WO1993004076A1 - 141 transcription factor and methods of isolating same - Google Patents

Abstract

Transcription factor YY1, which both represses and enhances transcription in adeno-associated virus type 2, herpes and Epstein-Barr viruses as well as oncogenes. YY1 has been isolated, cloned into appropriate expression vectors and its sequence determined.

Description

YY1 TRANSCRIPTION FACTOR AND METHODS OF ISOLATING SAME FIELD OF THE INVENTION

This application relates to the YY1 transcription factor, which is a eukaryotic protein regulating the expression of various genes. The application also relates to the isolation and cloning of the YY1

transcription factor.

BACKGROUND OF THE INVENTION

Transcription factors can be divided into two classes; those that activate and those that repress transcription. A variety of eukaryotic transcriptional activators has been described (reviewed in Johnson and McKnight, 1989; Mitchell and Tjian, 1989). Discrete domains that participate in transcriptional activation have been identified, including the acidic amino acid stretch in GAL4 (Gill and Ptashne, 1987), the glutaminerich sequence in Spl (Courey and Tjian, 1988) and the proline-rich region in CTF/NF-1 (Mermod et al., 1989). Relatively few transcriptional repressors have been described in eukaryotes. The Drosophila Krύppel protein is the only instance in which functional domains capable of mediating repression have been identified. An N- terminal, alanine-rich domain from the Krύppel protein fused to the DNA binding domain of the lac repressor can repress transcription of target genes containing lac operator sequences (Licht et al., 1990). For repression of some target genes, there is also a requirement for a function contained within or very close to the C-terminal zinc finger domain of the Krύppel protein, in addition to its role in DNA binding (Zuo et al., 1991). Recently, a human transcriptional repressor was cloned, termed GCF, that binds to GC-rich sequences (Kageyama and Pastan, 1989). Domains within the protein that mediate

transcriptional repression have not been mapped.

Regulation of eukaryotic mRNA transcription is governed by DNA sequence elements that serve as binding sites for sequence-specific transcription factors (Dynan et al., 1985; Ptashne, 1988; Mitchell et al., 1989).

These include upstream and downstream promoter-proximal elements, enhancers, repressors, and silencers, which modulate the rate of specific initiation by RNA

polymerase II. In addition, the promoter-proximal region between -45 to +30 (relative to the start of initiation) contains two highly conserved motifs, the TATA sequence at around -30 and CA at +1 (Bucher et al., 1986).

Although the TATA element-binding factor TFIID has been purified and cloned from several organisms and has provided invaluable insight into the process of

transcription initiation and its regulation, little is known about factors that interact at the +1 region.

SUMMARY OF THE INVENTION

A sequence located at -50 to -70 of the adeno- associated virus (hereinafter referred to as: AAV) P5 promoter mediates adenovirus EIA-induced transcriptional activation of the promoter (Chang et al., 1989). This same element mediates transcriptional repression in the absence of EIA. Although two distinct cellular proteins were found to interact with the sequence, only one of them, YY1, is involved in transcriptional repression. In addition to the sequence between -50 to -70 (P5-60 site),

YY1 binds to the sequence surrounding the transcription initiation region of the promoter (P5+1 site). Both binding sites are capable of repressing transcription directed by heterologous promoters. EIA not only relieves repression exerted by YY1 but stimulates transcription through the YY1 binding site. Thus, EIA can activate transcription through the same cis element that mediates repression.

The cDNA that encodes YY1 from HeLa cells has been cloned. YY1 was expressed as a GAL4-YY1 chimera. The GAL4-YY1 hybrid protein was able to direct both

repression (Figures 13-15) and ElA-mediated activation (Table 2) to a promoter containing a GAL4 binding site. Given the apparent ability of this factor to mediate opposite effects, i.e., repress or activate

transcription, depending on the intracellular milieu it is named Yin and Yang χ_m

YY1 repressed transcription directed by a TATA element plus initiator sequence both within transfected cells (Figure 6) and in cell-free transcription extracts (Figure 7). YY1 also repressed activity of the SV40 enhancer/promoter when its binding site was inserted into the 5' side of the enhancer element (Table 1), indicating that repression is not limited to a specific promoter or sequence context. A single copy of the YY1 binding site reduced transcription 10 to 75-fold (Figure 6 and

Table l) . The mechanism of repression is unclear, but it likely does not result from steric hindrance or

competition between YY1 and other factors for access to the DNA template (Glass et al., 1988) since the YY1 binding site was located to the 5' side of all known factor binding sites in both constructs tested for repression. The binding of YY1 to its cognate sites may direct positioning of nucleosomes downstream of its binding sites, thereby preventing formation of an active transcription complex, as is thought to be the case for the yeast α2 repressor (Roth et al., 1990). EIA protein relieved YY1-mediated repression of transcription directed by the TATA-initiator sequence (Figure 6). In fact, EIA not only relieved repression but also activated transcription through the YY1 site. The switch from repression to activation caused the activity of YY1-TATA-initiator promoter constructs to be increased 250 to 1,120-fold in response to EIA (Figure 6).

The same protein (YY1) mediates both the repression and activation responses. First, extracts of adenovirus- infected and uninfected cells contain the same amount of

YY1 DNA-binding activity, and the activities in the two extracts formed complexes with an oligonucleotide

containing a YY1 recognition site that co-migrated.

Second, methylation interference experiments indicated that the proteins from the two sources contact the same bases within the YY1 binding site. Third, only one protein, with an apparent molecular weight of 68kD, was purified from infected and uninfected cells by

oligonucleotide affinity chromatography. The 68kD factor might well be modified or associated with different accessory factors in the presence of EIA, but there is no evidence for a second protein with YY1 DNA-binding specificity in infected as compared to uninfected cells.

Approximately 2.3 kb of the YY1-specific cDNA, which includes the entire YY1 coding sequence has been

sequenced as shown in Figure 11 and identified as SEQ ID N0:1. RNA blot analysis was performed using the cDNA to probe HeLa cell mRNA, and a single 2.6 kb band was identified. A consensus sequence for initiation of translation (Kozak, 1984a; Kozak, 1984b) is present near the 5' end of the cDNA (position 241-243), although the reading frame remains open to its 5' side. It seems likely that this AUG serves as the normal translation start site since it is used at good efficiency in a reticulocyte lysate to produce a protein that migrates in SDS polyacrylamide gels as a polypeptide close to 68 kD in size. We do not yet know why the 414 amino acid protein that would be produced by initiation at this AUG site migrates in SDS polyacrylamide gels with an apparent molecular weight of 68kD. 414 amino acids would normally exhibit a molecular weight on the order of 42kD in the absence of post-translational modifications. It is common for nuclear proteins like YY1 to be

phosphorylated; some are also modified with 0-linked carbohydrates.

As used herein, transcription factor YY1 denotes a mammalian transcription factor that can both enhance and repress transcription and which is native to mammalian tissue. The term also refers to any bioactive portions of the YY1 factor that have either the repressor or enhancer functions of YY1, or other activities characteristic of YY1. Natural allelic variations of YY1 may exist in nature and may be distinguished by amino acid differences in the overall sequence or by deletions, substitutions, insertions, inversions or additions of one (or more) amino- acids in the sequence. In addition, the location of and degree of post translational modification might depend on the nature of the host cellular

environment.

The potential exists, through the use of recombinant DNA technology and post-translation modifications to prepare various YY1 derivatives, variously modified by a resultant single or multiple amino acids substitutions, deletions, additions or replacements, or by addition of modifying groups such as phosphates, carbohydrates, etc.

All such allelic variations and modifications resulting in derivatives of transcription factor YY1 are included within the scope of this invention, as long as it maintains either YY1's characteristic enhancement or repression of transcription activity. Alternatively, such derivatives may maintain the ability to bind to the characteristic YY1 binding sites.

YY1 contains four C₂H₂-type zinc fingers (underlined in Figure 11) that exhibit sequence similarity (73% identity) to those of the REX-1 protein (Hosier et al., 1989). REX-1 is a zinc finger protein whose expression is rapidly reduced by retinoic acid treatment of F9 teratocarcinoma cells. Its DNA recognition site, as well as its functions remain unclear. The YY1 zinc finger motifs, as well as those of REX-1, are related to those of the GLI-Krύppel family of genes (Ruppert et al., 1988). Three of the four fingers in YY1 belong to the GLI subgroup with the consensus amino acid sequence:

[Y/F]XCX₃GCX₃[F/Y]X₅LX₂HX_3-4H[T/S]]GEKP, and one belongs to the Krύppel subgroup with the

consensus sequence:

[Y/F]XCX₂CX₃ FX₅1X₂ HXRXHTGEKP. As discussed above, YY1 can either repress or activate depending on the intracellular milieu. The Drosophila Krύppe^'l protein can also repress or activate transcription, depending on the context of its binding site (Frasch and Levine, 1987; Licht et al., 1990;

Ruppert et al., 1988).

Two different domains in the Krύppel protein have been reported to exhibit repression activity when fused to a heterologous DNA-binding domain. The first is a sequence very close to or within the zinc finger domain (Zuo et al., 1991), the region of similarity between YY1 and Krύppel proteins. The second is an alanine-rich (32%) domain (Licht et al., 1990). YY1 as identified by SEQ ID NO:l does not contain an alanine-rich sequence, but it does contain a glycine-rich (42%) segment between amino acids 157 to 201 (Figure 11); which, given the similarity of glycine and alanine residues might serve the same function. It is not known whether the Krύppel protein contains an activation domain or performs this function by interacting with an adaptor protein; however we have identified an acidic domain between amino acid 12 to 53 of YY1 shown in SEQ ID NO:1 (Figure 11) that can activate transcription when removed from the context of the YY1 protein and fused to a heterologous DNA-binding domain. It is of note that YY1 contains a stretch of 11 consecutive histidine residues between amino acids 70-80. The basic histidine stretch might conceivably neutralize the putative activating function of the nearby acidic domain under repressing conditions.

As mentioned above, the YY1 protein can bind to its recognition site with equal efficiency whether present in uninfected or adenovirus-infected, EIA-containing

extracts and a GAL4-YY1 fusion protein can repress in the absence and activate in the presence of EIA (Figures 13- 15 and Table 2). Therefore YY1 is likely bound at its recognition site under both repressing and activating conditions. Given this assumption, two general classes of model can be envisioned (Figures 16 and 17). Both predict an alteration of the functional status of YY1 by masking or exposing its putative activation and

repression domains. The change could result from a conformational alteration due to a covalent modification (Figure 16). It is also possible that YY1 physically interacts with the EIA protein to bring EIA to the promoter (Figure 16). It has been shown that EIA stimulates transcription when brought to the vicinity of a promoter (Lillie and Green, 1989). Alternatively, the switch from repression to activation could be the consequence of an interaction between YY1 and an

accessory protein (Figure 17).

In the absence of a helper virus, adeno-associated virus normally integrates into the genome of its host cell and remains transcriptionally silent. YY1 binding within the P5 promoter presumably helps maintain the viral chromosome in its latent state. A domain that represses transcription within the Epstein-Barr virus BZLFl promoter includes a YY1 binding site. Apparently,

YY1 helps to regulate expression of the BZLFl gene, whose product, in turn, mediates the switch from latent to lytic infection (Rooney et al., 1988). Thus it is possible that YY1 plays a role in the maintenance of latency in several different virus systems.

We have identified putative YY1 binding sites in the promoters of several adenovirus early genes. This raises the possibility that one aspect of EIA function within an adenovirus-infected cell is to convert YY1 from its repressing to activating mode. These putative YY1 binding sites also suggest an explanation for the fact that adenovirus mutants unable to produce the EIA protein exhibit efficient transcriptional activation if used to infect cells at high input multiplicities (Shenk et al., 1980). Presumably, high viral DNA copy number could titrate out the repressor and facilitate activation in the absence of the EIA transactivator protein.

Some of the better characterized proto-oncogenes control transcription. For example, the dominant

oncogenes jun and fos are positive acting transcription factors which together form the activity known as AP-1. Recessive oncogene, pRB, appears to repress the activity of the EZF transcription factor. Since YY1 influences transcription, it is possible that YY1 might exhibit oncogenic properties if it were mutated, over expressed, or not expressed at all within a cell. If YY1 exhibits oncogenic properties, then antisense nucleic acids, antibodies and chemotherapeutic agents that specifically target the protein could prove valuable in cancer

diagnosis and treatment.

DESCRIPTION OF THE DRAWINGS

Figure 1. Identification of YY1 binding activity in HeLa cells. Band shift assay of YY1 binding activity in HeLa cells. ³²P-labeled P5-60 oligonucleotide, corresponding to the sequence from -49 to -71 of the AAV P5 promoter was used as substrate for binding with crude HeLa nuclear extracts. P5ML is an oligonucleotide containing an MLTF (major late transcription factor) binding site (Chang et al., 1989). Numbers above the lanes indicate the molar excess of unlabeled, competitor DNAs. Bands

corresponding to YY1 complex and free DNA are labeled.

Figure 2. DNase I footprint analysis of affinity

purified YY1 on the AAV P5 transcriptional control region. The left most lanes (lanes 1 and 2) contain the "P-labeled P5 promoter DNA fragment (+24 to -96)

subjected to chemical cleavage to produce a sequence ladder. Reaction mixes received no protein (lanes 3 and 4), aff inity purified factor 2 (lane 5) or affinity purified YY1 (lane 6). The arrows to the left of the autoradiogram designate the EIA-responsive element, and the bars to the right mark the protected sequences. Figure 3. Identification of specific sequences required for YY1 binding in the AAV P5 promoter. Band shift assay with a ³²P-labeled, partially methylated P5 promoter DNA fragment. Affinity purified YY1 protein was allowed to bind to the ³²P-labeled P5 promoter fragment (+24 to -96) . Numbers above the lanes indicate the relative amount of

YY1 protein used. Complexes corresponding to YY1 binding at the P5 +1 site only (BI), +1 and-60 (BII), as well as free probe DNA, are labeled.

Figure 4. Methylation interference analysis of YY1 binding sites in the P5 promoter. The methylation interference pattern on the non-coding strand of the P5 promoter is shown. The left-most lane represents

methylated free DNA cleaved by piperidine to generate a sequence ladder. Lanes 2 and 3 correspond to BI and BII in (Figure 3) after chemical cleavage. Interference at the +1 and the -60 positions are indicated at the right. Strong methylation interference bases are indicated by solid dots. Weak interference sites are indicated by solid rectangles.

Figure 5. Wild-type and mutant YY1 binding site

sequences at both P5-60 and P5+1 positions. Proteins that bind to the P5 transcriptional control region are schematically diagrammed. YY1 is shown as a stippled rod; factor 2 is represented by an open oval; MLTF is shown as a left-ward striped rectangle and TFIID is a right-ward striped rectangle. Wild-type and various mutant derivatives of P5-60 and P5+1 YY1 binding

sequences are listed. Strong methylation interference bases are indicated by solid dots. Weak interference sites are indicated by solid rectangles. Changed bases in the mutant oligonucleotides are underlined. Figures 6-8. Transcriptional repression and activation of heterologous synthetic promoters by YY1 binding elements and EIA. Figure 6. Assay results with different CAT plasmid constructs transfected into HeLa cells in the presence or absence of a plasmid encoding the EIA protein. The extent of acetylation in various reactions was determined relative to that for pTICAT, and results are presented as the mean + SD of three independent transfections.

Construction of each plasmid as well as the sequences of various wild-type and mutant YY1 binding sites are described in Figure 5 and pg 19 & 20 of the

specification.

Figure 7. Representative autoradiogram of in vitro transcription from synthetic promoters. Results are presented as means + SD of three independent reactions using at least two separate preparations of template DNA and nuclear extracts.

Figure 8. Band shift analysis of YY1 binding to

oligonucleotides containing wild-type and mutant YY1 recognition sites. ³²P-labeled P5-60 and P5+1

oligonucleotides were used as substrates for binding with partially purified YY1. P5ML is an oligonucleotide containing an MLTF binding site. Numbers above the lanes indicate the molar excess of unlabeled, competitor DNAs. Bands corresponding to YY1 complex and free DNA are labeled. Refer to Figure 5 for the exact sequences of both wild-type and mutant P5-60 and P5+1 YY1 binding sites. Figures 9-10. Visualization of purified YY1 protein and determination of its binding specificity.

Figure 9. Silver staining of YY1 protein at different purification stages. The left-most lane contains

molecular weight markers with sizes indicated (kD, kilodalton). Lane 1, crude HeLa nuclear extract; lane 2, flow through from HIC chromatography; lane 3, HIC column fraction containing no YY1 activity; lane 4, peak HIC fraction containing YY1; lane 5, eluate from the first round of affinity column chromatography; lanes 6 to 10, flow through and sequential washes from the first round affinity column chromatography; lanes 11 to 13, purified

YY1 protein eluted from second round affinity column purification.

Figure 10. Band shift assay with renatured YY1 protein.

³²P-labeled P5-60 oligonucleotide was reacted with

affinity purified YY1 (lane 1); flow through from the affinity column (lane 2); denatured and renatured YY1 protein (lanes 3 and 5); another protein of unknown identity (lanes 4 and 6). The bands representing YY1 complex and the free DNA are labeled. Figure 11. Nucleotide and protein sequences of YY1.

The cDNA sequence of YY1 and the predicted protein sequence are shown. Peptide sequence numbering starts at the putative translational initiation codon AUG

(nucleotide 241). The stop codon is denoted by an asterisk. A stretch of eleven consecutive acidic

residues (aa 43-52) and eleven histidine residues (aa 70- 80) is underlined as are the four zinc-finger sequences near the C-terminus of the protein and potential polyadenylation signals within the 3' untranslated region.

Figure 12. Bacterially synthesized YY1 protein binds specifically to its cognate sites. Ten ng of purified, renatured YY1-HIS fusion protein was incubated with the ³²P-labeled P5-60 oligonucleotide (YY1 site at -60).

Oligonucleotide P5+1 contains YY1 binding sequence at the cap site of the P5 promoter. AP-1 represents an

oligonucleotide containing an AP-1 binding site whose sequence is taken from the promoter of the human

collagenase gene (Mύller, et al., 1989). Numbers above the lanes indicate the molar excess of unlabeled

competitor DNA. Bands corresponding to YY1 complex and free DNA are labeled.

Figures 13-15. GAL4-YY1 fusion protein represses

transcription from a promoter with a GAL4 binding site. Figure 13. Construction of the target and effector plasmids. pGAL4-TKCAT contains five GAL4 binding sites placed upstream of the TK TATA box in plasmid pBL2CAT. pGAL4-YY1 contains the entire coding region of YY1 cDNA with GAL4 (1-147) fused to its N-terminus. pGAL4-YY1Δ is a derivative of pGAL4-YY1, lacking 83 amino acids at the C-terminus.

Figures 14-15. Assays were performed by transfecting either HeLa (Fig. 14) or NXE 3T3 (Fig. 15) cells. CAT assay results were plotted with mean±SD from three independent transfections. The line connected by solid dots represents data collected from co-transfeetion with GAL4 (1-147) only. The line connected by solid squares represents data collected from co-transfection with GAL4 YY1. The line connected by solid triangles represents data collected from co-transfection with pGAL4-YY1Δ. The line connected by open squares represents co-transfectio of pGAL4 YY1 with TKCAT. Each co-transfection assay contained lOμg of target plasmids.

Figures 16-17. Models for YY1-mediated transcriptional repression and activation. Figure 16. YY1 protein as a repressor is represented by a shaded rectangle. YY1 as an activator is shown as a shaded oval. EIA protein is represented by a dark square. In the absence of EIA, YY1 represses

transcription. When EIA is present, YY1 is converted from a repressor to an activator by either modifying it or by physically associating with it.

Figure 17. YY1 protein exists as a transcriptional activator with the activation domain being blocked by another cellular protein, represented by a dark circle.

EIA dissociates the cellular protein and unmasks the transcriptional activation domain of YY1.

Figures 18-19. P5+1 element can function as a

transcription Inr.

Figure 18. In vitro transcription of the F5 promoter.

Wildtype and mutant templates, illustrated at the top of each panel, were transcribed in HeLa nuclear extracts, and product RNAs were assayed by reverse transcription.

The expected size of the reverse transcripts are shown on the left side of the autoradiogram. Molecular weight markers are derived from ³²P-labelled Msp I-digested pBR322 fragments. Figure 19. Specific transcription with the P5+1 element.

Wildtype and mutant templates, illustrated at the top of each panel, were transcribed in HeLa nuclear extracts, and product RNAs were assayed by reverse transcription. The expected size of the reverse transcripts are shown on the left side of the autoradiogram. Molecular weight markers are derived from ³²P-labelled Msp I-digested pBR322 fragments. Figures 20-23. Transcription directed by the F5+1 element in the absence and presence of YY1 activities.

Figure 20. HeLa nuclear extracts were depleted for YY1 activities by two sequential passages through a YY1- specific DNA affinity column. Ξlectrophoretic mobility shift assay (EMSA) was used to monitor YY1 activity.

Figure 21. HeLa nuclear extracts depleted for YY1.were used to transcribe template P5+1. Anti-YY1 or preimmune antibodies were added where indicated. ,Primer extension products specific to the template are indicated by arrow.

Figure 22. Drosophila embryo extracts devoid of YY1 were used to transcribe template pP5+l. Anti-YY1 or preimmune antibodies were added where indicated. Primer extension products specific to the template are indicated by arrow.

Figure 23. Western blot analysis of YY1. The arrow indicates the expected size of YY1 protein from Hela extract. Prestained high-molecular-weight markers (BRL) were used as standards.

Figures 24-26. A YY1-related factor regulates

transcription of the Tdt and LeZF-J Inr. Figure 24. Similarity of initiation sequences from the P5, Tdt, and LeIF-J promoters. Nucleotides outlined indicate sequences conserved between the P5 Inr and the Tdt Inr or between the P5 Inr and the LeIF-J Inr.

Sequences underlined are inverted repeats. +1 represents the major transcription start site.

Figure 25. In vitro transcription of the LeIF-J+1 sequence elements. Reverse transcription products of the RNA are indicated by arrow.

Figure 26. EMSA of YY1-related factor binding to the Tdt and LeIF-J Inr. Formation of complex I is specifically inhibited by addition of excess Tdt Inr and the P5 Inr oligonucleotides but not by the addition of an API oligonucleotide. Complex II and III are nonspecific.

Complex IV is specifically competed by addition of excess LeIF-J Inr and the P5 Inr oligonucleotides but not by the addition of an API oligonucleotide. Complex V is

nonspecific.

DETAILED DESCRIPTION OF THE INVENTION

A. Transcription Repression by YY1, Identification of Two Cellular Proteins That Interact with the

P5-60 Element

We have studied the cellular DNA binding proteins that interact with the element centered at -60 of the adeno-associated virus P5, an element which was

previously shown to confer EIA responsiveness to a heterologous promoter. A 22 bp oligonucleotide containing this sequence, shown in Figure 5 and identified as SEQ ID NO:3 (P5-60), was used to search for specific DNA binding proteins by band shift assays. Assays for Protein-DNA Interaction and for CAT Expressio

Band shift assays and footprint analysis were performed as described (Chang, et al., 1989). For methylation interference analysis (Ausubel et al., 1988), the end-labeled P5 promoter fragments (Chang et al.,

1989) were partially methylated and used as substrates for band shift assays with highly purified YY1

preparations. The complexes formed, as well as the free DNA fragments, were excised from the gel. DNA was recovered by elution, cleaved with piperidine and assaye on a 6% acrylamide gel containing 8 M urea.

As shown in Figure 1, when analyzed with crude HeLa nuclear extract (Dignam et al., 1983), a sequence- specific DNA protein complex (complex I) was detected. The formation of the complex was prevented by addition o excess unlabeled P5-60 oligonucleotide but not by an oligonucleotide (P5ML) containing the binding site for the major late transcription factor. The cellular protein that is responsible for the formation of this complex has been purified (described below) and will be referred to as YY1 in this application. A second

sequence-specific DNA protein complex was present but was masked by a co-migrating complex formed by a non-specifi DNA binding activity in HeLa cells (complex II, Figure 1). This protein (termed factor 2) was detected after removal of the non-specific DNA-binding activity by chromatography.

To further characterize the specific DNA-protein interaction between the YY1 protein and its binding sites, DNase I footprinting experiments were performed with DNA affinity purified YY1 protein. The substrate used was the P5 transcriptional control region (-96 to +24) taken from plasmid P5-CAT190 (Chang et al., 1989). A footprint covering the P5-60 sequence element was readily visible (Figure 2, lane 6). In addition, a second footprint was also detected at the transcription initiation region (P5+1 sequence) of the promoter

(Figure 2, lane 6). To test whether YY1 binds to P5+1 sequence, an oligonucleotide corresponding to the sequence between -10 to +13 of the P5 promoter was synthesized (P5+1), as shown in Figure 5 and hereinafter identified as SEQ ID NO: 7. Both the ³²P-labeled P5-60 and P5+1 oligonucleotides were used in band shift assays. A sequence-specific DNA-protein complex formed with the P5- 60 oligonucleotide could be competed efficiently by unlabeled P5-60 as well as by P5+1 oligonucleotide.

Similarly, a co-migrating complex formed with P5+1 could be competed by unlabeled P5+1 and by P5-60

oligonucleotide. This was additional evidence that the same protein, namely YY1, binds to both the P5-60 and the P5+1 sequences.

To delineate the nucleotides within the P5-60 and P5+1 elements that are in close contact with bound YY1 protein, methylation interference experiments were performed (Siebenlist and Gilbert, 1980). As shown in Figure 3, two specific DNA-protein complexes (BI and BII) were formed when a partially methylated P5 promoter fragment (non-coding strand labeled) was incubated with DNA affinity purified YY1 protein. These complexes as well as the band corresponding to free probe DNA were excised from the gel. The DNA in the complexes was recovered by elution, cleaved by piperidine and analyzed on a 6% polyacrylamide gel containing 8 M urea. Figure 4 shows the methylation interference patterns observed on the non-coding strand of the P5 promoter. The DNA- protein complex BI in Figure 3 corresponds to YY1 binding at P5+1 site only (Figure 4, lane 2). Complex BII represents YY1 binding at both P5+1 and at P5-60 sites of the promoter. Additional complexes migrating between BI and free DNA (Figure 3) were also analyzed. No

interference with piperidine cleavage was detected suggesting the complexes did not result from a sequence- specific interaction of YY1 with the DNA. The

methylation interference data, together with the DNase I footprint experiment, define the P5-60 YY1 binding site as the 5'-CGACATTTT-3' portion of SEQ ID NO:3 and the P5+1 YY1 binding site as the 5'-CTCCATTTT-3' portion of SEQ ID NO: 7. The locations of factor binding sites in the P5 promoter are summarized in Figure 5.

YY1 Binding Sites Mediate Transcriptional Repression that Can Be Relieved by the Adenovirus EIA Protein

EXPERIMENT USING TATA BOX

The P5-60 element capable of mediating EIA-induced transcriptional activation contains partially overlapping binding sites for two cellular proteins, YY1 and factor 2. In order to determine whether the YY1 recognition sequence alone would confer EIA-responsiveness, we engineered a basal promoter construct (pTI) composed only of an initiator sequence (Smale and Baltimore, 1989) and a TATA box.

pTI contains a minimal synthetic promoter similar to that described by Smale and Baltimore (1989). It was constructed by inserting oligonucleotides containing the Tdt initiator and the TATA sequence from the adenovirus major late promoter into BamHI/SacI and EcoRI/SacI sites of pSP72, respectively. Oligonucleotides containing the P5-60 or P5+1 sequence, and various mutant derivatives thereof were inserted into the EcoRV site of pTI (-50 from initiation of transcription). This same set of promoter constructs was used for in vitro transcription analysis or in vivo transfection experiments with the CAT reporter gene at +52, downstream of the transcription initiation site.

Oligonucleotides containing the P5-60 or P5+1 sequence, as well as various mutant versions of the two sequences (as shown in Fig. 5) were placed directly upstream of the TATA box in construct pTI. To test the functions of these binding sites within transfected cells, the chloramphenicol acetyltransferase (CAT) reporter gene was introduced downstream of the initiator sequence in all the constructs described above.

Surprisingly, the presence of the P5-60 element caused a 10-fold reduction in CAT activity within transfected HeLa cells (Figure 6, compare -EIA column, lines 1 and 3). This reduction in CAT activity was specifically mediated by binding of the YY1 protein since mutations that prevented YY1 binding also alleviated the repression. P5-60 (mtl), shown in Figure 5 and identified as SEQ ID NO: 4, contains only half of the P5-60 element, and P5-60 (mt3), shown in Figure 5 and identified as SEQ ID NO: 6, has two base substitutions (Figure 5). -Neither mutant element caused repression (Figure 6, -EIA column, lines 2 and 4), and band shift assays indicated that both of these mutated sites were defective for YY1 binding

(Figure 8), indicating that YY1 is responsible for repression.

EXPERIMENT WITH SV40 PROMOTER

The repression is not likely to be due to simple steric hindrance by YY1 of TFIID binding since insertion of Spl binding sites at the identical position as the YY1 binding site has been shown previously (Smale and

Baltimore, 1989) to activate the artificial promoter studied here. Further, as shown below, the YY1 binding site can mediate repression when inserted upstream of th SV40 promoter/enhancer domain.

A 10-fold reduction in CAT activity was also

observed for pP5+1CAT which carries the P5+1 sequence (Figure 6, compare -EIA column, lines 5 and 1).

Mutations in the P5+1 sequence (Figure 5) that abolished

YY1 binding (Figure 8) relieved repression and fully restored CAT activity (Figure 6, -EIA column, lines 6 an 7). This reinforces the notion that YY1 binding is necessary and sufficient to repress transcription.

In vitro transcription experiments with nuclear extracts derived from HeLa cells were performed to confirm the in vivo CAT assay results and to demonstrate directly that the repression occurred at the level of transcription (Figure 7).

Each cell free transcription reaction (25 μl) contained 100 ng template DNA, 12 mM HEPES (pH 7.9), 12% (v/v) glycerol, 60 mM KCl, 0.12 mM EDTA, 0.3 mM PMSF, 0.3 mM DTT, 10 mM MgCl₂, 500 ng poly [dG-dC], 0.5 mM ATP, 0.5 mM GTP, 0.5 mM CTP, 0.5 mM DTP, 1 mM creatine phosphate, and approximately 72-96 μg of HeLa cell nuclear extracts. Reactions were incubated for 1 hr at 30°C, and terminate by the addition of 225 μl of stop buffer (10 mM Tris-HCl [pH 7.4], 10 mM EDTA, 1% SDS, 20 μg/ml yeast tRNA).

Following proteinase K treatment, phenol/chloroform extraction and ethanol precipitation, template DNAs were degraded with RNase-free DNase I for 1 hr at 37°C in the presence of ribonuclease inhibitors. The RNA products were analyzed by primer extension assay (McKnight and Kingsbury, 1982).

Reaction products were assayed by primer extension and the expected product of 79 nucleotides is denoted by an arrow at the right as shown in Figure 7. Reaction products were quantified relative to those produced by pTI, and results are presented as mean±SD of three independent reactions using at least two separate preparations of template DNA and nuclear extracts.

Wild-type P5-60 and P5+1 sequences repressed

transcription about 5 and 2 fold, respectively (Figure 7, lanes 3 and 5) while mutated derivatives incapable of binding YY1 did not (Figure 7, lanes 2, 4, 6 and 7). The results obtained from the in vitro transcription assay differed from those of the CAT assay in one aspect; while the two elements exhibited similar levels of repression in vivo, the P5+1 sequence mediated repression less well than the P5-60 element in vitro .

To determine if EIA protein can relieve

transcriptional repression caused by YY1, the pXC15 plasmid which expresses the Ad5 EIA and E1B genes

(Hearing and Shenk, 1983) was co-transfected with the same set of CAT constructs into HeLa cells (Figure 6, +E1A column). The TATA sequence in the constructs was taken from the major late promoter of adenovirus

(Sawadogo and Roeder, 1985), and it has been implicated in mediating EIA transcriptional activation of the adenovirus major late gene (Leong et al., 1988).

Consistent with this observation, pTICAT carrying this TATA element exhibited about a 7-fold induction of CAT activity (Figure 6, line 1) when co-transfected with the EIA-expressing plasmid. Subtracting the effect of EIA through the TATA sequence, we observed about a 4-fold increase in CAT activity when the YY1 binding site was present (Figure 6, compare lines 5 and 1). This suggests that YY1 alone can mediate EIA-induced transcriptional activation. In contrast, factor 2., which was not

involved in repression, mediated EIA-induced

transcription about 2-3 fold (Figure 6, lines 2 and 4). Together, YY1 and factor 2 exhibited a 17-fold induction of CAT activity in the presence of EIA (Figure 6, line 3). The ability of EIA to activate transcription through the P5-60 YY1 binding site is consistent with our earlier study (Chang et al., 1989) in which the P5-60 element mediated an 8 fold induction of specific transcription directed by the SV40 early promoter and initiated at the appropriate start site.

To summarize, binding of YY1 to its cognate sites causes transcriptional repression. The transcriptional repression can be relieved by adenovirus EIA protein. In addition, YY1 responds to EIA by activating transcription since a functional YY1 binding site, in the presence of EIA, stimulates transcription to a greater extent than observed in constructs carrying mutant sites unable to bind the factor. This implies that YY1 is a

transcriptional repressor and can be converted into a transcriptional activator by EIA.

To ascertain that the transcriptional repression mediated by YY1 in the context of the initiator sequence plus TATA box is not a promoter-specific phenomenon, we constructed another set of recombinant CAT plasmids carrying either the P5-60 or the P5+1 or mutant sequence upstream of the entire SV40 enhancer/promoter domain (pSVECAT) (see Figure 5 for the sequences of the wild- type and mutant YY1 binding sites). The pSV40ECAT recombinants carrying wild-type and mutant P5-60 or P5+1 sequences were constructed by inserting double stranded oligonucleotides into pSV40ECAT at the BamHI site (5' to the SV40 enhancer). Two μg of each plasmid was

transfected into HeLa cells. The results are expressed as mean ± SD of three independent transfections. The relative CAT activity reported as 1 represents an average CAT conversion of 65%. The results are shown in Table 1 below.

pP5-60SV40ECAT contained multiple copies of the wild-type P5-60 sequence, identified as SEQ ID N0:3; pP5- 60(mt2)SVECAT, identified as SEQ ID NO:5, contained multiple base-pair substitutions in the P5-60 sequence and showed substantially diminished binding for both YY1 and factor 2 as determined by band shift assays. The wild-type P5-60 sequence caused at least a 15-fold reduction of CAT activity compared to pSV40ECAT with no insert (Table 1, compare pP5-60SVECAT and pSVECAT). In contrast, the mutant P5-60 sequence pP5-60(mt2)

identified as SEQ ID NO:5 failed to repress CAT activity, suggesting that either YY1, factor 2 or both contributed to repression. To determine if YY1 binding alone can repress the SV40 enhancer activity, the P5+1 sequence that only binds YY1 was tested.

A single copy of the P5+1 sequence (pP5+lSVECAT) repressed expression by a factor of 75 while mutant derivatives of the P5+1 sequence [pP5+l(mtl)SVECAT identified as SEQ ID NO:8 or pP5+l(mt2)SVECAT identified as SEQ ID NO:9 and shown in Figure 5], that were unable to bind YY1 (Figure 8) failed to repress.

When the P5+1 sequence was inserted 3' to the CAT sequence in pSV40ECAT, no reduction in CAT expression was detected (data not shown). Taken together, these data demonstrate that YY1 can repress activity of the SV40 promoter/enhancer when its binding site is located 5' to the transcriptional control region.

Isolation of a YY1 cDNA Clone

YY1 activity was purified by hydrophobic interaction column chromatography followed by two rounds of

oligonucleotide affinity chromatography. Nuclear extract was prepared from HeLa cells grown in spinner culture supplemented with 10% calf serum as described (Dignam et al., 1983). YY1 activity was partially purified from HeLa nuclear extract by chromatography on a TSK phenyl- 5PW matrix in an LKB UltroPac HPLC column (Pharmacia LKB Biotechnology, Inc., Piscataway, N.J.). Peak fractions containing YY1 activity were pooled and dialyzed to a final concentration of 0.1 M KCl.

To produce an oligonucleotide affinity matrix for purification of YY1, a linker containing a BamHI cleavage site was added to the ends of an oligonucleotide

containing the P5-60 sequence and self-ligated to

generate multimers of 10-15 copies. They were then coupled to CNBr-activated Sepharose beads (Pharmacia CNBr-activated Sepharose 4B). Pooled, partially purifie

YY1 activity (peak fractions from TSK phenyl-5PW

chromatography) was passed through the column at a flow rate of 15 ml/hr. The column was washed extensively with loading buffer containing 25 mM HEPES-KOH, 12.5 mM MgCl₂, 10% glycerol, 0.1% NP40, 0.1 mM KCL, 1 mM DTT. YY1 activity was eluted with loading buffer containing 0.6 M KCl. For the second round of purification through the DNA-affinity column, the eluate from the first round was diluted to final KCl concentration of 0.1 M before it was loaded onto the column. The column was washed and YY1 activity was eluted as describe above. The most highly purified material contained a major polypeptide with an apparent molecular weight of 68 kD (Figure 9). The 68 kD protein band was excised from the gel, protein was recovered by elution, denatured in 6 M guanidine and renatured in DNA binding buffer (Wang, et al., 1987).

Renaturation of Affinity Purified YY1 from a Silver- Stained Gel

Approximately 200 ng of affinity purified YY1 was loaded onto a 12.5% SDS polyacrylamide gel. The 68 kD

YY1 protein was visualized by silver staining (Oakley et al., 1980). The band was then excised out of the gel, minced and incubated with 2 ml of elution buffer

containing 50 mM Tris-HCl (pH 7.9), 0.1 mM EDTA, 5 mM DTT, 0.2 M KCl, 0.2 mg/ml BSA and 0.1% SDS at room temperature for 2 hrs (Wang et al., 1987). The eluted protein was precipitated by addition of 4 volumes of acetone at -80°C and the precipitate was resuspended in 100 μl of 6 M guanidine, 0.1% NP40 and 0.1 M KCl at room temperature. After the purified protein was

resolubilized, guanidine was removed by passage through a G-50 column equilibrated with the binding buffer used for band shift assays. The protein was allowed to renature at room temperature for l hr in the band shift assay buffer before it was used for band shift analysis. As a control, another protein band was chosen randomly and subjected to the same treatment. Band shift assays were performed using ³²P-labeled P5-60 oligonucleotide. The renatured 68 kD protein, but not the control protein, formed a complex co-migrating with that formed with DNA affinity purified YY1 (Figure 10), indicating that we had successfully purified YY1 from HeLa cells to near

homogeneity.

Microsequencing of Purified YY1 Protein and Preparation of Probes.

About 10 μg of purified YY1 protein was subjected to trypsin digestion to generate peptides for

microsequencing.

Affinity-purified YY1 proteins were pooled and concentrated by precipitation with acetone at -80°C overnight. The protein pellet was resuspended in sample buffer, boiled and subjected to electrophoresis on a 12.5% SDS polyacrylamide gel. The protein was

electrophoretically transferred to nitrocellulose paper and visualized by staining with Ponceau S dye (Arbersold, et al., 1987). The region that contained the 68 kD YY1 protein was carefully excised within a one cm² piece of nitrocellulose paper. Subsequent trypsin digestion, recovery of peptides and microsequencing were performed by W. Lane (Harvard University Microsequencing

Facility).

A 21 amino acid sequence, a portion of the sequence identified as SEQ ID NO:2, (DIDHETWEEQIIGENSPPDY) was obtained from one tryptic peptide. Two degenerate oligonucleotides were designed based on the amino acid sequence corresponding to the ends of the sequence: a 17- mer [5'GA(TC)AT(ATC)GA(TC)CA(TC)GA(AG)AC3'], identified as SEQ ID NO: 10, corresponding to the coding strand sequence and a 27-mer corresponding to the opposite strand at the other end of the sequence with inosine replacing nucleotides at the wobble position (5' GTA ATC IGG IGG GTT GTT ITC ICC IAT 3') identified as SEQ ID NO: 11. These two oligonucleotides resulted in a combined degeneracy of greater than 16,000 fold.

To clone the 63 bp DNA fragment by PCR, first strand cDNA synthesized from HeLa poly A⁺ RNA with reverse transcriptase was used as template and primed with the two degenerate oligonucleotides. The PCR cycle

parameters were: 95°C 1 min, 48°C 1 min, 72°C 1 min.

After 40 cycles, a 63 bp DNA fragment was evident, and it was subsequently cloned and sequenced. Analysis of its sequence demonstrated that it matched the peptide from which the oligonucleotide primers were derived.

To obtain a homologous probe for screening a HeLa cDNA library, a 17-mer oligonucleotide (5'-

TGTT CTTCAACCACTGT-3') identified as SEQ ID NO:12 was synthesized based on the internal sequence of the 63 bp DNA fragment and was used as a probe to screen a HeLa λgtll cDNA library (D98/AH-2, a generous gift of T.

Kadesch, University of Pennsylvania). Hybridization to plaques on filters was performed in Denhardt' s buffer at 42°C for 20 hrs. The filters were then washed twice each at 25°C, 37°C, and 42°C, each time for 20 min in 2X SSC, 0.1% SDS. The final wash was done at 42°C for 10 min. in 0.2X SSC, 0.1% SDS. Five positive plaques were isolated by screening 6x10⁵ plaques. Recombinant phage DNA was isolated from plates according to the plate-lysis method (Maniatis, et al., 1982). The inserts were isolated by EcoRI digestion and subcloned into the EcoRI site of pGEM7zf(+) vector (Promega Corp., Madison, WI.). To sequence the inserts, nested deletion constructs were made from both directions using the Erase-a-Base kit (Promega Corp., Madison, WI) with minor modifications. Briefly, the recombinant plasmid was linearized by two restriction enzymes by cleavage at two restriction sites adjacent to each other in the polylinker region to generate both 3' and 5' overhangs. After exonuclease IIl digestion, the DNA was subjected to electrophoresis on a 1% low melting agarose gel. The shortened DNA fragments were excised from the gel and ligation was performed after the low melting agarose was dissolved and diluted. Overlapping deletion constructs were sequenced by the dideoxy method (Sanger et al., 1977).

Figure 11 shows the YY1 cDNA as part of the sequenc identified as SEQ ID NO:l and its amino acid sequence (identified as SEQ ID:2). The cDNA encodes a 414 amino acid zinc finger protein, and it includes the 21 amino acids determined by peptide sequencing of purified YY1.

The identity of the protein encoded by the putative

YY1 cDNA was verified by expressing it as a fusion protein in E. coli, purifying it and testing its ability to bind to YY1 recognition sites. Expression and Purification of a HIS-YYl Fusion Protein Which Binds to YYl Recognition Sites

To construct a HIS-YY1 plasmid, a DNA fragment containing the entire YY1 coding region beginning from the putative translation initiation AUG was isolated by digestion of clone 14-1 with Styl/Hindlll. The ends of this fragment were blunted by Klenow polymerase and cloned into the pDS56-6xHIS vector (Abate et al., 1990; a generous gift of R. Gentz, Hoffman-LaRoche & Co., Basel, Switzerland) in frame at its BamHI site (also blunt ended). The resulting fusion protein contains 12 additional amino acids, MRGSHHHHHHGS, at the N-terminus of the YY1 protein and the six histidine residues allow the protein to be purified by chromatography on a nickel chelate matrix (Hochuli et al., 1987; Abate et al., 1990).

To prepare bacterial lysate containing the HIS-YY1 fusion protein, the fusion plasmid was introduced into a bacterial strain RR (Maniatis et al., 1982). Usually a liter of bacterial culture was grown to optical density of 0.7 to 0.8 units. Synthesis of the fusion protein was induced by 1 mM IPTG. Bacteria were allowed to grow in the presence of IPTG for 2 hrs. They were pelleted by centrifugation and lysed in 12 ml guanidine-HCl (pH 7.8) by incubating overnight at 4°C. The insoluble debris were removed by centrifugation at 18 krpm in a SS34 rotor.

The nickel chelate column was processed as follows: a 1 ml (bed volume) column was first washed with 10 ml distilled H₂O, followed by 8 ml of 0.1 M NiSO₄ and another 2 ml of distilled H₂O the column was then equilibrated with 10 ml of 6 M guanidine-HCl (pH 8.0). Bacterial lysate in the same buffer was loaded and the column washed first with 10 ml of 6 M guanidine-HCl (pH 8.0) and then with 10 ml of 6 M guanidine-HCl (pH 6.0). The fusion protein was eluted with 7 ml of 6 M guanidine-HCl (pH 5.0) and quantified by BioRad protein analysis

(Ausubel et al., 1988). To renature the fusion protein, guanidine was removed by a stepwise dialysis procedure. The eluate was first dialyzed against 1 M guanidine and then 0.1 M guanidine for 2 hours. Subsequently, it was dialyzed against PBS for 2 hours at 4°C twice. The purity of the eluate was assessed by Coomassie Blue staining (Ausubel et al., 1988). The yield of the purified HIS-YY1 fusion protein was about 4-5 mg/liter of bacteria.

Figure 12 displays a band shift assay in which a ³²Plabeled oligonucleotide containing the P5-60 sequence was used as probe. The YY1 cDNA was cloned in frame into the histidine fusion protein expression vector pDS56-6xHIS with the putative translation initiation AUG encoding the first amino acid of YY1 in the fusion protein.

Translation of the fusion protein starts from the AUG in the vector and adds 12 additional amino acids to the Nterminus of the YY1 protein. The fusion protein bound to the labeled probe, and the binding could be competed by addition of excess unlabeled P5-60 or P5+1

oligonucleotide. Thus, the cDNA encodes a protein that binds to YY1 recognition sites.

The Cloned YY1 Protein Represses Transcription and

Responds to EIA

The ability of YY1-specific cDNA to repress

transcription when bound near a promoter was shown by constructing YY1 chimeric proteins with an added DNA binding specificity to distinguish them from endogenous

YY1 activity. Construction of GAL4-YY1 and GAL4-YY1Δ recombinant plasmids

To construct GAL4-YY1 and GAL4-YY1Δ recombinant plasmids, the YY1 cDNA insert was excised from subclone 14-1 by Apal/Clal and Apal/Hindlll digestion. Both Apal and Clal sites are located in the polylinker regions of the vector flanking the cDNA insert, while the Hindlll site is present in the YY1 cDNA at nucleotide position 1235 (Figure 11). The Apal/Clal and the Apal/Hindlll DN fragments, after their ends were blunted by treatment with the Klenow polymerase, were cloned into the EcoRI site (also blunted with Klenow polymerase) of GAL4 (1- 147) expression plasmid. Due to the presence of the polylinker sequence, the fusion proteins encoded by the chimeric genes carry an additional 14 amino acids that connect the C-terminus of the GAL4 sequence to the N- terminus of the YY1 sequence.

All the above recombinants were verified by sequence analysis. The oligonucleotides used for constructing the recombinants are described in Figure 5.

Plasmids containing adenovirus 12S or 13S EIA cDNAs expressed under control of the cytomegalovirus immediate early promoter (Bagchi et al.. 1990) were provided by J. Nevins (Duke University).

The full length YY1 cDNA and a derivative lacking the sequence encoding the C-terminal 83 amino acids of

YY1 were each cloned into the GAL4 fusion vector, pG4 (Ma and Ptashne, 1987). The fusion proteins (GAL4-YY1 and GAL4 YY1Δ) contain the GAL4 DNA binding domain (amino acids 1-147) at their N-terminus (Figure 13), directing the fusion protein to promoters containing GAL4 binding sites. To construct a target gene for the fusion

protein, a plasmid with the CAT gene under the control of TK promoter was chosen. Five GAL4 binding sites (Lillie and Green, 1989) were placed upstream of the TATA box in plasmid pBL2CAT (Luckow and Schutz, 1987) to create pGAL4-TKCAT (Figure 13). When GAL4-YY1 was cotransfected with the target plasmid pGAL4-TKCAT, a marked repression of CAT activity was observed in both HeLa and NIH 3T3 cells (Figure 14 and 15). For transfection assays (Wigler et al., 1978), HeLa cells and NIH3T3 cells were grown on 10 cm dishes in Dulbecco's modified Eagle's medium supplemented with 10% fetal calf serum. Fortyeight hours after transfection, cells were lysed by repeated freeze/thaw cycles and extracts assayed for CAT activity (Gorman et al., 1982).

Repression was dependent on the presence of the GAL4 binding sites because CAT expression was not affected when pTKCAT, lacking GAL4 binding site, was used as a target (Figure 14). The GAL4-YY1Δ protein failed to repress, indicating that the terminal 83 amino acid domain of YY1 includes an element essential for

repression. Both fusion proteins could be detected by immunofluorescence using a polyclonal antiserum to the GAL4 domain, ruling out the possibility that the

truncated fusion protein failed to repress because it was grossly unstable and failed to accumulate within

transfected cells. To ascertain that repression was not caused by the truncated GAL4 protein (amino acids 1-147), the target GAL4-TKCAT gene was co-transfected with the parent plasmid pG4 which only expresses the truncated GAL4 protein. No repression was detected in HeLa cells (Figure 14) and a modest activation was observed in

NIH3T3 cells (Figure 15).

The ability of adenovirus EIA proteins to relieve repression and activate through the fusion protein was examined by cotransfecting HeLa cells with the plasmid encoding the GAL4-YY1 protein, the plasmid carrying the GAL4-responsive reporter construct, and plasmids encoding either the 12S or 13S EIA protein. 10 μg of target plasmids (ρGAL4-TKCAT or pTKCAT) were transfected into HeLa cells with plasmids encoding either the EIA 12S or the EIA 13S mRNAs. The results are expressed as mean ± SD of three independent transfections. (Table 2).

The relative CAT activity reported as 1 represents an average CAT conversion of 15%. As before, the GAL4-YY1 protein inhibited CAT expression by a factor of about 12. The 12S EIA protein relieved the repression, and the 13S EIA protein relieved the repression and further activated CAT expression by a factor of about 3.

In sum, these results indicate that the cloned

YY1 protein, when fused to the GAL4 DNA binding domain, can repress transcription. The repression can be

relieved by either EIA protein, and the 13S EIA protein can activate to a modest extent through the fusion protein. The cloned YY1 protein mimics the behavior of the endogenous activity in all respects.

B. YY1 is an Initiator Sequence-Binding

Protein that Directs and Actuates

Transcription In Vitro

The adeno-associated virus type 2 P5 promoter +1 region (P5+1 element) sequence is necessary and sufficient for accurate basal transcription. Further, partially purified YY1 can restore basal level transcription from a P5+1 element in a HeLa extract depleted for YY1 or a Drosophila embryo extract devoid of

YY1 activity, while a YY1-specific antibody can block the reactivation. Finally, using electrophoretic mobility shift assay, we have identified YY1 related factors that bind to two other transcription initiators in cellular genes.

YY1 interacts with DNA elements centered at -60 and at the transcription initiation region of the adenoassociated virus P5 promoter. Since association of YY1 at the +1 region may point to novel means of

transcriptional control, the effects of P5 promoter deletions that define the 5' and 3' boundaries minimally required for accurate and efficient initiation of

transcription were examined. Using an in vitro

transcription assay, DNA templates extending from

nucleotide -50 or -39 to +24, both of which include only the TATA and P5+1 elements, were actively transcribed (Fig. 18, lanes 2, 3).

With the exception of pΔ-20/+24, plasmids shown in Figure 18 were previously described (Chang et al., 1989). Plasmid pΔ-20/+24 was created by deleting the DNA sequences between the Ava I and Kpn I sites in plasmid p- 50. Plasmids pMLTATA and pSpl (referred to as plasmids II and III, respectively in Smale, Schmidt, Berk and Baltimore, 1990) were provided by Stephen Smale

(University of California, Los Angeles). All other plasmids illustrated in Figure 19 were constructed as described for Tdt plasmids used in similar studies (Smale and Baltimore, 1989; Smale, Schmidt, Berk and Baltimore, 1990) but with the following oligonucleotides instead: (the complements of which are shown in Fig. 5), P5+1 5'- GATCCCGCTTCAAAATGGAGACCGAGCT-3' identified as SEQ ID NO:7; P5+1 mt, 5'GATCCCGCTTCCAACTTTAGACCGAGCT-3' identi- fied as SEQ ID NO: 13 and their complements to produce Bam HI and Sac I ends; and P5TATA, 5 ' -

AATTCTGGGTATTTAAGCCCGAGTGAGGAGCT-3' and its complement to produce Sac I and Eco R1 ends. All constructions were confirmed by DNA sequence analysis (Sanger et al., 1977). HeLa nuclear extracts were prepared by the method of Dignam et al., (1983). In vitro transcription and primer extension (Smale and Baltimore, 1989) were performed as described, except that 400 ng of template DNA was used in each reaction instead. Analysis of RNA in Figure 18 utilized a synthetic oligonucleotide primer complementary to the CAT gene sequence that was present downstream of promoter elements, 5'-GCCATTGGGATATATCAACGGTGG-3'.

Primer extension analysis of RNA synthesized in Figure 19 used a synthetic oligonucleotide complementary to the SP6 promoter sequence in the vector (Promega). All experiments were carried out independently a minimum of three times with at least two separate plasmid and nuclear extract preparations to ensure data reproducibility.

An approximately 5 fold decrease in

transcription was consistently observed when the TATA element was deleted (Fig. 18, lane 4). As expected, no transcription was detected from a promoterless construct (Fig. 18, lane 5). Deletion of sequences from -20 to +24, which removed the P5+1 element but left the TATA box intact, also eliminated detectable transcription (Fig. 18, lane 6). We conclude that the sequence surrounding the P5 transcription start site is sufficient to

independently direct specific initiation of transcription in vi tro . This minimal sequence is reminiscent of the previously defined terminal deoxynucleotidyltransferase (Tdt) gene initiator (Inr) motif (Smale and Baltimore, 1989), which contains the transcription start site. This sequence is sufficient to direct transcription initiation, and its activity is enhanced in the presence of a TATA box.

To confirm that the P5+1 element indeed

possesses the ability to function as an Inr, we repeated experiments with the P5+1 sequence similar to those carried out by Smale and Baltimore (1989) and Smale, Schmidt, Berk and Baltimore (1990). Oligonucleotides bearing sequences from -7 to +11 in the P5 promoter were inserted in vector pSP72 and examined for the ability to initiate transcription in vitro.

As shown in Fig. 19, lane 2, plasmid template pP5+1 produced a transcript of the expected length as detected by primer extension analysis. However, plasmid template with a mutation in the P5+1 sequence so that it no longer binds YY1, did not produce detectable specific product (Fig. 19, lane 3). In agreement with previous studies (Smale, Schmidt, Berk & Baltimore, 1990), we found that the adenovirus type 2 major late promoter TATA box or the SV40 Spl binding sites alone promoted a low level of transcription initiation (Fig. 19, lanes 7,10), but the promoter strength and fidelity increased in the presence of an Inr element (Fig. 19, lanes 8, 11).

Mutation of the P5+1 element such that it no longer binds

YY1 also diminished transcription to its original level s (Fig. 19, lanes 9, 12). In addition, unlike the major late promoter TATA box, we found that the P5 promoter TATA, which is identical to the SV40 promoter TATA sequence, directed a detectable level of transcription only in the presence of the P5+1 element (Fig. 19, lanes 4, 5).

Transcription from the p5+1 Element Requires YY1

To determine whether YY1 is required to direct transcription from the P5+1 element, we used DNA affinity chromatography to prepare a HeLa extract depleted of YY1 DNA binding activity (Fig. 20), and asked whether this depleted extract was still capable of transcribing the pP5+1 template.

Preparation of DNA-affinity column and

purification of YY1 closely followed the method of

Kadonaga and Tjian (1986). Briefly, synthetic

oligonucleotides SEQ ID NO: 14 (5'- CGGGAGGGTCTCCATTTTGAAG-3' and SEQ ID NO: 15 5'- CCCTCCCGCTTCAAAATGGAGA-3') were annealed, ligated, then coupled to sepharose columns. Nuclear extracts were prepared from 36 liters of HeLa cells grown in spinner culture in SMEM medium supplemented with 5% calf serum and antibiotics. After loading the crude fraction (input), the column was washed and the bound protein eluted as described (Kadonaga et al., 1986). EMSA was used to track YY1 activity. Flow-through from the first passage was reapplied to a fresh column to further deplete YY1 activity. Eluate fractions containing high levels of YY1 activity, as monitored by electrophoretic mobility shift assay (EMSA) (data not shown), were pooled and diluted with 6 volumes of buffer Z (20mM HEPES pH 7.8, 12.5 mM MgCl₂, ImM DTT, 20% glycerol, 0.1% NP40) (Kadonaga et al., 1986); and reapplied to a second column to further purify YY1. Fractions highly enriched in YY1 (Figure 20, eluate fractions 7, 8, 9) from this second pass were pooled, dialyzed against buffer D (20 mM HEPES pH 7.9, 20% glycerol, 0.1 M KCl, 0.2 mM EDTA, 0.5 mM PMSF, 0.5 mM DTT) (Dignam et al., 1983) and quantified with the BioRad protein assay for use in subsequent in vitro

transcription and western blot experiments.

EMSA using the P5+1 oligonucleotides were performed as previously described (Chang et al., 1989) with the exception that the binding reactions contained 12 mM HEPES pH 7.9, 10% glycerol, 5 mM MgCl₂, 60 mM KCl, 1 mM DTT, 50 μg/ml BSA, 0.5 mM EDTA, and 0.05% NP40.

Eluate fractions were diluted 1:48 before use for EMSA. For in vitro transcription, HeLa nuclear extracts were prepared as described (Dignam et al., 1983), Drosophila embryo extracts were obtained from Stratagene.

Transcription and primer extension were performed as described in Fig. 19 with the following modifications: Ih vitro transcription was performed by preincubating template pP5+l at 0°C with either buffer D alone ( Figure 21, lane 2 and Figure 22, lane 1), various amounts of protein fractions (Figure 21, lanes 3-11; Figure 22, lanes 3-9), 70 μg of HeLa extract (Figure 21, lane 2), and 1 μl of either pre_lιπmune or anti-YY1 antibody (Figure 21, lanes 10, 11; Figure 22, lanes 8, 9) for 15 min before addition of either HeLa nuclear extract (Figure 21, lane 1), or YY1-depleted extract (Figure 21, lanes 2- 11), or Drosophila embryo extract (Figure 22, lanes 1-9). Transcription was initiated by addition of ribonucleoside triphosphates and was allowed to proceed for 1 hr at 30°C (Figure 21) or 30 min at 25°C (Figure 22).

Polyclonal antibody against YY1 was prepared by immunizing rabbits with gel-purified YY1 fusion protein (HIS-YY1) produced in E. coli as previously described. Preimmune and anti-YY1 antibodies were purified using Econo-Pac Protein A columns (Bio-Rad) before use.

Proteins (30 μg each, Figure 23 , lanes 1 , 2 ; 0.625 μg, Figure 23, lane 3) were prepared in SDS denaturation buffer before being separated on a 12.5% SDS- polyacrylamide gel and electroblotted onto a

nitrocellulose membrane. The presence of YY1 was

determined using the alkaline phosphatase procedure

(Harlow and Lane, 1988) with anti-YY1 primary antibodies and a biotinylated anti-rabbit secondary antibody. As shown in Fig. 21, lane 2, transcription from the P5+1 element was barely detectable when the depleted fraction was used as the source of polymerase and transcription factors, although this identical extract supported transcription from an adenovirus 5 major late promoter template (data not shown). With the addition of fractions enriched in YY1, a much higher level of

transcription was observed (Fig. 21, lanes 3-8). This activation of transcription is highly specific and requires YY1, since fractions with very low YY1 activity were unable to activate transcription (Fig. 21, lane 9).

A YY1-specific antibody was employed to be certain that the key factor removed by the DNA affinity chromatography was, indeed, YY1. The antibody was made by immunization of a rabbit with a YY1 fusion protein produced in E.coli, and it can specifically block

formation of a YY1-DNA complex in a band shift assay. By western blot analysis, this antibody detects a single polypeptide migrating at relative molecular mass of

68,000 (M.68K), as expected for YY1, in the fraction retained on the DNA affinity matrix but not in the depleted extract (Fig. 23, lanes 2, 3). Purified YY1- specific antibody, but not preimmune antibody, was able to reverse the activation effect (Fig. 21, lanes 10, 11). Taken together, these data strongly suggest that YY1 is an Inr regulatory protein that can activate transcription by binding to the P5+1 element.

Initiator Sites Besides P5+1 Can Compete for YYl Binding

While general RNA polymerase II transcription factors from Drosophila and HeLa cells are functionally interchangeable (Heiermann and Pongs, 1985), many

sequence-specific DNA-binding proteins that interact with promoter and enhancer elements are not conserved between insects and mammals. YY1 activity is undetectable in Drosophila embryo extracts and, as shown in Fig. 22, lane 1, Drosophila embryo extract alone is unable to support transcription of the P5+1 element. However, when

complemented with either a HeLa nuclear extract or with fractions enriched in YY1, transcription was readily observed (Fig. 22, lanes 2-7). Again, transcription required a factor immunologically related to YY1, since

YY1-specific antibody blocked the activation (Fig. 22, lane 9).

Cellular factors that bind to the immediate vicinity of the +1 region have previously been identifie in several promoters (Cordingley et al._f 1987; Parks et al., 1988; Means et al., 1990; Boyer et al., 1990; Jones et al., 1988; Kato et al., 1991; Stenlund et al., 1987;

Sawadago et al., 1985; Van Dyke et al. 1988; Nakajima et al., 1988 and Garfinkel et al., 1990). The mouse mammary tumor virus (Buetti et al., 1983; Majors et al., 1983) and (Lee et al., 1984), the adenovirus E1B9, the

dihydrofolate reductase (Means et al., 1990), the

triosephosphate isomerase (Boyer et al., 1990), and the human immunodeficiency virus type (Jones et al., 1988) and (Okamoto et al., 1990) promoter +1 regions bear no obvious homology to the P5+1 sequence and unlike the P5+1 element, these sequences are insufficient for basal transcription in the absence of upstream or downstream elements. The bovine papilloma virus PI promoter

contains a novel Spl binding site (Li et al., 1991) immediately downstream from the start of transcription, and has not yet been tested for its ability to

independently direct transcription. The +1 region of the adenovirus major late promoter, by itself, was able to direct transcription initiation in one report (Smale and Baltimore, 1989), but not another (Hen et al., 1982). It has been shown that a DNase footprint spanning the major late promoter TATA box and +1 sequence resulted from the binding of TFIID (Sawadogo et al., 1985; Van Dyke et al., 1988 and Nakajima et al., 1988).

More recent data, however, suggest that a novel protein designated CAP-site binding factor interacts with sequences immediately downstream from the start of transcription in this promoter (Garfinkel et al., 1990). The ability of this factor to stimulate transcription in the absence of any upstream or downstream sequence remains unknown. In contrast, comparison of the

initiation site of lymphocyte-specific terminal

deoxynucleotidyltransferase gene (Tdt Inr) (Smale and Baltimore, 1989) and the initiation site of the human leukocyte interferon gene (LeIF-J) (Ullrich et al., 1982) revealed striking similarity to the P5+1 element

(Fig. 24). For this reason, we examined the ability of the LeIF-J +1 region to function as an Inr and further tested whether YY1 binds to either the Tdt or the LeIF-J +1 sequences.

Construction of plasmids for in vitro transcription were similar to those described in Fig.

19. Oligonucleotides used were: LeIF-J+1, 5'- GATCCCTAGGTTTTCTGGAGACTGAGCT-3' and its complement to produce Bam HI and Sac I ends. EMSA was carried out identical to Fig. 20, and probes used for the assay were ³²P-5' end-labelled double stranded oligonucleotides whose sequences are given in Fig. 24. Molar excess of

unlabelled competitor was added as indicated. Non- specific API competitor consisted of oligonucleotides 5'- GGATGTTATAAAGCATGAGTCAGACACCTCTGGCT-3' and its

complement. No proteins were added in reactions

presented in Fig. 26, lanes 1,12. As shown in Fig. 25, the LeIF-J+1 sequence functions in initiation of transcription as do the Tdt and the P5 Inr. Moreover, an oligonucleotide consisting of the P5+1 sequence can effectively inhibit the binding of a nuclear factor to the Tdt (Fig. 26, lanes 6-8, complex I) as well as the LeIF-J Inr sequences (Fig. 26, lanes 17-19, complex IV). Since a variety of

electrophoretic mobility shift complexes were observed on comparison of a series of Inr sequences, and since the complexes generated using binding sites corresponding to P5, Tdt, and LeIF-J did not migrate identically even though they were each able to compete for YYl binding (Fig. 26), we believe that there may be multiple types of Inr elements, with the P5, Tdt, and the LeIF-J elements belonging to the same class.

This is the first identification of a factor that can mediate transcription through the transcription start sites called an Inr element. YY1 can direct the general transcription machinery to initiate RNA synthesis at its binding site. Previously, we have shown that the P5+1 element when placed upstream of either a synthetic promoter or the SV40 early promoter/enhancer can repress transcription. Here, we demonstrated that this same element can activate transcription when present alone or downstream from the TATA or Spl sites.

YY1 when altered, either in its amino acid sequence or in the levels at which it is expressed, may have prognostic value, by predicting how a tumor might respond to various treatments. YY1 and YY1 from tumor cells or other altered states may be used to screen for natural biological products or organic chemical reagents that reverse or alleviate the oncogenic effects of the quantitatively or qualitatively abnormal YY1 produced in tumor cells. Alternatively, YY1 may regulate the replication of viruses (e.g. Epstein-Barr viruses) to whose DNA it binds. Thus, agents that influence or modify the activity of YY1 may also alter the behavior of the viral pathogens.

Although the present invention has been

described in relation to particular embodiments thereof, many other variations and modifications and other uses will become apparent to those skilled in the art. It is preferred, therefore, that the present invention be limited not by the specific disclosure herein, but only by the appended claims.

REFERENCES

Abate, C, Luk, D., Gentz, R., Rauscher III, F. J., and Curran. T. (1990). Expression and purification of the leucine zipper and DNA-binding domains of Fos and Jun: Both Fos and Jun contact DNA directly.

Proc. Natl. Acad. Sci. USA 87, 1032-1036.

Arbersold, R. H., Leavitt, J., Saavedra, P. A., Hood,

L. E., and Kent, S. B. H. (1987). Internal amino acid sequence analysis of proteins separated by one- or two-dimensional gel electrophoresis after in sit protease digestion on nitrocellulose. Proc. Natl. Acad. Sci. USA 84, 6970-6974.

Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., and Struhl, K. (1988). Current Protocols in Molecular Biology, eds. (New York: Current Protocols).

Bagchi, S. Raychaudhuri, P. and Nevins, J. R. 1990.

Adenovirus EIA proteins can dissociate heteromeric complexes involving the E2F transcription factor: A novel mechanism for EIA trans-activation. Cell 62, 659-669.

Boyer, T. G. and Maquat, L. E., J. Biol. Chem. 265,

20524-20532 (1990).

Bucher, P. and Trifonov, E. N., Nucleic Acids Res. 14, 10009-10026 (1986).

Buetti, E. and Diggelmann, H., Embo. J. 2, 1423-1429

(1983).

Chang, L. S., Shi, Y., and Shenk, T. (1989). Adenoassociated virus P5 promoter contains an adenovirus ElA-inducible element and a binding site for the major late transcription factor. J. Virol. 63, 3479-3488.

Cordingley, M. G., Riegel, A. T. and Hager, G. L., Cell 48, 261-270 (1987). Courey, A. J., and Tjian, R. (1988). Analysis of Spl in vivo reveals multiple transcriptional domains, including a novel glutamine-rich activation motif. Cell 55, 887-898.

Dignam, J. D., Lebovitz, R. M., and Roeder, R. G. (1983).

Accurate transcription initiation of RNA Polymerase II in a soluble extract from isolated mammalian nuclei. Nucleic Acids Res. 11, 1475-1489.

Dynan, W. S. and Tjian, R., Nature 316, 774-778 (1985). Frasch, M., and Levine, M. (1987). Complementary

patterns of even- skipped and fushi tarazu expression involve their differential regulation by a common set of segmentation genes in Drosophila. Genes Dev. 1, 981-995.

Garfinkel, S., Thompson, J. A., Jacob, W. F., Cohen, R. and Safer, B. J., Biol. Chem. 265, 10309-10319

(1990).

Gill, G., and Ptashne, M. (1987). Mutants of GAL4

protein altered in an activation function. Cell 51, 121-126.

Glass, C. K., Holloway, J. M., Devary, O. V., Rosenfeld, M. G. (1988). The thyroid hormone receptor binds with opposite transcriptional effects to a common sequence motif in thyroid hormone and estrogen response elements. Cell 54, 313-323.

Gorman, C. M., Moffat, L. F., and Howard, B. H. (1982).

Recombinant genomes which express chloramphenicol acetyltranferase in mammalian cells. Mol. Cell.

Biol. 2, 1044-1051.

Harlow, E. and Lane D., Antibodies: A Laboratory Manual (Cold Spring Harbor Laboratory, New York, 1988).

Hearing, P., and Shenk, T. (1983). The adenovirus type 5 EIA transcriptional control region contains a duplicated enhancer element Cell 33, 695-703. Heiermann, R. and Pongs, O., Nucleic Acids Res. 13, 2709-

2730 (1985).

Hen, R. Sassone-Corsi, P., Corden, J., Gaub, M. P. and

Chambon, P., Proc. Natn. Acad. Sci. U.S.A. 79, 7132- 7136 (1982).

Hochuli, E., Dobeli, H., and Schacher, A. (1987). New metal chelate adsorbent selective for proteins and peptides containing neighboring histidine residues. J. Chromatography 411, 177-184.

Hosier, B. A., LaRoas, G. J., Grippo, J. F., and Gudas, L. J. (1989). Expression of REX-1, a gene

containing zinc finger motifs, is rapidly reduced by retinoic acid in F9 teratocarcinoma cells. Mol.

Cell. Biol. 9, 5623-5629.

Johnson, P. F., and McKnight, S. L. (1989). Eukaryotic transcriptional regulatory proteins. Annual Rev. Biochem. 58, 799-839.

Jones, K. A., Luciw, P. A. and Duchange, N., Genes Dev.

2, 1101-1114 (1988).

Kadonaga, J. T. and Tjian, R., Proc. Natn. Acad. Sci. 83, 5889-5893 (1986).

Kageyama, R., and Pastan, I. (1989). Molecular cloning and characterization of a human DNA binding factor that represses transcription. Cell 59, 815-825.

Kato, H., Horikoshi, M. and Roeder, R. G., Science 251, 1476-1479 (1991).

Kozak, M. (1984a). Point mutations close to the AUG

initiator codon affect the efficiency of translatio of rat preproinsulin in vivo. Nature 308, 241-246. Kozak, M. (1984b). Compilation and analysis of sequences upstream from the translational start site in

eukaryotic mRNAs. Nucl. Acids Res. 12, 857-872. Lee, F . , Hall , C . V. , Ringold, G. M. , Dobson, D . E. , Luh, J. and Jacob, P. E., Nucleic Acids Res. 12, 4191-4206 (1984).

Leong, K., Brunet, L. and Berk, A. J. (1988). Factors responsible for the higher transcriptional activity of extracts of adenovirus-infected cells fractionate with the TATA box transcription factor. Mol. Cell. Biol. 8, 1765-1774.

Li, R., Knight, J. D., Jackson, S. P., Tjian, R. and

Botchan, M. R., Cell 65, 493-505 (1991).

Licht, J. D., Grossel, M. J., Figge, J., and

Hansen, U. M. (1990). Drosophila Krύppel protein is a transcriptional repressor. Nature 346, 76-79. Lillie, J. W., and Green. M. R. (1989). Transcription activation by the adenovirus Ela protein. Nature 338, 39-44.

Luckow, B., and Schutz, G. (1987). CAT constructions

with multiple unique restriction sites for the functional analysis of eukaryotic promoters and regulatory elements. Nucl. Acids Res. 15, 5490.

Ma, J., and Ptashne, M. (1987). The carboxy-terminal 30 amino acids of GAL4 are recognized by GAL80. Cell 50,137-142.

Majors, J. and Varmus, H. E., Proc. Natn. Acad. Sci.,

U.S.A. 80, 5866-5870 (1983).

Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982).

Molecular Cloning, A Laboratory Manual, eds. (Cold Spring Harbor: Cold Spring Harbor Laboratory).

McKnight. S. L., and Kingsbury, R. (1982). Transcription control signals of a eukaryotic protein-coding gene. Science 217, 316-324.

Means, A. L. and Farnham, P. J., Mol. Cell. Biol. 10,

653-661 (1990). Mermod, N., O'Neill, E. A., Kelly, T. J., and Tjian, R.

(1989). The proline-rich transcriptional activator of CTF/NF-1 is distinct from the replication and DNA binding domain. Cell 58, 741-753.

Mitchell, P. J., and Tjian, R. (1989). Transcriptional regulation in mammalian cells by sequence-specific

DNA binding proteins. Science 245, 371-378.

Mϋller, U., Robert. M. P., Engel, D. A., Doerfer, W., and

Shenk, T. (1989). Induction of transcription factor AP-1 by adenovirus EIA protein and cAMP. Genes

Dev. 3, 1991-2002.

Nakajima, N., Norikoshi, M. and Roeder, R. G., Mol. Cell.

Biol. 8, 4028-4040 (1988).

Oakley, B. R., Kirsch, D. R., and Morns, N. R. (1980). A simplified ultrasensitive silver stain for detecting proteins in polyacrylamide gels. Anal. Biochem.

105, 361-363.

Okamoto, T., Benter, T., Josephs, S. F., Sadaie, M.R. and Wong-Staal, F., Virology 177, 606-614 (1990).

Parks, C. L., Banerjee, S. and Spector, D. J.,

J. Virol. 62, 54-67 (1988).

Ptashne, M., Nature 335, 683-689 (1988).

Rooney, C, Taylor, N., Countryman, J., Jensen, H.,

Kolman, J., and Miller, G. (1988). Genome

rearrangements activate the Epstein-Barr virus gene whose product disrupts latency. Proc. Natl. Acad. Sci. USA 85", 9801-9805.

Roth, S. Y., Dean, A., and Simpson, R. T. (1990). Yeast α2 repressor positions nucleosomes in TRP1/ARS1 chromatin. Mol. Cell. Biol. 10, 2247-2260.

Ruppert, J. M., Kinzler, K. W., Wong, A. J., Bigner, S.

H., Kao, F. T., Law, M. L., Seuanez, H. N., O'Brian, S. J., and Vogelstein, B. (1988). The GLI-Kruppel family of human genes. Mol. Cell. Biol. 8, 3104- 3113.

Sanger, F., Nicklen, S., and Coulson, A. R. (1977). DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74, 5463-5467.

Sawadogo, M., and Roeder, R. G. (1985). Interaction of a gene-specific transcription factor with the

adenovirus major late promoter upstream of the TATA box region. Cell 43, 165-175.

Shenk, T., Jones, N., Colby, C., and Fowkles, D. (1980).

Functional analysis of adenovirus 5 host range deletion mutants defective for transformation of rat embryo cells. Cold Spring Harbor Symp. Quant. Biol. 44, 367-375.

Siebenlist, U., and Gilbert. W. (1980). Contacts between E. coli RNA polymerase and an early promoter of phage T7. Proc. Natl. Acad. Sci. USA 77, 122-126. Smale, S. T., and Baltimore, D. (1989). The "initiator" as a transcription control element. Cell 57, 103- 113.

Smale, S. T., Schmidt, M. C, Berk, A. J. and Baltimore, D., Proc. Natn. Acad. Sci. U.S.A. 87, 4509-4513 (1990).

Stenlund, A., Bream, G. L. and Botchan, M. R., Science 236, 1666-1671 (1987).

Ullrich, A., Gray, A., Goeddel, D. V. and Dull, T.J., J.

Mol. Biol. 36, 467-486 (1982).

Van Dyke, M. W., Roeder, R. G. and Sawadago, M., Science 241, 1335-1338 (1988).

Wang, L., Nishiyama, K., Araki, K, Kitamura, D., and

Watanabe, T. (1987). Purification of an octomer sequence (ATGCAAAT) -binding protein from human B cells. Nucl. Acids Res. 15, 10105-10116. Wigler, M., Pellicer, A., Silverstein, S., and Axel, R. (1978). Biochemical transfer of a single-copy eucaryotic genes using total cellular DNA as donors. Cell 14, 725.

Zuo, P., Stanojevic, D., Colgan, J., Han, K., Levine, M., and Manley, J. L. (1991). Activation and repression of transcription by the gap proteins hunchback and Krüppel in cultured Drosophila cells. Genes Dev. 5, 254-264.

SEQUENCE LISTING (1) GENERAL INFORMATION:

(i) APPLICANT: Shi, Yang

Seto, Edward

Shenk, Thomas

(ii) TITLE OF INVENTION: YY1 TRANSCRIPTION FACTOR AND METHODS OF

ISOLATING SAME

(iii) NUMBER OF SEQUENCES: 10

(iv) CORRESPONDENCE ADDRESS:

(A) ADDRESSEE: Ostrolenk, Faber, Gerb & Soffen

(B) STREET: 1180 Avenue of the Americas - 7th Floor

(C) CITY: New York

(D) STATE: New York

(E) COUNTRY: USA

(F) ZIP: 10036-8403

(V) COMPUTER READABLE FORM:

(A) MEDIUM TYPE: Floppy disk

(B) COMPUTER: IBM PC compatible

(C) OPERATING SYSTEM: PC-DOS/MS-DOS

(D) SOFTWARE: Patentln Release #1.0, Version #1.25

(vi) CURRENT APPLICATION DATA:

(A) APPLICATION NUMBER: US 07/746,485

(B) FILING DATE: 16-AUG-1991

(C) CLASSIFICATION:

(viii) ATTORNEY/AGENT INFORMATION:

(A) NAME: Dennis, Manette

(B) REGISTRATION NUMBER: 30,623

(C) REFERENCE/DOCKET NUMBER: M-12594 CIP (1570-8)

(ix) TELECOMMUNICATION INFORMATION:

(A) TELEPHONE: (212) 382-0700

(B) TELEFAX: (212) 382-0888

(C) TELEX: 236925

(2) INFORMATION FOR SEQ ID NO:l:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 2353 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS : double

(D) TOPOLOGY: linear (ii) MOLECULE TYPE: DNA (genomic)

(iii) HYPOTHETICAL: NO

(iv) ANTI-SENSE: NO

(vi) ORIGINAL SOURCE:

(A) ORGANISM: Homo sapiens

(F) TISSUE TYPE: HeLa cells derived from cervical

carcinoma

(G) CELL TYPE: tumor cells

(H) CELL LINE: HeLa

(vii) IMMEDIATE SOURCE:

(A) LIBRARY: D98/AH-2

(B) CLONE: p14-1 or pYY1

(ix) FEATURE:

(A) NAME/KEY: CDS

(B) LOCATION: 241..1485

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:

CGCCGAGACG AGCAGCGGCC GAGCGAGCGC GGGCGCGGGC GCACCGAGGC GAGGGAGGCG 60

GGGAAGCCCC GCCGCCGCCG CCCCGCCCGC CCCTTCCCCC GCCGCCCGCC CCCTCTCCCC 120

CCGCCCGCTC GCCGCCTTCC TCCCTCTGCC TTCCTTCCCC ACGGCCGGCC GCCTCCTCGC 180

CCGCCCGCCC GCAGCCGAGG AGCCGAGGCC GCCGCGGCCG TGGCGGCGGA GCCCTCAGCC 240

ATG GCC TCG GGC GAC ACC CTC TAC ATC GCC ACG GAC GGC TCG GAG ATG 288 Met Ala Ser Gly Asp Thr Leu Tyr lle Ala Thr Asp Gly Ser Glu Met

1 5 10 15

CCG GCC GAG ATC GTG GAG CTG CAC GAG ATC GAG GTG GAG ACC ATC CCG 336 Pro Ala Glu lle Val Glu Leu His Glu lle Glu Val Glu Thr lle Pro

20 25 30

GTG GAG ACC ATC GAG ACC ACA GTG GTG GGC GAG GAG GAG GAG GAG GAC 384 Val Glu Thr lle Glu Thr Thr Val Val Gly Glu Glu Glu Glu Glu Asp

35 40 45

GAC GAC GAC GAG GAC GGC GGC GGT GGC GAC CAC GGC GGC GG6 GGC GGC 432 Asp Asp Asp Glu Asp Gly Gly Gly Gly Asp His Gly Gly Gly Gly Gly

50 55 60

CAC GGG CAC GCC GGC CAC CAC CAC CAC CAC CAT CAC CAC CAC CAC CAC 480 His Gly His Ala Gly His His His His His His His His His His His

65 70 75 80 CCG CCC ATG ATC GCT CTG CAG CCG CTG GTC ACC GAC GAC CCG ACC CAG 528 Pro Pro Met lle Ala Leu Gln Pro Leu Val Thr Asp Asp Pro Thr Gln

85 90 95

GTG CAC CAC CAC CAG GAG GTG ATC CTG GTG CAG ACG CGC GAG GAG GTG 576 Val His His His Gln Glu Val lle Leu Val Gln Thr Arg Glu Glu Val

100 105 110

GTG GGC GGC GAC GAC TCG GAC GGG CTG CGC GCC GAG GAC GGC TTC GAG 624 Val Gly Gly Asp Asp Ser Asp Gly Leu Arg Ala Glu Asp Gly Phe Glu

115 120 125

GAT CAG ATT CTC ATC CCG GTG CCC GCG CCG GCC GGC GGC GAC GAC GAC 672 Asp Gln lle Leu lle Pro Val Pro Ala Pro Ala Gly Gly Asp Asp Asp

130 135 140

TAC ATT GAA CAA ACG CTG GTC ACC GTG GCG GCG GCC GGC AAG AGC GGC 720 Tyr lle Glu Gln Thr Leu Val Thr Val Ala Ala Ala Gly Lys Ser Gly

145 150 155 160

GGC GGC GGC TCG TCG TCG TCG GGA GGC GGC CGC GTC AAG AAG GGC GGC 768 Gly Gly Gly Ser Ser Ser Ser Gly Gly Gly Arg Val Lys Lys Gly Gly

165 170 175

GGC AAG AAG AGC GGC AAG AAG AGT TAC CTC AGC GGC GGG GCC GGC GCG 816 Gly Lys Lys Ser Gly Lys Lys Ser Tyr Leu Ser Gly Gly Ala Gly Ala

180 185 190

GCG GGC GGG CGC GGC GCC GAC CCG GGC AAC AAG AAG TGG GAG CAG AAG 864 Ala Gly Gly Arg Gly Ala Asp Pro Gly Asn Lys Lys Trp Glu Gln Lys

195 200 205

CAG GTG CAG ATC AAG ACC CTG GAG GGC GAG TTC TCG GTC ACC ATG TGG 912 Gln Val Gln lle Lys Thr Leu Glu Gly Glu Phe Ser Val Thr Met Trp

210 215 220

TCC TCA GAT GAA AAA AAA GAT ATT GAC CAT GAG ACA GTG GTT GAA GAA 960 Ser Ser Asp Glu Lys Lys Asp lle Asp His Glu Thr Val Val Glu Glu

225 230 235 240

CAG ATC ATT GGA GAG AAC TCA CCT CCT GAT TAT TCA GAA TAT ATG ACA 1008 Gln lle lle Gly Glu Asn Ser Pro Pro Asp Tyr Ser Glu Tyr Met Thr

245 250 255

GGA AAG AAA CTT CCT CCT GGA GGA ATA CCT GGC ATT GAC CTC TCA GAT 1056 Gly Lys Lys Leu Pro Pro Gly Gly lle Pro Gly lle Asp Leu Ser Asp

260 265 270

CCC AAA CAA CTG GCA GAA TTT GCT AGA ATG AAG CCA AGA AAA ATT AAA 110 4 Pro Lys Gln Leu Ala Glu Phe Ala Arg Met Lys Pro Arg Lys lle Lys

275 280 285

GAA GAT GAT GCT CCA AGA ACA ATA GCT TGC CCT CAT AAA GGC TGC ACA 1152 Glu Asp Asp Ala Pro Arg Thr lle Ala Cys Pro His Lys Gly Cys Thr

290 295 300

AAG ATG TTC AGG GAT AAC TCG GCC ATG AGA AAA CAT CTG CAC ACC CAC 1200 Lys Met Phe Arg Asp Asn Ser Ala Met Arg Lys His Leu His Thr His

305 310 315 320 GGT CCC AGA GTC CAC GTC TGT GCA GAA TGT GGC AAA GCT TTT GTT GAG 1248 Gly Pro Arg Val His Val Cys Ala Glu Cys Gly Lys Ala Phe Val Glu

325 330 335

AGT TCA AAA CTA AAA CGA CAC CAA CTG GTT CAT ACT GGA GAG AAG CCC 1296 Ser Ser Lys Leu Lys Arg His Gln Leu Val His Thr Gly Glu Lys Pro

340 345 350

TTT CAG TGC ACG TTC GAA GGC TGT GGG AAA CGC TTT TCA CTG GAC TTC 1344 Phe Gln Cys Thr Phe Glu Gly Cys Gly Lys Arg Phe Ser Leu Asp Phe

355 360 365

AAT TTG CGC ACA CAT GTG CGA ATC CAT ACC GGA GAC AGG CCC TAT GTG 1392 Asn Leu Arg Thr His Val Arg lle His Thr Gly Asp Arg Pro Tyr Val

370 375 380

TGC CCC TTC GAT GGT TGT AAT AAG AAG TTT GCT CAG TCA ACT AAC CTG 1440 Cys Pro Phe Asp Gly Cys Asn Lys Lys Phe Ala Gln Ser Thr Asn Leu

385 390 395 400

AAA TCT CAC ATC TTA ACA CAT GCT AAG GCC AAA AAC AAC CAG TGAAAAGAAG 1492 Lys Ser His lle Leu Thr His Ala Lys Ala Lys Asn Asn Gln

405 410 415

AGAGAAGACC CTTCTCGACC ACGGGAAGCA TCTTCCAGAA GTGTGATTGG GAATAAATAT 1552

GCCTCTCCTT TGTATATTAT TTCTAGGAAG AATTTTAAAA ATGAATCCTA CACACCTAAG 1612

GGACATGTTT TGATAAAGTA GTAAAAATTA AAAAAAAAAA ACTTTACTAA GATGACATTG 1672

CTAAGATGCT CTATCTTGCT CTGTAATCTC GTTTCAAAAA CACAGTGTTT TTGTAAAGTG 1732

TGGTCCCAAC AGGAGGACAA TTCATGAACT TCGCATCAAA AGACAATTCT TTATACAACA 1792

GTGCTAAAAA TGGGACTTCT TTTCACATTC TTATAAATAT GAAGCTCACC TGTTGCTTAC 1852

AATTTTTTTA ATTTTGTATT TTCCAAGTGT GCATATTGTA C.ACTTTTTTG GGGATATGCT 1912

TAGTAATGCT ACGTGTGATT TTTCTGGAGG TTGATAACTT TGCTTGCAGT AGATTTTCTT 1972

TAAAAGAATG GGCAGTTACA TGCATACTTC AAAAGTATTT TCCTGTAAAA AAAAAAAAAG 2032

TTATATAGGT TTTGTTTGCT ATCTTAATTT TGGTTGTATT CTTTGATGTT AACACATTTT 2092

GTATAATTGT ATCGTATAGC TGTATTGAAT CATGTAGTAT CAAATATTAG ATGTGATTTA 2152

ATAGTGTTAA TCAATTTAAA CCCATTTTAG TCACTTTTTT TTTCCAAAAA AATACTGCCA 2212

GATGCTGATG TTCAGTGTAA TTTCTTTGCC TGTTCAGTTA CAGAAAGTGG TGCTCAGTTG 2272

TAGAATGTAT TGTACCTTTT AACACCTGAT GTGTACATCC CATGTAACAG AAAGGGCAAC 2332

AATAAAATAG CAATCCTAAA G 2353 (2) INFORMATION FOR SEQ ID NO:2:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 414 amino acids

(B) TYPE: amino acid

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: protein

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:2:

Met Ala Ser Gly Asp Thr Leu Tyr lle Ala Thr Asp Gly Ser Glu Met 1 5 10 15

Pro Ala Glu lle Val Glu Leu His Glu lle Glu Val Glu Thr lle Pro

20 25 30

Val Glu Thr lle Glu Thr Thr Val Val Gly Glu Glu Glu Glu Glu Asp

35 40 45

Asp Asp Asp Glu Asp Gly Gly Gly Gly Asp His Gly Gly Gly Gly Gly 50 55 60

His Gly His Ala Gly His His His His His His His His His His His 65 70 75 80

Pro Pro Met lle Ala Leu Gln Pro Leu Val Thr Asp Asp Pro Thr Gln

85 90 95

Val His His His Gln Glu Val lle Leu Val Gln Thr Arg Glu Glu Val

100 105 110

Val Gly Gly Asp Asp Ser Asp Gly Leu Arg Ala Glu Asp Gly Phe Glu

115 120 125

Asp Gln lle Leu lle Pro Val Pro Ala Pro Ala Gly Gly Asp Asp Asp 130 135 140

Tyr lle Glu Gln Thr Leu Val Thr Val Ala Ala Ala Gly Lys Ser Gly 145 150 155 160

Gly Gly Gly Ser Ser Ser Ser Gly Gly Gly Arg Val Lys Lys Gly Gly

165 170 175

Gly Lys Lys Ser Gly Lys Lys Ser Tyr Leu Ser Gly Gly Ala Gly Ala

180 185 190

Ala Gly Gly Arg Gly Ala Asp Pro Gly Asn Lys Lys Trp Glu Gln Lys

195 200 205

Gln Val Gln lle Lys Thr Leu Glu Gly Glu Phe Ser Val Thr Met Trp 210 215 220

Ser Ser Asp Glu Lys Lys Asp lle Asp His Glu Thr Val Val Glu Glu 225 230 235 240 Gln lle lle Gly Glu Asn Ser Pro Pro Asp Tyr Ser Glu Tyr Met Thr 245 250 255

Gly Lys Lys Leu Pro Pro Gly Gly lle Pro Gly lle Asp Leu Ser Asp

260 265 270

Pro LysGlu Leu Ala Glu Phe Ala Arg Met Lys Pro Arg Lys lle Lys

275 280 285

Glu Asp Asp Ala Pro Arg Thr lle Ala Cys Pro His Lys Gly Cys Thr 290 295 300

Lys Met Phe Arg Asp Asn Ser Ala Met Arg Lys His Leu His Thr His 305 310 315 320

Gly Pro Arg Val His Val Cys Ala Glu Cys Gly Lys Ala Phe Val Glu

325 330 335

Ser Ser Lys Leu Lys Arg His Gln Leu Val His Thr Gly Glu Lys Pro

340 345 350

Phe Gln Cys Thr Phe Glu Gly Cys Gly Lys Arg Phe Ser Leu Asp Phe

355 360 365

Asn Leu Arg Thr His Val Arg lle His Thr Gly Asp Arg Pro Tyr Val 370 375 380

Cys Pro Phe Asp Gly Cys Asn Lys Lys Phe Ala Gln Ser Thr Asn Leu 385 390 395 400

Lys Ser His lle Leu Thr His Ala Lys Ala Lys Asn Asn Gln

405 410

(2) INFORMATION FOR SEQ ID NO:3:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS : single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA (genomic)

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:3;

GTTTTGCGAC ATTTTGCGAC AC 22 (2) INFORMATION FOR SEQ ID NO:4:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 12 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS : single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA (genomic)

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: GTTTTGCGAC AC 12

(2) INFORMATION FOR SEQ ID NO:5:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 12 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS : single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA (genomic)

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:5: GTTTTGCGAC AC 12

(2) INFORMATION FOR SEQ ID NO:6:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA (genomic)

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:6: GTTTTAAGAC ATTTTAAGAC AC 22 (2) INFORMATION FOR SEQ ID NO:7:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA (genomic)

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:7: GTTTTGTGAT ATTTTGCGAC AC 22

(2) INFORMATION FOR SEQ ID NO:8:

(i) SEQUENCE CHARACTERISTICS :

(A) LENGTH: 22 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS : single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA (genomic)

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:8: AGGGTCTCCA TTTTGAAGCG GG 22

(2) INFORMATION FOR SEQ ID NO:9:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS : single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA (genomic)

(xi) SEQUENCE DESCRIPTION: SEQ ID NO:9: AGGGTCTAAA TTTTGAAGCG GG 22 (2) INFORMATION FOR SEQ ID NO: 10:

(i) SEQUENCE CHARACTERISTICS:

(A) LENGTH: 22 base pairs

(B) TYPE: nucleic acid

(C) STRANDEDNESS: single

(D) TOPOLOGY: linear

(ii) MOLECULE TYPE: DNA (genomic)

(xi) SEQUENCE DESCRIPTION: SEQ ID NO: 10: AGGGTCTCCA GTTGGAAGCG GG 22

Claims

WHAT IS CLAIMED IS:

1. A purified and isolated mammalian

transcription factor which both represses and enhances transcription.

2. The transcription factor according to claim 1, which is substantially similar to human

transcription factor YY1.

3. The transcription factor according to claim 1, which interacts with the transcription

regulatory regions of adeno-associated virus type 2.

4. The transcription factor according to claim 1, which interacts with the transcription

regulatory regions of viruses that infect humans and/or animals, including those selected from the group Epstein- Barr virus.

5. The transcription factor according to claim 1, which interacts with the transcription

regulatory regions of oncogenes.

6. The transcription factor according to claim 1, which has an apparent molecular weight of about 68kD in SDS-polyacrylamide gels.

7. The transcription factor according to claim 1, which is capable of binding to a sequence surrounding a transcription initiation region of the promoter (P5+1 site) of a adeno-associated virus.

8. The transcription factor according to claim l, which is capable of binding to a sequence located at a -50 to -70 (P5-60 site) of an adeno- associated virus P5 promoter.

9. The transcription factor according to claim 1, wherein the transcription factor has an effect on transcription and wherein the effect on transcription can be modified by an adenovirus EIA protein.

10. The transcription factor of claim 1, wherein the transcription factor binds to a promoter of an adenovirus early gene.

11. The transcription factor according to claim 1, wherein the transcription factor represses transcription directed by a TATA element plus initiator sequence.

12. The transcription factor of claim 1, wherein the transcription factor is capable of affecting the latency of viruses.

13. A eukaryotic transcription factor YY1, having an apparent molecular weight of about 68 kD and which can bind to a sequence surrounding a transcription initiation region of a promoter (P5+1 site) of an adeno- associated virus.

14. A eukaryotic transcription factor YY1 having an apparent molecular weight of about 68kD, which can bind to a sequence located at a -50 to -70 (P5-60 site) of an adeno-associated virus P5 promotor.

15. The transcription factor according to claim l, wherein the transcription factor is capable of mediating transcription through an initiator (Inr) sequence.

16. A transcription factor having the characteristics of YY1, which is capable of mediating transcription through an initiator (Inr) element.

17. The transcription factor according to claim 16, wherein the initiation site is an initiation site of a lymphocyte-specific terminal

deoxynucleotidyltransferase gene (Tdt Inr).

18. The transcription factor according to claim 16, wherein the initiation site is the initiation site of a human leukocyte interferon gene (LeIF-J Inr).

19. A transcription factor having the characteristics of YY1.

20. A transcription factor according to claim 19, which is a zinc finger protein.

21. A transcription factor according to claim 1, wherein the transcription factor is a zinc finger protein.

22. A transcription factor according to claim 21, which has a glycine-rich sequence.

23. A transcription factor according to claim 21 which has an acidic domain.

24. A cellular factor present in a protein complex, the protein complex including YY1, wherein the cellular factor contributes to the ability of YY1 to activate or repress transcription.

25. The protein complex of claim 24, wherein the cellular factor is EIA.

26. A method of isolating transcription factor

YY1 comprising the steps of:

a) preparing a nuclear extract from a human cell culture; and

b) running the nuclear extract through a DNA affinity column, wherein the DNA affinity column has oligonucleotides comprising a sequence corresponding to a

YY1 binding site.

27. The method for isolating transcription factor YY1 according to claim 26, wherein the nuclear extract of step (a) is run through a HPLC column before it is run through the DNA affinity column of step (b).

28. The method of isolating transcription factor YY1 according to claim 26, wherein step (b) is repeated twice.

29. The method of isolating YY1 according to claim 26, wherein the DNA affinity column has

oligonucleotides comprising a sequence corresponding to a sequence surrounding a transcription initiation region of a promoter P5+1 site of an adeno-associated virus.

30. The method of isolating YY1 according to claim 26, wherein the DNA affinity column has oligonucleotides comprising a sequence corresponding to a P5-60 site of an adeno-associated virus P5 promoter.

31. A DNA isolate consisting essentially of a DNA sequence encoding eukaryotic transcription factor

YY1.

32. The DNA isolate according to claim 31, wherein the eukaryotic transcription factor YY1 comprises an amino acid sequence identified as SEQ ID NO:1 and shown in Figure 11.

33. A DNA isolate consisting essentially of a

DNA sequence encoding a domain of eukaryotic

transcription factor YY1 having a repressor function.

34. The DNA isolate according to claim 33, wherein the domain of eukaryotic transcription factor YY1 having a repressor function comprises the carboxy- terminal 100 amino acids of the amino acid sequence identified as SEQ ID NO:2 and shown in Figure 11.

35. A recombinant expression vector containing a DNA sequence encoding YY1.

36. A recombinant expression vector containing a DNA sequence encoding eukaryotic transcription factor

YY1, wherein the vector is capable of expressing

eukaryotic transcription factor YY1 in a transformed microorganism or cell culture.

37. The recombinant expression vector according to claim 35, wherein the eukaryotic transcription factor YY1 has an amino acid sequence identified as SEQ ID NO:2 and shown in Figure 11.

38. A microorganism transformed with the vector of claim 35, the microorganism being capable of expressing eukaryotic transcription factor YY1.

39. The microorganism according to claim 38, wherein the eukaryotic transcription factor YY1 comprises an amino acid sequence identified as SEQ ID NO:2 and shown in Figure 11.

40. A cell culture capable of expressing an eukaryotic transcription factor YY1, obtained by

transforming an eukaryotic cell line with the vector of claim 35.

41. The cell culture according to claim 40, wherein the eukaryotic transcription factor YY1 comprises an amino acid sequence identified as SEQ ID NO:2 and shown in Figure 11.

42. A recombinant expression vector comprising a DNA sequence comprising the GAL4 DNA binding domain and encoding eukaryotic transcription factor YY1.

43. A recombinant expression vector comprising a DNA sequence comprising the GAL4 DNA binding domain and a derivative of the full length YY1 cDNA lacking a

sequence encoding the 83 C- terminal amino acids of YY1.

44. A DNA isolate containing a YY1 binding site.

45. A DNA isolate containing a P5-60 YY1 binding site according to claim 44, 5' CGACATTTT-3', and identified as SEQ ID NO:3.

46. A DNA isolate having a P5+1 YY1 binding site according to claim 44, comprising 5'-CTCCATTTT 3', and identified as SEQ ID NO: 7.

47. The use of a repressor activity of eukaryotic transcription factor YY1 as a means to effect a repression of a gene.

48. The use of a repressor activity of eukaryotic transcription factor YY1 according to

claim 47, wherein the repression is not limited to a specific sequence.

49. The use of a repressor activity of eukaryotic transcription factor YY1 according to

claim 47, wherein the gene is a viral gene.

50. The use of a repressor activity of

eukaryotic transcription factor YY1 according to

claim 49, wherein the gene is an oncogene.

51. The use of a nucleotide sequence encoding a repressor function of eukaryotic transcription factor

YY1 as a means to effect a repression of a gene.

52. The use of a nucleotide sequence according to claim 51, wherein the repression is not limited to a specific sequence.

53. The use of a nucleotide sequence according to claim 51, wherein the gene is a viral gene.

54. The use of a nucleotide sequence according to claim 51, wherein the gene is an oncogene.

55. The use of eukaryotic transcription factor

YY1 to activate a gene.

56. The use of eukaryotic transcription factor

YY1 according to claim 55, further comprising the use of EIA protein.

57. The use of a nucleotide sequence encoding an activator function of eukaryotic transcription factor

YY1 to activate a gene.

58. The use of a nucleotide sequence according to claim 57, further comprising the use of EIA protein.

59. A YY1-specific antibody.

60. The antibody of claim 59, wherein said antibody blocks formation of a YY1-DNA complex in a band shift assay.