DEVELOPMENTAL
BIOLOGY
153,172-175 (1992)
Differential Regulation of the Transforming Growth Factor Type-02 Gene Promoter in Embryonal Carcinoma Cells and Their Differentiated Cells DAVID KELLY,*,? MICHAEL A. O’REILLY,$ AND ANGIE RIZZINO**~ *Eppley Institute for Research in Cancer and Allied Diseases, +Department of Path,ology and Microbiology, University of Nebraska Medical Center, 600 South &?nd Street, Omaha, Nebrasku 68198-6805;and $-Laboratory of Chemopreventim, National Cancer Institute, National Institutes qf Heulth Accepted May 1.3,1992 Previous studies have shown that EC cells do not express detectable levels of TGF-02 or its mRNA until they differentiate. This suggested that differentiation influences the transcription of the TGF-P2 gene in this model system. To address this possibility, we have examined the activity of the TGF-P2 promoter in EC cells and their differentiated cells using gene constructs containing various portions of the TGF-P2 promoter inserted upstream of the reporter gene, chloramphenicol acetyltransferase (CAT). We determined that the level of CAT increases approximately ninefold when EC cells were induced to differentiate. Our studies also indicate that the TGF-02 promoter contains at least two positive regulatory elements that are separated by a negative regulatory element. Finally, we have identified a CRE/ATF-like site that appears to be responsible for a positive regulatory element located between -77 and -40. a 1992 Academic Press. Inc.
EC cells and their differentiated cells, but TGF-02 (Kelly et ah, 1990; Mummery et al, 1990) and TGF-P receptors (Rizzino, 1987) do not appear until EC cells are induced to differentiate. Interestingly, the increases in TGF-02 which accompanies differentiation appear to be the result of a large increase in the steady-state levels of TGF-02 mRNA rather than to an enhancement of its release (Kelly et al., 1990; Mummery et ah, 1990). To gain a better understanding of the mechanisms that regulate expression of the TGF-P2 gene during differentiation, we prepared TGF-P2 promoter/reporter gene constructs and examined their expression in two mouse EC cell lines and their differentiated counterparts. Our findings suggest strongly that the increase in the steady-state levels of TGF-P2 mRNA that accompanies differentiation is due, at least in part, to increased transcription of the TGF-/32 promoter.
INTRODUCTION
Transforming growth factor type-/3 (TGF-P)’ refers to a complex family of at least five genetically distinct polypeptides that exert potent effects on growth and differentiation (reviewed in Rizzino, 1988; Roberts and Sporn, 1990). Although TGF$l, TGF-02, and TGF-P3 appear to have similar effects in several in vitro systems and appear to be functionally equivalent, several studies have shown that the three TGF-P species can exert quantitatively or qualitatively different effects (Ohta et al., 1987; Rosa et al., 1988; Graycar et al., 1989; Roberts et al., 1990). Furthermore, the expression and production of TGF-/31, TGF-P2, and TGF-P3 appear to be regulated both spatially and temporally during the earliest stages of mouse embryogenesis (Campbell et ah, 1990; Kelly et al, 1990; Mummery et al., 1990; Slager et al., 1991), suggesting that the different forms of TGF-P have distinct roles during mammalian embryogenesis. The expression and production of the different forms of TGF-P has been examined previously in embryonal carcinoma (EC) cells, which serve as a model system for studying the expression of growth factors and their receptors during early development (reviewed in Rizzino, 1989). In this model system, TGF-@l is produced in both
MATERIALS
Copyright All rights
Q 1992 by Academic Press, Inc. of reproduction in any form reserved.
METHODS
Stock cultures of F9 EC cells and PC-13 EC cells were maintained as reported previously (Rizzino et al., 1983). Differentiation of EC cells was induced by a 72-hr treatment with 5 puMRA. The F9 EC cells were transfected by a modified calcium phosphate precipitation method designed for nonadherent cell lines (Davis et ab, 1986). The F9- and PC-13-differentiated cells were transfected in monolayer by calcium phosphate precipitation. For both transfection techniques, 20 pg of TGF-82 promoter-CAT plasmid DNA was cotransfected with 2 pg of the P-galactosidase expression plasmid, pCH-110 (Pharmacia).
1 To whom correspondence should be addressed. ’ Abbreviations used: transforming growth factor-p, TGF-P; embryonal carcinoma, EC; retinoic acid, RA; chloramphenicol acetyltransferase, CAT; polymerase chain reaction, PCR; fetal bovine serum, FBS. 0012-1606/92 $5.00
AND
172
KELLY, O’REILLY, AND RIZZINO
+z@ r, *3 pB2 - 770 ~02 - 520 pB2 - 347 pB2 - 187 pB2-77 -
pB2-40
FIG. 1. Structure of TGF-132 promoter-CAT gene constructs. TGF-p2 promoter fragments were generated by polymerase chain reaction using different Y-specific oligonucleotide primers between ~778 to -40 and a common 3’.specific oligonucleotide primer at +63 relative to the transcription initiation site. DNAs were amplified with Taq polymerase (Perkin-Elmer-Cetus) using cycling temperatures of annealing at 50°C for 2 min, elongating at 72°C for 4 min, and denaturing at 94°C for 1 min. The amplified DNA fragments were cloned into the promoterless chloramphenicol acetyltransferase (CAT) expression plasmid pGEM4SVOCAT, using Hind111 and Kp?lI restriction sites that were built into the oligonucleotide primers. The identity of each clone was confirmed by sequence analysis. The constructs were named pp2-n, Lvhere )i is the distance in nucleotides upstream of the transcription initiation site.
CAT activities were determined by the method of Seed and Sheen (1988) and normalized to ,&galactosidase activity by the method of Rosenthal (1987) to adjust for any differences in transfection efficiency (Hall et al., 1983). All promoter/reporter gene constructs were prepared using the polymerase chain reaction (PCR) and amplifying the 5’-flanking region and the first 63 bp of the first exon of the human TGF-/32 gene from genomic DNA (Noma et ah, 1991). Details regarding the construction of the TGF-/32 promoter/CAT constructs are provided in Fig. 1. All plasmids were purified by Qiagen tip-500 columns according to the manufacturer’s instructions. RESULTS
AND
DISCUSSION
In this study, we employed chimeric gene constructs in which various amounts of the human TGF-P2 promoter region were inserted upstream of the reporter gene, CAT (Fig. 1). To compare the activity of the TGF02 promoter in EC cells and their differentiated cells, the TGF-D2 promoter-CAT constructs, pb2-778, ~02-7’7, and ~02-40, were transfected into F9 EC cells. Following transfection, half of the cultures were treated with RA to induce differentiation of the F9 EC cells. In F9 EC cells, only low levels of CAT activity were detected for each construct (Fig. 2A). In contrast, CAT activity was substantially greater in the F9 EC cells treated with RA (Fig. 2A). The average increases in CAT activity observed for the constructs p/?2-778, p@2-77, and ~02-40 in three experiments were 9.4-, 4.8-, and 3%fold, respectively. A similar pattern of differentiation-dependent
TGF-/32 Gene Expressinn in EC Cells
173
TGF-P2 expression was observed when PC-13 EC cells were transfected with p/32-778 (data not shown). Thus, up-regulation of TGF-82 promoter activity when F9 EC and PC-13 EC cells are induced to differentiate suggests that the increase in steady-state levels of TGF-/32 mRNA is due, at least in part, to increased transcription of the TGF-P2 gene. To understand further the mechanism by which TGF1132 promoter expression is up-regulated during differentiation, we attempted to identify the DNA regulatory elements that are required for normal TGF-P2 promoter expression in the differentiated cells. Therefore, we transfected the various TGF-P2 promoter-CAT gene constructs into the differentiated cells and determined CAT activity. F9- and PC-13-differentiated cells had very similar patterns of TGF-P2 promoter activity (Fig. 2B). The shortest construct, pp2-40, which contains the TATA box and very little of the 5’-flanking region was expressed at only slightly higher levels than the plasmid pSVO-CAT (data not shown), which lacks a eukaryotic promoter (Langner et al., 1986). In contrast, pp2-‘77, which is only slightly larger than pp2-40, was expressed at a level similar to the largest plasmid, p@2-778. ~02-77 and p@2-778 were always the most highly expressed TGF-82 promoter-CAT constructs in the F9- and PC13-differentiated cells. The high level of CAT activity expressed by p/32-77 contrasted with the low level of expression by p/32-187. The patterns of expression shown in Fig. 2B suggest, that there are at least three regions that influence the expression of the TGF-P2 promoter in the differentiated cells: (1) A positive regulatory element between -77 and -40, (2) a negative regulatory element in the region between -187 and -77, and (3) a second positive regulatory element in the region between -347 and -187. Interestingly, the TGF-02 promoter-CAT constructs have also been transfected into CCL-64 cells and the pattern of expression observed in these cells is very similar to that observed in F9- and PC-13-differentiated cells (Michael O’Reilly, unpublished results). The large difference in the level of CAT activity expressed by pp2-77 and pfl2-40 in the differentiated cells suggests the presence of a positive regulatory element in the region between -77 and -40. Sequence analysis of the TGF-P2 promoter identified a CRE/ATF-like element between -74 and -67 (Noma et al., 1991). This element contains six of the eight consensus nucleotides observed in other gene promoters regulated by the CRE/ ATF family of transcription factors (Deutsch et ul., 1988; Roesler et ub, 1988). To determine if this CRE/ ATF-like element influences the basal activity of the TGF-02 promoter, a mutant construct, pp2-77M, was transfected into the differentiated cells. p/Y2-77M was mutated at two nucleotides (-71 and -69) that have
174
DEVELOPMENTAL BIOLOGY CAT
VOLUME 153,1992
activity (qm)
~82.778
pB2al
PEP-77
Relative CAT actbity r15
30
-14
26 T
26
-17.is
24
-13 -11
p
22
-10
g
M
C
18 16
if
pa2-776
pm52a
pB2-347
InI pw-18-r
@2-n
ti
14
-6
5
-5
t
-4 -3
$
12 10 6 6
2
4
1
2
0
0
p&MO
FIG. 2. Effect of differentiation on TGF-&Z promoterCAT activity. (A) F9 EC cells were transfected with the TGF-p2 promoterCAT plasmids, p@-778, ~62-77, and ~02-40. For this study, F9 EC cells (1 X 106) were resuspended and combined with each plasmid-CaPO, precipitate before being distributed equally into two loo-mm culture dishes containing DME supplemented with 10% FBS. The following day, the transfected cells were washed and refed with fresh medium and either left untreated (hatched bar) or treated (solid bar) with RA (5 @f). CAT activity of the cell lysates was determined 48 hr later as described under Materials and Methods. (B) The basal activity of the TGF-82 promoter was examined in FS-differentiated and PC-13-differentiated cells. FS-differentiated (solid bar) and PC-13-differentiated (open bar) cells were derived from F9 EC and PC-13 EC cells (1 X lo5 per loo-mm culture dish) cultured in DME supplemented with lo’% FBS and 5 #RA. After 72 hr, the differentiated cells were refed with medium lacking RA and cotransfected in monolayer with TGF-02 promoter-CAT plasmids and the normalizing plasmid, pCH-110. CAT activity of each cell lgsate was determined 48 hr after transfection and was normalized to D-galactosidase expression, as described under Materials and Methods. The bars represent ratios of CAT activity relative to the activity of the p@2-40 construct, which was designated as 1.00. CAT activity of p/32-40 in F9- and PC-13-differentiated cells was 3978 and 714 cpm, respectively. (C)To identify a sequence necessary for TGF-02 promoter activity, F9- and PC-13-differentiated cells were cotransfected in monolayer with the normalizing plasmid, pCH-110, and the TGF-fl2 promoter-CAT plasmids, p/12-77, pfi2-77M (a mutated form of p/32-77), and p/32-40. The mutant construct, p@2-77M, was generated by PCR using a Y-oligonucleotide primer that had been modified at ~71 (C to T) and -69 (T to G) within a CREIATFlike sequence located between -74 and -67 of the TGF-m promoter. The mutant DNA fragment was amplified and cloned into pGEM4-SVOCAT as described in the legend for Fig. 1. CAT activity of each cell lysate was determined 48 hr after transfection and was normalized to fi-galactosidase expression, as described under Materials and Methods. The bars (p@2-77, hatched; pfi2-77M, solid; and p/Z-40, open) represent ratios of CAT activity relative to the activity of the ~02-40 construct, which was designated as 1.00. CAT activity of p/%40 in F9- and PC-13-differentiated cells was 4832 and 2308 cpm, respectively. All experiments shown in this figure were repeated twice with similar results.
been shown to be critical for the biological activity of CRE/ATF elements (Deutsch et al., 1988). Specifically, the sequence at -74 to -67 in ~02-77 was modified from GCACGTCA to GCATGGCA. CAT activity was reduced by approximately 65% when F9- and PC-13-differentiated cells were transfected with p/3277M instead of p/Z-77 (Fig. 2C). Thus, it appears that
the CRE/ATF-like element located at -74 to -67 exerts a significant influence on the basal activity of the TGF82 promoter in the differentiated cells. Interestingly, the addition of dbcAMP, which elevates the expression of properties characteristic of parietal extraembryonic endoderm (Strickland et al., 1980), resulted in a 3.1-fold increase in p/32-778 expression in FS-differentiated cells
KELLY, O’REILLY, AND RIZZINO
(data not shown). Analogs of CAMP have been shown to up-regulate a number of gene promoters that contain CRE-like elements (Roesler et al., 1988). These data suggest that a member(s) of the CRE/ATF family of transcription factors interacts with the TGF-j32 promoter and regulates its expression. In summary, the differential expression of the TGF$2 promoter in EC cells and their differentiated counterparts provides a useful system for studying transcription factors and in cis regulatory elements that control the expression of this gene. Our findings have identified at least three important regions in the TGF132promoter that control its expression in EC cells that have been induced to differentiate. Future studies will focus on the identification of the member(s) of the CRE/ATF family that binds to this site as well as mapping precisely the location of the negative and the upstream positive regulatory elements in the TGF-02 promoter. Solon Rhode is thanked for his advice during the early stages of this work. Keith Miller and Phillip Wilder are thanked for their comments on this manuscript. This work was supported bg a grant from the Council of Tobacco Research (2.520) and by core grants from the National Cancer Institute (Laboratory Cancer Research Center Support Grant, CA 36727) and the American Cancer Society (ACS SIG-16). David Kelly was supported by a graduate fellowship provided by the Nebraska Research Initiative in Biotechnology. REFERENCES Campbell, W. J., Kelly, D., and Rizzino, A. (1990). Expression of transforming growth factor-03 by emhryonal carcinoma cell, parietal endodcsrm-like cells and early mouse embryos. 1?! liitro Cell L&r: Biol. 26, 1181-1185. Davis, L. G., Dihner, M. D., and Battep, J. F. (1986). Calcium phosphate transfection of nonadherent and adherent cells with purified plasmids. I)/ “Basic Methods in Molecular Biology,” pp. 286-289. Elsevicr, New York. Deutsch, P. J., Hocffler, J. P., Jameson, J. L., Sin, J. C.. and Habener, J. F. (1988). Structural determinants for transcriptional activation by CAMP-responsive DNA elements. .I. Riol. Cl,em. 263, 1846618472. Graycar, J. L., Miller, D. A., Arrick, B. A., Lyons, R. M., Moses, H. L., and Dergnck, R. (1989). Human transforming growth factor+3: Recombinant expression, purification, and biological activities in comparison with transforming growth factors-01 and -$2. Mol. 6&ocrircol. 3, 1977-1986. IIall, C. V., Jacobs, P. E., Ringold, G. M., and Lee, F. (1983). Expression and regulation of Eschc~richicl coli IacZ gene fusions in mammalian cells. .J. Mol. AppI. Grwt. 2, 101-109. Kelly, I)., Campbell, W. J., Tiesman, J., and Rizzino, A. (1990). Regulation and expression of transforming growth factor type-D during earls mammalian development. (‘!/tote~knolon!/ 4, 227-242.
Langner, K.-D., Weyer, U., and Doerfler, W. (1986). Trans effect of the El region of adenoviruses on the expression of a prokargotic gene in mammalian cells: Resistance to 5’.CCGG-3’ mrthglation. I’m:. Nufl. Acad. Sci. USA 83, 1598-1602. Mummery, C. L., Slager, H., Kruijer, W., Feijen, A., Freund, E., Koornneef, I., and van den Eijnden-van Raaij, A. J. M. (1990). Expression of transforming growth factor 82 during the differentiation of murine embryonal carcinoma and embryonic stem cells. Urr?. Bid. 137, 161-170. Noma, T., Glick, A. B., Geiser, A. G., O’Reilly, M. A., Miller, J., Roberts, A. B., and Sporn, M. B. (1991). Molecular cloning and structure of the human transforming growth factor+2 gene promoter. Growth Fwtors 4, 247-255. Ohta, M., Greenberger, J. S., Anklesaria, P., Bassols, A., and Massayue, J. (1987). Two forms of transforming growth factor-ij distinguished by multipotential haematopoictic progenitor cells. ,Vut/rrcJ 329, 539-541. Rizzino, A. (1987). Appearance of high aflinitg receptors for type B transforming growth factor during differentiation of murine embryonal carcinoma cells. Co ttwr Krs. 47, 4386-4390. Rizzino, A. (1988). Transforming growth factor-/? Multiple efIects on cell differentiation and extracellular matrices. L)rc,. Biol. 130, 41 I422. Rizzino, A. (1989). Use of embryonal carcinoma cells to study growth factors during early mammalian development. 11, “Growth Factors in Mammalian Development” (I. Y. Rosenblum and S. Htgner, Eds.), pp. 113-134. CRC Press, Inc., Bora Raton, FL. Rizzino, A., Orme, L. S., and De Larco, J. E. (1983). Emhryonal carcinoma cell growth and differentiation: Production of and response to molecules Lvith transforming growth factor activity. &,rp Co// Rcs. 143, 1431.52. Roberts, A. B., Kondaiah, P., Rosa, F., Wanatabe, S., Good, P., Rochc, N. S., Rebbert, M. L., David, I. B., and Sporn, M. B. (1990). Mesoderm induction in Xwol~/r.s /trc>t.i.sdistinguishes between the various TGF@isoforms. C;r.owtll Ftrctors 3, 27?2XG. Roberts, A. B., and Sporn, M. B. (1990). The transforming growth factor-$‘s It/ “Handbook of Experimental Pharmacology” (Sporn, M. B., and Roberts, A. B.. Eds. ), Vol. 9% 1 ), pp. 419472. SpringerVerlag, Heidelberg. Roesler, VV’.J., Vandenbark, G. R., and Hanson, R. 1%‘.(1988). Cyclic AMP and the induction of eukaryotic gene transcription. ,I Bid. Chc~ttc.263, 9063-9066 Rosa, F., Roberts, A. B., Danielpour, D., Dart, L. L., Sporn, M. B., and David, I. 8. (1988). Mesoderm induction in amphibians: The role of TGF-da-like factors. Sckrtw 239, 783-785. Rosenthal, N. (1987). Identification of regulatory elements of cloned genes with functional assays. I,, “Methods in Enzymolog,ry: Guide to Molecular Cloning Techniques” (S. L. Berger and A. R. Kimmcl, Eds.), Vol. 152, pp. 701-720. Academic Press, San Diego, CA. Seed, B., and Sheen, J. Y. (1988). A simple phase-extraction assay for chloramphenicol acyltransferase activity. &r/e 67, 271-277. Slager, II. G., Las-son, K. A.. van den Eijnden-van Raaij, A. J. M., delaat, S. W.. and Mummery, C. L. (1991). Differential localization of TGF-@2 in mouse preimplantation and earl;\ postimplantation development. L)et,. Biol. 145, 205-218. Strickland, S., Smith, K. K., and Marotti, K. R. (1980). Hormonal induction of differentiation in trratocarcinoma stem cells: Gencration of parietal cndoderm hv rctinoic acid and dihutgr~l CAMP. (‘(al/ 21, 347-355.
BIOLOGY
153,172-175 (1992)
Differential Regulation of the Transforming Growth Factor Type-02 Gene Promoter in Embryonal Carcinoma Cells and Their Differentiated Cells DAVID KELLY,*,? MICHAEL A. O’REILLY,$ AND ANGIE RIZZINO**~ *Eppley Institute for Research in Cancer and Allied Diseases, +Department of Path,ology and Microbiology, University of Nebraska Medical Center, 600 South &?nd Street, Omaha, Nebrasku 68198-6805;and $-Laboratory of Chemopreventim, National Cancer Institute, National Institutes qf Heulth Accepted May 1.3,1992 Previous studies have shown that EC cells do not express detectable levels of TGF-02 or its mRNA until they differentiate. This suggested that differentiation influences the transcription of the TGF-P2 gene in this model system. To address this possibility, we have examined the activity of the TGF-P2 promoter in EC cells and their differentiated cells using gene constructs containing various portions of the TGF-P2 promoter inserted upstream of the reporter gene, chloramphenicol acetyltransferase (CAT). We determined that the level of CAT increases approximately ninefold when EC cells were induced to differentiate. Our studies also indicate that the TGF-02 promoter contains at least two positive regulatory elements that are separated by a negative regulatory element. Finally, we have identified a CRE/ATF-like site that appears to be responsible for a positive regulatory element located between -77 and -40. a 1992 Academic Press. Inc.
EC cells and their differentiated cells, but TGF-02 (Kelly et ah, 1990; Mummery et al, 1990) and TGF-P receptors (Rizzino, 1987) do not appear until EC cells are induced to differentiate. Interestingly, the increases in TGF-02 which accompanies differentiation appear to be the result of a large increase in the steady-state levels of TGF-02 mRNA rather than to an enhancement of its release (Kelly et al., 1990; Mummery et ah, 1990). To gain a better understanding of the mechanisms that regulate expression of the TGF-P2 gene during differentiation, we prepared TGF-P2 promoter/reporter gene constructs and examined their expression in two mouse EC cell lines and their differentiated counterparts. Our findings suggest strongly that the increase in the steady-state levels of TGF-P2 mRNA that accompanies differentiation is due, at least in part, to increased transcription of the TGF-/32 promoter.
INTRODUCTION
Transforming growth factor type-/3 (TGF-P)’ refers to a complex family of at least five genetically distinct polypeptides that exert potent effects on growth and differentiation (reviewed in Rizzino, 1988; Roberts and Sporn, 1990). Although TGF$l, TGF-02, and TGF-P3 appear to have similar effects in several in vitro systems and appear to be functionally equivalent, several studies have shown that the three TGF-P species can exert quantitatively or qualitatively different effects (Ohta et al., 1987; Rosa et al., 1988; Graycar et al., 1989; Roberts et al., 1990). Furthermore, the expression and production of TGF-/31, TGF-P2, and TGF-P3 appear to be regulated both spatially and temporally during the earliest stages of mouse embryogenesis (Campbell et ah, 1990; Kelly et al, 1990; Mummery et al., 1990; Slager et al., 1991), suggesting that the different forms of TGF-P have distinct roles during mammalian embryogenesis. The expression and production of the different forms of TGF-P has been examined previously in embryonal carcinoma (EC) cells, which serve as a model system for studying the expression of growth factors and their receptors during early development (reviewed in Rizzino, 1989). In this model system, TGF-@l is produced in both
MATERIALS
Copyright All rights
Q 1992 by Academic Press, Inc. of reproduction in any form reserved.
METHODS
Stock cultures of F9 EC cells and PC-13 EC cells were maintained as reported previously (Rizzino et al., 1983). Differentiation of EC cells was induced by a 72-hr treatment with 5 puMRA. The F9 EC cells were transfected by a modified calcium phosphate precipitation method designed for nonadherent cell lines (Davis et ab, 1986). The F9- and PC-13-differentiated cells were transfected in monolayer by calcium phosphate precipitation. For both transfection techniques, 20 pg of TGF-82 promoter-CAT plasmid DNA was cotransfected with 2 pg of the P-galactosidase expression plasmid, pCH-110 (Pharmacia).
1 To whom correspondence should be addressed. ’ Abbreviations used: transforming growth factor-p, TGF-P; embryonal carcinoma, EC; retinoic acid, RA; chloramphenicol acetyltransferase, CAT; polymerase chain reaction, PCR; fetal bovine serum, FBS. 0012-1606/92 $5.00
AND
172
KELLY, O’REILLY, AND RIZZINO
+z@ r, *3 pB2 - 770 ~02 - 520 pB2 - 347 pB2 - 187 pB2-77 -
pB2-40
FIG. 1. Structure of TGF-132 promoter-CAT gene constructs. TGF-p2 promoter fragments were generated by polymerase chain reaction using different Y-specific oligonucleotide primers between ~778 to -40 and a common 3’.specific oligonucleotide primer at +63 relative to the transcription initiation site. DNAs were amplified with Taq polymerase (Perkin-Elmer-Cetus) using cycling temperatures of annealing at 50°C for 2 min, elongating at 72°C for 4 min, and denaturing at 94°C for 1 min. The amplified DNA fragments were cloned into the promoterless chloramphenicol acetyltransferase (CAT) expression plasmid pGEM4SVOCAT, using Hind111 and Kp?lI restriction sites that were built into the oligonucleotide primers. The identity of each clone was confirmed by sequence analysis. The constructs were named pp2-n, Lvhere )i is the distance in nucleotides upstream of the transcription initiation site.
CAT activities were determined by the method of Seed and Sheen (1988) and normalized to ,&galactosidase activity by the method of Rosenthal (1987) to adjust for any differences in transfection efficiency (Hall et al., 1983). All promoter/reporter gene constructs were prepared using the polymerase chain reaction (PCR) and amplifying the 5’-flanking region and the first 63 bp of the first exon of the human TGF-/32 gene from genomic DNA (Noma et ah, 1991). Details regarding the construction of the TGF-/32 promoter/CAT constructs are provided in Fig. 1. All plasmids were purified by Qiagen tip-500 columns according to the manufacturer’s instructions. RESULTS
AND
DISCUSSION
In this study, we employed chimeric gene constructs in which various amounts of the human TGF-P2 promoter region were inserted upstream of the reporter gene, CAT (Fig. 1). To compare the activity of the TGF02 promoter in EC cells and their differentiated cells, the TGF-D2 promoter-CAT constructs, pb2-778, ~02-7’7, and ~02-40, were transfected into F9 EC cells. Following transfection, half of the cultures were treated with RA to induce differentiation of the F9 EC cells. In F9 EC cells, only low levels of CAT activity were detected for each construct (Fig. 2A). In contrast, CAT activity was substantially greater in the F9 EC cells treated with RA (Fig. 2A). The average increases in CAT activity observed for the constructs p/?2-778, p@2-77, and ~02-40 in three experiments were 9.4-, 4.8-, and 3%fold, respectively. A similar pattern of differentiation-dependent
TGF-/32 Gene Expressinn in EC Cells
173
TGF-P2 expression was observed when PC-13 EC cells were transfected with p/32-778 (data not shown). Thus, up-regulation of TGF-82 promoter activity when F9 EC and PC-13 EC cells are induced to differentiate suggests that the increase in steady-state levels of TGF-/32 mRNA is due, at least in part, to increased transcription of the TGF-P2 gene. To understand further the mechanism by which TGF1132 promoter expression is up-regulated during differentiation, we attempted to identify the DNA regulatory elements that are required for normal TGF-P2 promoter expression in the differentiated cells. Therefore, we transfected the various TGF-P2 promoter-CAT gene constructs into the differentiated cells and determined CAT activity. F9- and PC-13-differentiated cells had very similar patterns of TGF-P2 promoter activity (Fig. 2B). The shortest construct, pp2-40, which contains the TATA box and very little of the 5’-flanking region was expressed at only slightly higher levels than the plasmid pSVO-CAT (data not shown), which lacks a eukaryotic promoter (Langner et al., 1986). In contrast, pp2-‘77, which is only slightly larger than pp2-40, was expressed at a level similar to the largest plasmid, p@2-778. ~02-77 and p@2-778 were always the most highly expressed TGF-82 promoter-CAT constructs in the F9- and PC13-differentiated cells. The high level of CAT activity expressed by p/32-77 contrasted with the low level of expression by p/32-187. The patterns of expression shown in Fig. 2B suggest, that there are at least three regions that influence the expression of the TGF-P2 promoter in the differentiated cells: (1) A positive regulatory element between -77 and -40, (2) a negative regulatory element in the region between -187 and -77, and (3) a second positive regulatory element in the region between -347 and -187. Interestingly, the TGF-02 promoter-CAT constructs have also been transfected into CCL-64 cells and the pattern of expression observed in these cells is very similar to that observed in F9- and PC-13-differentiated cells (Michael O’Reilly, unpublished results). The large difference in the level of CAT activity expressed by pp2-77 and pfl2-40 in the differentiated cells suggests the presence of a positive regulatory element in the region between -77 and -40. Sequence analysis of the TGF-P2 promoter identified a CRE/ATF-like element between -74 and -67 (Noma et al., 1991). This element contains six of the eight consensus nucleotides observed in other gene promoters regulated by the CRE/ ATF family of transcription factors (Deutsch et ul., 1988; Roesler et ub, 1988). To determine if this CRE/ ATF-like element influences the basal activity of the TGF-02 promoter, a mutant construct, pp2-77M, was transfected into the differentiated cells. p/Y2-77M was mutated at two nucleotides (-71 and -69) that have
174
DEVELOPMENTAL BIOLOGY CAT
VOLUME 153,1992
activity (qm)
~82.778
pB2al
PEP-77
Relative CAT actbity r15
30
-14
26 T
26
-17.is
24
-13 -11
p
22
-10
g
M
C
18 16
if
pa2-776
pm52a
pB2-347
InI pw-18-r
@2-n
ti
14
-6
5
-5
t
-4 -3
$
12 10 6 6
2
4
1
2
0
0
p&MO
FIG. 2. Effect of differentiation on TGF-&Z promoterCAT activity. (A) F9 EC cells were transfected with the TGF-p2 promoterCAT plasmids, p@-778, ~62-77, and ~02-40. For this study, F9 EC cells (1 X 106) were resuspended and combined with each plasmid-CaPO, precipitate before being distributed equally into two loo-mm culture dishes containing DME supplemented with 10% FBS. The following day, the transfected cells were washed and refed with fresh medium and either left untreated (hatched bar) or treated (solid bar) with RA (5 @f). CAT activity of the cell lysates was determined 48 hr later as described under Materials and Methods. (B) The basal activity of the TGF-82 promoter was examined in FS-differentiated and PC-13-differentiated cells. FS-differentiated (solid bar) and PC-13-differentiated (open bar) cells were derived from F9 EC and PC-13 EC cells (1 X lo5 per loo-mm culture dish) cultured in DME supplemented with lo’% FBS and 5 #RA. After 72 hr, the differentiated cells were refed with medium lacking RA and cotransfected in monolayer with TGF-02 promoter-CAT plasmids and the normalizing plasmid, pCH-110. CAT activity of each cell lgsate was determined 48 hr after transfection and was normalized to D-galactosidase expression, as described under Materials and Methods. The bars represent ratios of CAT activity relative to the activity of the p@2-40 construct, which was designated as 1.00. CAT activity of p/32-40 in F9- and PC-13-differentiated cells was 3978 and 714 cpm, respectively. (C)To identify a sequence necessary for TGF-02 promoter activity, F9- and PC-13-differentiated cells were cotransfected in monolayer with the normalizing plasmid, pCH-110, and the TGF-fl2 promoter-CAT plasmids, p/12-77, pfi2-77M (a mutated form of p/32-77), and p/32-40. The mutant construct, p@2-77M, was generated by PCR using a Y-oligonucleotide primer that had been modified at ~71 (C to T) and -69 (T to G) within a CREIATFlike sequence located between -74 and -67 of the TGF-m promoter. The mutant DNA fragment was amplified and cloned into pGEM4-SVOCAT as described in the legend for Fig. 1. CAT activity of each cell lysate was determined 48 hr after transfection and was normalized to fi-galactosidase expression, as described under Materials and Methods. The bars (p@2-77, hatched; pfi2-77M, solid; and p/Z-40, open) represent ratios of CAT activity relative to the activity of the ~02-40 construct, which was designated as 1.00. CAT activity of p/%40 in F9- and PC-13-differentiated cells was 4832 and 2308 cpm, respectively. All experiments shown in this figure were repeated twice with similar results.
been shown to be critical for the biological activity of CRE/ATF elements (Deutsch et al., 1988). Specifically, the sequence at -74 to -67 in ~02-77 was modified from GCACGTCA to GCATGGCA. CAT activity was reduced by approximately 65% when F9- and PC-13-differentiated cells were transfected with p/3277M instead of p/Z-77 (Fig. 2C). Thus, it appears that
the CRE/ATF-like element located at -74 to -67 exerts a significant influence on the basal activity of the TGF82 promoter in the differentiated cells. Interestingly, the addition of dbcAMP, which elevates the expression of properties characteristic of parietal extraembryonic endoderm (Strickland et al., 1980), resulted in a 3.1-fold increase in p/32-778 expression in FS-differentiated cells
KELLY, O’REILLY, AND RIZZINO
(data not shown). Analogs of CAMP have been shown to up-regulate a number of gene promoters that contain CRE-like elements (Roesler et al., 1988). These data suggest that a member(s) of the CRE/ATF family of transcription factors interacts with the TGF-j32 promoter and regulates its expression. In summary, the differential expression of the TGF$2 promoter in EC cells and their differentiated counterparts provides a useful system for studying transcription factors and in cis regulatory elements that control the expression of this gene. Our findings have identified at least three important regions in the TGF132promoter that control its expression in EC cells that have been induced to differentiate. Future studies will focus on the identification of the member(s) of the CRE/ATF family that binds to this site as well as mapping precisely the location of the negative and the upstream positive regulatory elements in the TGF-02 promoter. Solon Rhode is thanked for his advice during the early stages of this work. Keith Miller and Phillip Wilder are thanked for their comments on this manuscript. This work was supported bg a grant from the Council of Tobacco Research (2.520) and by core grants from the National Cancer Institute (Laboratory Cancer Research Center Support Grant, CA 36727) and the American Cancer Society (ACS SIG-16). David Kelly was supported by a graduate fellowship provided by the Nebraska Research Initiative in Biotechnology. REFERENCES Campbell, W. J., Kelly, D., and Rizzino, A. (1990). Expression of transforming growth factor-03 by emhryonal carcinoma cell, parietal endodcsrm-like cells and early mouse embryos. 1?! liitro Cell L&r: Biol. 26, 1181-1185. Davis, L. G., Dihner, M. D., and Battep, J. F. (1986). Calcium phosphate transfection of nonadherent and adherent cells with purified plasmids. I)/ “Basic Methods in Molecular Biology,” pp. 286-289. Elsevicr, New York. Deutsch, P. J., Hocffler, J. P., Jameson, J. L., Sin, J. C.. and Habener, J. F. (1988). Structural determinants for transcriptional activation by CAMP-responsive DNA elements. .I. Riol. Cl,em. 263, 1846618472. Graycar, J. L., Miller, D. A., Arrick, B. A., Lyons, R. M., Moses, H. L., and Dergnck, R. (1989). Human transforming growth factor+3: Recombinant expression, purification, and biological activities in comparison with transforming growth factors-01 and -$2. Mol. 6&ocrircol. 3, 1977-1986. IIall, C. V., Jacobs, P. E., Ringold, G. M., and Lee, F. (1983). Expression and regulation of Eschc~richicl coli IacZ gene fusions in mammalian cells. .J. Mol. AppI. Grwt. 2, 101-109. Kelly, I)., Campbell, W. J., Tiesman, J., and Rizzino, A. (1990). Regulation and expression of transforming growth factor type-D during earls mammalian development. (‘!/tote~knolon!/ 4, 227-242.
Langner, K.-D., Weyer, U., and Doerfler, W. (1986). Trans effect of the El region of adenoviruses on the expression of a prokargotic gene in mammalian cells: Resistance to 5’.CCGG-3’ mrthglation. I’m:. Nufl. Acad. Sci. USA 83, 1598-1602. Mummery, C. L., Slager, H., Kruijer, W., Feijen, A., Freund, E., Koornneef, I., and van den Eijnden-van Raaij, A. J. M. (1990). Expression of transforming growth factor 82 during the differentiation of murine embryonal carcinoma and embryonic stem cells. Urr?. Bid. 137, 161-170. Noma, T., Glick, A. B., Geiser, A. G., O’Reilly, M. A., Miller, J., Roberts, A. B., and Sporn, M. B. (1991). Molecular cloning and structure of the human transforming growth factor+2 gene promoter. Growth Fwtors 4, 247-255. Ohta, M., Greenberger, J. S., Anklesaria, P., Bassols, A., and Massayue, J. (1987). Two forms of transforming growth factor-ij distinguished by multipotential haematopoictic progenitor cells. ,Vut/rrcJ 329, 539-541. Rizzino, A. (1987). Appearance of high aflinitg receptors for type B transforming growth factor during differentiation of murine embryonal carcinoma cells. Co ttwr Krs. 47, 4386-4390. Rizzino, A. (1988). Transforming growth factor-/? Multiple efIects on cell differentiation and extracellular matrices. L)rc,. Biol. 130, 41 I422. Rizzino, A. (1989). Use of embryonal carcinoma cells to study growth factors during early mammalian development. 11, “Growth Factors in Mammalian Development” (I. Y. Rosenblum and S. Htgner, Eds.), pp. 113-134. CRC Press, Inc., Bora Raton, FL. Rizzino, A., Orme, L. S., and De Larco, J. E. (1983). Emhryonal carcinoma cell growth and differentiation: Production of and response to molecules Lvith transforming growth factor activity. &,rp Co// Rcs. 143, 1431.52. Roberts, A. B., Kondaiah, P., Rosa, F., Wanatabe, S., Good, P., Rochc, N. S., Rebbert, M. L., David, I. B., and Sporn, M. B. (1990). Mesoderm induction in Xwol~/r.s /trc>t.i.sdistinguishes between the various TGF@isoforms. C;r.owtll Ftrctors 3, 27?2XG. Roberts, A. B., and Sporn, M. B. (1990). The transforming growth factor-$‘s It/ “Handbook of Experimental Pharmacology” (Sporn, M. B., and Roberts, A. B.. Eds. ), Vol. 9% 1 ), pp. 419472. SpringerVerlag, Heidelberg. Roesler, VV’.J., Vandenbark, G. R., and Hanson, R. 1%‘.(1988). Cyclic AMP and the induction of eukaryotic gene transcription. ,I Bid. Chc~ttc.263, 9063-9066 Rosa, F., Roberts, A. B., Danielpour, D., Dart, L. L., Sporn, M. B., and David, I. 8. (1988). Mesoderm induction in amphibians: The role of TGF-da-like factors. Sckrtw 239, 783-785. Rosenthal, N. (1987). Identification of regulatory elements of cloned genes with functional assays. I,, “Methods in Enzymolog,ry: Guide to Molecular Cloning Techniques” (S. L. Berger and A. R. Kimmcl, Eds.), Vol. 152, pp. 701-720. Academic Press, San Diego, CA. Seed, B., and Sheen, J. Y. (1988). A simple phase-extraction assay for chloramphenicol acyltransferase activity. &r/e 67, 271-277. Slager, II. G., Las-son, K. A.. van den Eijnden-van Raaij, A. J. M., delaat, S. W.. and Mummery, C. L. (1991). Differential localization of TGF-@2 in mouse preimplantation and earl;\ postimplantation development. L)et,. Biol. 145, 205-218. Strickland, S., Smith, K. K., and Marotti, K. R. (1980). Hormonal induction of differentiation in trratocarcinoma stem cells: Gencration of parietal cndoderm hv rctinoic acid and dihutgr~l CAMP. (‘(al/ 21, 347-355.
