basal levels, tRA potentiated the RU 28362 (GR- mediated) response to 46.7 times basal levels, which was 2.4-fold the calculated additive value, tRA poten- tiated the aldosterone (MR-mediated) response to 13.1 times basal levels, which was 1.7-fold the calculated additive value, tRA thus potentiates not only GR- but also MR-induced transactivation. Interpretat ion of the R U 38486 data are complicated because of its partial agonist activity. It is interesting to note that the partial agonist activity of RU 38486

136 R.D. Medh and T. J. Schmidt

was also potentiated to 2.7 times the calculated additive value by tRA.

1 5 0 0

1000

500

0

0

4000

3000

2 0 0 0

I000

P r e t r e a t m e n t s : Et0H RU 38486 ZK 9 8 2 9 9

A

E t 0 H Aldo

-I;RA F--I+tRA B

Fig. 4. Effect o f tRA on M R - m e d i a t e d t ranscr ip t iona l ac t i - va t ion o f the M M T V p r o m o t e r : A. K293 ce l l s w e r e t rans i en t l y t rans f ec t ed wi th p H H - L u c and p R S h G R ~ a n d p r e t r e a t e d for 4 h wi th e i ther e t h a n o l (sofid bars ) or 2 # M o f the a n t a g o n i s t s R U 38486 (open bars) or ZK 98299 ( h a t c h e d bars ) . Ce l l s w e r e further i n c u b a t e d for 2 4 h after the add i t i on o f e t h a n o l or 1 pM o f a l d o s t e r o n e . T h e abi l i ty o f e a c h G R a n t a g o n i s t to b lock a l d o s t e r o n e - i n d u c e d t r a n s c r i p t i o n a l ac t iv i ty o f G R w a s d e t e r m i n e d by m e a s u r i n g r e p o r t e r g e n e ac t iv i ty as d e s c r i b e d in the l e g e n d to Fig . 1. B. K293 ce l l s t rans i en t ly t rans f ec t ed s i m u l t a n e o u s l y wi th p H H - L u c , a n d G R a n d M R e x p r e s s i o n ve c tor s ( p R S h G I ~ a n d p R S h M R , r e s p e c t i v e l y ) , w e r e p l a t ed in 12-wel l d i s h e s a n d p r e t r e a t e d for 4 h wi th e i ther e t h a n o l or the i n d i c a t e d G R a n t a g o n i s t at 2 / tM. Cel l s w e r e i n c u b a t e d for a further 24 h after the add i t i on o f e thano l , 1 ~ R U 28362 or 1 ~M a l d o s t e r o n e . Cel l ex t rac t s w e r e e v a l u a t e d for luc i f erase ac t iv i ty a n d pro te in c o n t e n t as d e s c r i b e d in the M e t h o d s sec t ion . RU28362 a n d a l d o s t e r o n e (in the p r e s e n c e o f G R antagon i s t s RU38486 or ZK98299) w e r e u s e d to de t er - m i n e G R - and M R - s p e c i f i c r e s p o n s e s , r e s p e c t i v e l y . D a t a rep - r e s e n t a v e r a g e + S E M of dup l i ca te t r e a t m e n t g r o u p s f r o m a

r e p r e s e n t a t i v e e x p e r i m e n t .

Mapping of tRA responsive sequences within the M M T V promoter

The ability of G C and tRA to modulate transcrip- tional responses via distinct sequences within the M M T V promoter was investigated. Reporter con- structs containing either 220 or 151 bp of the 3' end of the M M T V promoter and the first 5 bp of the tran- scribed region ( - 2 2 0 to ÷5 or -151 to +5; Fig. 5(D)) were generated. Both constructs differ from p H H - L u c in that they contain only 5 bp downstream of the transcription start site. Sequences imparting steroid hormone responsiveness have previously been mapped within 202 bp ups t ream from the transcription start site. The construct p l 9 M T V (225)-Luc contains the entire steroid hormone responsive sequence, whereas p19MTV(156) -Luc lacks GRE1, the G R E known to be essential for G C responsiveness [11]. The RU28362- induced response from p H H - L u c , p19MTV(225) -Luc and p19MTV(156) -Luc was 16.6, 5.4 and 1.4 times the basal level, respectively. This is in agreement with previous reports suggesting that removal of sequences between - 1 8 7 and - 1 6 5 (GRE 1) abolishes G C responsiveness. All three con- structs responded comparably to various concen- trations of tRA, with luciferase activity that was 10.5, 6.8 and 6.1 times basal levels, respectively, when cells transfected with p H H - L u c , p19MTV(225) -Luc and p19MTV(156) -Luc were treated with 1 pmol tRA. Combined t reatment with 1 #mol RU 28362 and 1 pmol tRA resulted in a synergistic response from p H H - L u c and pMTV(225) -Luc , with luciferase ac- tivity that was 94 and 23.4 times basal levels, respect- ively (3.7- and 2.2-fold the activity calculated based on an additive response; Fig. 5(A and B)). The G C non-responsive construct p l 9MTV(156) -Luc did not exhibit any synergism, or R U 28362-induced poten- tiation of the tRA-induced response (Fig. 5(C)). Neither G C - or tRA-induced transcriptional activity of the promoterless vector p 19Luc (data not shown).

Identification of a consensus R A R E sequence within the M M TV-L TR

tRA by itself is capable of activating transcription from the M M T V promoter , and it may therefore do so via ligand-activated retinoid receptor binding to a target D N A sequence within the M M T V - L T R . The M M T V promoter sequence between -151 and +5 was screened for consensus RARE sequences, and a half-site ( C C A A G T ) at position - 6 5 to - 6 0 was identified (Fig. 6). This half-site, which could be the D N A sequence that mediates tRA responses, is located adjacent to four previously identified GC-re - sponsive elements within the M M T V - L T R . This con- sensus RARE half-site also overlaps with the proximal NF-1 site between - 6 6 and - 6 2 , which has been

tRA-induced Transcription from the MMTV Promoter 137

I0000

0 8000

:~ 6000

. ~ 4 0 0 0

-~ 2 0 0 0

0

I pHHLuc

362(obs.) ILX~+Ru 362(calc.)

0 10 - 8 10 - 7 10 - 6 10 - 5

:500

A

10170

5 0 0

0 0 0

5OO

0

p 19MTV(225)Luc

-8 - 7 0 10 10

B 4OO

I 300

: ~'00

0 - 6 - 5 - 8 - 7 - 8 - 5 10 10 0 10 10 10 10

t r a n s - R e t i n o i c Acid (M)

D -220 +110 - p H H - L u c

-2:

=m MMTV promoter

0 HRE O consensus RARE half-site

[] TATA box

p 1 9 M T V ( 2 2 5 ) - L u c

-1! p 1 9 M T V ( 1 5 6 ) - L u c

Fig. 5. 156 bp of MMTV-LTR are sufficient for dose-dependent RA-induced transcriptional activation: K293 cells were cotransfected with pRShGR~ and pHH-Luc (A), plgMTV(225)-Luc (B) or p19MTV(156)-Luc (C). Transfected cells were plated into six-well dishes and treated with the indicated concentrations of tRA in the absence (solid bars) or presence (open bars) of 1/~M RU 28362 for 24 h. Luciferase activity in cell extracts was normalized for protein content. Data represent average +SEM of agonist- induced activities of duplicate treat- ments from a representative experiment. Basal activity in response to ethanol vehicle was subtracted from agonist- induced ~tctivity in each case to facilitate comparison of observed and calculated responses. Hatched bars represent the calculated luciferase activity based on an additive response, which is the s u m of the activity

induced by each agonist individually. Panel D is a schematic representation of the constructs used.

shown by site-directed mutagenesis to be essential for maximal GC-induced transcriptional activation [11].

Role of the consensu;~ RARE half-site in mediating tRA-induced transcriptional responses from the M M T V promoter

To determine whether tRA-induced transcriptional responses were mediated via the consensus RARE half-site (CCAAGT) at position - 6 5 to - 6 0 of the M M T V promoter, these sequences were mutated to CCAtGg as described in the Methods section. This eliminated the conserved region of the consensus half-site (AGT), which has been shown to be essential for retinoid receptor-DNA interactions [21]. A comparison of transcriptional ac- tivity from the wild-type and mutated sequences revealed no significant differences in tRA-induced re- sponses or synergism with RU 28362 (Fig. 7). Ten micromolar of tRA induced transcription from p19MTV(225)Luc and p l 9 M T V ( R A R E m ) L u c by 4.1- and 3.0-fold, respectively, tRA also potentiated 1 pmol RU 28362-induced responses from the wild-

type and mutated M M T V promoter sequences by 2.1 and 1.8 times the calculated additive values, respect- ively. These data suggest that sequences other than the consensus RARE half-site mediate tRA-induced responses from the M M T V promoter.

D I S C U S S I O N

The ability of several different transcription factors to modulate GC-induced transcription through the M M T V promoter has been well documented [11-15]. In the present studies, we have demonstrated for the first time the ability of tRA to induce basal transcrip- tion and potentiate GC-induced transcription from the hormone-responsive sequences of the M M T V promoter (Fig. 1). This synergism between tRA and GC could result from one or more of the following mechanisms: (a) tRA-mediated potentiation of GR binding affinity and/or capacity; (b) mutual stabiliz- ation of the binding of respective ligand-activated receptors (transcription factors) on adjacent cis-acting D N A sequences; (c) specific protein-protein inter-

138 R.D. Medh and T. J. Schmidt

-201 gtttaaataa gtttaTGGTT ACAAACTGTT caagtggttt cctgacttgg tttggtatCA tAGTGTTCTa ttttcctatG TTCTTTTGGA aatgcttatg taaacqataa tataaa~gag ttgcaacagt cctaac

+15

CTTAAAAcga AATGTTCTGA AtctatCCAA tgctgatttt

ggatgtgaga TCTGAGCtct GTcttatgta ttgagtaaac

Fig. 6. H o r m o n e r e s p o n s e e l e m e n t s o f the M M T V - L T R s e q u e n c e : S e q u e n c e of the M M T V - L T R wi th the k n o w n H R E s in uppe rca se . G R E 1 : - 1 8 7 to -165; G R E 2 : - 1 2 3 to -105; G R E 3 : - 1 0 0 to -93 a n d G R E 4 : - 8 2 to -71. The p r e d i c t e d c o n s e n s u s RARE half -s i te is in u p p e r c a s e a n d is u n d e r l i n e d ( -65 to -60) . The T A T A box

s e q u e n c e is boxed.

actions between GR and RAR/RXR that result in complexes with higher transcriptional activity; or (d) tRA-mediated recruitment of another factor(s) that potentiates GR-induced transcriptional responses. Precedence for several of these mechanisms can be found in recently published reports. For example, tRA has been shown to upregulate GR binding ca- pacity in osteosarcoma cells with no change in bind- ing affinity [27]. In rat liver cells tRA induces translocation of GR into the nuclear compartment,

thus increasing the transcriptionally active fraction of G R [29]. Our data suggest that tRA does not signifi- cantly affect the binding affinity or maximal binding capacity of endogenous or transfected GR in K293 cells (Fig. 2). Several of our experiments have been performed with cells transfected with pRShGRa, and as there are reports [43] suggesting that DNA sequences within the GR coding sequence regulate GR gene expression thus enabling upregulation of transfected GR, it was important to determine

3000

2500

0

2000

1500

1000

~-3 500

p 19MTV(E25)Luc

~ - R U 382

r---] + RU 362

...T_

- 8 - 7 - 6 0 10 10 I0

i io

A p 19MTV(RAREm)Luc

it I .,ii

- 8 - 7 - 6 - 5 0 t0 10 10 10

t r a n s - R e t i n o i c Acid (M)

B

i

p19MTV(225)Luc: -71 ATCTATCCAAGTCTT-57

pl9MTV(RAREm)Luc: -71 ATCTATCCAtGgCTT-57 Fig. 7. T h e c o n s e n s u s R A R E hal f -s i te o f M M T V - L T R does n o t m e d i a t e r e t i n o i c ac id r e s p o n s i v e n e s s : K293 cells were co t rans fec ted wi th p R S h G R ~ a n d p lgMTV(225) -Luc (A) o r a s i te -d i rec ted m u t a n t p l 9 M T V ( R A R E m ) - L u c (B) tha t has two subs t i t u t i ons wi th in the c o n s e n s u s R A R E hal f -s i te ( C C A A G T to CCAtGg) . T rans f ec t ed cells were p la ted in six-well d i shes a n d t r ea ted wi th i n d i c a t e d c o n c e n t r a t i o n s o f tRA in t h e a b s e n c e (solid ba r s ) o r p r e s e n c e ( o p e n bars ) o f 1 / tM R U 28362. D a t a are a v e r a g e _+SEM o f dupl ica te t r e a t m e n t s f r o m a r e p r e s e n t a t i v e

expe r imen t .

tRA-induced Transcription from the MMTV Promoter 139

whether tRA enhanced G R binding in cells trans- fected with GR. Our data demonstrate that in K293 cells tRA-induced upregulation (as measured by an increase in maximal binding capacity) of G R is not a mechanism by which it potentiates GC- induced re- sponses. Similarly, the data presented in Table 1 indi- cate that tRA does not enhance the ability of GR- agonist complexes to translocate into the nucleus. Also, if tRA-induced re, sponses occurred via upregula- tion of G R or increase, d nuclear translocation of G R complexes, the same transcriptional effects would be seen irrespective of the G C responsive promoter used. However, as shown in Fig. 3(A), tRA does not affect GC- induced transcription from minimal G R E sequences transfected into K293 cells.

Transcript ional COOl:,erativity between G C and tRA has been reported for the growth hormone [32] and P E P C K [31] genes. The G C response unit of the P E P C K gene p romoter contains two G R binding sites and two accessory factor (AF) binding elements. One of these, AF1, corresponds to an RARE and has been shown to potentiate basal and GC- induced responses via RXR. Our studies demonstrate that the sequences imparting tRA responsiveness lie within 151 bp upstream from the transcription start site of the M M T V promoter , in the same region that contains four GC-responsive sequences as well as other tran- scription factor binding sites. We determined that G C - and RA-mediated synergism was specific for the M M T V promoter . Figure 3 demonstrates that tRA and G C did not affect GC- induced transcriptional re- sponses f rom pure GKE and tRA-induced transcrip- tional responses from pure RARE sequences, respectively. These data suggest that in addition to G R E sequences, other specific D N A sequences within the M M T V - L T R may be required for transcriptional interactions between G C and tRA. These results also suggest that direct prote in-prote in interactions between G R and RAR/RXR are probably not involved in the synergistic transcriptional responses. This is unlike the interaction :reported between G R and c-jun [8]. C-jun inhibits GC- induced responses from purely G C responsive sequences by direct prote in-prote in interactions with G R daat interfere with the ability of G R to bind to its response element. Cell specific cooperative responses between G R and c-jun have also been observed in T cells [44]. Similar interactions between c-jun and RAR/RXR have also been reported [45].

T o determine whether synergism between G C - and tRA-induced transcription from the M M T V promoter was specific for GR, we determined the potential abil- ity of tRA to synergize with M R - (which acts via the same cis-acting D N A response elements on the M M T V promoter) induced responses, tRA poten- tiated MR-induced transcriptional responses, although somewhat less effectively compared to its effect on GR-media ted responses (Fig. 4). The more potent

G R antagonist, R U 38486, functioned as a partial agonist. This is in agreement with previous reports that R U 38486 is capable of inducing nuclear translo- cation and specific D N A binding of GR, and can be converted to an agonist under certain conditions [46]. The weak agonist activity of RU 38486 was also potentiated significantly by RA (Fig. 4). These data suggest that the tRA-induced transcriptional syner- gism from the hormone-responsive sequences of the M M T V promoter is not specific to GR.

T o map D N A sequences that mediate tRA-induced transcriptional responses, the hormone-responsive region of the M M T V promoter ( - 2 2 0 to +5) was cloned into a promoterless luciferase reporter vector p19Luc to generate p19 MTV(225)Luc . G C and tRA induced a significant transcriptional activation and synergism from M M T V promoter sequences spanning - 2 2 0 to +5 bp from the transcription start site. A shorter fragment of the promoter ( -151 to +5), lack- ing GRE1 elicited a tRA-induced response that was indistinguishable f rom that induced by the 225 bp fragment. As expected, G C did not induce transcrip- tion f rom this sequence. We conclude that RA- induced transcriptional activation of the M M T V pro- moter is mediated via sequences within 151 bp of the transcription start site, and is independent of G C responsiveness. A hexanucleotide sequence corre- sponding to the consensus RARE half-site (C/T C A/ C A G T ) has been identified within the tRA respon- sive fragment, between - 6 5 and - 6 0 bp ups t ream from the transcription start site. This sequence ( C C A A G T ) may be the site of retinoid receptor bind- ing. Although retinoid receptors have been shown to bind to their response element as monomers , whether such an interaction yields a transcriptionally active complex is not known. Dimeric receptor may bind to the consensus RARE half-site and adjacent sequences. RXR modulates its activity primarily via heterodimeri- zation with RAR/T3R/VDR [24, 25], but has not been shown to heterodimerize with GR. Site-directed mutagenesis of the RARE half-site to C C A T G G did not abolish tRA-induced transcriptional responses, in- dicating that this consensus RARE half-site is not involved in tRA-mediated transcriptional responses. It is unlikely that the mutant RARE is also capable of modulat ing a tRA-induced transcriptional response because the trinucleotide A G T is highly conserved in all the known RARE sequences [21], and mutat ions within this sequence have previously been shown to abolish retinoid responsiveness [31]. Overlapping N F - 1 binding sequences prevented construction of mu- tations within the first three bases of the RARE half- site.

Data presented in this paper suggest that tRA- induced transcriptional responses f rom the M M T V promoter may be mediated by: (a) activation or upre- gulation of a factor (Fa in Fig. 8(A)) that directly binds to DNA; or (b) the activation/upregnlation of a

140 R.D. Medh and T. J. Schmidt

A B

÷

+ GC ;

4,

Fig. 8. Proposed models for RA-media ted transcriptional responses and synerg i sm with GC for MMTV pro- moter activation: Model A represents a scenario where an RA- induced factor (Fa) interacts directly with D N A sequences within the MMTV promoter . Model B represents a s ituation where Fa indirectly affects transcrip- tional responses via another transcription factor (TFx). In both cases cooperat ive interact ions with GR bound

to GRE sequences modulate synergist ic responses .

factor (Fa) that acts as an activator/potentiator of other transcription factor(s) (TFx in Fig. 8(B)). In the model proposed in Fig. 8(A), Fa could be a reti- noid receptor binding to a non-canonical RARE. In this model, synergism with G C could result f rom the simultaneous binding of Fa and G R to adjacent cis- acting sequences on the M M T V promoter , resulting in mutual stabilization of D N A - p r o t e i n interactions and an increased transcriptional efficiency. In the model proposed in Fig. 8(B), where Fa does not directly bind to D N A but functions as an activa- tor of TFx , Fa could potentially serve as a coacti- vator of GR, thereby facilitating the formation of a stable transcription complex modulat ing a syner- gistic response.

A candidate transcription factor meeting the cri- teria of Fa in model A is Oct-1. Synergistic effects of Oct-1 on GC- induced transcriptional ac- tivation of the M M T V promoter are well docu- mented [12]. There have been several reports suggesting that RA plays an important role in the regulation of octamer transcription factors (OTFs, including Oct - l ) [47-50]. Negative regulation (sup- pression) of Oct-4 gene expression has been impli- cated in the tRA-induced downregulation of k F G F expression in differentiated embryonal carcinoma cells [48]. Similarly, in Jurkat T cells RA decreases the levels of IL-2 expression via an octamer mot i f on the IL-2 promoter [49]. In a human embryonal carcinoma cell line, N T 2 / D 1 , RA induces a stimulation of the

synthesis of Oct-1 [50]. There are reports that suggest a transcriptional interference between Oct-1 and GR, which is brought about by direct prote in-prote in in- teractions [51]. However, it is well known that maxi- mal transcriptional activation of the M M T V promoter involves a synergistic interaction between G R and Oct-1 [12]. Moreover, Oct-1 has been demonstra ted to stimulate basal transcription f rom the M M T V pro- moter by binding to cis-acting octamer D N A sequences. Upregulat ion of Oc t - l , rather than downregulation, is thus likely to mediate RA- induced transcriptional responses from the M M T V promoter . However, there might be other unknown factors that may play the role of Fa in model A of Fig. 8.

The other hypothesis, schematically depicted in model B, proposes that RA induces Fa, which does not directly bind to D N A but interacts with another D N A binding protein (transcription factor TFx) to potentiate transcriptional responses. Fa may be a reti- noid receptor that interacts with another M M T V pro- moter -bound transcription factor. The association of ER with another D N A - b o u n d factor has been pro- posed as a mechanism of the raloxifene/ER-mediated activation of the TGF-f l3 p romoter [52]. In this model, T F x may be Oc t - l , an unknown factor, or a component of the basal transcriptional machinery ( T F I I D complex). Fa would then serve as an ®additional factor that couples G R with T F x to stabilize the entire transcription complex, resulting in

tRA-induced Transcription from the MMTV Promoter 141

a synergistic response to G C and RA. Th i s mode l is

analogous to that p roposed for C R E B - b i n d i n g pro te in

(CBP) , which has be, en shown to utilize dis t inct domains to interact with mul t ip le nuc lea r receptors and serve as an in tegrator of mul t ip le signal t r ansduc- t ion pathways [53]. C o m p e t i t i o n for CBP b ind i ng has been proposed as a m e c h a n i s m for the an t agon i sm be tween G R / R A R and AP1 [53]. However , we pro- pose that Fa modula tes cooperat ive in teract ions be tween G R and other t ranscr ip t ion factor(s).

Nuc lea r receptors are k n o w n to interact with ad-

di t ional factors to media te the act ivat ion of gene ex- pression [54]. Coactiva'cors that associate with re t inoid receptors, es trogen receptors and other nuc lear recep- tors in a l i g a n d - d e p e n d e n t m a n n e r have been ident - ified, and could potent ia l ly media te G R and RAR interact ions. Ano the r potent ia l media to r of RA-

induced responses f rom the M M T V p romote r is E1A. E1A or an analogous activity has been shown to serve

as a br idge be tween re t inoid receptors and the basal t ranscr ip t ional mach ine ry and modu la t e t ranscr ip- t ional cooperat ivi ty [55]. A n in terac t ion be tween E1A and oc tamer factors has also been repor ted [56].

In conclus ion , we have demons t r a t ed that in K293 cells tRA induces t ranscr ip t ion f rom the M M T V pro-

mote r f rom sequences located wi th in 151 bp of the t ranscr ip t ion start site. T h e consensus R A R E half-site located be tween - 6 5 and - 6 0 does no t appear to modu la t e tRA- induced responses, tRA and G C syner- gistically induce a t ranscr ip t ional response that is ap- proximately twice the additive response. This

synergism does no t appear to involve t R A - i n d u c e d e n h a n c e m e n t of G R b i n d i n g affinity, max imal b i nd i ng

capacity or nuc lea r t ranslocat ion. T h e inabil i ty of tRA to modu la t e GC-indu~ced t ranscr ip t ion f rom a con- struct con ta in ing only a G R E sequence suggests that

sequences wi th in the M M T V promoter , b u t dis t inct f rom the G R E sequences , media te the observed effects. These data also suggest that direct p r o t e i n - p ro te in in teract ions be tween G R and RAR/RXR, irre- spective of D N A b ind ing , are p robab ly no t respon- sible for the observed synergism. We propose that the observed t R A - i n d u c e d responses are indi rec t effects

resul t ing from the act ivat ion or i nduc t i on of a fac- tor(s) that either directly b inds to D N A or enhances

the D N A b i n d i n g abili ty of an addi t ional t ranscr ip t ion factor. T h e synergy be tween tRA and G C may result f rom a m u t u a l s tabil izat ion of D N A b ind ing by indi - vidual factors inc lud ing GR, and hence an enhance - m e n t of t ranscr ip t ional response. Regardless of the mechan i sm, these data demons t r a t e for the first t ime that the M M T V p romote r is responsive to tRA by itself, and that tRA synergizes with the G C - m e d i a t e d t ranscr ip t ional responses.

Acknowledgements--We thank Richard Lay for technical assistance and Drs Andrew F. Russo and Frederick E. Domann for many helpful discussions and adv:ice. We also thank Dr Andrew F. Russo

for the gifts of pHH-Luc and pRARE-TK-Luc, Dr Ronald M. Evans for pRShGR~ and pRShMR, and Dr John Cidlowski for pGRE1-CAT and pGRE2-CAT. This work was supported in part by a grant from the Iowa Heart Association (T.J.S.) and an Institutional Seed Grant from the American Cancer Society (R.D.M.).

R E F E R E N C E S

1. Leung K. and Munck A., Peripheral actions of glucocorticoids. Annual Review of Physiology 37 (1975) 245-272.

2. Weinberger C. and Bradley D. J., Gene regulation by receptors binding lipid-soluble substances. Annual Review of Physiology 52 (1990) 823-840,

3. Schmidt T. J. and Litwack G., Activation of the glucocorti- cold-receptor complex. Physiological Reviews 62 (1982) 1131- 1192.

4. Beato M., Gene regulation by steroid hormones. Cell 56 (1989) 335-344.

5. Ribeiro R. C., Kushner P. J. and Baxter J. D., The nuclear hormone receptor gene superfamily. Annual Review of Medicine 46 (1995) 443-453.

6. Aberblom I. E., Slater E. P., Beato M., Baxter J. D. and Mellon P. L., Negative regulation by glucocorticoids through interference with a cAMP responsive enhancer. Science 241 (1988) 350-353.

7. Okret S., Dong Y., Tanaka H., Cairns B. and Gustafsson J. A., The mechanism for glucocorticoid resistance in a rat hepatoma cell variant that contains functional glucocorticoid receptor. Journal of Steroid Biochemistry and Molecular Biology 40 (1991) 353-361.

8. Schule R. and Evans R. M., Cross-coupling of signal transduc- tion pathways: zinc finger meets leucine zipper. Trends in Genetics 7 (1991) 377-381.

9. Jonat C., Rahmsdorf H. J., Park K., Cato A. C. B., Gebel S., Ponta H. and Herrlich P., Antitumor promotion and antiin- flammation: down-modulation of AP-1 (Fos/Jun) activity by glucocorticoid hormone. Cell 62 (1990) 1189-1204.

10. Zhang X., Dong J. and Chiu J., Regulation of alpha-fetoprotein gene expression by antagonism between AP-1 and the gluco- corticoid receptor at their overlapping binding site. Journal of Biological Chemistry 266 (1991) 8248-8254.

11. Buetti E. and Kuhnel B., Distinct sequence elements involved in the glucocorticoid regulation of the mouse mammary tumor virus promoter identified by linker scanning mutagenesis. Journal of Molecular Biology 190 (1986) 379-390.

12. Bruggemeier U., Kalff M,, Franke S., Scheidereit C. and Beato M., Ubiquitous transcription factor OTF-1 mediates induction of the MMTV promoter through synergistic interaction with hormone receptors. Cell 64 (1991) 565-572.

13. Majors J. and Varmus H. E., A small region of the mouse mammary tumor virus long terminal repeat confers glucocorti- cold hormone regulation on a linked heterologous gene. Proceedings of the National Academy of Sciences of the U.S.A. 80 (1983) 5866-5870.

14. Cato A. C. B., Skroch P., Weinmann J., Butkeritis P. and Ponta H., DNA sequences outside the receptor-binding sites differentially modulate the responsiveness of the mouse mam- mary tumor virus promoter to various steroid hormones. EMBOJournal 7 (1988) 1403-1410.

15. Brnggemeier U., Rogge L., Winnacker E.-L and Beato M., Nuclear factor I acts as a transcription factor on the MMTV promoter but competes with steroid hormone receptors for DNA binding. EMBO Journal 9 (1990) 2233-2239.

16. Yang-Yen H., Chambard J., Sun Y., Smeal T., Schmidt T. J., Drouin J. and Karin M., Transcriptional interference between c-Jun and the glucocorticoid receptor: mutual inhibition of DNA binding due to direct protein-protein interaction. Cell 62 (1990) 1205-1215.

17. Levin A. A., Sturzenbecker L. J., Kazmer S., Bosakowski T., Huselton C., Allenby G., Speck J., Kratzeisen C., Rosenberger M., Lovey A. and Grippo J. F., A new pathway for vitamin A. Understanding the pleiotropic effects of retinoids. Annals of the New York Academy of Sciences 669 (1992) 70-85.

18. Gudas, L. J., Sporn, M. B. and Roberts, A. B., Cellular biology and biochemistry of the retinoids. In The Retinoids: Biology,

142 R. D. Media and T. J. Schmidt

Chemistry and Medicine, eds M. B. Spom, A. B. Roberts and D. S. Goodman. Raven Press, New York, 1994, pp. 444-520.

19. Petkovich M., Regulation of gene expression by vitamin A: the role of nuclear retinoic acid receptors. Annual Review of Nutrition 12 (1992) 443-471.

20. Mangelsdorf, D. J., Umesono, K. and Evans, R. M., The reti- noid receptors. In The Retinoids: Biology, Chemistry and Medicine, eds M. B. Sporn, A. B. Roberts and D. S. Goodman. Raven Press, New York, 1994, pp. 319-349.

21. Leid M., Kasmer P. and Chambon P., Multiplicity generates diversity in the retinoic acid signalling pathways. Trends in Biochemical Sciences 17 (1992) 427-433.

22. Pemrick S. M., Lucas D. A. and Grippo J. F., The retinoid receptors. Leukemia 8 (1994) 1797-1806.

23. Forman B. M., Casanova J., Raaka B. M., Ghysdael J. and Samuels H. H., Half-site spacing and orientation determines whether thyroid hormone and retinoic acid receptors and re- lated factors bind to DNA response elements as monomers, homodimers, or heterodimers. Molecular Endocrinology 6 (1992) 429-442.

24. Leid M., Kasmer P., Lyons R., Nakshatri H., Saunders M., Zacharewski T., Chen J.-Y., Staub A., Gamier J.-M., Mader S. Chambon P., Purification, cloning, and RXR identity of the HeLa cell factor with which RAR or TR heterodimerizes to bind target sequences efficiently. Cell 68 (1992) 377-395.

25. Bugge T. H., Pohl J., Lonnoy O. and Stunnenberg H. G., RXR~, a promiscuous partner of retinoic acid and thyroid hor- mone receptors. EMBO Journal 11 (1992) 1409-1418.

26. Ng K. W., Manji S. S., Yotmg M. F. and Findlay D. M., Opposing influences of glucocorticoid and retinoic acid on transcriptional control in preosteoblasts. Molecular Endocrinology 3 (1989) 2079-2085.

27. Song L.-N., Effects of retinoic acid and dexamethasone on pro- liferation, differentiation, and glucocorticoid receptor ex- pression in cultured human osteosarcoma cells. Oncology Research 6 (1994) 111-118.

28. Wan Y. J., Wang L. and Wu T. C., Dexamethasone inreases the expression of retinoid X receptor genes in rat hepatoma cell lines. Laboratory Investigation 70 (1994) 547-552.

29. Audouin-Chevallier I., Pallet V., Coustaut M., Alfos S., Higueret P. and Garcin H., Retinoids modulate the binding ca- pacity of the glucocorticoid receptor and its translocation from cytosol to nucleus in liver cells. Journal of Steroid Biochemistry and Molecular Biology 52 (1995) 321-328.

30. Imai E., Stromstedt P., Quinn P., Carlstedt-Duke J., Gustafsson J. and Granner D. K., Characterization of a com- plex glucocorticoid response unit in the phsophoenolpyruvate carboxykinase gene. Molecular and Cellular Biology 10 (1990) 4712-4719.

31. Lucas P. C., O'Brien R. M., Mitchell J. A., Davis C. M., Imai E., Forman B. M., Samuels H. H. and Granner D. K., A reti- noic acid response element is part of a pleiotropic domain in the phosphoenolpymvate carboxykinase gene. Proceedings of the National Academy of Sciences of the U.S.A. 88 (1991) 2184- 2188.

32. Bedo G., Santisteban P. and Aranda A., Retinoic acid regulates growth hormone gene expression. Nature 333 (1989) 669-672.

33. Nordeen S. K., Luciferase reporter gene vectors for analysis of promoters and enhancers. Biotechniques 6 (1988) 454-458.

34. Hollenberg M., Weinberger C., Ong E. S., Cerelli G., Oro A., Lebo R., Thompson E. B., Rosenfeld M. G. and Evans R. M., Primary structure and expression of a functional human gluco- corticoid receptor cDNA. Nature 318 (1985) 635-641.

35. Arriza J. L., Weinberger C., Cerelli G., Glaser T. M., Handelin B. L., Housman D. E. and Evans R. M., Cloning of human mineralocorticoid receptor complementary DNA: struc- tural and functional kinship with the glucocorticoid receptor. Science 237 (1987) 268-275.

36. Allgood V. E., Oakley R. H. and Cidlowski J. A., Modulation by vitamin B6 of glucocorticoid receptor-mediated gene ex- pression requires transcription factors in addition to the gluco- corticoid receptor. Journal of Biological Chemistry 268 (1993) 20870-20876.

37. Chen C. and Okayama H., High efficiency transformation of mammalian cells by plasmid DNA. Molecular and Cellular Biology 7 (1987) 2745-2752.

38. Wood, K. V., Recent advances and prospects for use of Beetle Luciferases as genetic reporters. In Bioluminescence and Chemilumniscence: Current status, eds P. Stanley and L. Kricka. John Wiley and Sons Ltd., Chichester, 1991.

39. Gorman C. M., Moffat L. F. and Howard B. H., Recombinant genomes which express chloramphenicol acetyl transferase in mammalian cells. Molecular and Cellular Biology 2 (1982) 1044-1051.

40. Bradford M. M., A rapid and sensitive method for the quanti- tation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72 (1976) 248 254.

41. Schmidt T. J. and Diehl E. E., In vitro activation of rat cardiac glucocorticoid antagonist versus agonist-receptor complexes. Biochimica et Biophysica Acta 970 (1988) 212-221.

42. Scatchard G., The attraction of proteins for small molecules and ions. Annals of the New York Academy of Sciences 51 (1949) 660-672.

43. Bumstein K. L., Jewell C. M. and Cidlowski J. A., Human glu- cocorticoid receptor contains cDNA sequences sufficient for receptor downregulation. Journal of Biological Chemistry 265 (1990) 7284-7291.

44. Shemshedini L., Knauthe R., Sassone-Corsi P., Pornon A. and Gronemeyer H., Cell-specific inhibitory and stimulatory effects of Fos and Jun on transcription activation by nuclear receptors. EMBOJournal 10 (1991) 3839-3849.

45. Pfahl M., Nuclear receptor/AP-1 interaction. Endocrine Reviews 14 (1993) 651-658.

46. Wehle H., Moll J. and Cato A. C. B., Molecular identification of steroid analogous with dissociated antiprogestin activities. Steroids 60 (1995) 368-374.

47. Kim K. Y., Kweon K. R., Lee M. S., Kwak S. T., Kim K. E., Hwang 13. D. and Lim K., Reduced level of octamer binding transcription factor [Oct-l] is correlated with H2B histone gene repression during differentaition of HL60 cells by all-trans retinoic acid. Biochemical and Biophysical Research Communications 213 (1995) 616-624.

48. Schoorlemmer J. and Kruijer W., Octamer-dependent regu- lation of kFGF gene in embryonal carcinoma and embryonic stem cells. Mechanisms of Development 36 (1991) 75-86.

49. Felli M. P., Vaca A., Meco D., Screpanti I., Farina A. R., Maroder M., Martinotti S., Petrangeli F., Frati I. and Gulinio A., Retinoic acid-induced down-regulation of the interleukin-2 promoter via cis-regulatory sequences containing an octamer motif. Molecular and Celhdar Biology 11 (1991) 4771-4778.

50. Hsu W. and Chen-Xiang S., Convergent regulation of NF-IL6 and Oct-1 synthesis by interleukin-6 and retinoic acid signaling in embryonal carcinoma cells. Molecular and Cellular Biology 13 (1993) 2515-2523.

51. Kutoh E., Stromstedt P. E. and Poellinger L., Functional inter- ference between the ubiquitous and constitutive octamer tran- scription factor OTF-1 and glucocorticoid receptor by direct protein-protein interactions involving the homeo subdomain of OTF-1. Molecular and Cellular Biology 12 (1992) 4960-4969.

52. Yang N. N., Venugopalan M., Hardikar S. and Glasebrook A., Identification of an estrogen response element activated by metabolites of 17/~-estradiol and raloxifene. Science 276 (1996) 1222-1225.

53. Kamei Y., Xu L., Heinzel T., Torchia ]., Kurokawa R., Gloss B., Lin S.-C., Heyman R. A., Rose D. W., Glass C. K. and Rosenfeld M. G., A CBP integrator complex mediates tran- scriptional activation and AP1 inhibition by nuclear receptors. Cell 85 (1996) 403-414.

54. Halachmi S., Marden E., Martin G., MacKay I., Abbondanza C. and Brown M., Estrogen receptor-associated proteins: poss- ible mediators of hormone-induced transcription. Science 264 (1994) 1455-1458.

55. Berkinstam A., Ruiz M. M., Barettino D., Horikoshi M. and Stunnenberg H. G., Cooperativity in transactivation between retinoic acid receptor and TFIID requires an activity analogous to E1A. Cell 69 (1992) 401-412.

56. Scholer H. R., Ciesiolka T. and Gross P., A nexus between Oct-4 and E1A: implications for gene regulation in embryonic stem cells. Cell66 (1991) 291 304.