Identification of preferred actinomycin-DNA binding sites by the combinatorial method REPSA

An important question in the study of ligand-DNA interactions is the determination of binding specificity. Here, we used the combinatorial method restriction endonuclease protection, selection, and amplification (REPSA) to identify the preferred duplex DNA-binding sites of the antineoplastic agent actinomycin D. After 10 rounds of REPSA, over 95% of the cloned DNAs exhibited significantly reduced FokI restriction endonuclease cleavage in the presence of 1 microM actinomycin. A chi(2) statistical analysis of their sequences found that 39 of the 45 clones contained one or more copies of the sequence 5'-(T/A)GC(A/T)-3', giving a p<0.001 for this consensus. A DNase I footprinting analysis of the cloned DNAs found that all possessed relatively high affinity actinomycin-binding sites with apparent dissociation constants between 12 and 258nM (average 98nM). The average footprint encompassed 7.6 bases and in most cases (90%) included one or more consensus sequences. Interestingly, several of the selected clones contained overlapping consensus sequences (e.g., 5'-TGCTGCT-3'), suggesting that such close proximity DNA-binding sites may actually be preferred by actinomycin under physiological conditions.

Characterization of a triplex DNA-binding protein encoded by an alternative reading frame of loricrin

In an attempt to identify genes encoding triple-helical DNA-binding proteins, we performed South-Western screening of a human keratinocyte cDNA expression library using a purine (Pu)-rich triplex DNA probe. We isolated two independent clones containing part of the loricrin gene. Both were translated with a different reading frame than that of the loricrin protein, the major component of the cell envelope during normal keratinocyte cornification. The affinity of the encoded polypeptide for Pu-triplex DNA was confirmed by electrophoretic mobility shift assays using a bacterially expressed N-terminal loricrin deletion fused with the maltose-binding protein (MBP-LOR3ARF). Interactions between Pu-triplex DNA and MBP-LOR3ARF are characterized by a distribution of four increasingly slower mobility complexes, suggesting that multiple MBP-LOR3ARF molecules can recognize a single triplex. Binding was also observed between MBP-LOR3ARF and a pyrimidine-motif triplex DNA, although at lower affinity than Pu-triplex DNA. No apparent binding was observed between MBP-LOR3ARF and double-stranded DNA, suggesting that MBP-LOR3ARF is a bona fide Pu-triplex binding protein. Finally, purified specific rabbit antibodies against LORARF detected four human proteins with apparent molecular masses of 210, 110, 68, and 66 kDa on Western blot analysis. The 210-, 110-, and 68-kDa proteins also showed specific Pu-triplex DNA binding in a South-Western experiment, suggesting that LORARF shares common domains with other human Pu-triplex DNA-binding proteins.

The yeast CDP1 gene encodes a triple-helical DNA-binding protein

The formation of triple-helical DNA has been implicated in several cellular processes, including transcription, replication and recombination. While there is no direct evidence for triplexes in vivo, cellular proteins that specifically recognize triplex DNA have been described. Using a purine-motif triplex probe and southwestern library screening, we isolated five independent clones expressing the same C-terminal 210 amino acids of the Saccharomyces cerevisiae protein Cdp1p fused with beta-galactosidase. In electrophoretic mobility shift assays, recombinant Cdp1pDelta1-867 bound Pu-motif triplex DNAs with high affinity (K:(d) approximately 5 nM) and bound Py-motif triplex, duplex and single-stranded DNAs with far lower affinity (0.5-5.0 microM). Genetic analyses revealed that the CDP1 gene product was required for proper chromosome segregation. The possible involvement of triplex DNA in this process is discussed.

The yeast STM1 gene encodes a purine motif triple helical DNA-binding protein

The formation of triple helical DNA has been evoked in several cellular processes including transcription, replication, and recombination. Using conventional and affinity chromatography, we purified from Saccharomyces cerevisiae whole-cell extract a 35-kDa protein that avidly and specifically bound a purine motif triplex (with a K(d) of 61 pM) but not a pyrimidine motif triplex or duplex DNA. Peptide microsequencing identified this protein as the product of the STM1 gene. Confirmation that Stm1p is a purine motif triplex-binding protein was obtained by electrophoretic mobility shift assays using either bacterially expressed, recombinant Stm1p or whole-cell extracts from stm1Delta yeast. Stm1p has previously been identified as G4p2, a G-quartet nucleic acid-binding protein. This suggests that some proteins actually recognize features shared by G4 DNA and purine motif triplexes, e.g. Hoogsteen hydrogen-bonded guanines. Genetically, the STM1 gene has been identified as a multicopy suppressor of mutations in several genes involved in mitosis (e.g. TOM1, MPT5, and POP2). A possible role for multiplex DNA and its binding proteins in mitosis is discussed.

Differential effects of cyclopolyamines on the stability and conformation of triplex DNA

Linear polyamines are excellent promoters of triplex DNA formation. The effects of structural rigidization of polyamines on triplex DNA stability are not known at present. We wished to develop a series of polyamine analogs as secondary ligands for triplex DNA stabilization for antigene applications. To accomplish this goal, we synthesized cyclopolyamines by interconnecting the two amino or imino groups of linear polyamines with a --(CH2)n-bridge (n=3,4,5). Melting temperature (Tm) data showed that [4,3]-spermine and [4,4]-spermine stabilized poly(dA) x 2poly(dT) triplex at >25 microM concentrations (Tm = 71 degrees C at 100 microM). The dTm/dlog [polyamine] values for these compounds were 26 and 40, respectively. [4,3]-Spermine and [4,4]-spermine also stabilized triplex DNA formed by a purine-motif triplex-forming oligonucleotide, TG3TG4TG4TG3T with its target duplex, as determined by Tm, circular dichroism (CD) spectroscopy, and electrophoretic mobility shift assay (EMSA). In contrast, [4,4]-putrescine and [4,5]-putrescine as well as [4,5]-spermine had no triplex DNA stabilizing effect. CD spectra also showed triplex DNA aggregation and psi-DNA formation at >100 microM [4,3]-spermine. These data demonstrate that structural rigidization of linear polyamines has a profound effect on their ability to stabilize triplex DNA and provoke conformational transitions.

Multiple rounds of transcription by RNA polymerase II at covalently cross-linked templates

An important control point for gene regulation is the frequency of initiations leading to different numbers of RNA polymerases simultaneously transcribing the same gene. To date, the only direct assay for multiple-round transcription by RNA polymerase II in vitro required G-free cassette-containing templates and GTP-free conditions and was thus restricted in application. Here we used instead templates containing a triplex-directed interstrand psoralen-DNA cross-link to block RNA polymerase II elongation at a specific location. Covalently cross-linked templates allowed simultaneous detection of both specific initiation and reinitiation with any combination of promoter and transcribed sequence. In reconstituted systems, identical stacking of RNA polymerases was observed when the first polymerase was halted by GTP deprivation at the end of a G-free cassette or by a covalent cross-link downstream of different transcribed sequences. In contrast to transcription of G-free cassettes, reinitiation was unaffected by the transcription factor SII on sequences containing all four nucleotides. In crude nuclear extracts, transcription of covalently cross-linked templates yielded a reinitiation pattern with a wider spacing than in more purified fractions, indicating that the elongation complexes from nuclear extract contained a different form of RNA polymerase II or a different complement of associated factors.

Self-association of G-rich oligodeoxyribonucleotides under conditions promoting purine-motif triplex formation

Efficient purine-motif triple-helix formation with guanosine/thymidine-rich oligodeoxyribonucleotides requires the presence of divalent cations (e.g., Mg2+) or polyamines at physiologic concentrations. However, under such conditions, we found that G-rich oligonucleotides were capable of self-association. Mixing experiments indicated a stoichiometry of two G-rich oligonucleotide strands in each complex. Dimerization was proportional to the oligonucleotide length, facilitated by increasing concentrations of multivalent cations, and inhibited by monovalent cations that promote G-quartet formation (e.g., K+, Rb+ NH4+). Although dimer formation was relatively slow (t(1/2) approximately 20 minutes), these species were quite stable, with dissociation rates on the order of days. Methylation protection experiments indicated that these dimers exhibited protected N7 position on most all guanines consistent with Hoogsteen base pairing, although this pattern differed from that observed under conditions favoring intramolecular quadruplex formation. Most important, G-rich oligonucleotide dimers were less capable of purine-motif triplex formation than were their denatured counterparts. Thus, these data indicated that G-rich oligodeoxyribonucleotides can form alternate self-associated structures under conditions that do not favor standard quadruplex formation and that these species can have altered properties with regard to their recognition of biologic targets.

Plasmids for the in vitro analysis of RNA polymerase II-dependent transcription based on a G-free template

Described are a series of plasmids containing combinations of adenovirus-2 major late promoter elements, including consensus TATA box and initiator, upstream of G-free transcription cassettes of various lengths. These provide an assortment of tools for investigating both basal and regulated transcription mechanisms by in vitro transcription methods.

Characterization of purine-motif triplex DNA-binding proteins in HeLa extracts

One piece of evidence indicating that triple-helical DNAs exist in vivo would be the demonstration of cellular proteins that recognize such structures. Using oligonucleotide probes containing a GT-rich purine-motif triplex, proteins from either HeLa nuclear or cytoplasmic extracts and electrophoretic mobility shift assays, we identified four specific human protein-triplex complexes. Proteins in these complexes did not recognize an analogous homopurine/homopyrimidine duplex DNA or a pyrimidine-motif triplex but did recognize purine-motif triplexes regardless of whether they possessed a phosphodiester or phosphorothioate backbone in the third strand or involved A*AT instead of T*AT base triplets. For each of these proteins, binding affinity increased with increasing triplex length. For some triplex-binding proteins, a weak affinity was noted for individual G-rich oligonucleotides, though this may actually reflect an affinity for quadruplex structures, which these oligonucleotides are prone to adopt. Ion exchange chromatographic fractionation of HeLa nuclear extracts indicated that at least three different proteins were responsible for the observed electrophoretic mobility shifts. Southwestern blotting methods identified three major polypeptides, with apparent molecular masses of 100, 60, and 15 kDa, that preferentially recognized purine-motif triplexes. These data demonstrate the existence of eukaryotic proteins that specifically recognize one triplex motif and support the idea of a biological role for triple helical DNA.

Oligodeoxyribonucleotide length and sequence effects on intramolecular and intermolecular G-quartet formation

The potential of guanine-rich oligodeoxyribonucleotides (oligos) as nucleic acid drugs is increasingly being investigated, for example, as aptamers against heparin-binding proteins and as purine-motif triplex-forming oligos. However, G-rich oligos can be very polymorphic under physiological conditions, often with the resulting structures possessing vastly different functional capabilities. To better understand the intrinsic oligo parameters that affect their structure, we used nondenaturing gel electrophoresis to investigate a series of G-rich oligos derived from the sequence 5'-TGGGTGGGGTGGGGTGGGT for their abilities to self-associate through G-quartet formation. From these studies the following observations could be made: (1) oligos containing four clusters of three or more contiguous Gs readily associated intramolecularly but did not associate intermolecularly; (2) intermolecular dimerization was the preferred mode of interaction when one of the oligos contained only two G clusters; and (3) T-rich extensions promoted multimerization of oligos into still higher-order species.

Identification of preferred distamycin-DNA binding sites by the combinatorial method REPSA

The combinatorial method restriction endonuclease protection, selection, and amplification (REPSA) was used to determine the preferred duplex DNA binding sites of the peptide N-methylpyrrolecarboxamide antibiotic distamycin A. After 12 rounds of REPSA, several sequences were identified that bound distamycin with an apparent affinity of 2-20 nM. Among these, the highest-affinity sites averaged 10 bp in length, suggesting that these sites may be occupied by multiple, cooperatively interacting distamycin molecules. Presently, REPSA is the only combinatorial approach that allows the identification of preferred DNA targets for small molecule ligands at physiologically relevant concentrations in solution. As such, it should prove useful in the design and screening of sequence-specific DNA-binding molecules.

Identification of preferred hTBP DNA binding sites by the combinatorial method REPSA

Here we describe the application of a novel combinatorial method, restriction endonuclease protection selection and amplification (REPSA), to identification of a consensus DNA binding site for the TATA binding subunit (hTBP) of the human general transcription factor TFIID. Unlike most combinatorial methods, REPSA is based on inhibition of an enzymatic template inactivation process by specific ligand-DNA complexes. The mild conditions of this method allow examination of proteins with atypical binding characteristics (e.g. limited discrimination between specific and non-specific binding sites), such as those found with hTBP. Analysis of 57 emergent sequences identified 47 sequences containing consensus 6 bp TATA elements as previously defined. However, further examination of these sequences indicated that a larger consensus, 5'-TATAAATA-3', could be supported by the data. Studies of the binding affinities and transcriptional activities of these four consensus TATA sequences demonstrated that hTBP binding affinity correlated directly with transcriptional activity in vitro and that the TATAAATA sequence was the best among the TATA sequences investigated.

Effects of chain length modification and bis(ethyl) substitution of spermine analogs on purine-purine-pyrimidine triplex DNA stabilization, aggregation, and conformational transitions

The natural polyamines--putrescine, spermidine, and spermine--are known to stabilize pyrimidine-purine-pyrimidine and purine-purine-pyrimidine triplex DNA formation. We studied the ability of two tetramine and two pentamine analogs of spermine and their bis(ethyl) derivatives to stabilize triplex DNA formation between 5'-TG3TG4TG4TG3T-3' and its target duplex probe, consisting of the oligonucleotides 5'-TCGAAG3AG4AG4AG3A-3' and 5'-TCGATC3TC4TC4TC3T-3'. We used electrophoretic mobility shift assay (EMSA), melting temperature (Tm) measurements, and circular dichroism (CD) spectroscopy to evaluate the effects of these novel polyamine analogs on triplex DNA stability, dissociation constants, aggregation, and conformation. In general, pentamines were more efficacious than tetramines in stabilizing triplex DNA, although most of the polyamines with pendant free amino groups caused DNA aggregation below 50% conversion to triplex DNA. Ethyl substitution of these pendant amino groups lowered their efficacy approximately 2-fold in stabilizing triplex DNA; however, this effect was more than compensated for by the lack of DNA aggregation in the presence of bis(ethyl)polyamines. A concentration-dependent increase in the Tm of triplex DNA was observed in the presence of polyamines. CD spectral measurements showed distinct differences in the conformation of triplex DNA stabilized in the presence of polyamines compared to the CD spectra of the oligonucleotides alone. Temperature-dependent CD spectra of triplex DNA showed monophasic melting in the absence and presence of polyamines, suggesting duplex/triplex --> single-stranded DNA transition. These results indicate that structural modifications of polyamines is an effective strategy to develop triplex DNA-stabilizing ligands, with potential applications in antigene therapeutics.

In vivo persistence of DNA triple helices containing psoralen-conjugated oligodeoxyribonucleotides

Triple helices represent an attractive method for modulating specific gene expression. In particular, cross-linking between a triplex-forming oligonucleotide (TFO) and its duplex DNA target, typically through the formation of psoralen photoadducts, allows efficient blocking of elongation by RNA polymerases in vitro. However, in vivo, this approach is limited by DNA repair of the photoadduct. Here we describe the use of an oligodeoxyribonucleotide 19mer psoralen-modified TFO to form covalent linkages between an oligonucleotide and both strands of the targeted duplex DNA, thereby efficiently blocking expression of a luciferase reporter gene. Most importantly, we demonstrate that both the psoralen cross-link and the purine-motif triplex remained intact for at least 72 h post-transfection, indicating that such species can persist for an extended period of time in vivo. These findings support the feasibility of an antigene approach for the therapeutic regulation of specific gene expression.

Sequence specificity of triplex DNA formation: Analysis by a combinatorial approach, restriction endonuclease protection selection and amplification

We have devised a combinatorial method, restriction endonuclease protection selection and amplification (REPSA), to identify consensus ligand binding sequences in DNA. In this technique, cleavage by a type IIS restriction endonuclease (an enzyme that cleaves DNA at a site distal from its recognition sequence) is prevented by a bound ligand while unbound DNA is cleaved. Since the selection step of REPSA is performed in solution under mild conditions, this approach is amenable to the investigation of ligand-DNA complexes that are either insufficiently stable or not readily separable by other methods. Here we report the use of REPSA to identify the consensus duplex DNA sequence recognized by a G/T-rich oligodeoxyribonucleotide under conditions favoring purine-motif triple-helix formation. Analysis of 47 sequences indicated that recognition between 13 bases on the oligonucleotide 3' end and the duplex DNA was sufficient for triplex formation and indicated the possible existence of a new base triplet, G.AT. This information should help identify appropriate target sequences for purine-motif triplex formation and demonstrates the power of REPSA for investigating ligand-DNA interactions.

Torsionally-strained DNA and intermolecular purine-purine-pyrimidine triple-helix formation

A potentially powerful pharmacological approach to modulating the expression of specific, disease-related genes involves the inhibition of transcription factor binding to promoter or enhancer elements through oligonucleotide-mediated triple-helix formation. In vivo, the typical target for intermolecular triplex formation would most likely be torsionally-strained rather than relaxed duplex DNA. To determine the effects of strained DNA on triplex formation, we investigated the interactions between a G/T-rich oligonucleotide and both supercoiled and relaxed plasmid DNA using a restriction endonuclease protection assay. Both the kinetics of formation and dissociation of purine-motif triplexes were unaffected by the conformational state of the duplex DNA. Similarly, the topological state of the plasmid targets was not affected by triplex formation. Taken together, these observations suggest that stable intermolecular triplexes can form in vivo under conditions of moderate torsional strain.

Polyamine effects on purine-purine-pyrimidine triple helix formation by phosphodiester and phosphorothioate oligodeoxyribonucleotides

Utilization of oligodeoxyribonucleotides to inhibit specific gene transcription in vivo (antigene strategy) requires the efficient formation of triple helices under physiological conditions. However, pyrimidine-motif triplexes are not favored at physiological pH, and physiological concentrations of potassium cations hamper purine-motif triplex formation. Here we investigated the effects of polyamines on promoting triplex formation by G/T-rich oligodeoxyribonucleotides containing either phosphodiester or a diastereomeric mixture of phosphorothioate linkages. Compared with Mg2+, equimolar concentrations of polyamines greatly facilitated purine-motif triplex formation with the following order of effectiveness: spermine > spermidine > putrescine. At low polyamine concentrations, phosphorothioate oligonucleotides were better at triplex formation than the corresponding phosphodiester oligonucleotides. Kinetic studies indicated that polyamines facilitated triplex formation by increasing the rate of oligonucleotide-duplex DNA association. However, triplex accumulation with either oligonucleotide was still low under physiological conditions (140 mM K+, 10 mM Mg2+, 1 mM spermine). The inhibitory effects of K+ could be partially overcome with high concentrations of Mg2+ or spermine, with phosphodiester oligonucleotides being better able to form triplexes than phosphorothioates under these conditions.

Oligodeoxyribonucleotide length and sequence effects on intermolecular purine-purine-pyrimidine triple-helix formation

The binding of guanosine/thymidine-rich oligodeoxyribonucleotides containing various deletions, extensions, and point mutations to polypurine DNA targets was investigated by DNase I footprinting. Intermolecular purine-purine-pyrimidine triple-helical DNA formation was best achieved using oligonucleotides 12 nucleotides in length. Longer oligonucleotides were slightly weaker in binding affinity, whereas shorter oligonucleotides were considerably weaker. Oligonucleotide extensions had a slight effect on triplex formation, while single point mutations located near the oligonucleotide ends had a greater effect. In the cases of extensions and point mutations, changes to the 3' end of the oligonucleotide had a consistently greater effect on triplex formation than changes to the 5' end. Such differences in triplex-forming ability were not caused by an intrinsic property of these oligonucleotides, since the same point mutated oligonucleotides could bind with high affinity to duplex DNAs containing complementary sites. Taken together, our data suggest that there may be an asymmetry involved in the process of purine-motif triplex formation, with interactions between the 3' end of the oligonucleotide and complementary sequences on the target duplex DNA being dominant.

Sp1, USF, and GAL4 activate transcription independently of TFIID-downstream promoter interactions

One way specific transcription factors are thought to activate transcription initiation is by facilitating interactions between the general transcription factor TFIID and DNA sequences downstream of the TATA element. Examples supporting this model include transcription activation from the core adenovirus E4 promoter by either the human gene-specific transcription factor ATF or the acidic-domain fusion protein GAL4-AH. In these cases, appearance of downstream promoter binding by TFIID correlated directly with transcription activation by these proteins. Previously we had shown that downstream promoter binding by TFIID involved recognition of the initiator DNA control element and that the extent of this binding correlated directly with initiator-dependent transcription activation in vitro. We now report our use of DNase I footprinting and in vitro transcription to investigate the effects of various stimulatory transcription factors on TFIID binding and transcription efficiency from different initiator-containing promoters. Transcription factors investigated included Sp1, USF, and several GAL4-acidic domain fusion proteins. We found that none of these transcription factors appreciably affected downstream promoter binding by TFIID, whether qualitatively or quantitatively. In fact, all of these transcription factors stimulated transcription in vitro regardless of the strength of the initiator element present. When both elements were present, transcription stimulation mediated by proximally bound transcription factors and by TFIID-initiator interaction seemed to be synergistic. Taken together, our data would suggest that transcription activation by these two means occurs through different steps within the transcriptional process.

Monovalent cation effects on intermolecular purine-purine-pyrimidine triple-helix formation

The binding of a 19-mer guanosine-rich oligodeoxyribonucleotide, TG3TG4TG4TG3T (ODN 1), to a complementary polypurine DNA target was investigated by DNase I footprinting and restriction endonuclease protection assays. Monovalent cations inhibited intermolecular purine-purine-pyrimidine triple-helical DNA formation, with K+ and Rb+ being most effective, followed by NH4+ and Na+. Li+ and Cs+ had little to no effect. Similar results were observed with the G/A-rich oligonucleotide AG3AG4AG4AG3AGCT. Kinetic studies indicated that monovalent cations interfered with oligonucleotide-duplex DNA association but did not significantly promote triplex dissociation. The observed order of monovalent cation inhibition of triplex formation is reminiscent of their effect on tetraplex formation with G/T-rich oligonucleotides. However, using electrophoretic mobility shift assays we found that the oligonucleotide ODN 1 did not appear to form a four-stranded species under conditions promoting tetraplex formation. Taken together, our data suggest that processes other than the self-association of oligonucleotides into tetraplexes might be involved in the inhibitory effect of monovalent cations on purine-pyrimidine-purine triplex formation.

Initiator sequences direct downstream promoter binding by human transcription factor IID

Whereas the human transcription factor IID generally interacts with only the TATA box element on most class II gene promoters, on certain promoters (e.g., the adenovirus 2 major late promoter) TFIID protects DNA sequences up to +35 base pairs downstream of the start site of transcription from DNase I cleavage. In this study, we show that Ad2 MLP sequences from -3 to +5 were sufficient to direct downstream promoter binding by TFIID when introduced into the human hsp70 gene promoter. These sequences correspond to the initiator transcription control element. Initiator mutations resulting in a loss of downstream binding demonstrated a diminished transcription efficiency in vitro. Likewise, initiator-dependent transcription stimulation required TFIID fractions capable of downstream promoter binding. Given the recent findings that immunopurified, human TFIID exhibited downstream promoter binding on the Ad2 MLP (Zhou, Q., Lieberman, P.M., Boyer, T.G. and Berk, A.J. (1992) Genes Dev. 6, 1964-1974), our data suggest that human TFIID can recognize the initiator element and that this interaction is required for maximally efficient transcription initiation.

An alternatively spliced mRNA from the AP-2 gene encodes a negative regulator of transcriptional activation by AP-2

AP-2 is a retinoic acid-inducible and developmentally regulated activator of transcription. We have cloned an alternative AP-2 transcript (AP-2B) from the human teratocarcinoma cell line PA-1, which encodes a protein differing in the C terminus from the previously isolated AP-2 protein (AP-2A). This protein contains the activation domain of AP-2 and part of the DNA binding domain but lacks the dimerization domain which is necessary for DNA binding. Analysis of overlapping genomic clones spanning the entire AP-2 gene proves that AP-2A and AP-2B transcripts are alternatively spliced from the same gene. Both transient and stable transfection experiments show that AP-2B inhibits AP-2 transactivator function, as measured by an AP-2-responsive chloramphenicol acetyltransferase reporter plasmid. Furthermore, constitutive AP-2B expression in PA-1 cells causes a retinoic acid-resistant phenotype, anchorage-independent growth in soft agar, and tumorigenicity in nude mice, in a fashion similar to transformation of these cells by oncogenes. To determine the mechanism by which AP-2B exerts its inhibitory function, we purified bacterially expressed AP-2A and AP-2B proteins. While bacterial AP-2B does not bind an AP-2 consensus site, it strongly inhibits binding of the endogenous AP-2 present in PA-1 cell nuclear extracts. However, DNA sequence-specific binding of bacterially expressed AP-2A cannot be inhibited by bacterially expressed AP-2B. Therefore, inhibition of AP-2 activity by the protein AP-2B may require an additional factor or modification supplied by nuclear extracts.

In vitro inhibition of c-myc transcription by mithramycin

Mithramycin is a DNA-binding antibiotic that has been reported to selectively affect c-myc expression [Snyder, R. C. et al., (1991) Biochemistry 30, 4290-4297]. We used in vitro transcription to investigate the specificity of mithramycin action. We found that mithramycin inhibited transcription from the human c-myc P1 and P2 promoters, as well as from a minimal adenovirus-2 major late promoter, with equal efficiencies. Mithramycin also inhibited transcription elongation by creating kinetic blockades to the passage of RNA polymerase II. These data suggest that mithramycin may inhibit transcription non-specifically by affecting general processes such as transcription elongation.

Single-step purification of bacterially expressed polypeptides containing an oligo-histidine domain

Plasmid expression vectors have been constructed that direct the synthesis in Escherichia coli of fusion proteins containing a stretch of six histidine residues at either the N or C terminus. This oligo-histidine domain allows the single-step purification of the fusion proteins, under nondenaturing conditions, by immobilized metal affinity chromatography on Ni2+ bound to iminodiacetic acid-agarose. Several eukaryotic transcription factors (e.g., the upstream stimulatory factor for the adenovirus major late promoter) have been successfully purified, in an active state, by this method.

DNA-binding and transcriptional properties of human transcription factor TFIID after mild proteolysis

The existence of separable functions within the human class II general transcription factor TFIID was probed for differential sensitivity to mild proteolytic treatment. Independent of whether TFIID was bound to DNA or free in solution, partial digestion with either one of a variety of nonspecific endoproteases generated a protease-resistant protein product that retained specific DNA recognition, as revealed by DNase I footprinting. However, in contrast to native TFIID, which interacts with the adenovirus major late (ML) promoter over a very broad DNA region, partially proteolyzed TFIID interacted with only a small region of the ML promoter immediately surrounding the TATA sequence. This novel footprint was very similar to that observed with the TATA factor purified from yeast cells. Partially proteolyzed human TFIID could form stable complexes that were resistant to challenge by exogenous templates. It could also nucleate the assembly of transcription complexes on the ML promoter with an efficiency comparable to that of native TFIID, yielding similar levels of transcription initiation. These results suggest a model in which the human TFIID protein is composed of at least two different regions or polypeptides: a protease-resistant "core," which by itself is sufficient for promoter recognition and basal transcriptional levels, and a protease-sensitive "tail," which interacts with downstream promoter regions and may be involved in regulatory processes.

Stability of transcription complexes on class II genes

Commitment of a TATA box-driven class II gene to transcription requires binding of only one transcription factor, TFIID. Additional factors (TFIIB, TFIIE, and RNA polymerase II) do not remain associated with the TFIID-promoter complex during the course of transcription. This indicates that there are two intermediates along the transcription reaction pathway which may be potential targets for the regulation of gene expression.

Physical analysis of transcription preinitiation complex assembly on a class II gene promoter

Transcription of protein-encoding genes by human RNA polymerase II requires multiple ancillary proteins (transcription factors). Interactions between these proteins and the promoter DNA of a viral class II gene (the major late transcription unit of adenovirus) were investigated by enzymatic and chemical footprinting. The experiments indicated that the assembly of functionally active RNA polymerase II-containing transcription preinitiation complexes requires a complete set of transcription factors, and that both specific protein-DNA and protein-protein interactions are involved. This allows individual steps along the transcription reaction pathway to be tested directly, thus providing a basis for understanding basic transcription initiation mechanisms as well as the regulatory processes that act on them.

Multiple forms of the human gene-specific transcription factor USF. I. Complete purification and identification of USF from HeLa cell nuclei

The human gene-specific upstream stimulatory transcription factor (USF) is required, both in vivo and in vitro, for maximal expression of the major late promoter (MLP) of adenovirus. We report here the complete purification and identification of USF from HeLa cell nuclei. The protein was followed throughout its purification using a quantitative filter binding assay. With a combination of classical purification techniques and fast-flow protein liquid chromatography, USF can be purified to homogeneity starting either with a standard HeLa cell nuclear extract or with a higher salt extract from (lysed) HeLa cell nuclei (nuclear pellet extract). Approximately 20,000-fold purification from the nuclear pellet extract and 80,000-fold from the nuclear extract are necessary to obtain homogeneous preparations of the transcription factor. A maximum of 20,000 molecules of USF appear to be present in HeLa cells. Two major forms of the USF protein can be distinguished both by their slightly different mobilities in sodium dodecyl sulfate gel electrophoresis (apparent molecular weights 44,000 and 43,000, respectively) and by different electrophoretic mobilities of the corresponding protein-DNA complexes. Both forms of USF are heat-stable and interact with the MLP as monomers. Antibodies elicited against purified HeLa USF interact with the transcription factor bound to the MLP upstream element.

Two forms of transcription factor TFIIIC in extracts from HeLa cells

Partially-purified preparations of transcription factor TFIIIC derived from either cytoplasmic (S100) or nuclear (NE) HeLa cell extracts possessed comparable activities when assayed by in vitro transcription, although it was possible to detect sequence-specific binding to the VAI RNA gene promoter by DNase I footprinting only with TFIIIC (NE). On the basis of mixing experiments, this difference could not be ascribed to either transcriptional or DNA-binding inhibitors. These findings provide further evidence for the existence of two forms of TFIIIC: one which is present in both extracts and is transcriptionally active; another which predominates in nuclear extracts and is transcriptionally inactive.

Novobiocin interferes with the binding of transcription factors TFIIIA and TFIIIC to the promoters of class III genes

Novobiocin has been shown to inhibit class III gene transcription from both chromatin and non-chromatin templates. Since novobiocin is a well characterized inhibitor of type II DNA topoisomerases, it has been postulated that a gyrase activity is necessary for transcription. Using DNase I footprinting, we show here that novobiocin inhibits the specific binding of polymerase III transcription factors TFIIIA and TFIIIC to the promoters of the 5S RNA and VA RNA genes, respectively. Concentrations of novobiocin employed were comparable to those necessary to inhibit HeLa topoisomerase II. In vitro transcription assays, performed under equivalent conditions, demonstrated that similar novobiocin concentrations were necessary for transcription inhibition. These results strongly suggest that novobiocin interferes with transcription by inhibiting specific protein-DNA interactions.

Multiple proteins bind to VA RNA genes of adenovirus type 2

Using fractionated HeLa cell nuclear extracts and both nuclease (DNase I) cleavage and chemical cleavage (methidiumpropyl-EDTA X Fe(II) protection methodologies, we demonstrated the presence of three proteins which interacted specifically, yet differentially, with the two VA genes of adenovirus type 2. One, previously identified as transcription initiation factor TFIIIC, bound to a site centered on the transcriptionally essential B-block concensus element of the VAI gene and, with a lower affinity, to the analogous site in the VAII gene. Another, identified as the cellular protein involved in adenovirus replication, nuclear factor I, bound to sites immediately downstream from the two VAI terminators (at approximately +160 and +200). The third, a previously unrecognized VA gene binding protein termed VBP, bound immediately upstream of the B-block element in the VAI gene but showed no binding to VAII. Possible roles for these proteins in VA gene transcription were investigated in in vitro assay systems reconstituted with partially purified transcription factors (RNA polymerase III, TFIIIB, and TFIIIC). Although TFIIIC activity was present predominantly in fractions containing B-block binding activity, there was not complete correspondence between functional and DNA binding activities. The nuclear factor I-like protein had no effect when added to a complete transcription reaction. The presence of VBP appeared to depress the intrinsic ratio of VAI-VAII synthesis, thereby simulating the relative transcription levels observed early in adenovirus infection of HeLa cells. These observations suggest a model, involving both intragenic binding factors (VBP and TFIIIC) and variable template concentrations, for the differential regulation of VA transcription during the course of adenovirus infection.

Methidiumpropyl-EDTA.Fe(II) and DNase I footprinting report different small molecule binding site sizes on DNA

DNase I and MPE.Fe (II) footprinting both employ partial cleavage of ligand-protected DNA restriction fragments and Maxam-Gilbert sequencing gel methods of analysis. One method utilizes the enzyme, DNase I, as the DNA cleaving agent while the other employs the synthetic molecule, methidium-propyl-EDTA (MPE). For actinomycin D, chromomycin A3 and distamycin A, DNase I footprinting reports larger binding site sizes than MPE.Fe (II). DNase I footprinting appears more sensitive for weakly bound sites. MPE.Fe (II) footprinting appears more accurate in determining the actual size and location of the binding sites for small molecules on DNA, especially in cases where several small molecules are closely spaced on the DNA. MPE.Fe (II) and DNase I report the same sequence and binding site size for lac repressor protein on operator DNA.

Chromomycin, mithramycin, and olivomycin binding sites on heterogeneous deoxyribonucleic acid. Footprinting with (methidiumpropyl-EDTA)iron(II)

The DNA binding sites for the antitumor, antiviral, antibiotics chromomycin, mithramycin, and olivomycin on 70 base pairs of heterogeneous DNA have been determined by using the (methidiumpropyl-EDTA)iron(II) [MPE x Fe(II)] DNA cleavage inhibition pattern technique. Two DNA restriction fragments 117 and 168 base pairs in length containing the lactose operon promoter-operator region were prepared with complementary strands labeled with 32P at the 3' end. MPE x Fe(II) was allowed to partially cleave the restriction fragment preequilibrated with either chromomycin, mithramycin, or olivomycin in the presence of Mg2+. The preferred binding sites for chromomycin, mithramycin, and olivomycin in the presence of Mg2+ appear to be a minimum of 3 base pairs in size containing at least 2 contiguous dG x dC base pairs. Many binding sites are similar for the three antibiotics; chromomycin and olivomycin binding sites are nearly identical. The number of sites protected from MPE x Fe(II) cleavage increases as the concentration of drug is raised. For chromomycin/Mg2+, the preferred sites on the 70 base pairs of DNA examined are (in decreasing affinity) 3'-GGG, CGA greater than CCG, GCC greater than CGA, CCT greater than CTG-5'. The sequence 3'-CGA-5' has different affinities, indicating the importance of either flanking sequences or a nearly bound drug.

Map of distamycin, netropsin, and actinomycin binding sites on heterogeneous DNA: DNA cleavage-inhibition patterns with methidiumpropyl-EDTA.Fe(II)

We report a direct technique for determining the binding sites of small molecules on naturally occurring heterogeneous DNA. Methidiumpropyl-EDTA.Fe(II) [MPE.Fe(II) cleaves double helical DNA with low sequence specificity. Using a combination of MPE.Fe(II) cleavage of drug-protected DNA fragments and Maxam-Gilbert gel methods of sequence analysis, we have determined the preferred binding sites on a Rsa I-EcoRI restriction fragment from pBR322 for the intercalator actinomycin D and the minor groove binders netropsin and distamycin A. Netropsin and distamycin A gave identical DNA cleavage-inhibition patterns and bound preferentially to A+T-rich regions with a minimal protected site of four base pairs. We were able to observe the effect of increasing concentration on site selection by netropsin and distamycin A. Actinomycin D afforded a completely different cleavage-inhibition pattern, with 4- to 16-base-pair-long protected regions centered around one or more G.C base pairs.