INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Transcription factors that bind to DNA immediately adjacent to or partially overlapping RNA polymerase binding sites can easily be imagined to assist either the initial binding of RNA polymerase or subsequent steps of the initiation process by means of direct contact with RNA polymerase. Indeed, a number of prokaryotic factors are known to work in precisely these ways (17). Transcription factors that bind to DNA sites located some distance away from RNA polymerase can also be imagined to interact with polymerase through DNA looping or reaching of the C-terminal domain of the alpha subunit of RNA polymerase (2). In the case of looping systems, a DNA-binding protein might also assist or hinder loop formation by other proteins and therefore affect transcription without making direct contact with RNA polymerase or other proteins of the initiation complex. Additionally, a non-DNA-binding protein might mediate the interaction of two DNA-binding proteins.

The mechanism of action of the transcription factor AraC from Escherichia coli at the promoter for the high-affinity arabinose uptake proteins, p_FGH, may be different from the general possibilities mentioned above and therefore different from its mechanism at the well-studied promoter of the araBAD operon, p_BAD (9, 10, 16). At p_BAD AraC protein binds adjacent to and partially overlapping the RNA polymerase 35 region (4) (Fig. 1), where in vitro it stimulates both the binding and the isomerization steps of open-complex formation (26) and where in vivo it, with the catabolite activator protein (CAP) protein, yields maximal transcription initiation (15). At p_FGH, however, it is the CAP protein and not AraC protein that is adjacent to and partially overlapping the RNA polymerase 35 region (Fig. 1). Although at p_FGH AraC protein possesses two binding sites, centered at positions 79.5 and 154.5, placing both upstream from the CAP binding site, only the CAP-proximal site is necessary for significant activation of transcription from p_FGH (10, 16).

View larger version (29K): [in this window] [in a new window]   FIG. 1.   AraC protein and CAP binding sites in relationship to RNA polymerase at the pBAD and pFGH promoters. Arrows indicate the positions and orientations of the AraC protein recognition half-sites. The bottom half shows the sequence of pFGH with the binding sites in bold and indicated by arrows. The RNA polymerase 35 and 10 regions are underlined.

Unlike the binding sites of many transcription factors, the two half-sites of the known AraC binding sequences are in direct repeat orientation rather than inverted repeat orientation (4). This means that the full binding site can possess either of two orientations. The orientation of both of the sites at p_FGH with respect to the RNA polymerase site and the direction of transcription is opposite to that found at the araBAD binding site.

Experiments have shown that, for AraC protein alone to activate transcription at promoters similar to p_BAD, its binding site must both be in the orientation used at p_BAD and partially overlap the promoter's 35 region (20). How, then, can AraC assist in the activation of transcription at p_FGH when pointed in what is apparently the wrong direction and when located in what are apparently the wrong places? Does RNA polymerase contact CAP and, in addition, contact AraC so that both proteins directly assist binding or isomerization, or does AraC activate transcription by some mechanism other than direct contact with RNA polymerase?

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Oligonucleotides. Oligonucleotides used as primers (Table 1) for cloning, sequencing, and PCR were synthesized on an Applied Biosystems 381A synthesizer, deprotected (22), and purified as previously described (6).

DNA for in vitro assays. Wild-type p_FGH, including nucleotides from positions 310 to +158, was cloned by using PCR with chromosomal DNA as the template and oligonucleotides 1327 and 1321 as primers. These primers introduced EcoRI and BamHI cleavage sites that were used in the cloning. Linear DNA molecules were generated by PCR from plasmid DNA containing the wild-type p_FGH promoter or derivatives using primers 826 and either 1360 or 1362. Primer 826 was ³²P 5' end labeled, while the other primer was not. The radioactive linear DNA was then purified from a 10% acrylamide gel, phenol extracted, precipitated, and resuspended in water. The DNA concentration was determined by ethidium bromide staining and comparison with known concentrations of standard size markers.

Purification of proteins. AraC protein was purified by Jeff Withey (23). CAP and RNA polymerase holoenzyme were both purified by Steve Hahn (8). AraC and CAP were both titrated with DNA to determine the minimal amount of protein required for 100% of the DNA to be bound in the absence of competition by heparin. RNA polymerase activity was determined by incubating RNA polymerase with excess galP1 DNA fragment from p19T/121 (3). Active RNA polymerase was considered equal to the amount of RNA polymerase capable of forming an open complex at this promoter. RNA polymerase holoenzymes containing truncated alpha subunits were generously donated by Akira Ishihama, and activity was determined as described above for the wild-type RNA polymerase.

Open-complex formation. Radioactive, linear DNA was diluted to a concentration of 0.06 nM in 1× binding buffer (100 mM KCl, 25 mM Na-HEPES [pH 7.4], 0.5 mM MgCl₂, 2.5 mM dithiothreitol, 0.1 mM cyclic AMP, 0.1 mg of bovine serum albumin per ml, 0.1 mM K-EDTA [pH 7.4], 5% glycerol). A 250-µl volume of reaction mixture containing 0.06 nM DNA was incubated at 37°C with CAP and with or without AraC protein for 10 min. RNA polymerase, already diluted in 1× binding buffer, was added at active concentrations of 0.06 to 20 nM and allowed to incubate with the DNA at 37°C. For dissociation of open complex, heparin was added to a final concentration of 2 µg/ml, and 20-µl samples were taken at various intervals. For all other experiments, 20-µl samples were removed from the reaction mixture and added to a tube containing heparin. Polyacrylamide gels containing 6% (wt/vol) acrylamide, 0.1% bis-acrylamide, 0.1% ammonium persulfate, 0.2% (wt/vol) N,N,N',N'-tetramethylethylenediamine, and 1× electrophoresis buffer (10 mM Tris-acetate, pH 7.4, and 1 mM K-EDTA, pH 7.4) were soaked prior to electrophoresis in 1× electrophoresis buffer for 1 h and were then prerun in fresh 1× electrophoresis buffer containing 0.05 mM cyclic AMP for 30 min. All samples, without DNA dye, were loaded onto the gel 1 min after the addition of heparin. Ten minutes after the last sample was loaded, circulation of the buffer through an ice bath was begun. After electrophoresis, gels were dried under vacuum and exposed to PhosphorImager plates. Band intensities were quantitated with the Molecular Dynamics PhosphorImager.

DNase footprinting. Proteins were allowed to bind to 1.2 nM DNA as described above. Samples (50 µl) were removed from the reaction and were added to 1.5-ml microcentrifuge tubes containing 1 µl of 50 mM CaCl₂ and 1 µl of DNaseI (10 µg/ml). The reaction mixtures were incubated at 37°C for 30 s and then quenched with 50 µl of a DNase quenching mixture. The samples were precipitated with 500 µl of 95% ethanol. Sequencing reactions with template DNA from pCJF210 and oligonucleotide 826 as the primer were used to generate size standards, which were run on the same gels.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Open-complex formation in vitro. In preliminary studies, we confirmed prior work of Hendrickson et al. and Lu et al. (10, 16) showing that the upstream AraC site, araFGH2, does not play a significant role in induction. Only the CAP-proximal AraC site, araFGH1, is required for normal activation of p_FGH, and all upstream DNA can be deleted without significant effect. We also found that inverting araFGH1 inactivated the promoter. Thus, in vitro studies need take into consideration DNA containing only araFGH1 in its native orientation and downstream sequences.

View larger version (67K): [in this window] [in a new window]   FIG. 2.   In vitro assays of promoter activity. Accumulation of open complex over time after the addition of RNA polymerase (RNAP). Samples were 0.06 nM DNA incubated with 0.14 nM AraC protein and 0.20 nM CAP for 5 min prior to the addition of 0.6 nM RNA polymerase. After RNA polymerase was added, samples were removed at the indicated times, incubated with heparin for 1 min, and then loaded onto a gel.

Kinetic measurements of open-complex formation. The DNA migration retardation assay was used to measure the kinetics of the accumulation of open complex. In these experiments DNA was at a concentration of 0.06 nM, and when both AraC and CAP had been added first, varying the RNA polymerase concentration from 0.6 to 1.8 nM (Fig. 3A), or even 6 nM (data not shown), had no discernable effect on the initial rates of open-complex formation. This means that the concentration of the closed complex, from which open complex is derived, was virtually the same at all the RNA polymerase concentrations used. These data permit estimating an upper limit for the K_d of RNA polymerase binding to p_FGH as 0.2 nM.

View larger version (13K): [in this window] [in a new window]   FIG. 3.   Kinetics of open complex formation in the presence (A) and in the absence (B) of AraC protein. Concentrations were as follows: DNA, 0.06 nM; AraC, 0.14 nM; and CAP, 0.2 nM. RNA polymerase concentrations were as indicated in the figure.

Competing, nonproductive RNA polymerase binding site. Measurements of open complex when CAP, but not AraC protein, was present on the DNA yielded surprising kinetics. We expected that the rate of open-complex formation in the absence of AraC protein would be slow and perhaps proportional to the RNA polymerase concentration but that eventually at least 80% of the DNA molecules would form open complexes as long as RNA polymerase was in excess over the DNA. Instead, changes in the concentration of RNA polymerase only moderately changed the rate of open complex formed (Fig. 3B) and only moderately changed the rate of open-complex formation. Even at RNA polymerase excesses of 300-fold above the DNA, the maximum amount of open complex formed was only 30% of the input DNA (data not shown). At both high and low concentrations of RNA polymerase, the half-time for open-complex formation was 0.7 min, almost the same as that seen when AraC was present. The amount of open complex detected, however, ranged from a maximum of 5% open complex at low RNA polymerase concentrations to a maximum of 30% open complex at high RNA polymerase concentrations.

View larger version (22K): [in this window] [in a new window]   FIG. 4.   Binding order. Concentrations were as follows: DNA, 0.06 nM; AraC, 0.14 nM; RNA polymerase (RNAP), 0.6 nM; and CAP, 0.2 nM. Proteins were added in the order shown, and each arrow represents a 5-min incubation step at 37°C.

RNA polymerase likely contacts both CAP and AraC protein. To determine whether the alpha subunit is essential for open-complex formation at p_FGH, as it is at many promoters, we measured the amount of open complex formed when an RNA polymerase containing truncated alpha subunits (12) is used in the presence or absence of AraC protein. In the presence of bound CAP, the RNA polymerase containing truncated alpha subunits shows much less of a response to AraC than the wild-type RNA polymerase (Fig. 5). This result suggests that at p_FGH, RNA polymerase interacts with AraC protein via the C-terminal domain of the  subunit.

View larger version (16K): [in this window] [in a new window]   FIG. 5.   Open-complex formation with wild-type (filled boxes) and -truncated RNA polymerases (open boxes) in the presence and absence of AraC protein. Concentrations were as follows: RNA polymerase, 1.2 nM; DNA, 0.6 nM; AraC, 0.14 nM; and CAP, 0.2 nM.

View larger version (88K): [in this window] [in a new window]   FIG. 6.   Footprinting of the alpha subunit of RNA polymerase in the presence and absence of CAP. Concentrations were as follows: DNA, 0.6 nM; CAP, 2 nM; RNA polymerase (RNAP), 0, 0.6, 1.2, 2.4, and 4.8 nM in experiments with wild-type RNAP and 0, 0.6, 1.2, 2.4, and 4.8 nM in experiments with the RNA polymerase containing truncated alpha subunits.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

An important question in the field of transcription is determining mechanisms by which two different activators can function together to stimulate transcription at a promoter. It seems likely that understanding multiple-activator systems in prokaryotes, where they can be conveniently studied, will assist understanding of the much more commonly found multiple-activator systems of eukaryotes.

One prokaryotic multiple-activator system is the malP-lamB-malM operon, whose promoter requires CAP and MalT (19). Here, the roles of both activators seem to be understood. MalT activates transcription only when bound at the correct sites. The role of CAP is to block access of MalT to the incorrect sites, thus ensuring occupancy of the correct sites (21). Another multiple-activator system is the ansB promoter (5, 24). In E. coli, both CAP and FNR are required for its full activation, and in Salmonella, two dimers of CAP are required. The effects of mutations in the activator proteins and DNase footprinting suggest that the C-terminal domain of the alpha subunit of RNA polymerase may contact both CAP and FNR and simultaneously come close to the DNA between the two activators. Several artificial multiple-activator systems have also been studied (2, 5, 13). In these, the lambda phage activator cI and CAP or CAP and FNR or two molecules of CAP act cooperatively to stimulate a promoter's activity. Such systems have been shown to display cooperative action between the multiple activators as well as behavior in response to mutant activators consistent with interactions between the C-terminal domain of the alpha subunit and both activators, similar to that seen for the natural system at the ansB promoter.

The araFGH promoter, p_FGH, is a good candidate for studies on multiple activators. Its activity requires AraC and CAP, but its structure is different from that of the araBAD promoter, p_BAD, which displays only a strong AraC dependence in vitro (26). Indeed, in the work described here, we have been able to generate CAP and AraC dependence in vitro for the formation of open complexes at p_FGH. The mechanistic studies which then became possible revealed that AraC stimulates open-complex formation at p_FGH by functioning together with CAP to assist the binding of RNA polymerase to the promoter. Not only does this joint action of the proteins increase activity of the p_FGH promoter, but it appears to increase activity at the expense of a nonfunctional and potentially interfering RNA polymerase binding site located a short distance downstream from p_FGH. A similar close-by RNA polymerase binding site that is not an in vivo promoter is also seen in the lac operon (18). Additionally, our work suggests that AraC at p_FGH makes important contacts with the C-terminal domain of the alpha subunit of RNA polymerase.

The rate of open complex formation at p_FGH was surprisingly insensitive to the concentration of RNA polymerase added to the reactions. This insensitivity means that at even the lowest RNA polymerase concentration used, all the promoter sites were occupied by RNA polymerase in closed complexes. To estimate limits on the possible affinity for RNA polymerase binding, if the K_d for RNA polymerase binding were 0.6 nM, then the rate at an RNA polymerase concentration of 0.6 nM would have been 50% of that seen at the highest RNA polymerase concentration. Since a 25% effect would have been just detectable, these data mean that the K_d for RNA polymerase binding to p_FGH is substantially less than 0.6 nM at the conditions used (100 mM KCl, 0.5 mM MgCl₂; 37°C). Typically, for E. coli promoters under similar conditions, the K_d for RNA polymerase ranges from 1 nM to 1 µM (1), meaning that RNA polymerase binds very tightly to the AraC-CAP-p_FGH promoter complex. The rate of open-complex formation at p_FGH is given by the initial slope in the accumulation of open complexes, 0.7 min, which can be compared to rates that are often found, which range from 10 to 1,000 s.

Our studies on the kinetics of open-complex formation did not address the possibility that either CAP or AraC also act at transcription steps after formation of the open complex. Also, our studies did not permit determination of the order of action of CAP and AraC at p_FGH. Because cooperative action was seen in vitro, we can exclude the absolute necessity for additional proteins in the initiation process. A simultaneous interaction seems more compatible with the very high affinity with which RNA polymerase binds to the AraC-CAP-DNA complex, but certainly does not prove it. Simultaneous interaction of RNA polymerase with both activators is possible via the C-terminal domains of the two alpha subunits contained within RNA polymerase.

The CAP and AraC sites are reversed in p_FGH and p_BAD, that is, inversion of the segment of DNA and its bound proteins lying between positions 31 and 102 converts the binding positions of the proteins and their orientations at p_FGH into those of p_BAD. Thus, since the roles of AraC and CAP in stimulating transcription are highly similar (7, 26, 28) and the two proteins can activate transcription from similar positions with respect to RNA polymerase, perhaps these two proteins activate transcription by identical mechanisms and therefore are interchangeable.

The binding site for CAP is centered at position 41.5 at p_FGH as well as at the gal promoter. Why, then, is CAP sufficient to activate the gal operon but not p_FGH? Apparently, the sequence of the RNA polymerase binding site at gal is such that interaction with CAP alone is sufficient to provide adequate binding and rate of isomerization. The sequence at ara p_FGH must require the additional binding energy or strain provided by AraC.

In summary, a minimal araFGH promoter consisting of the RNA polymerase binding site, the partially overlapping CAP binding site, and a binding site for AraC protein located just upstream, has been studied in vitro. CAP and AraC act cooperatively to stimulate open-complex formation by assisting very tight binding of RNA polymerase to p_FGH, and AraC appears to contact the C-terminal domain of the alpha subunit of RNA polymerase. Additionally, the stimulation by AraC likely prevents interference from a nearby nonproductive RNA polymerase binding site.