sign in sign up
<<
Home , Transcription factor

Proc. Nati. Acad. Sci. USA Vol 85, pp. 5016-5020, July 1988 Biochemistry A factor active on the epidermal (DNA-binding //A431 cells/ receptor) RYoICHIRo KAGEYAMA, GLENN T. MERLINO, AND IRA PASTAN Laboratory of Molecular , National Cancer Institute, National Institutes of Health, 9000 Rockville Pike, 37/4E16, Bethesda, MD 20892 Contributed by Ira Pastan, March 31, 1988

ABSTRACT We have developed an in vitro transcription that binds to the region of the EGFR gene and system for the epidermal growth factor receptor (EGFR) specifically stimulates transcription of the EGFR gene, oncogene by using nuclear extracts ofA431 human epidermoid termed EGFR-specific transcription factor (ETF). These carcinoma cells, which overproduce EGFR. We found that a results suggest that ETF might be a key factor for the nuclear factor, termed EGFR-specific transcription factor transcriptional activation of this cellular oncogene. (ETF), specifically stimulated EGFR transcription by 5- to 10-fold. In this report, ETF, purified by using sequence- specific oligonucleotide affinty chromatography, is shown by MATERIALS AND METHODS renaturing material eluted from a NaDodSO4/polyacrylamide Preparation and Fractionation of A431- Nuclear Ex- gel to be a protein with a molecular mass of 120 kDa. ETF binds tract. Preparation of A431-cell nuclear extracts and their to the promoter region, as measured by DNase I "footprinting" fractionation by -agarose and DEAE-Sepharose and gel-mobility-shift assays, and specifically stimulates the CL-6B columns were done as described (25). The buffer used transcription of the EGFR gene in a reconstituted in vitro was HM, which consists of 20 mM Hepes (pH 7.9), 1 mM transcription system. These results suggest that ETF could play MgCl2, 2 mM dithiothreitol, and 17% (vol/vol) glycerol, to a role in the overexpression of the cellular oncogene EGFR. which KCl was added as indicated. The fraction BA eluted with 0.12 M KCl/HM from a DEAE-Sepharose column The epidermal growth factor receptor (EGFR) has extensive contained ETF. Fraction BB eluted with 0.25 M KCI/HM homology with the erbB oncogene product of the avian and fraction BC eluted with 0.5 M KCI/HM contained erythroblastosis virus (1, 2). Overproduction of the EGFR general transcription factors. Most ofthe RNA polymerase II has been detected in several types of cancers (3-12), and it activity was present in fraction BC (see Fig. 3). Our minimal has been shown that in vitro transcription system consisted of fractions BB, BC, overexpression of the EGFR leads to and 0.1 M KCI/HM-eluted heparin-agarose fraction A, which epidermal growth factor-dependent transformation (13, 14). contains a stimulating activity. These results indicate that the EGFR gene can function as an Fraction BA was purified by sequence-specific oligonucle- oncogene. otide affinity chromatography. The affinity resin was pre- Expression of the EGFR gene is regulated at a variety of pared as described by Wu et al. (23) by annealing two levels, including gene amplification (3-11) and mRNA sta- oligonucleotides, 5' CCCGCGCGAGCTAGACGTCCGGG- bility (ref. 9; Y. Jinno, G.T.M., and I.P., unpublished data). CAGCCCCCGGCGCAGCGCGGCCG 3' and 5' CGGCCG- Very often, the production of EGFR mRNA correlates CGCTGCGCCGGGGGCTGCCCGGACGTCTAG 3'. The directly with EGFR levels on the cell surface (9). Thus, 0.5 M KCI fraction was diluted to 0.1 M KCI with HM buffer, transcriptional control undoubtedly plays a major role in the reapplied to the column, and eluted as described (25). The regulation of the EGFR gene. Therefore, it is possible that a eluates were analyzed with a 10% NaDodSO4/polyacrylam- trans-acting factor may transcriptionally activate the EGFR ide gel. gene in some tumors. Evidence that the protooncogene c-jun In Vitro Transcription. Promoter deletion mutants were encodes the transcription factor AP-1 (15) suggests that made by using restriction enzymes or BAL-31 nuclease and transcription factors could contribute to transformation. fusing the promoter fragments to the chloramphenicol ace- Although many transcription factors have been isolated and tyltransferase (CAT) gene (26) as described (25). The bound- shown to stimulate transcription (16-23), very little is known aries of the promoter contained in these plasmids were about what transcription factors are involved with the acti- determined by DNA (27). The in vitro reconsti- vation of the expression of . To understand the tution reaction mixture contained supercoiled templates, mechanism by which the EGFR gene is regulated, we have heparin-agarose fraction A, DEAE-Sepharose fractions BB developed an in vitro transcription system that uses nuclear and BC in the presence or absence of DEAE-Sepharose extracts of A431 cells, which overproduce EGFR. fraction BA (see Fig. 3A). After incubation for 60 min at 30°C, The promoter ofthe EGFR gene lacks a "TATA box" and RNA was prepared and hybridized with 5'-end-labeled CAT- "CAAT box" but contains multiple "GC boxes" (24). In specific primer. The CAT primer is a synthetic single- vitro transcription studies showed that purified transcription stranded 24-mer that hybridizes to the region between resi- factor Spl binds to the GC boxes and stimulates EGFR dues 4920 and 4943 of pSV2cat (25). The primer extension transcription -4-fold (25). We also have detected another product was analyzed on a 5% polyacrylamide sequencing factor that specifically enhances EGFR transcription but not gel. The in vitro major transcription initiation site is at simian virus 40 (SV40) early transcription (25). This obser- position -48 relative to the initiation site. vation prompted us to characterize and purify this factor. In DNase I "Footprinting." For the coding strand probe, the this study, we describe the purification of a 120-kDa protein HindIll fragment of the EGFR promoter region was excised

The publication costs of this article were defrayed in part by page charge Abbreviations: EGFR, epidermal growth factor receptor; CAT, payment. This article must therefore be hereby marked "advertisement" chloramphenicol acetyltransferase; SV40, simian virus 40; ETF, in accordance with 18 U.S.C. §1734 solely to indicate this fact. EGFR-specific transcription factor. 5016 Downloaded by guest on September 25, 2021 Biochemistry: Kageyama et al. Proc. Natl. Acad. Sci. USA 85 (1988) 5017 from the template plasmid containing sequence between To determine whether a specific region of the promoter is positions - 297 and - 20, treated with calf intestinal alkaline necessary for ETF activity, we analyzed ETF responsiveness phosphatase, 5'-end-labeled with T4 polynucleotide with several deletion mutants. The minimal reconstituted tran- and [-y-32P]ATP, digested with Ava I, and purified from a scription system was assembled with and without the addition polyacrylamide gel. For the noncoding strand probe, the of ETF-containing fraction BA (see Fig. 3A). As shown in Fig. Hinfl fragment of the EGFR promoter region was excised 1B, transcription from deletion mutants that contain sequence from the plasmid containing sequence between positions from position -388 to position -20 was stimulated by the -388 and -20, 5'-end-labeled, digested with HindIII, and addition ofETF. Deletion ofthe region between positions - 388 purified from a polyacrylamide gel. and - 256 resulted in a diminished but still detectable response. DNase I-footprinting reactions were as described by Dy- However, additional deletion of sequence from position - 256 nan and Tjian (28). The DNA-binding reaction mixtures to position -222 resulted in no stimulation by ETF (Fig. 1B, contained 5 ng ofthe end-labeled probe, 1 gg ofsonicated calf lanes 7 and 8). These results indicate that the region between thymus DNA, and the protein sample in a final volume of 50 positions - 256 and - 222 is necessary for ETF activity and also A.l. Calf thymus DNA was omitted for the assay of purified suggest that multiple binding sites may exist in the region factors. between positions - 388 and - 222. As a control, each reaction Gel-Mobility-Shift Assay of Renatured ETF. One micro- mixture also contained SV40 DNA; SV40 early transcription gram of the affinity-purified ETF was separated on a was not affected by fraction BA. NaDodSO4/polyacrylamide gel, and several protein bands To investigate whether a specific nuclear factor binds to were excised, eluted, and subjected to renaturation as de- the region between positions - 256 and - 222, we carried out scribed by Briggs et al. (16). We added 50 pug ofbovine serum DNase I-footprinting assay on fraction BA with the EGFR albumin in the elution step. Gel-mobility-shift assays were as promoterfragment as a probe. Fig. 2A shows that fraction BA described by Fried and Crothers (29). We used the 5'-end- contained a factor that bound to the coding and noncoding labeled oligonucleotides as a strands ofthe region between positions - 248 and - 233 (Fig. hybridized probe. 2B), with protection ofthe coding strand being more evident. A DNase I-hypersensitive site was also observed near the RESULTS protected region (Fig. 2A, as shown by arrow). These data Experiments using crude A431-cell nuclear extracts indicated indicate that ETF binds to the region between positions - 248 that the promoter region between positions -151 and - 20 and - 233 to specifically stimulate EGFR transcription. The ( + 1 is the translation initiation site) was sufficient for in vitro DNA sequence ofthe region is 5' CAGCCCCCGGCGCAGC EGFR transcription (25), although in vivo transient CAT 3' (Fig. 2B). To purify the factor, we made a sequence-specific oligo- expression experiments suggested that regions further up- nucleotide column stream were also involved with stimulatory and inhibitory affinity (23, 31). We used short double- effects stranded oligonucleotides consisting of a 36-base-pair seg- on EGFR transcription (30). We observed that tran- ment containing the ETF-binding site to make an affinity scription from the plasmids containing the upstream region resin. Fraction BA was applied to this column twice; each could be stimulated by reconstitution of fractionated ex- time it was eluted with 0.5 M KCL. Purified fractions were tracts. When the fractions of heparin-agarose and DEAE- analyzed by NaDodSO4/polyacrylamide gel electrophoresis. Sepharose columns were reconstituted (fractions A + BA + As shown in Fig. 3B, only a few protein bands remained after BB + BC) (see Fig. 3A), the template containing sequence up this procedure; major bands are 120, 100, and 80 kDa. We to position - 775 showed about five times more transcription then examined the DNA-binding activity and the in vitro than the template containing sequence up to position -151 transcription activity of affinity-purified ETF. As shown in (Fig. 1A, lanes 2 and 8). In addition, it was found that optimal Fig. 3C, the affinity-purified ETF still protected the region activity of the extended promoter was only achieved in the between positions - 248 and - 233 on a DNase I-footprinting presence of DEAE-Sepharose fraction BA, which contains assay. In addition the affinity-purified ETF stimulated EGFR ETF (Fig. LA, lanes 1 and 2). These results suggest that the transcription but had no effect on SV40 early transcription upstream regions are required for ETF activity. (Fig. 3D). These data clearly demonstrated that the purified A B -775 -573 -388 -151 -388 -331 -256 -222 _-+ _-+ - + - + _+ _-+ - +

SV40-

"e N EGFR- - .4m WI _w -_

1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8

FIG. 1. In vitro transcription analysis of EGFR promoter. (A) Transcription from deletion mutants of EGFR promoter was measured in the presence (+) or absence (-) of DEAE fraction BA by a primer-extension assay. The nucleotide numbers of endpoints of the deletion mutants are shown above each lane. Transcripts from EGFR and control SV40 early promoters are indicated on the left ofthe autoradiograph. Templates were made by fusing promoter fragments ofEGFR to the CAT gene. The in vitro reconstitution reaction mixture contained supercoiled templates, heparin-agarose fraction A, and DEAE-Sepharose fractions BB and BC in the presence (+) or absence (-) of DEAE-Sepharose fraction BA (see Fig. 3A). After incubation for 60 min at 30°C, RNA was prepared and hybridized with 5'-end-labeled CAT-specific primer. The primer extension product was analyzed on a 5% polyacrylamide sequencing gel. (B) More precise localization of the region necessary for ETF activity was determined by a primer-extension assay as described in A. Downloaded by guest on September 25, 2021 5018 Biochemistry: Kageyama et al. Proc., Nad. Acad Sci. USA 85 (1988) A CODING NONCODING

G C 0 2 5 10 0 G 0 2 5 10 0 4*- -240 - IL

-240- L wU -260-

B -260 -240 -220 AGCTAGACGTCCGGGCAGCCCCCGG-- -P - -- GC A^^UAGCGCGC CUCAUu^ FIG. 2. DNase I-footprinting analysis with EGFR promoter. (A) DNA-binding activities of DEAE-Sepharose fraction BA were analyzed by using a coding or noncoding strand of the EGFR probe. The coding or noncoding probes were 5'-end-labeled at positions - 331 or - 168, respectively. DNase I-footprinting reactions were carried out as described by Dynan and Tjian (28). The volume (in sul) offraction BA (6.5 ,ug/lul) used is indicated above each lane. Lane 0 shows the control DNase-digestion pattern with no protein factors added. The sequence ladder patterns (lanes G and C) were produced from each probe. The nucleotide residues ofthe EGFR promoter are indicated on the left ofthe autoradiographs, and the region protected by fraction BA is shown on the right. The DNase I-hypersensitive band is indicated by an arrow. (B) Promoter region of EGFR gene. The protected sequence by fraction BA on a DNase I-footprinting assay is underlined. ETF specifically acted on the EGFR transcription by inter- taining sequences beginning at position - 256 showed weaker acting with the promoter region. stimulation by ETF than the template containing sequences This factor also binds with approximately the same affinity beginning at position - 388 (Fig. 1B). Transient expression to several other regions spanning positions - 388 and - 256, assays with various CAT constructs also indicated that these suggesting that these regions are also necessary for the regions were important for EGFR in vivo maximal effect ofETF (data not shown). The hypothesis that (G.T.M., Y. Jinno, and A. C. Johnson, unpublished data). ETF interacts with upstream DNA sequences was also These results indicate that ETF plays an important role in supported by our observation that a mutant template con- EGFR transcription in vivo as well as in vitro. C D A B CODING NONCODING - 4 8 - ETF - ETF A431 NE -220 - Heparin Agaoe 200- -260- _ -SV40 0.M 0.4M A B 116- -. DEAE Sepharos 97- .;p:-,i* -EGFR 0.12M 0.25M I 0.5M -240- *-AV~ 11 BA BB BC - Ofigo Affinity x2 ( pl 11 ) 66 I-240-

0.5M i ETF

-260 - 43- ,., _011 -......

-220 s

FIG. 3. Purification and characterization of ETF. (A) ETF was purified from an A431-cell nuclear extract as described in this scheme. Sequence-specific oligonucleotide affinity resin was prepared as described by Wu et al. (23) by annealing two oligonucleotides, 5' CCCGCGCGAGCTAGACGTCCGGGCAGCCCCCGGCGCAGCGCGGCCG 3' and 5' CGGCCGCGCTGCGCCGGGGGCTGCCCG- GACGTCTAG 3'. (B) Eluate from the affinity step was analyzed by silver staining of NaDodSO4/polyacrylamide gel. The sizes of the protein is indicated by an arrow. (C) The DNA-binding activity of the affinity-purified ETF markers are shown on the left in kDa. The 120-kDa protein DNase was analyzed by DNase l-footprinting assay with a coding or noncoding strand of the EGFR probe. The protected region and I-hypersensitive site are shown on the right. (D) In vitro transcriptional analysis was done in the absence (-) or presence ofthe affinity-purified ETF. The added volume of ETF (in ,l) is shown above each lane. Downloaded by guest on September 25, 2021 Biochemistry: Kageyama et al. Proc. Natl. Acad. Sci. USA 85 (1988) 5019 Although affinity-purified ETF still contained several bound to the promoter regions and stimulated EGFR tran- bands, the presence of a 120-kDa protein appeared to be best scription independently. Nonetheless, the possibility of correlated with the ETF activities (data not shown). To cooperation needs to be examined more carefully because determine ifthe 120-kDa protein was ETF, regions containing our minimal in vitro transcription system still contained very it and other protein bands were excised from an NaDodSO4/ low but detectable levels of these factors. polyacrylamide gel, and the present were eluted and The ETF-binding site has an extremely high G + C content renatured. The ability of the renatured proteins to bind to the (88%). In fact, the center of the sequence consists of only EGFR promoter was analyzed by a gel-mobility-shift assay guanosine and cytidine (CCCCCGGC). An identical se- (Fig. 4). In the presence of the native affinity-purified ETF, quence oriented in the opposite direction is found in the a protein-DNA complex migrated with retarded mobility promoter region ofthe human insulin receptor gene (between relative to unbound DNA (lane 1). Control experiments residues -43 and -36) (32). This implies that ETF might showed that the DNA fragment containing the ETF-binding regulate other growth factor receptor and play an site was able to effectively compete for formation of the important role in the control of . complex but nonspecific DNA fragments were not (data not Imagawa et al. (22) reported that the shown). As shown in Fig. 4, only the 120-kDa protein recognized by transcription factor AP-2 is YCSCCMNSSS, exhibited DNA-binding activity after renaturation (lane 2). where Y is thymidine or cytidine, S is cytidine or guanosine, These data demonstrate that the 120-kDa protein is ETF. M is adenosine or cytidine, and N is any nucleoside. This site is one of several that appear to be involved in the induction of gene expression by phorbol ester tumor promoters. The DISCUSSION sequence of this site matches well with the central sequence We have shown that a purified factor, termed ETF, binds to ofthe ETF-binding site, raising the possibility that ETF might the promoter region of the EGFR gene and specifically mediate transcriptional activation by some tumor promoters. stimulates EGFR transcription in vitro. This factor did not However, ETF differs from AP-2 in several important re- have any effect on SV40 early transcription, and we could not spects. The molecular mass of ETF is 120 kDa, whereas the find any sequence similar to ETF-binding site in the SV40 molecular mass of AP-2 is 50 kDa. Furthermore, ETF has promoter regions. ETF has a molecular mass of 120 kDa on little or no effect on the SV40 promoter, but AP-2 has been a NaDodSO4/polyacrylamide gel. Our data indicate (25) that shown to stimulate SV40 transcription (21, 22). Now that a the native form of ETF has a molecular mass of 270 kDa, highly purified factor has been obtained it may be possible to measured by gel filtration. The transcriptionally active form understand the precise mechanism of transcriptional activa- of ETF may be a dimer or associate with other factors. It is tion of the EGFR oncogene. possible that other factors intervene between ETF and the initiation because the We thank P. Marino, A. Johnson, and C. McKeon for useful transcription complex ETF-binding site discussions; C. Wu for technical advice; Michael Gottesman and is relatively distant from the in vitro major transcription John Brady for reading this manuscript; B. Lovelace and A. Harris initiation site (-190 base pairs separate the sites). Between for help with cell culture; and S. Neal for photographic assistance. these sites there are two Spl-binding sites that are important for EGFR expression (25, 30). There are additional Spl- 1. Downward, J., Yarden, Y., Mayes, E., Scrace, G., Toffy, N., binding sites in the region between positions - 388 and - 256 Stockwell, P., Ullrich, A., Schlessinger, J. & Waterfield, M. D. (ref. 30; G.T.M., Y. Jinno, and A. C. Johnson, unpublished (1984) Nature (London) 307, 521-527. data). This suggests that ETF and Spl could interact and 2. Yamamoto, T., Nishida, T., Miyajima, N., Kawai, S., Ooi, T. cooperate with each other in some manner. However, our & Toyoshima, K. (1983) Cell 35, 71-78. preliminary in vitro transcription studies did not show any 3. Merlino, G. T., Xu, Y.-h., Ishii, S., Clark, A. J. L., Semba, K., between the two Toyoshima, K., Yamamoto, T. & Pastan, I. (1984) Science 224, cooperation factors. When added back to 417-419. the minimal reconstituted transcription system, both factors 4. Ullrich, A., Coussens, L., Haytlick, J. S., Dull, T. J., Gray, A., 0 0 LO 0 Tam, A. W., Lee, J., Yarden, Y., Libermann, T. A., Schles- C0 Lo singer, J., Downward, J., Mayes, E. L. V., Whittle, N., Wa- terfield, M. D. & Seeburg, P. H. (1984) Nature (London) 309, 418-425. 5. Lin, C. R., Chen, W. S., Kruiger, W., Stolarsky, L. S., Weber, W., Evans, R. M., Verma, I. M., Gill, G. N. & Rosenfeld, M. G. (1984) Science 224, 843-848. 6. Cowley, G., Smith, J. A., Gusterson, B., Hendler, F. & Ozanne, B. (1984) Cancer Cells 1, 5-10. 7. Merlino, G. T., Xu, Y.-h., Richert, N., Clark, A. J. L., Ishii, S., Banks-Schlegel, S. & Pastan, I. (1985) J. Clin. Invest. 75, 1077-1079. 8. Yamamoto, T., Kamata, N., Kawano, H., Shimizu, S., Kuroki, T., Toyoshima, K., Rikimaru, K., Nomura, N., Ishizaki, R., Pastan, I., Gamou, S. & Shimizu, N. (1986) Cancer Res. 46, 414-416. 1 2 3 4 5 6 7 9. Xu, Y.-h., Richert, N., Ito, S., Merlino, G. T. & Pastan, I. (1984) Proc. Natl. Acad. Sci. USA 81, 7308-7312. FIG. 4. Gel-mobility-shift assay of renatured ETF. One micro- 10. Libermann, T. A., Nasbaum, H. R., Razon, N., Kris, R., Lax, gram of the affinity-purified ETF was separated by NaDodSO4/ I., Soreq, H., Whittle, N., Waterfield, M. D., Ullrich, A. & polyacrylamide gel electrophoresis and several protein bands were Schlessinger, J. A. (1985) Nature (London) 313, 144-147. excised, eluted, and renatured as described by Briggs et al. (16). The 11. Wong, A. J., Bigner, S. H., Bigner, D. D., Kinzler, K. W., hybridized oligonucleotides, which were the same as used for the Hamilton, S. R. & Vogelstein, B. (1987) Proc. Natl. Acad. Sci. preparation of the oligonucleotide affinity resin, were labeled and USA 84, 6899-6903. used as a probe. Protein-DNA complex migrated with retarded 12. Fabricant, R. N., Delarco, J. E & Todaro, G. J. (1977) Proc. mobility relative to unbound DNA, as indicated by an arrow. Lane Natl. Acad. Sci. USA 74, 565-569. 1 shows the control experiment with the affinity-purified native ETF. 13. Velu, T. J., Beguinot, L., Vass, W. C., Willingham, M. C., The molecular masses in kDa of the renatured proteins are shown Merlino, G. T., Pastan, I. & Lowy, D. R. (1987) Science 238, above each lane. 1408-1410. Downloaded by guest on September 25, 2021 5020 Biochemistry: Kageyama et al. Proc. Natl. Acad. Sci. USA 85 (1988)

14. DiFiore, P. P., Pierce, J. H., Fleming, T. P., Hazon, R., Zimarino, V. & Ueda, H. (1987) Science 238, 1247-1253. Ullrich, A., King, C. R., Schlessinger, J. & Aaronson, S. A. 24. Ishii, S., Xu, Y.-h., Stratton, R. H., Roe, B. A., Merlino, G. T. (1987) Cell 51, 1063-1070. & Pastan, I. (1985) Proc. Nadl. Acad. Sci. USA 82,4920-4924. 15. Bohmann, D., Bos, T. J., Admon, A., Nishimura, T., Vogt, 25. Kageyama, R., Merlino, G. T. & Pastan, I. (1988) J. Biol. P. K. & Tjian, R. (1987) Science 238, 1386-1392. Chem. 263, 6329-6336. 16. Briggs, M. R., Kadonaga, J. T., Bell, S. P. & Tjian, R. (1986) 26. Gorman, C. M., Moffat, L. F. & Howard, B. H. (1982) Mol. Science 234, 47-52. Cell. Biol. 2, 1044-1051. 17. Chodosh, L. A., Carthew, R. W. & Sharp, P. A. (1986) Mol. 27. Maxam, A. M. & Gilbert, W. (1980) Methods Enzymol. 65, Cell. Biol. 6, 4723-4733. 499-560. 18. Jones, K. A., Kadonaga, J. T., Rosenfeld, P. J., Kelley, T. J. 28. Dynan, W. S. & Tjian, R. (1983) Cell 35, 79-87. & Tjian, R. (1987) Cell 48, 79-89. 29. Fried, M. & Crothers, D. M. (1981) NucleicAcids Res. 9,6505- 19. Angel, P., Imagawa, M., Chiu, R., Stein, B., Imbra, R. J., 6525. Rahmsdorf, H. J., Jonat, C., Herrlich, P. & Karin, M. (1987) 30. Johnson, A. C., Ishii, S., Jinno, Y., Pastan, I. & Merlino, G. T. Cell 49, 729-739. (1988) J. Biol. Chem. 263, 5693-5699. 20. Lee, W., Mitchell, P. & Tjian, R. (1987) Cell 49, 741-752. 31. Kadonaga, J. T. & Tjian, R. (1986) Proc. NatI. Acad. Sci. USA 21. Mitchell, P. J., Wang, C. & Tjian, R. (1987) Cell 50, 847-861. 83, 5889-5893. 22. Imagawa, M., Chiu, R. & Karin, M. (1987) Cell 51, 251-260. 32. Araki, E., Shimada, F., Uzawa, H., Mori, M. & Ebina, Y. 23. Wu, C., Wilson, S., Walker, B., Dawid, I., Paisley, T., (1987) J. Biol. Chem. 262, 16186-16191. Downloaded by guest on September 25, 2021