THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1991 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 266, No. 32, Issue of November 15, PP. 21991-21996,1991 Printed in U. S. A .

Identification of a Thyroid Hormone Response Element in the Phosphoenolpyruvate Carboxykinase (GTP) Gene EVIDENCE FOR SYNERGISTIC INTERACTION BETWEEN THYROID HORMONE AND cAMP cis-REGULATORY ELEMENTS*

(Received for publication, January 18, 1991)

Marta GiraltS, Edwards A. Park, Austin L. Gurneys, Jinsong Liu, Parvin Hakimi, and Richard W. Hansonn From the Department of Biochemistry, Case Western Reserve University School of Medicine, Cleveland, Ohio 44106

Transcription of the gene for the cytosolic form of phosphoenolpyruvate carboxykinase (GTP) (EC 4.1.1.32) (PEPCK) in the liver is regulated by many hormones including thyroid hormone (T3). In order to identify the elements in the promoter which are re- quired for transcriptional induction by T3, we co- transfected a T3 receptor expression vector with a PEPCK-CAT reporter gene into HepG2 cells. Using vectors with deletions in the PEPCK promoter, we identified a single T3 response element (TRE) between positions -332 and -308. This element binds [‘2SI]T3- labeled T3 receptor contained in nuclear extracts pre- pared from rat liver. Furthermore, the P3(I) element (-250 to -234), a previously described cis-sequence involved in mediating the induction of PEPCK gene transcription by CAMP, is also required for the T3 responsiveness of the promoter. In the absence of either the TRE or the P3(I) binding sites, no stimulation of transcription from the PEPCK promoter by T3 was observed, indicating that both elements are required for the T3 transcriptional regulation. Finally, a syn- ergistic induction of PEPCK gene transcription by T3 and cAMP is described. This interaction requires both T3- and CAMP-responsive cis-acting elements.

Cytosolic phosphoenolpyruvate carboxykinase (GTP) (PEPCK)’ is a key enzyme in gluconeogenesis. PEPCK is present primarily in the liver and kidney, but enzyme activity has also been noted in adipose tissue, the jejunum (Hanson and Garber, 1972) and the mammary gland (Garcia-Ruiz et al., 1983). In the liver, transcription of this gene is stimulated

* This work was supported by Grants DK-21859 and DK-24451 from the National Institutes of Health. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ Recipient of a Postdoctoral Fellowship from the “Plan de For- macion de Profesorado y Personal Investigator,” Ministerio de Edu- cation y Ciencia, Spain. Present address: Unitat de Bioquimica i Biologia Molecular, Departament de Biquimica i Fisiologia, Univer- sitat de Barcelona, Avda Diagonal, 645, 08028 Barcelona, Spain.

~~

§ Trainee on the Metabolism Training Program. T To whom correspondence and reprint requests should be ad-

dressed. The abbreviations used are: PEPCK, phosphoenolpyruvate carb-

oxykinase; T:,, 3,3’,5-~-triiodothyronine, thyroid hormone; TRE, thy- roid hormone response element; CRE, cyclic AMP response element; CREB, CRE-binding protein; C/EBP, CCAAT/enhancer binding protein; 8-Br-cAMP, 8-bromo-cyclic AMP; CAT, chloramphenicol acetyltransferase; tp , base pair(s); Hepes, 4-(2-hydroxyethyl)-l-pi- perazineethanesulfonic acid.

by several hormones including glucagon (via CAMP), gluco- corticoids, and thyroid hormone (Lamers et al., 1982; Loose et al., 1985), whereas it is inhibited by insulin (Granner et al., 1983). Recent studies using transgenic mice have demon- strated that a relatively small region of the PEPCK promoter (nucleotides -460 to +73) contains the essential information to confer the appropriate pattern of developmental, tissue- specific, and dietary regulation characteristic of the PEPCK gene (McGrane et al., 1988, 1990). Multiple cis-acting ele- ments in this promoter-regulatory region have been found by footprint analysis using proteins prepared from rat liver nuclei and purified transcription factors (Roesler et al., 1989; Park et al., 1990). &Acting elements which account for the CAMP, glucocorticoid, and insulin responses have already been de- scribed in this region (Short et al., 1986; Quinn et al., 1988; Bokar et al., 1988; Imai et al., 1990; O’Brien et al., 1990). Furthermore, the recent characterization of the transcrip- tional induction of the PEPCK gene by cAMP has indicated that two cis-acting elements in the PEPCK promoter, CRE- 1 and P3(I), are required (Liu et al., 1991). The CRE-1 and the P3(I) elements both bind C/EBP (CCAAT/enhancer binding protein) while CRE-1 also binds CREB (CAMP- responsive element binding protein) (Park et al., 1990).

Previous studies in this and other laboratories have shown that T3 stimulates hepatic PEPCK gene expression. It has been demonstrated that T3 increases the rate of transcription of the PEPCK gene in vivo (Loose et al., 1985) and that the time course of this effect parallels the nuclear T3 receptor occupancy (Hartong et al., 1987). On the other hand, in vitro studies using hepatocytes in culture have shown that T3 modestly increases PEPCK mRNA levels and can greatly augment the effect of cAMP on the PEPCK mRNA expres- sion (Iynedjian and Salavert, 1984; Hoppner et al., 1986). More recently, it has been shown that T3 treatment of rat FA0 hepatoma cells expressing the chicken thyroid hormone receptor (chicken c-erbAa) causes a specific increase in the level of PEPCK mRNA. In that system, T3 also potentiates the cAMP induction of PEPCK mRNA expression (Muiioz et al., 1990). These data suggest a direct effect of thyroid hormone on PEPCK gene transcription as well as an interac- tion of T3 with the cAMP transcriptional stimulation of the gene.

The aim of the present study was to identify the cis- elements responsible for T3 induction of PEPCK gene expres- sion. The effects of T3 on gene transcription are mediated by nuclear receptors which are structurally related to the viral oncogene v-erbA and are members of the steroid hormone receptor superfamily of ligand-responsive transcriptional fac- tors (Sap et al., 1986; Weinberger et al., 1986; Evans, 1988).

21991

21992 Thyroid Hormone Regulates PEPCK Gene Transcription

Two major groups of T3 receptors, designated as a and P forms, have been described, although the functional signifi- cance of these different forms is not yet understood (Hodin et al., 1989). However, the two forms display distinct tissue- specific expression, and the P-T3 receptor type has been shown to be the predominant form expressed in rat liver (Murray et al., 1988). The specific DNA sequences that mediate transcrip- tional activation or inhibition in response to T3 are known as thyroid hormone response elements (TREs). TRE sequences bind T3 receptors with high affinity (Glass et al., 1987, 1988) and have been mainly studied in pituitary-specific genes, such as the genes for rat growth hormone (Ye et al., 1988; Glass et al., 1987; Wight et al., 1988; Brent et al., 1989a and 1989b; Norman et al., 1989), prolactin (Forman et al., 1988; Day and Mauer, 1989; Stanley, 1989) and a- and P-thyrotropin (Chat- terjee et al., 1989; Carr et al., 1989). Recently, TREs have been found in the 5'-flanking region of two T3-responsive genes expressed in liver malic enzyme (Petty et al., 1990), and the S14 gene (Zilz et al., 1990).

In this paper, we report the presence of a unique TRE in the PEPCK promoter which acts together with P3(I), another cis-acting element in the promoter, to confer T3 regulation to the gene. Finally, we show that T3 and cAMP act synergisti- cally in stimulating transcription of the PEPCK gene and that this effect depends on functional cis-acting elements.

EXPERIMENTAL PROCEDURES

Materials-L-Triiodothyronine was purchased from Sigma. DNA- modifying enzymes and poly(d1). (dC) were obtained from Boehringer Mannheim. ['251]T:l NEX-11OX (2,200 Ci/mmol), [-y-"P]ATP (6,000 Ci/mmol), and [3H]chloramphenicol were purchased from Du Pont- New England Nuclear.

Plasmids and Probes-Construction of the 5"deletion mutants of the PEPCK promoter ligated to the chloramphenicol acetyltransfer- ase (CAT) gene has been previously described (Short et al., 1986; Park et al., 1990). Introduction of specific block mutations into the PEPCK promoter was carried out as described in Liu et al. (1990). The CAT expression vector (DelP4)PEPCK-CAT contains an inter- nal deletion between nucleotides -330 and -272 in the PEPCK promot,er. RSV-hT3RP (kindly provided by M. G. Rosenfeld) is an expression vector that contains the human-erbA p form of the T S

receptor driven by the Rous sarcoma virus promoter (Glass et al., 1989). Oligonucleotides were chemically synthesized using an Applied Biosystems 380A DNA Synthesizer. The optimal-TRE oligonucleo- tide contains the palindromic TRE defined by Glass et al. (1988) flanked by XbaI-compatible ends. The (-332/-308)PEPCK is a 25- bp oligonucleotide which corresponds to positions -332 to -308 of the PEPCK promoter. The P3(I) oligonucleotide contains the nucle- otides -251 to -234 of the PEPCK promoter, and the CRE-1 oligo- nucleotide corresponds to nucleotides -94 to -77 of the PEPCK promoter. The SV1-CAT vector which contains the enhancerless SV40 promoter driving the CAT gene has been described previously (Bokar et al., 1988). Two copies of the (-332/308)PEPCK-TRE were ligated into the XbaI site which is 5' to the SV1 promoter.

Cell Culture and Transfection Assays"HepG2 human hepatoblas- toma cells were grown in a 1:l mixture of Ham's F-12:Dulbecco's modified Eagle's medium supplemented with 5% calf serum and 5% fetal calf serum. DNA was transfected by calcium phosphate precip- itation (Park et al., 1990), and to each plate were added 5 pg of CAT vector, 5 pg of RSV-hT3R6, and 2.5 pg of RSV-Pgal. The latter was included as a control for the efficiency of separate transfections. In order to ensure equivalent transfection efficiency, cells (80-90% confluent) were treated with trypsin, pooled, and added to the precip- itated DNA. The mixture was plated out and left for 4-5 h. The cells were then washed twice, and the medium was changed to F- 12:Dulbecco's modified Eagle's medium containing charcoal-treated 10% fetal calf serum (Honvitz and McGuire, 1978 Brent et al., 1988a, 1988b). The cells were incubated for 36-38 h with or without the addition of 100 nM T,. When the cAMP treatment was performed, 0.5 mM 8-Br-CAMP plus 1 mM theophylline were added 16-18 h before harvesting the cells. Analysis of CAT activity in freeze-thaw lysates of the cells was performed as described by Ausubel et al. (1989), using equal amounts of cellular protein for each sample. ['HI

Chloramphenicol and butyryl-SCoA were added to each extract. The butyrylated chloramphenicol was extracted by xylenes and quantified by scintillation counting. The CAT activity of different PEPCK-CAT vectors (separate transfection experiments) was corrected for trans- fection efficiency using the ,&galactosidase activity as a standard.

DNA Binding Experiments-Nuclear extracts from rat liver were prepared by the method of Gorsky et al. (1986). For the gel retardation assays, oligonucleotides were labeled using [-y-32P]ATP and T4 poly- nucleotide kinase. The DNA probe (20,000-30,000 cpm) was incu- bated for 15-20 min at 22 "C with 5 pg of protein from the nuclear extract and 3 pg of poly(d1). (dC) in a final volume of 15 pl containing 20 mM Hepes (pH 7.6), 0.1 mM EDTA, 1 mM dithiothreitol, 10% glycerol, 50 mM NaCl, and 20 nM T3. Reactions were loaded onto 5% polyacrylamide gels in 0.5 X TBE (44.5 mM Tris, 44.5 mM borate, 1 mM EDTA). Electrophoresis was performed at 175 V for 90 min at 4 "C. Gels were analyzed by autoradiography. In the competition experiments, unlabeled oligonucleotides (40 ng/lane) were included in each respective binding reaction.

The receptor-DNA complex was also demonstrated by labeling the T, receptor with ['251]T,. Binding reactions were performed by incu- bating aliquots of rat liver nuclear extract (30 pg/lane) with 2 nM [1251]T3 for 60 min at 4 'C. DNA-binding reactions were subsequently performed using the same conditions described above but with 10 ng of unlabeled oligonucleotide per lane and without adding unlabeled T3 in the binding buffer. Electrophoresis was carried out for 90 min at 4 "C using 5% polyacrylamide gels in 6.7 mM Tris, 1 mM EDTA, and 3.3 mM sodium acetate (Lavin et al., 1988).

The DNase I footprinting assay was performed as previously de- scribed (Roesler et al., 1989). The DNA probe was prepared by end- labeling one strand of DNA using [-y-32P]ATP and T4 polynucleotide kinase and was incubated with 30 pg of rat liver nuclear extract. The oligonucleotide competition was performed by adding 10 or 40 ng of each unlabeled oligonucleotide to the binding reaction.

RESULTS

Identification of a Thyroid Hormone Response Element in the PEPCK Promoter-To determine the cis-acting sequences of the PEPCK promoter responsible for conferring T3 tran- scriptional regulation to the gene, a series of 5'-deletions of the PEPCK promoter (-490/+73), linked to the CAT struc- tural gene, were tested by transient transfection into HepG2 cells. HepG2 cells are a human hepatoblastoma cell line that has been used to study hepatic functions including the effect of thyroid hormone on liver metabolic processes (Javitt, 1990). However, we noted only a marginal T3-dependent increase in transcription from the PEPCK promoter in cells transfected with the (-490)PEPCI(-CAT vector (data not shown). We then co-transfected the PEPCK-CAT vectors with a mam- malian expression vector for the human P-T3 receptor (RSV- hT3RP) which has been used in several cell lines to identify T3 response elements (Glass et al. 1989; Chatterjee et al., 1989; Thompson and Evans, 1989). As is shown in Fig. l.4, co- transfection of (-490)PEPCK-CAT with RSV-hT3RP re- sulted in a 3-fold induction of CAT activity in the presence of 100 nM T3 when compared with cells incubated in the absence of the hormone. Dose-response experiments indicated that maximal T3 stimulation of the CAT activity occurred at 100 nM T3, and all subsequent experiments were performed at this concentration. Concentrations of T3 of 0.1 nM, 1 nM, and 10 nM were also tried yielding inductions of 45%, 70%, and 90% of that achieved with 100 nM T3. A similar T3 responsiveness was observed with the (-355)PEPCK-CAT chimeric gene (Fig. LA). However, deletion of 78 bp from -355 to -277 resulted in a complete loss of the T3 induction of transcription from the PEPCK promoter. Similar results were obtained using progressive downstream deletions. To further assess these data, we constructed a PEPCK-CAT vector that contained an internal deletion between positions -330 and -272 in the PEPCK promoter. As is shown in Fig. LA, the (DelP4)PEPCK-CAT vector showed no significant T3 acti- vation. These observations are consistent with the localization

Thyroid Hormone Regulates PEPCK Gene Transcription 21993 A.

Fold lnductlon

E.

CAT 2.7 f 0.3

CAT 0.6 rt 0.2

CAT 0.8 f 0.2

CAT 1.0 rt 0.3

- 6 8 CAT 0.8 rt 0.2

del P4 CAT 0.7 0.1 - 5 3 0 1 - 2 7 2

8 V l

FIG. 1. Transcriptional induction by thyroid hormone of PEPCK-CAT constructs. A, PEPCK-CAT expression vectors, con- taining several deletions of the PEPCK promoter, were co-transfected with the T3 receptor expression vector RSV-hT3RP into HepG2 cells. DNA transfection and CAT assays were performed as described under “Experimental Procedures.” Results are presented as -fold induction in the presence or absence of 100 nM T, and are the average at least three experiments, each done in duplicate. The basal activity of each CAT vector did not vary more than 20% in the absence of T3. No difference was observed in the activity of vectors transfected with or without the T, receptor expression vector in the absence of T,. There

defined in the PEPCK promoter (Roesler et al., 1989). B, a SV1-CAT is also a schematic diagram of the protein binding domains previously

expression vector containing two copies of the (-332/-308)PEPCK- TRE was transfected into HepG2 cells in the presence or absence of 100 nM T,. The experiments were performed three times in duplicate.

of a TRE between positions -330 and -277 in the PEPCK promoter.

To test the ability of the (-332/-308) sequence from the PEPCK promoter to function as an independent thyroid response element, we ligated two copies of this sequence into a vector containing the enhancerless SV40 promoter driving the CAT gene. The (-332/-308) sequence was subsequently identified as the T3 receptor binding site (Figs. 3 and 5). When the parent SV1-CAT and (PEPCK-TRE)SVl-CAT genes were transfected into HepG2 cells, transcription from the PEPCK-TRE gene was stimulated 2.6-fold by 100 nM T3, while the SV1-CAT gene was unresponsive (Fig. 1B). These results indicate that the isolated PEPCK-TRE can respond

Synergistic Induction of PEPCK Gene Transcription by Thyroid Hormone and CAMP-In order to analyze whether the addition of T3 had any effect on the induction of PEPCK transcription by CAMP, we treated co-transfected cells either with T3, 8-Br-CAMP, or both effectors together. Addition of 8-Br-CAMP caused a 4-fold induction in the CAT activity of the (-490)PEPCK-CAT vector (Fig. 2), which is consistent with the range of induction we have previously observed (Liu et al., 1991). Incubation of the cells in the presence of both T3 and 8-Br-CAMP caused a 19-fold induction, suggesting a synergistic interaction between the effects of TI and CAMP. To determine if the observed synergism was dependent upon cis-elements within the PEPCK promoter, various chimeric genes containing the PEPCK promoter with specific muta- tions linked to the structural gene for CAT were tested (Fig. 2). First, using (DelP4)PEPCK-CAT, a T3 unresponsive vec-

to T3.

2 1 -

2 0 -

1 9 - 1 8 - 1 7 -

-

0 u . 3

2

1

0 Controt

T3

B CAMP

T3 +CAMP

T

-490 PEPCK detP4 PEPCK P3(1) PEPCK CRE1 PEPCK

FIG. 2. Synergistic enhancement by thyroid hormone and cAMP of PEPCK gene transcription. Transfection protocol and PEPCK-CAT chimeric genes used in this experiment are described under “Experimental Procedures.” The cells were not treated (con- trol) or treated by the addition of 100 nM T3 (T3), 0.5 mM 8-Br- cAMP plus 1 mM theophylline (CAMP), or a combination of both agents (7’3 + CAMP). Results are presented as -fold induction relative to the control value and are expressed relative to the control value of the (-49O)PEPCK-CAT, corrected for its transfection efficiency. Values are the means & S.E.M. for three different transfection experiments, each done in duplicate.

tor, we observed a 4-fold induction by 8-Br-CAMP in either the presence or absence of TS. We also analyzed the response of two PEPCK-CAT vectors in which specific block mutations were introduced into the CRE-1 and P3(I) elements, which impair the cAMP responsiveness of the promoter (Liu et al., 1991). The induction of transcription from the PEPCK pro- moter by 8-Br-CAMP was about 2-fold and 1.5-fold with the P3(I) and CRE-1 block mutation, respectively, in agreement with the previous observations. When we analyzed the T3- cAMP interaction using the PEPCK promoter with a muta- tion at CRE-1, the effect was additive. Surprisingly, the PEPCK promoter with a mutation at P3(I) not only exhibited an impaired transcriptional response to cAMP induction but also was not induced by T3. These results indicate that the synergistic effect of T3 and cAMP on inducing transcription from the PEPCK promoter is mediated by a synergistic inte- raction of the T3 and cAMP cis-acting elements. In addition, the lack of TS inducibility of the P3(I) block mutation suggests that two cis-sequence elements are required for the T3 tran- scriptional regulation of the PEPCK gene.

Identification of Thyroid Hormone Receptor Binding Sites- We next used gel mobility shift assays with nuclear extracts from rat liver to determine whether the T3 receptor bound to one or both of the sequences required for conferring TB responsiveness to the PEPCK promoter. A 25-bp oligonucle- otide corresponding to positions -332 to -308 of the PEPCK promoter was labeled with 32P and incubated with rat liver nuclear extracts. A single DNA-protein complex was formed (Fig. 3A). To determine whether this binding was sequence- specific and corresponded to the binding of T3 receptor pres-

21994

A. - Compelltor

Thyroid Hormone Regulates PEPCK Gene Transcription

. .

FIG. 3. Binding of TS receptor to the PEPCK promoter. A, :"P-labeled (-332/-308)PEPCK oligonucleotide was incubated with nuclear extracts from rat liver. Where indicated, unlabeled competitor DNA (40 ng/lane) was included in the binding reaction. B, binding of [1Z511]Tn-labeled TR receptors present in rat liver nuclear extracts to the indicated unlabeled oligonucleotides. The 32P-labeled (-332/ -308)PEPCK probe was incubated with rat liver nuclear extracts and run in parallel. Gel shift assays were performed as described under "Experimental Procedures."

ent in the nuclear extract, a TRE oligonucleotide containing the well defined optimal palindromic T3 receptor binding sequence (Glass et al., 1988) was used as a competitor. A 100- fold molar excess of unlabeled optimal TRE competed specif- ically, whereas no competition was detected using a 100-fold molar excess of either the CRE-1 or the P3(I) DNA probes.

To directly identify the protein within the retarded com- plex, Ts receptors in nuclear extracts from rat liver were labeled with [ 9 ] T 3 and then incubated with unlabeled DNA probes. A single ['251]T3-labeled complex was identified using either the optimal TRE of the (-332/-308)PEPCK-TRE sequences (Fig. 3B). The ['251]T3-labeled band co-migrated with the R2P-labeled band monitored in parallel on the same gel. Competition of nonradioactive T3 (100-fold molar excess) demonstrated the specificity of the [12'I]T3 binding to the receptor. [ '2sI]T3-labeled receptors did not enter efficiently into the gel in the absence of DNA (no DNA lane). As expected, no ['2sI]T3-labeled band was detected with the non- specific sequences CRE-1 or P3(I).

Finally, we assessed the location of T3 receptor binding sites in the intact PEPCK promoter by oligonucleotide com- petition DNase I footprinting. The optimal TRE and the (-332/-308)PEPCK-TRE oligonucleotides competed for binding at a single region of the PEPCK promoter which corresponded to the defined -3321-308 site (Fig. 4). In con- trast, the P3(I) oligonucleotide did not alter the footprinting at that site, although it competed for binding at other sites, indicating that the PEPCK promoter contains a unique site capable of binding the T3 receptor.

DISCUSSION

Thyroid hormone regulates many aspects of hepatic metab- olism including a direct effect by T3 on the synthesis of several regulatory enzymes in lipogenesis and gluconeogenesis (Good- ridge, 1987). T3 also amplifies other hormonal or nutritional signals to assist in adapting to starvation, cold, and stress (for review, see Oppenheimer and Samuels, 1983). Previous re-

P 6

".. - - FIG. 4. DNase I footprinting with oligonucleotide competi-

tion. A 564-bp XbaI-BgnI fragment of the PEPCK promoter (nucle- otides -490 to +73) was prepared from pBH1.2 by end-labeling the noncoding strand with T 4 polynucleotide kinase and [(J""'P]ATP. The probe was incubated with 30 pg of rat liver nuclear extract. The amount of the indicated competitor oligonucleotides added is marked in the figure. Detailed description of the regions protected (showed by boxes in the figure) are provided in Roesler et al., 1989.

ports indicated that T3 stimulates PEPCK gene transcription in rat liver (Loose et al., 1985; Hoppner et al., 1986; Hartong et al., 1987). In this study, we have demonstrated a T3 induc- tion of the PEPCK gene transcription that requires both T3 and its nuclear receptor and have identified a TRE in the PEPCK promoter at -332 to -308. Moreover, a direct inte- raction between T3 receptors present in nuclear extracts of rat liver and the (-332/-308)PEPCK oligonucleotide was found by gel shift assay. The demonstration that the DNA- protein complex formed with both the optimal and the (-332/ -308)PEPCK TRE oligonucleotides was specifically labeled with [12'I]T3 provides further evidence that this sequence binds the T3 receptor.

Several studies involving mutations in the promoter for the rat growth hormone gene defined a TRE consensus sequence, which contained either direct or inverted repeats of the "half- site" sequence 5' AGGT(C/A)A 3' (Glass et al., 1988; Brent et al., 1989a, 1989b). The sequence of the T3 receptor-binding site in the PEPCK promoter (antisense strand) contains an imperfect direct repeat of the TRE consensus sequence (see Fig. 5A) . Furthermore, extensive sequence homology was ob- served when comparing TRE sequences identified in other promoters with that in the PEPCK promoter (Fig. 5B).

Thyroid Hormone Regulates PEPCK Gene Transcription 21995

A.

-316 TGGGGGTCAAGGACAGG -332 rPEPCK-TRE antisense strand ""->""-> .*....*. *. AGGTCAAGGTCA consensus sequence

(d i rect repeats)

B .

rPEPCK-TRE TGGGGGTCAAGGACAGG (-316/-332) antisense strand

rS14-TREl TTGGGGCCTGGCAGC (-2697/-2683) raMHC-TRE TGGAGGTOACAGGAGGA (-131/-147) antisense strand

rME-TRE TTGGGGTTAGGGGAGGA (-281/-265)

rGH-TRE TCAGGOACGTOACCGCA (-181/-165) rPRL-TRE TTGGGGTCAOAAGAGGC (-1556/-1542)

FIG. 5. Comparison of the nucleotide sequence of various TB receptor binding sites with that of the PEPCK promoter. A, the antisense strand sequence between -316 and -332 of the PEPCK- TRE is compared to the consensus TRE sequence (Brent et al., 1989a, 1989b). Homologous nucleotides are indicated by asterisks. B, the sequences of various native TREs are aligned uersus the PEPCK- TRE. The TREs are from the promoters of the malic enzyme ( M E ) (Petty et al., 1990), Spot 14 (Zilz et al., 1990), a-myosin heavy chain ((uMHC) (Izumo and Mahdavi, 1988), growth hormone ( G H ) (Normal et al., 1989), and prolactin (Day and Maurer, 1989) genes from rat. Numbers inparentheses indicate the position of the sequences in each promoter.

The location of the TRE in the PEPCK promoter does not coincide with any of the previously identified cis-acting ele- ments described in the promoter. As we have already men- tioned, CAMP-responsive sites at CRE-1 and P3(I) are located 3' to the TRE, while the glucocorticoid response unit defined by Imai et al. (1990) is 5'. However, the PEPCK-TRE se- quence overlaps the 5' end of the protein binding domain of P4 (-330 to -270) which was defined by footprinting analysis using nuclear extracts prepared from rat liver (Roesler et al., 1989) (see schematic diagram of the protein binding domains in the PEPCK promoter in Fig. 1). The P4 site contains two low affinity C/EBP binding sites, P4(I) and P4(II) (Park et al., 1990). Both previous studies indicated the presence of another binding site at the 5' end of P4 which displayed different binding properties to the defined P4(I) and P4(II) regions. Our results show that this site P4(III) is the T3 receptor binding site in the PEPCK promoter. Furthermore, the oligonucleotide competition DNase I footprinting (Fig. 4) demonstrates that there is a unique TRE site in the promoter and also that the P3(I) oligonucleotide, a well defined C/EBP binding site, is unable to compete for the binding to the TRE site, whereas it does compete for the binding of proteins to the P4(I) and P4(II) domains.

The PEPCK-TRE can act as an independent thyroid re- sponse element when it is ligated to a neutral promoter. However, within the context of the PEPCK promoter, a second element P3(I) is also required for the T3 induction even though this element does not bind the thyroid receptor. The requirement of more than a TRE to confer transcrip- tional responsiveness to T3 has been previously described only with promoters from pituitary specific genes. Ye et al. (1988) reported that both the TRE and two cell-specific basal ele- ments are needed for T3 responsiveness of the rat growth hormone promoter. Likewise, the involvement of cell-specific basal elements has also been suggested for the prolactin promoter (Day and Mauer, 1989). The results of this study indicate that the PEPCK promoter also requires cis-elements

which interact with the TRE to induce gene transcription. Thus, the requirement of cis-elements other than the TREs may be a widespread rather than a pituitary-specific charact- eristic of the molecular mechanisms of T3-regulated transcrip- tion. A mechanism in which T3 acting via its receptor enhances the affinity and/or binding of trans-acting factors to other cis-regulatory sequences is possible. Support for this suggestion comes from the finding that the levels and/or affinities of liver nuclear proteins bound to the two cis- sequence elements of the malic enzyme promoter are influ- enced by the thyroid status of the rat (Petty et al., 1989).

Several previous studies have indicated a potentiating effect of T3 on the cAMP induction of PEPCK gene expression. This effect has been described both in vivo (Loose et al., 1985) and in hepatocytes and hepatoma cells (Hoppner et al., 1986; Muiioz et al., 1990). In the present study, we have demon- strated that T3 and cAMP act synergistically to induce PEPCK gene transcription. To our knowledge, cooperative interactions between T3 and cAMP have only been described for transcription from the promoter of the rat growth hormone gene (Copp and Samuels, 1989). We have further demon- strated that this effect requires functional cis-acting elements for both the T3 and cAMP responses, which excludes the possibility that the stimulation of transcription from the T3

promoter was mediated by alterations in the pathway of cAMP by T3 or vice versa. We suggest that the synergistic effect of T3 and cAMP is due to interactions between tran- scription factors bound to the promoter. The fact that the P3(I) site is involved in both T3 and cAMP responsiveness of the PEPCK promoter points to a key role for the trans-acting factor(s) bound to this site. Footprinting analysis has shown that the P3(I) element binds proteins enriched in liver nuclei when compared to kidney, spleen, and brain (Roesler et al., 1989) and that it can bind C/EBP and may also bind other members of the C/EBP family (Park et al., 1990). Further- more, recent experiments with transgenic mice that contain the bovine growth hormone gene linked to the PEPCK pro- moter with a block mutation in the P3(I) element have suggested an important role for this element in enhancing liver-specific expression of the PEPCK promoter.' Thus, P3(I) plays an essential role in conferring both tissue speci- ficity and hormone inducibility to the PEPCK promoter. Further investigations will be required to analyze possible interactions between the T3 receptor and members of the C/ EBP family. Likewise, the synergistic cooperation between T3 and cAMP makes the PEPCK gene an interesting model for analyzing how several hormones converge to regulate the expression of a single gene.

Acknowledgments-We thank M. G. Rosenfeld for the generous gift of RSV-hT3Rb. We also thank Teiko Kimura for her technical assistence.

REFERENCES

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Siedman, J. G., Smith, J. A., and Struhl, K. (1989) Current Protocols in Molecular Biology, Vol. 1, Wiley Interscience, New York

Baker, J. A., Roesler, W. J., Vanderbark, G. R., Kaetzel, D. M., Hanson, R. W., and Nilson, J. H. (1988) J. Biol. Biol. 263, 19740- 19747

Brent, G. A., Harney, J. W., Moore, D. D., and Larsen, P. R. (1988) Mol. Endocrinol. 2, 792-798

Brent, G. A., Larsen, P. R., Harney, J. W., Koenig, R. J., and Moore, D. D. (1989a) J. Biol. Chem. 264, 178-182

Brent, G. A., Harney, J. W., Chen, Y., Warne, R. L., Moore, D. D., and Larsen, P. R. (1989b) Mol. Endocrinol. 3, 1996-2004

Y. Pate1 and R. W. Hanson, unpublished results.

21996 Thyroid Hormone Regulates PEPCK Gene Transcription Carr, F. E., Burnside, J., and Chin, W. W. (1989) Mol. Endocrinol. 3,

Chatterjee, V. K. K., Lee, J., Rentoumis, A., and Jameson, J. L. (1989)

Copp, R. P., and Samuels, H. H. (1989) Mol. Endocrinol. 3 , 790-796 Day, R. N., and Maurer, R. A. (1989) Mol. Endocrinol. 3 , 931-938 Evans, R. M. (1988) Science 240,889-895 Forman, B. M., Yang, C. R., Stanley, F., Casanova, J., and Samuels,

Garcia-Ruiz, J. P., Lobato, M. F., Ros, M., and Moreno, F. J. (1983)

Glass, C. K., Franco, R., Weinberger, C., Albert, V. R., Evans, R. M.,

Glass, C. K., Holloway, J. M., Devary, 0. V., and Rosenfeld, M. G.

Glass, C. K., Lipkin, S. M., Devary, 0. V., and Rosenfeld, M. G.

Goodridge, A. G. (1987) Annu. Reu. Nutr. 7, 157-185 Gorman, C. M., Moffat, L. F., and Howard, B. H. (1982) Mol. Cell.

Gorski, K., Carneiro, M., and Schibler, U. (1986) Cell 4 7 , 767-776 Granner, D. K., Andreone, T, Sasaki, K., and Beale, E. (1983) Nature

305,549-551 Hanson, R. W., and Garber, A. J. (1972) Am. J . Clin. Nutr. 25,1010-

1021 Hartong, R., Wiersinga, W. M., and Lamers, W. H. (1987) Endocri-

nology 120,2460-2467 Hodin, R. A., Lazar, M. A., Wintman, B. I., Darling, D. S., Koening,

R. J., Larsen, P. R., Moore, D. D., and Chin, W. W. (1989) Science

Hoppner, W., Sussmuth, W., O’Brien, C., and Seitz, H. J. (1986) Eur.

Horwitz, K. B., and McGuire, W. L. (1978) J. Bwl. Chem. 253,8185-

Imai, E., Stromstedt, P., Quinn, P. G., Carlstedt-Duke, J., Gustafsson,

Iynedjian, P. B., and Salavert, A. (1984) Eur. J . Biochem. 145,489-

Izumo, S., and Mahdavi, V. (1988) Nature 334, 539-542 Javitt, N. B. (1990) FASEB J. 4,161-168 Lamers, W. H., Hanson, R. W., and Meisner, H. M. (1982) Proc.

Natl. Acad. Sci. U. S. A. 79,5137-5141 Lavin, T. N., Baxter, J. D., and Horita, S. (1988) J. Biol. Chem. 2 6 3 ,

Liu, J., Roesler, W. J., and Hanson, R. W. (1990) BioTechniques 9 ,

709-716

Proc. Natl. Acad. Sci. U. S. A. 8 6 , 9114-9118

H. H. (1988) Mol. Endocrinol. 2, 902-911

Enzyme 30,265-268

and Rosenfeld, M. G. (1987) Nature 3 2 9 , 738-741

(1988) Cell 5 4 , 313-323

(1989) Cell 59,697-708

Biol. 2 , 1044-1051

244.76-79

J. Biochem. 159,399-405

8191

J., and Granner, D. K. (1990) Mol. Cell. Biol. 10,4712-4719

497

9418-9426

738-742

Liu, J., Park, E. A., Gurney, A. L., Roesler, W. J., and Hanson, R. W. (1991) J. Biol. Chem. in press

Loose, D. S., Cameron, D. K., Short, H. P., and Hanson, R. W. (1985) Biochemistry 24,4509-4512

McGrane, M. M., devente, J., Yun, J., Bloom, J., Park, E. A., Wynshaw-Boris, A., Wagner, T., Rottman, F. M., and Hanson, R. W. (1988) J. Biol. Chem. 263,11443-11451

McGrane, M. M., Yun, J., Moorman, A. F. M., Lamers, W. H., Hendrich, G. K., Arafa, B. M., Park, E. A., Wagner, T., and Hanson, R. W. (1990) J. Biol. Chem. 2 6 5 , 22371-22379

Mufioz, A., Hoppner, W., Sap., J., Brady, G., Nordstrom, K., Seitz, H. J., and Vennstrom, B. (1990) Mol. Endocrinol. 4 , 312-320

Murray, M. B., Zilz, N. D., McCreary, N. L., MacDonald, M. J., and Towle, H. C. (1988) J. Biol. Chem. 263,12770-12777

Norman, M. F., Lavin, T. N., Baxter, J. D., and West, B. L. (1989) J. Bwl. Chem. 264,12063-12073

O’Brien, R. M., Lucas, P. C., Forest, C. D., Magnuson, M. A., and Granner, D. K. (1990) Science 249,533-537

Oppenheimer, J. H., and Samuels, H. H. (1983) Molecular Basis of Thyroid Hormone Action, Academic Press, New York

Park, E. A., Roesler, W. J., Liu, J., Klemm, D. J., Gurney, A. L., Thatcher, J. D., Shuman, J., Friedman, A., and Hanson, R. W. (1990) Mol. Cell. Biol. 10 , 6264-6272

Petty, K. J., Morioka, H., Mitsuhashi, T., and Nikodem, V. M. (1989) J. Biol. Chem. 264,11483-11490

Petty, K. J., Desvergne, B., Mitsuhashi, T., and Nikodem, V. M. (1990) J. Biol. Chem. 265,7395-7400

Quinn, P. G., Wong, T. W., Magnuson, M. A., Shabb, J. B., and Granner, D. K. (1988) Mol. Cell. Biol. 8, 3467-3475

Roesler, W. J., Vanderbark, G. R., and Hanson, R. W. (1989) J. Bwl. Chem. 264,9657-9664

Sap, J., Muiioz, A., Damm, K., Goldberg, Y., Ghysdael, J., Leutz, A., Beug, H., and Vennstrom, B. (1986) Nature 324,635-640

Short, J. M., Wynshaw-Boris, A., Short, H. P., and Hanson, R. W. (1986) J. Biol. Chem. 261,9721-9726

Stanley, F. (1989) Mol. Endocrinol. 3 , 1627-1633 Thompson, C. C., and Evans, R. M. (1989) Proc. Natl. Acad. Sci. U.

Weinberger, C., Thompson, C. C., Ong, E. S., Lebo, R., Gruol, D. J.,

Wight, P. A., Crew, M. D., and Spindler, S. R. (1988) Mol. Endocrinol.

Ye, Z.-S., Forman, B. M., Aranda, A., Pascual, A., Park, H.-Y., Casanova, J., and Samuels, H. H. (1988) J. Biol. Chem. 263,7821- 7829

Zilz, N. D., Murray, M. B., and Towle, H. C. (1990) J. Bwl. Chem.

S. A. 8 6 , 3494-3498

and Evans, R. M. (1986) Nature 3 2 4 , 641-646

2,536-542

265,8136-8143