DNA AND CELL BIOLOGY Volume 10, Number 2, 1991 Mary Ann Lieber!, Inc., Publishers Pp. 113-118
Multiple Sequence-Specific Single-Strand-Binding Proteins for the Promoter Region of the Rat Albumin Gene MICHELLE
FLAVIN*>t
and FRANÇOIS
STRAUSSt
ABSTRACT Previous work from our laboratory described a protein that binds to single-stranded DNA in the early promoter of simian virus 40 in a sequence specific fashion. We have now used the gel retardation assay to search for similar sequence-specific single-strand-binding proteins for the promoter region of the rat albumin gene in nuclear extracts of rat hepatoma cells. Several proteins of this kind were detected, three of which are described in the present paper. Two of them bind specifically to the noncoding strand and the third one binds to the coding strand. The most abundant of these proteins binds to a pyrimidine stretch inside the coding region of the gene and appears to be homologous to the previously observed SV40-binding protein. Possible functions for sequence-specific single-strand-binding proteins in transcription are discussed.
INTRODUCTION
Thetensively lecular
regulatory
sequences of genes have been used in-
purify components of the momachinery responsible for activation and transcription in higher eukaryotes. These studies have led to the discovery of a number of factors that interact specifically with regulatory sequences of viral or cellular genomes (for review, see Mitchell and Tjian, 1989). Among the cellular genes studied, the liver-specific genes, most notably the albumin gene, are a particularly good system in which to study gene regulation, due to their inducibility and high level of expression in liver relative to other tissues. DNA sequence comparison between genes coding for different liver-specific proteins has shown similarities that are thought to correspond to important regulatory elements in liver, and sequence comparisons of a given gene in different species has revealed homologous sequences that probably have regulatory functions (see references in Frain et al, 1989). In addition, in vivo and in vitro studies have further defined these regulatory elements. For the rat albumin gene, the DNA sequence extending to position -170 upstream from the transcription initiation site is important for tissue-specific expression (Ott et al, 1984; Gorski et al, 1986; Heard et al, 1987; Maire et al, 1989). In addition, several factors that interact specifically with this sequence have been detected, some liver-specific, to
others ubiquitous (Cereghini et al, 1987; Lichtsteiner et al, 1987). One of them designated HNF1 or LF-B1, has recently been cloned (Frain et al, 1989) and complements
detect and
•Unité INSERM-U56,
du Kremlin-Bicêtre, 94270 Le 75251 Paris 05, France.
Hôpital
tlnstitut Jacques Monod,
nuclear extract from spleen to give liver-specific transcripts in vitro (Lichtsteiner and Schibler, 1989). Despite these advances in our knowledge of regulatory sequences and of the protein factors that interact with a
them,
we still have few data and ideas about the actual molecular mechanisms of gene activation and transcription in
higher eukaryotes.
In recent work from our laboratory using the control region of simian virus 40 (SV40) to purify sequence-specific proteins, we observed a protein that binds specifically to single-stranded DNA sequences. This protein, designated H16, binds to two specific sites on the late-coding strand of the SV40 early promoter, but shows no detectable binding to the early-coding strand, to double-stranded DNA, or to RNA (Gaillard et al, 1988; Gaillard and Strauss, 1990). We wished to determine whether this protein represented an isolated case specific to SV40, or was an example of a general family of sequence-specific single-strand-binding proteins. Therefore, we looked for such proteins in nuclear extracts of cultured hepatoma cells. We found that three proteins bind specifically to single-stranded sequences in the promoter region of the rat albumin gene, one of these proteins being apparently homologous to the protein that binds to the SV40 early promoter.
Kremlin-Bicêtre, France,
113
FLAVIN AND STRAUSS
114
MATERIALS AND METHODS The DNA sequence extending from the Alu I restriction site at position -175 to the Hpa II site at +76 of the rat albumin gene was taken from the plasmid SubJB (Sargent et al, 1981), a kind gift of Dr. Jean-Louis Danan. Nuclei were purified from the differentiated hepatoma cell line Fao, or, for some experiments, from the dedifferentiated cell line C2 (Deschatrette et al, 1980). DNA labeling, strand separation, and nuclear protein extraction were performed as described previously (Gaillard et al, 1988). Nonradioactive E. coli DNA, sonicated to an average length of 1 kb, was heat-denatured and used as competitor DNA. The synthetic DNA homopolymers used as competitor (see Fig. 3, below) were from Pharmacia. For gel retardation assays, DNA (5,000 cpm, about 0.1 ng) was incubated with proteins (1 /d) at room temperature for 30 min in 25 id of 50 mM NaCl, 20 mM Tris-HCl pH 7.5, 1 mM EDTA, 1 mM dithiothreitol (DTT) and 100 fig/m\ of bovine serum albumin. Electrophoresis was on a 4% polyacrylamide gel in 6.7 mMTris, 3.3 mM sodium acetate, 1 mM EDTA pH 7.8, as described elsewhere (Strauss and
Varshavsky, 1984). For purification of proteins g8 and d20, nuclear proteins were extracted in 0.4 M NaCl, 10 mM potassium phosphate pH 7.5, 25 mM Tris-HCl pH 7.5, and 2 mM DTT. They were loaded directly, without dialysis, on a 6-ml hydroxyapatite column and eluted with a linear gradient of 0 to 0.4 M potassium phosphate pH 7.5 in the same buffer. Two pools were made with the eluted proteins, the first
composed of fractions eluting between 0 and 0.2 M potassium phosphate and the second composed of fractions eluting between 0.2 and 0.4 M potassium phosphate. Each pool was dialyzed against 25 mM Tris-HCl pH 7.5, 1 mM DTT and fractionated further by HPLC on a mono Q ion exchange column (Pharmacia). Proteins were eluted with a linear gradient of 0-0.4 M NaCl in the same buffer followed by a 1 M NaCl step elution at the end of the gradient. Protein g8 was present in the 0-0.2 M pool from hydroxyapatite and eluted at 0.15 M NaCl from the mono Q
column. Protein d20 was present in the 0.2 to 0.4 M pool from hydroxyapatite and eluted at the 1 M NaCl step from the mono Q column. For purification of protein n7, a nuclear extract, prepared as described above, was diluted 10-fold with 25 mM Tris-HCl pH 7.5, 1 mM DTT and loaded on a mono Q column. Proteins were eluted with a linear gradient of 0-0.4 M NaCl in the same buffer. Fractions eluting between 0.24 and 0.31 M NaCl contained DNA-binding activity and were pooled. They were then diluted fivefold with 50 mM HEPES-NaOH pH 7.5, 2 mM DTT, loaded on a mono S column, and eluted with a linear gradient of NaCl from 0 to 0.5 M in the same buffer. Protein n7 eluted at 0.17 M NaCl.
RESULTS To search for sequence-specific single-strand-binding proteins for the promoter region of the rat albumin gene,
Non-coding strand
Coding strand
m
m
BP
V
12 3 4 5 6 7 8 9 10111213141516171819 20
PHP* WOTOTVMMW
• WWW™" "ils • m 1 2 3 4 5 6 7 8 9 1011121314151617181920 .
Assay of nuclear protein fractions for sequence-specific single-strand-binding activities. Nuclear extracts of rat hepatoma cells were fractionated by chromatography on hydroxyapatite, and the proteins eluting between 0 and 0.2 M potassium phosphate were pooled, dialyzed, and fractionated by HPLC on a mono Q column. Elution of proteins was with a linear gradient of NaCl from 0 (fraction 1) to 0.4 M (fraction 19), followed by a step elution (fraction 20). The 20 fractions from the mono Q column were assayed by gel retardation with the labeled coding and noncoding strands of the
FIG. 1.
promoter region of the
ples.
An
rat albumin gene. The amount of nonradioactive E. coli
autoradiogram of the gel is shown.
competitor
DNA
was
50 ng in all
sam-
SEQUENCE-SPECIFIC SINGLE-STRAND-BINDING PROTEINS
coding strand
non-coding strand
115
gs
gs
12 3 4
5
6
7
8
9 10 11 12
12 3 4
5
6
7
8
9 10 11 12
d20 1 23456789 10 1 23456789 10
d20
t..
à
ff
UN»
n7
123456789 10 123456789 10
WwW **
n7
123456789 10 123456789 10 Detection of three sequence-specific singlefor the promoter region of the rat albumin gene. Protein fractions were assayed with the labeled coding and noncoding strands of the promoter region of the rat albumin gene, in the presence of inreasing amounts of denatured E. coli competitor DNA. In the figure, the coding strand is the fast-migrating strand. Competitor DNA amounts were 1, 2, 4, 8, 15, 30, 60, 125, 250, and 500 ng in lanes 1-10, respectively. FIG. 2.
strand-binding proteins
we assayed nuclear extracts from rat hepatoma cells by a gel retardation assay using either the coding strand or the noncoding strand of the Alu l-Hpa II fragment that extends from position -175 to +76 of the gene. When such an experiment was performed with total crude nuclear extracts, all sequence-specific proteins were masked by pro-
If
°*
Wfliw
FIG. 3. Competition by different DNA sequences. Proteins were allowed to interact with the labeled coding strand (protein g8) or noncoding strand (proteins d20 and n7) of the promoter region of the rat albumin gene in the presence of the following nonradioactive competitor DNAs. Lanes 1-4, Denatured E. coli DNA, 8, 15, 30, 60 ng, respectively; lanes 5-7, homologous strand (coding strand for g8, noncoding strand for d20 and n7), 4, 8, and 15 ng, respectively; lanes 8, 4 ng of single-stranded E. coli DNA plus 250 ng of double-stranded DNA of bacteriophage X; lanes 9-12, 4 ng of single-stranded E. coli DNA plus 250 ng of the synthetic polymers poly(dC), poly(dG), poly(dT), and poly(dA), respectively.
nonspecifically to the DNAs (data not shown). Therefore, we fractionated the nuclear extracts by several different Chromatographie procedures, assayed each protein fraction by gel retardation, and concentrated on the fractions with little nonspecific DNA-binding activity. Figure 1 shows the results obtained for a typical nuclear extract fractionation by chromatography on hydroxyapatite followed by HPLC on a mono Q column. The 20 fractions obtained by HPLC were assayed for binding to teins that bound
FLAVIN AND STRAUSS
116
B
280
+47 +76
-175
BstEII Hpal
Alu
Hinfl
+1 A coding B non-coding C non-coding D non-coding E non-coding
ir
im
123C123C123C123C123C
Hpal
BstEII
GTAACCTTTCTCCTCCTCCTCTTCATCTC FIG. 4. Determination of the binding site of protein n7. DNA strands of different fragments of the -175 to +280 segment of the rat albumin gene were labeled and assayed with protein n7 by gel retardation. A map of the region and the nucleotide sequence of the noncoding strand between the But EII and Hpa II sites are also shown. The fragments assayed were: A, Alu l-Hpa II coding strand; B, Alu l-Hpa II noncoding strand; C, Hpa ll-Hinf I noncoding strand; D, Alu IBst EII noncoding strand; E, Bst Ell-Hinf I noncoding strand. E. coli competitor DNA amounts were 1, 4, and 16 ng in lanes 1-3, respectively. Lanes C, Control, no protein added.
coding and noncoding strands of the Alu l-Hpa II fragment. Several retarded bands are present at comparable intensities with both of the DNA strands, notably in fractions 4, 5, and 6, and correspond to uncharacterized nonspecific single-strand binding proteins. A protein that elutes in fraction 7 and beyond, however, forms a specific complex with the noncoding strand, and a less abundant protein that elutes maximally in fraction 8 binds specifically to the coding strand. Using various combinations of Chromatographie procedures including hydroxyapatite and HPLC mono Q and mono S columns, we detected several proteins that formed specific complexes with either the coding or the noncoding strands of the Alu l-Hpa II fragment. To confirm the sequence-specificity of the complexes, all of these protein fractions were subsequently assayed in the presence of increasing amounts of nonradioactive, single-stranded competitor DNA from E. coli. Figure 2 shows results obtained for the three fractions containing such sequence-specific single-strand binding activities that gave the best signal-tonoise ratios: protein g8 which bound to the coding strand, and proteins d20 and n7 which bound to the noncoding the
strand. To study the sequence
specificity of the proteins further, competition experiment with different DNA sequences, as shown in Fig. 3. Upon addition of nonradioactive homologous strand, the complexes quickly disappear as expected for a dilution by an excess of substrate (lanes 5-7). Addition of denatured E. coli DNA slowly decreases the amount of complex formed (lanes 14). This is as expected because E. coli DNA would contain randomly distributed binding sites on a statistical basis. Addition of a large amount of native DNA of bacteriophage X does not decrease the formation of the complex (lane 8), showing the absence of affinity of the proteins fordouble-stranded DNA. Finally, addition of large amounts of the synthetic homopolymers poly(dC), poly(dG), poly(dT), and poly(dA) has little effect on the formation of the we
performed
a
complexes (lanes 9-12), except for protein g8 which shows a moderate affinity for poly(dT) (lane 11) and protein n7 which binds weakly to poly(dC) (lane 9). The most abundant of these three proteins was protein n7. We localized its binding site by preparing smaller fragments of the Alu l-Hini I fragment which extends from position -175 to +280 of the rat albumin gene. These fragments are shown in Fig. 4, as are the results of the gel retardation assays. Protein n7 binds to the noncoding strand (B) but not to the coding strand (A) of the Alu IHpa II fragment. It does not bind to the noncoding strand between the Hpa II and Hinf I sites (C) nor between the Alu I and Bst EII sites (D), but it does bind to the Bst Ell-Hinf I fragment (E), showing that the DNA segment from positions +47 to +76 is necessary for binding. The nucleotide sequence of this segment is also shown in Fig. 4. It is a pyrimidine-rich sequence that is part of the coding sequence of the albumin gene (see Discussion). Protein n7 presents several similarities with protein H16, a protein from cultured monkey cells that binds to the late strand of the SV40 early promoter. Both proteins elute from mono Q columns at the same NaCl molarity, bind to C-rich sequences and have some affinity for poly(dC), give similar shifts by gel retardation, and are fairly abundant. This similarity is further shown by gel retardation with different sequences in Fig. 5, showing that the rat protein n7 binds
to
the late strand of the SV40 promoter, and the
monkey protein H16 binds to the noncoding strand of the albumin gene. In addition, the shifts produced by both proteins binding to different fragments are identical. All these observations strongly suggest that proteins n7 and H16 are homologous proteins. DISCUSSION
By fractionating nuclear extracts of rat hepatoma cells, have detected several proteins that bind, in a sequence-
we
SEQUENCE-SPECIFIC SINGLE-STRAND-BINDING PROTEINS albumin non-coding strand I H16 I n7
DNA Protein
1
-
23123C1
SV40 late strand I H16 n7
23123C
Comparison between the rat protein, n7, and the monkey protein, H16. Both proteins were assayed by gel retardation for binding to the noncoding strand of the alFIG. 5.
bumin gene and to the late strand of the SV40 control reE. coli competitor DNA amounts were 1, 4, and 16 ng in lanes 1-3, respectively. Lanes C, Control, no protein added.
gion.
specific fashion, to single-stranded DNA in the promoter region of the rat albumin gene. The possibility that such sequence-specific single-strand-binding proteins might be present in nuclear extracts was suggested by our previous observation of a protein binding to the late strand of the SV40 early promoter (Gaillard et al, 1988). However, the multiplicity of such proteins was unexpected. It is important to note that none of these proteins could be detected in crude extracts by the gel retardation assay, as they were masked by the very abundant, nonspecific single-strandbinding proteins, even in the presence of competitor DNA. Only after fractionation by two Chromatographie steps did the three proteins described here become clearly evident. As these proteins were first found in differentiated hepa-
cells, we also looked for their presence in dedifferentiated rat hepatoma cells that do not express the albumin gene. We found that all three of the proteins were present in both cell lines in similar amounts (data not shown). This does not exclude, however, the possibility that they play a role in albumin gene expression. The dedifferentiated line can revert to the differentiated state and reexpress hepatic functions (Deschatrette et al, 1980), suggesting that most of the molecular machinery responsible for expressing liver-specific genes remains present in the dedifferentiated line. One of the three proteins, n7, is very similar to the previously characterized monkey protein, H16, that binds to the early promoter of SV40 and has recently been shown to stimulate transcription by RNA polymerase II in vitro (Gaillard and Strauss, 1990). Both proteins behave similarly on hydroxyapatite, mono Q and mono S columns, they bind to the same DNA fragments, and they give identical shifts in gel retardation assays. Therefore, protein n7 probably represents the homolog of monkey protein H16. toma
117
the DNA-binding sites in SV40 and the rat albumin gene have different nucleotide sequences, they both are pyrimidine-rich, and more particularly, C-rich. In this respect, it is interesting to note that the binding site of protein n7 contains the sequence CTCCTCCTCCTC that codes for four consecutive leucine residues at positions 6-9 in the amino acid sequence of rat albumin. In mammals, the frequency of the codon CTC among all leucine codons is only 22% (Lathe, 1985). Therefore, the probability that, in a cluster of four leucines, all of them are coded by CTC is only 0.002, suggesting that the presence of the four CTC triplets is not due to chance but has been selected for, consistent with a possible regulatory role for that sequence. Finding a protein binding site within the coding region of the albumin gene was unexpected. It has not been experimentally determined whether this particular sequence element is important for rat albumin gene control, as most of the work to define regulatory sequences has concentrated on the region upstream of the transcription initiation site. A number of other genes, however, contain DNA regulatory elements that extend 3' to the start of transcription (e.g., Mansour et al, 1986; Hultmark et al, 1986; Stenlund et al, 1987; Theill et al, 1987; Bourachot et al,
Although
1989).
The existence of proteins that bind to single-stranded DNA and exhibit sequence specificity has been suggested or documented by several recent articles. These proteins are thought to be involved in replication (Fry et al, 1988; Traut and Fanning, 1988; Roller et al, 1989), recombina-
(Edelmann et al, 1989), or transcription (Gaillard et al, 1988; Feavers et al, 1989; Lannigan and Notides, 1989; Rajavashisth et al, 1989; Wilkison et al, 1990). Indeed, during all these processes the DNA double helix is transiently opened, and proteins that recognize specific tion
single-stranded sequences can interact with DNA at that time. Such proteins might even play a part in regulating the opening of the double helix. The observation reported here of at least three proteins that bind to single-stranded DNA in the region of the promoter of the rat albumin gene reinforces the suggestion that such proteins might play important roles in the mechanisms of transcription by RNA polymerase II in higher eukaryotes. Thus far, characterization of gene regulatory factors has been concentrated almost exclusively on double-strand binding proteins. However, in contrast to the DNA double helix, which has a fairly uniform structure, single-stranded DNA can adopt a much larger variety of structures as a function of its nucleotide sequence, and therefore should be more suitable for highly specific interactions with regulatory proteins (Crick, 1971). It would not be surprising, therefore, if sequence-specific single-strand binding proteins played important role in the regulation of genome function.
an
ACKNOWLEDGMENTS We thank Jean-Louis Danan for the
gift of plasmid
subJB, Michèle Weber for technical assistance, and Claire Gaillard for stimulating discussions. M.F. gives special thanks to Jean Deschatrette and Michèle Meunier-Rotival
FLAVIN AND STRAUSS
118
for support and interest. We are grateful to Susan Elsevier for critical reading of the manuscript. This work was supported by grants from the Association pour la Recherche sur le Cancer, Ligue Nationale Française Contre le Cancer, and Fondation pour la Recherche Médicale.
REFERENCES BOURACHOT, B., YANIV, M., and HERBOMEL, P. (1989). Control elements situated downstream of the major transcriptional start site are sufficient for highly efficient polyomavirus late transcription. J. Virol. 63, 2567-2577. CEREGHINI, S., RAYMONDJEAN, M., CARRANCA, A.G., HERBOMEL, P., and YANIV, M. (1987). Factors involved in control of tissue-specific expression of albumin gene. Cell 50, 627-638. CRICK, F.
(1971). General model for the chromosomes of higher organisms. Nature 234, 25-27. DESCHATRETTE, J., MOORE, E.E., DUBOIS, M., and WEISS, M. (1980). Dedifferentiated variants of a rat hepatoma: Reversion analysis. Cell 19, 1043-1051. EDELMANN, W., KRÖGER, B., GOLLER, M., and HORAK, I. (1989). A recombination hotspot in the LTR of a mouse an in vitro system. Cell 57, 937946. FEAVERS, I.M., McEWAN, I.J., LIANG, H., and JOST, J.-P. (1989). An estradiol-dependent protein from chicken liver binds single-stranded DNA and RNA. J. Biol. Chem. 264, 91149117.
retrotransposon identified in
FRAIN, M., SWART, G., MONACI, P., NICOSIA, A., STAMPFLI, S., FRANK, R., and CÓRTESE, R. (1989). The liver-specific transcription factor LF-B1 contains a highly diverged homeobox DNA binding domain. Cell 59, 145-157. FRY, M., PERRINO, F.W., LEVY, A., and LOEB, L.A. (1988). Factor D is a selective single-stranded oligodeoxythymidine binding protein. Nucleic Acids Res. 16, 199-211. GAILLARD, C, and STRAUSS, F. (1990). Sequence-specific
single-strand-binding protein for the SV40 early promoter stimulates transcription in vitro. J. Mol. Biol. 215, 245-255. GAILLARD, C, WEBER, M., and STRAUSS, F. (1988). A sequence-specific single strand binding protein for the late-coding strand of the SV40 control region. J. Virol. 62, 2380-2385. GORSKI, K., CARNEIRO, M., and SCHIBLER, U. (1986). Tissue-specific in vitro transcription from the mouse albumin promoter. Cell 47, 767-776. HEARD, J.-M., HERBOMEL, P., OTT, M.-O., MOTTURAROLLIER, A., WEISS, M., and YANIV, M. (1987). Determinants of rat albumin promoter tissue specificity analyzed by an improved transient expression system. Mol. Cell. Biol. 7, 2425-
2434.
HULTMARK, D„ KLEMENZ, R., and GEHRING, W. (1986). Translational and transcriptional control elements in the untranslated leader of the heat-shock gene hsp22. Cell 44, 429438. LANNIGAN, D.A., and NOTIDES, A. (1989). Estrogen receptor selectively binds the "coding strand" of an estrogen responsive element. Proc. Nati. Acad. Sei. USA 86, 863-867. LATHE, R. (1985). Synthetic oligonucleotide probes deduced
from amino acid sequence data. Theoretical and practical considerations. J. Mol. Biol. 183, 1-12. LICHTSTEINER, S., and SCHIBLER, U. (1989). A glycosylated liver-specific transcription factor stimulates transcription of the albumin gene. Cell 57, 1179-1187. LICHTSTEINER, S., WUARIN, J., and SCHIBLER, U. (1987). The interplay of DNA-binding proteins on the promoter of the mouse albumin gene. Cell 51, 963-973. MAIRE, P., WUARIN, J., and SCHIBLER, U. (1989). The role of cis-acting promoter elements in tissue-specific albumin gene expression. Science 244, 343-346. MANSOUR, S.L., GRODZICKER, T., and TJIAN, R. (1986). Downstream sequences affect transcription initiation from the adenovirus major late promoter. Mol. Cell. Biol. 6, 2684-2694. MITCHELL, P.J., and TJIAN, R. (1989). Transcriptional regulation in mammalian cells by sequence-specific DNA binding proteins. Science 245, 371-378.
OTT, M.-O., SPERLING, L., HERBOMEL, P., YANIV, M., and WEISS, M. (1984). Tissue-specific expression is conferred by a sequence from the 5' end of the rat albumin gene. EMBO J. 3, 2505-2510. RAJAVASHISTH, T.B., TAYLOR, A.K., ANDALIBI, A., SVENSON, K.L., and LUSIS, A.J. (1989). Identification of a zinc finger protein that binds to the sterol regulatory element. Science. 245, 640-643. ROLLER, R.J., McCORMICK, A.L., and ROIZMAN, B. (1989). Cellular proteins specifically bind single- and doublestranded DNA and RNA from the initiation site of a transcript that crosses the origin of DNA replication of herpes simplex virus 1. Proc. Nati. Acad. Sei. USA 86, 6518-6522. SARGENT, T., JAGODZINSKI, L., YANG, M., and BONNER, J. (1981). Fine structure and evolution of the rat serum albumin gene. Mol. Cell. Biol. 1, 871-883. STENLUND, A., BREAM, G.L., and BOTCHAN, M.R. (1987). A promoter with an internal regulatory domain is part of the origin of replication in BPV-1. Science 236, 1666-1671. STRAUSS, F., and VARSHAVSKY, A. (1984). A protein binds to a satellite DNA repeat at three specific sites that would be brought into mutual proximity by DNA folding in the nucleosome. Cell 37, 889-901. THEILL, L.E., WIBORG, O., and VUUST, J. (1987). Cell-specific expression of the human gastrin gene: Evidence for a con-
trol element located downstream of the TATA box. Mol. Cell. Biol. 7, 4329-4336. TRAUT, W., and FANNING, E. (1988). Sequence-specific interactions between a cellular DNA-binding protein and the simian virus 40 origin of DNA replication. Mol. Cell. Biol. 8, 903-911. WILKISON, W.O., MIN, H.Y., CLAFFEY, K.P., SATTERBERG, B.L., and SPIEGELMAN, B.M. (1990). Control of the adipsin gene in adipocyte differentiation. Identification of distinct nuclear factors binding to single- and double-stranded DNA. J. Biol. Chem. 265, 477-482.
Address reprint requests to: Dr. François Strauss Institut Jacques Monod 2 Place Jussieu 75251 Paris 05, France Received for publication August 1, 1990, and in revised form November 14, 1990.
