Skip to main content
NCBI home page
Protein Science : A Publication of the Protein Society logoLink to Protein Science : A Publication of the Protein Society
. 2003 Aug;12(8):1613–1620. doi: 10.1110/ps.0305703

Surface plasmon resonance studies of wild-type and AV77 tryptophan repressor resolve ambiguities in super-repressor activity

Michael D Finucane 1, Oleg Jardetzky 1
PMCID: PMC2323948  PMID: 12876311

Abstract

The interactions of wild-type (WT) and AV77 tryptophan repressor (TR) with several operators have been studied using surface plasmon resonance. The use of this real-time method has been able to settle several outstanding issues in the field, in a way that has heretofore not been possible. We resolve the issue of the super-repressor status of the AV77 aporepressor and find that in contrast to early studies, which found no significant difference in the binding constants in vitro to those of the WT, that there is indeed a clear difference in the binding constant that can simply account for the phenotype. Accordingly, there is no need for alternative proposals invoking complex equilibria with in vivo components not found in the in vitro experiments. In addition, we find that the AV77 holorepressor–DNA complex is much more stable than the equivalent WT complex, which has not been apparent from either in vitro or equilibrium binding experiments.

Keywords: Tryptophan repressor, AV77, surface plasmon resonance, super-repressor, protein–DNA interactions, indirect readout, operator length

Tryptophan repressor (TR) is one of the smallest (25 kD) regulatory proteins known. It is also one of the most studied, and yet remains incompletely understood. The protein is a symmetrical dimer that binds to operator DNA in the presence of l-tryptophan. It regulates several pathways implicated in tryptophan and aromatic biosynthesis, including the trpEDCBA, AroH, AroL, and mtr operons. It is also autoregulatory, controlling transcription from the trpR operon.

Despite—or perhaps because of—the large number of studies done on the binding of TR to its operators, there is a wide range of conflicting binding data in the literature. This is perhaps not surprising, as there are a wide variety of methods available for measuring the binding of repressors to DNA. To date, most binding studies of the TR system have used gel retardation. To produce meaningful results, carefully optimized conditions must be used (Carey 1988; Cann 1989). Unfortunately, this often necessitates reducing the pH to nonphysiological levels (Carey 1988; Liu and Matthews 1994), and thereby altering the binding equilibrium. Binding constants obtained with gel retardation can differ even when the same pH is used (Klig et al. 1987; Hurlburt and Yanofsky 1990), depending on the salt composition of the buffers used, presumably because weak-interacting complexes do not remain intact during electrophoresis. Filter binding studies suffer from an inability to distinguish between complexes of different stoichiometries, which unfortunately occur in the TR system and probably have a major role in the discrimination between operators (Kumamoto et al. 1987; LeTilly and Royer 1993; Jeeves et al. 1999). Because of this, binding constants obtained using this method are unreliable. Fluorescence anisotropy suffers few of the disadvantages of gel retardation and filter-binding assays but gives only equilibrium constants, and does not yield binding and dissociation rates. Because of these limitations in the methodology used in previous studies, we set out to reconcile the various conflicts in the literature using surface plasmon resonance (SPR), a more recently developed technique, which offers several advantages, the most important of which is the ability to follow the binding and dissociation reactions in real time (Morton and Myszka 1998). SPR has been shown to be advantageous in studying protein–DNA interactions (e.g., Parsons et al. 1995; Oda et al. 1999; Tsoi and Yang 2002). Using SPR, we have been able to eliminate confusion as to the source of the enhanced activity of the apo form of the AV77 mutant protein. We also find that AV77 holorepressor forms an especially stable complex with operator DNA, which has not been previously reported.

Results

Real-time interaction analysis using SPR

After binding 100 RU of biotinylated 20-consensus DNA ([1] in Fig. 1 ▶) to a streptavidin-coated biacore chip, the surface was briefly washed, and then WT TR was pumped over the surface. This results in a very rapid rise in signal intensity, indicating that protein is binding to the DNA-coated surface. No response of comparable intensity is seen on an underivatized control surface.

Figure 1.

Figure 1.

Open in a new tab

DNA sequences discussed in the text. Bold type indicates the base pairs at positions +5 to +9 (from the center of symmetry) that are contacted by the repressor in the crystal structure (PDB: 1TRO).

TR binds to operator DNA in the presence of l-tryptophan. To verify that the response that we observed was ligand-specific, we diluted stock aporepressor into buffer containing no l-tryptophan. On passing the protein over surfaces containing DNA (Fig. 2 ▶, a1 and a2) no response was seen. The same protein stock was diluted into buffer containing 2 mM l-trp. As the KD of l-tryptophan binding to WT TR is ~60 μM (Arvidson et al. 1986, Jin et al. 1993), >99% of WT-TR is holorepressor in this buffer. When WT holorepressor was passed over the same DNA surfaces, there was a rapid increase in signal intensity, indicating that WT holorepressor was bound (Fig. 2 ▶, a3). As soon as buffer flow was resumed, there was a sharp decrease in signal intensity (Fig. 2 ▶, d3). This is expected, as there is no l-tryptophan ligand in the running buffer, and therefore the WT holorepressor is rapidly converted to the apo form. To eliminate the possibility that the sample of aporepressor had in some way been inactivated, we mixed the apo and holorepressor solutions together in equal proportions, and repeated the experiment. The addition of tryptophan ligand (present in the holorepressor solution) was sufficient to fully restore activity (Fig. 2 ▶, a4). Finally, we bound holorepressor to the operator-DNA surface, but instead of washing with tryptophan-free buffer, we washed with buffer containing 2 mM tryptophan, to observe the rate of dissociation of holorepressor (Fig. 2 ▶, d5). The presence of tryptophan in the wash buffer retarded the dissociation of TR from the surface. In this experiment, several different natural operators were used, and it is clear that the stability of the holorepressor complex depends on the sequence of the immobilized operator (Fig. 2 ▶, d5, A,B,C).

Figure 2.

Figure 2.

Open in a new tab

SPR traces of WT TR binding to DNA-coated chip surfaces. (a1–a5) Association phases 1–5; 1: 20 nM apo WT TR, 2: 1 μM apo WT TR, 3: 1 μM holo WT TR, 4: (50:50) mixture of the apo- and holo- proteins used in a2 and a3, respectively, 5: 1 μM holo WT TR. (d1–d5) Dissociation phases 1–5. No l-tryptophan was used in the dissociation buffer of 1–4. l-trytophan (2 mM) was included in the buffer during dissociation period 5. Three different operator surfaces are shown: A, 39-AroH; B, 40-EDCBA; and C, 40-trpR. The data show several large spikes as the traces result from the subtraction of data from a control surface on which no DNA was immobilized. These spikes have been truncated by the removal of several data points for presentation purposes.

Binding stoichiometry

Because the signal intensity is directly proportional to the amount of material bound to the surface, in theory it is possible to determine the stoichiometry of the protein–DNA interaction directly. In practice, this is less than straightforward for several reasons. First, the operator DNA is highly symmetrical, and has a high propensity to form hairpins and other nonfunctional structures. Although these will not bind protein, they will bind to the surface through the biotin/streptavidin interaction, and so an unknown fraction of the DNA bound may be nonfunctional. Second, some of the DNA may be bound in an inaccessible orientation, again making it difficult to determine the amount of functional DNA available. Finally, WT TR binds tandemly to DNA, depending on the operator sequence (Kumamoto et al. 1987), and above 100 μM forms increasingly large nonspecific aggregates bound to the DNA (Fig. 3 ▶; Reedstrom et al. 1997). It is therefore difficult to determine the response that is attributable to a single dimer binding. In an attempt to determine the stoichiometry of binding of repressor to DNA, the surface was titrated with protein over a wide concentration range (Figs. 3 ▶, 4 ▶). Although WT TR seems to approach a temporary saturation at concentrations in the range 150–250 nM, at higher concentrations the amount of protein bound to the surface continues to rise at all concentrations tested, up to 300 μM. In fitting the data to various models, we decided not to include oligomerization effects, as this would yield too many variable parameters for the fit to be meaningful. Instead, we observed carefully at which concentrations such an effect became significant for each operator surface, and limited our analysis to data obtained at concentrations below this threshold.

Figure 3.

Figure 3.

Open in a new tab

Binding response of WT holorepressor to three different operator sequences. Empty circles, 40-EDCBA (Fig. 1 ▶, [4]); filled circles, 40-trpR (Fig. 1 ▶, [4]); crossed circles, 39-AroH (Fig. 1 ▶, [5]). Resonance units were measured when the binding approached equilibrium, ~3 min after the start of the protein injection. The buffer included 2 mM l-tryptophan ligand.

Figure 4.

Figure 4.

Open in a new tab

Binding of WT holorepressor to two different operators. (A, squares) 40-trpR; (B, circles) 40-EDCBA. The data have been normalized on the Y-axis such that 1 unit represents 1 dimer bound per operator (as determined from the RMAX, fitted to an assumed 1:1 binding model). The symbols are used only to differentiate the lines; the lines are the actual data points.

Binding of WT holorepressor to 20-Consensus ([1] in Figure 1 ▶)

Mass transfer occurred even at low operator loading levels (see Materials and Methods); therefore, it was included in the model used for fitting in our analysis of the binding curves. We found in this case that we needed to limit our analysis to those experiments that had protein concentrations on the order of the KD (2, 5 nM), which yielded excellent fits to the data. Fitting the 2 nM and 5 nM data sets independently gave the same KD; 3.1 nM, but jointly analyzing these 2 curves, fitting to common values for all parameters, increased the fitted value of KD to 3.7 nM, with an accompanying worsening of fit (χ2 = 2.5). This indicates that there is already a slight deviation from simple 1:1 kinetics at 5 nM. Nevertheless, including data up to 50 nM does not change the KD much further; subsequent global analysis at higher concentrations (2, 5, 10, 20, 50 nM TR) yields a KD of 4.3 nM. When all the data (Fig. 5A ▶) are jointly analyzed, including data up to 1000 nM where nonspecific binding is significant, the KD remains the same, with additional nonspecific binding (beyond the fitted value of RMAX) being fitted by the analysis software as an increase in the refractive index parameter as it does not follow 1:1 binding kinetics.

Figure 5.

Figure 5.

Open in a new tab

Comparison of WT TR and AV77 binding to 20-consensus. (A) WT holorepressor, (B) AV77 holorepressor, (C) WT aporepressor, and (D) AV77 aporepressor. Protein concentrations are as shown in the legend to the right of each pair of graphs. The buffer in (A) and (B) contained 2 mM l-tryptophan; the buffer in (C) and (D) contained no added tryptophan.

Overall, the simplest explanation of the data is that the first repressor dimer binds with a KD of about 3.1 nM (RMAX ~230), with a second, subsequent dimer binding with a very similar or slightly higher KD. This would result in an increase of the fitted RMAX parameter at higher protein concentrations but very little change in the value of KD, as is observed. Although the value of the KD is well defined, with any attempt to fix it at a different value resulting in a significant worsening of fit, the values of ka and kd are not individually well-defined. Setting ka at any fixed value from 107 to 109 does not alter the fit, as the kd simply shifts to compensate. The fitted RMAX value at 2 nM (220 RU) indicates that about two dimers bind at concentrations below 500 nM, where a temporary plateau is reached. This assumes that 100% of the DNA bound to the chip is available for repressor binding, which has not been determined.

It should be noted that the apparent KD obtained from inspection of the Req versus [TR] plots (Fig. 6 ▶) is greater than that obtained from analysis of the direct binding curves. This indicates that the surface concentration of the DNA is not sufficiently low in comparison with the protein concentrations used for such an apparent KD to be valid.

Figure 6.

Figure 6.

Open in a new tab

Comparison of WT TR and AV77 binding to 20-consensus. Empty circles, WT; filled circles, AV77; solid line, holorepressor; broken line, aporepressor. Resonance units were measured when the binding approached equilibrium (Req), ~12 min after the start of the protein injection.

Binding of WT holorepressor to 40-EDCBA

Binding of WT holorepressor to the 40-EDCBA operator ([3] in Fig. 1 ▶) fit reasonably well to a simple 1:1 binding model, within the range 4–20 nM protein concentration. Fitting to this model, a KD of 14.1 nM was obtained. It has been shown previously at pH 6 (Liu and Matthews 1993) that binding of two dimers occurs at concentrations above ~0.1 nM, with apparent cooperativity, from which a single, apparent KD of 0.3 nM was obtained. Although the difference in KD can be expected, there is no reason to expect a difference in stoichiometry based on the pH difference (7.5 versus 6). Accordingly, we attempted fitting the data to a model in which a second dimer of TR binds after the first. This did not improve the fit, but did make the rates less well-defined, as is to be expected from the increase in the number of parameters. Estimated association rates of the presumed second dimer were varied in the fits from 0 to 108 (1/M.s), and all fit equally well to the data. The apparent KD did not vary much, however, only decreasing to ~12 nM. We can therefore only state that the binding of a second dimer is compatible with the data, but is not required to fit the data. This is reasonable if the binding constants for the first and second dimers are approximately the same.

Binding of WT holorepressor to 40-trpR

Gel-retardation studies have shown that WT holorepressor binds to 40-trpR ([4] in Fig. 1 ▶) in a different manner than to either 40-EDCBA or 40-AroH (Liu and Matthews 1993). Whereas the latter two operators show clearly defined bands corresponding to complexed DNA at all protein concentrations above 0.1 nM, 40-trpR shows no such band at any concentration tested (up to 100 nM). Despite this, they obtained a KD (2.2 nM under their conditions) by plotting the reduction in the intensity of the band corresponding to free DNA. This is about a 10-fold increase compared with their results for both the EDCBA and aroH operators. Using Biacore, we also find that the KD (165 nM) for 40-trpR is ~10-fold that of 40-EDCBA, although the KDs that we obtain at pH 7.5 are ~50-fold greater than those obtained at pH 6 by Liu and Matthews (1993). The half-life of the complex that we determine for the TR/40–trpR complex is ~4 sec, compared with ~30 sec for the 40-EDCBA operator (Fig. 4 ▶).

Binding of the AV77 super aporepressor

The binding of both the apo and holo forms of WT TR and AV77 is shown in Figures 5 and 6 ▶ ▶. It is immediately clear that in this in vitro study, AV77 aporepressor binds to 20-consensus at lower protein concentrations than does WT aporepressor (Fig. 5C,D ▶). Furthermore, at equivalent protein concentrations, approximately double the amount of AV77 aporepressor is bound to the consensus operator than is WT aporepressor (Fig. 5C,D ▶). The order is reversed for the ligand-bound holorepressor, with WT TR now binding at lower concentrations than the mutant. The KD may not be safely estimated by visual examination of the binding curves (Fig. 6 ▶) as this estimation depends on the DNA target concentration being at least an order of magnitude below the KD, which has not been shown. Nevertheless, the relative order of the KD’s can be seen from the figure, as the amount of operator immobilization is identical in each. A detailed kinetic analysis of the SPR curves gives: KD of the WT holorepressor is 3–4 nM; WT aporepressor, 5–20 μM; and AV77 aporepressor, 2–3 μM. The SPR curve of AV77 holorepressor is clearly bi-exponential or higher, and does not fit well to a 1:1-binding model (Fig. 5B ▶). In contrast with the aporepressor results, comparing, for example the 2 nM data set from Figure 5 ▶, panels A and B, shows that less AV77 holorepressor is bound at lower protein concentrations than for WT holorepressor.

As the on and off times are close to the instrumental dead time (1–2 sec) the values are not precise and ranges of values are given. In selecting the curves to be fitted, we chose the lowest concentrations that gave reasonable signal strength, to have the slowest, and hence most accurately measureable, rates. We also made sure that the concentrations chosen for the fitting were close to the KD indicated from a preliminary analysis as this yields more reliable data (BIAcore Instrument manual).

As the fitted KD’s are ~5-fold lower than would be estimated from the binding curves (Fig. 6 ▶), we conclude that the DNA concentration is not limiting and that the values obtained from a full analysis of the sensorgram are more accurate. The results agree very well with those obtained using an alkaline phosphase assay (Marmorstein et al. 1991). In contrast with all previous studies, however, this real-time assay was able to show that the binding and dissociation of AV77 holorepressor follows biphasic kinetics. Although the KD (app) of AV77 holorepressor appears weaker than that of the WT holorepressor, the lifetime of the AV77 holorepressor operator complex is much longer. This effect is not seen for the AV77 aporepressor (Fig. 5 ▶).

Discussion

Surface plasmon resonance studies of tryptophan repressor/operator interactions

In this study, we have remeasured the binding constants of TR to several natural operators (Fig. 1 ▶). The results we have obtained show that the trpR system is ideally suited for SPR measurement, with the system responding to corepressor and the different natural operators as one might expect (Fig. 2 ▶). The KD obtained for the WT holorepressor–20-consensus interaction from the analysis of the binding curves is in agreement with the literature (3 nM, Klig et al. 1987; 7 nM, Klig and Yanofsky 1988; 5 nM, Marmorstein et al. 1991) although gel retardation studies at lower pH give significantly lower values (0.2 nM, Carey et al. 1991; 0.1–0.3 nM, Liu and Matthews 1993,1994). The nonphysiological pH used in the gel studies was chosen because the complex is more stable at that pH, and therefore the values obtained, although less physiologically relevant, become more reliable. For the trpR operator, the complex with WT TR was insufficiently stable to be measured directly by gel retardation (Liu and Matthews 1993), and so the binding constant was inferred from the decrease in free operator alone. In contrast to gel retardation studies, instability of the complex is no obstacle to its direct visualization using SPR, and so the KD was able to be determined directly, rather than indirectly (Fig. 4 ▶). We have confirmed that the WT TR-trpR complex is indeed anomalously instable, with the half-life of the complex being approximately 8 × shorter than for the EDCBA operator.

Limitations that have restricted our analysis include instrumental factors related to the immobilization of the ligand, most notably the tradeoff between mass transfer limitation of the rates versus low signal strength. The greatest difficulty we have encountered, however, is unavoidable using any technique; TR binds in multiple tandem binding modes, and freely oligomerizes both on and off the DNA. Some of these binding events are of very similar affinity, leading to difficulty in extracting individual equilibrium constants. In addition, at higher concentrations, it appears that relatively nonspecific aggregation occurs, rendering an accurate estimation of RMAX impossible.

Super-repressor status of AV77

AV77 was originally named as a super-repressor on the basis that it repressed the trpEDCBA operon at lower concentrations of intracellular l-tryptophan than the WT (Kelley and Yanofsky 1985). The in vivo result was confirmed using a challenge-phage assay, which found that the apo- (Arvidson et al. 1993) but not the holo- (Shapiro et al. 1993) AV77 mutant protein had increased activity. Remarkably, the binding constants of the aporepressors were found to be identical in vitro (Hurlburt and Yanofsky 1990). The same was found to be true of the holorepressors (Hurlburt and Yanofsky 1990; Liu and Matthews 1994). In contrast, Marmorstein et al. (1991) determined that the binding constant of aporepressor to operator DNA was eightfold higher for AV77 than for WT TR using an alkaline phosphatase binding assay. This was dismissed as an explanation of the super-aporepressor status of the mutant on the grounds that the binding of AV77 holorepressor was determined to be 2.3-fold weaker than WT holorepressor in the same study. Because the ligand binding of the free repressor was unaltered by the mutation, it was deduced that reduced ligand affinity of the DNA-bound form should compensate, leaving still a 2.3-fold decrease in the assembly of the holorepressor/operator complex (Arvidson et al. 1993). This argument assumed that the formation of the aporepressor/operator complex should not have an effect on the rate of transcription. It is unclear to us why this assumption was made.

An alternative explanation was proposed by Gryk et al. (1996), in which it was suggested that AV77 aporepressor might bind less strongly to alternative operators, thereby increasing the population available to bind to the consensus operator sequence. We find that AV77 aporepressor binds more strongly than the WT aporepressor in vitro, and that the results of the original filter binding studies were in error. No alternative explanation is therefore needed.

Equilibrium results obtained using fluoresence anisotropy agree that the AV77 mutant protein binds more strongly to the operator (Grillo and Royer 2000). Our results show that not only does the AV77 aporepressor bind at lower concentrations of tryptophan than WT aporepressor under equilibrium conditions, but also that it does so rapidly and in a meaningful timeframe for transcriptional control. As we see no reason why the aporepressor complex should be unable to prevent transcription in the same way that the holorepressor complex does, we ascribe the apo-super-repressor status of AV77 to the lower concentration at which it binds (Figs. 5 ▶, 6 ▶). Our findings do not rule out that this may be caused in part by a lower tendency of AV77 aporepressor to oligomerize in solution (Grillo and Royer 2000).

Real-time data that we have obtained for the AV77 holorepressor show that equilibrium analysis only tells half the story. Whereas the WT holorepressor appears to bind more strongly to consensus operator on the basis of an equilibrium KD, this does not reflect the reality that the lifetime of the AV77 complex is appreciably longer (Fig. 5B ▶). This has not been observed previously. As the protein is operator-bound, this reflects an activity that is not accounted for by its availability in solution. Rather, there is some feature of the operator-bound AV77 holorepressor that slows down its release from the DNA considerably. At this point, we prefer not to speculate on its origin.

In summary, we have been able to definitively discriminate between conflicting results found with more traditional methods in the literature by using SPR. Despite these limitations, we have been able to conclusively confirm results in the literature, and more importantly, to correct several major errors.

Materials and methods

Materials

Oligonucleotides were obtained (highly purified salt-free) from MWG Biotech (High Point, NC) and used without further purification. Double-stranded operators were annealed by mixing both strands at a concentration of 10 μM, before heating at 95°C for 5 min, followed by cooling to 20°C at 1°C/min. The concentration was then reduced to 4 μM for storage. The operators were diluted further to 200 nM just before immobilization on the Biacore SA chips. Restriction enzymes were obtained from New England Biolabs (Cambridge, MA). Surface plasmon resonance chips were obtained from BIAcore.

Bacterial strains and plasmids

The plasmid PJPR2, which overexpresses WT TR, was obtained as a gift from Professor Yanofsky at Stanford University (Stanford, CA), and expressed in E. coli CY15071, also obtained from his laboratory.

Site-directed mutagenesis and subcloning

The AV77 mutant protein was prepared using the Stratagene Quickchange mutagenesis kit, and the oligonucleotides 5′-AAT GAACTCGGCGTAGGCATCGCGACG and 5′-CGTCGCGAT GCCTACGCCGAGTTCATT.

Overexpression and purification of proteins

WT TR was prepared as described previously (Paluh and Yanofsky 1986). Protein was buffer-exchanged into 100 mM potassium phosphate, 20 mM KCl, pH 7.5 and stored frozen in aliquots until use.

Binding studies

Surface plasmon resonance (SPR) was carried out on a Biacore 3000 instrument, using streptavidin chips from the manufacturer. Running buffer was 100 mM potassium phosphate, 20 mM KCl, 0.005% Tween 20, pH 7.5, either without (aporepressor) or with (holorepressor) 2 mM l-tryptophan. All running buffers were filtered through a 20 μM membrane to remove debris and degassed at running temperature before use. Protein solutions were thawed, filtered, then assayed for protein concentration using an extinction coefficient of 15000 cm−1 M−1 per monomeric subunit. All concentrations reported here are for the dimeric form of the repressor. The protein was then diluted into the running buffer, so as to match the refractive index of the solutions as closely to that of the running buffer as possible. After each binding experiment, the DNA surface was regenerated by stripping the protein from the surface with 1 M NaCl for at least 2 × 30 sec. Occasionally, this was increased to 2 × 90 sec if the surface was not completely regenerated. This is dependent on the stability of the protein–DNA complexes.

During the SPR experiments, the flow rate was routinely set to 50 μL/min to reduce mass transport, and the surface immobilization of the DNA was usually <150 RU (see Morton and Myszka 1998 or the Biacore instrument manual for a discussion of mass transport effects). We attempted to use lower immobilization levels (50 RU) but even at this low level, we detected mass transport. As the instrumental noise and drift were significant at these low immobilization levels, we decided against reducing the DNA immobilization levels further, but instead introduced mass transport into the model used to fit the data. Control surfaces with no DNA attached were used to correct for refractive index changes between the samples and the running buffer.

The results were analyzed using BIAevaluation 3.0, supplied by the instrument manufacturer. As the refractive index of the samples and controls were matched, no correction for this was included in the models except as described in the results section.

Acknowledgments

This work was funded under grant 5RO1 GM33385–16 from the NIH. We thank the Davis Laboratory at Stanford for the use of their BIAcore 3000.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 USC section 1734 solely to indicate this fact.

Abbreviations

  • TR, tryptophan repressor

  • WT, wild-type

  • RU, resonance units

  • RMAX, RU when all the operator bound to the chip is occupied 1:1 with protein

Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.0305703.

References

  1. Arvidson, D.N., Bruce, C., and Gunsalus, R.P. 1986. Interaction of the Escherichia coli trp aporepressor with its ligand, L-tryptophan. J. Biol. Chem. 261 238–243 [PubMed] [Google Scholar]
  2. Arvidson, D.N., Pfau, J., Hatt, J.K., Shapiro, M., Pecoraro, F.S., and Youderian, P. 1993. Tryptophan super-repressors with alanine 77 changes. J. Biol. Chem. 268 4362–4369. [PubMed] [Google Scholar]
  3. Cann, J.R. 1989. Phenomenonological theory of gel electrophoresis of protein–nucleic acid complexes. J. Biol. Chem. 26417032–17040. [PubMed] [Google Scholar]
  4. Carey, J. 1988. Gel retardation at low pH resolves trp-repressor-DNA complexes for quantitative study. Proc. Natl. Acad. Sci. 85 975–979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Carey, J., Lewis, D.E.A., Lavoie, T.A., and Yang, J. 1991. How DOES trp repressor bind to its operator? J. Biol. Chem. 266 24509–24513. [PubMed] [Google Scholar]
  6. Gryk, M.R., Jardetzky, O., Klig, L.S., and Yanofsky, C. 1996. Flexibility of DNA binding domain of trp repressor required for recognition of different operator sequences. Protein Sci. 5 1195–1197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Grillo, A.O. and Royer, C.A. 2000. The basis for the super-repressor phenotypes of the AV77 and EK18 mutants of trp repressor. J. Mol. Biol. 295 17–28. [DOI] [PubMed] [Google Scholar]
  8. Hurlburt, B.H. and Yanofsky, C. 1990. Enhanced operator binding by trp super-repressors of Escherichia coli. J. Biol. Chem. 265 7853–7858. [PubMed] [Google Scholar]
  9. Jeeves, M., Evans, P.D., Parslow, R.A., Jaseja, M., and Hyde, E.I. 1999. Studies of the Escherichia coli Trp repressor binding to its five operators and to variant operator sequences. Eur. J. Biochem. 265 919–928. [DOI] [PubMed] [Google Scholar]
  10. Jin, L., Yang, J., and Carey, J. 1993. Thermodynamics of ligand binding to trp repressor. Biochemistry 32 7302–7309. [DOI] [PubMed] [Google Scholar]
  11. Kelley, R.L. and Yanofsky, C. 1985. Mutational studies with the trp repressor of Escherichia coli support the helix-turn-helix model of repressor recognition of operator DNA. Proc. Natl. Acad. Sci 82 483–487. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Klig, L.S. and Yanofsky, C. 1988. Increased binding of operator DNA by trp super-repressor EK49. J. Biol. Chem. 263 243–246. [PubMed] [Google Scholar]
  13. Klig, L.S., Crawford, I.P., and Yanofsky, C. 1987. Analysis of trp repressor-operator interaction by filter binding. Nucleic Acids. Res. 15 5339–5351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kumamoto, A.A., Miller, W.G., and Gunsalus, R.P. 1987. E. coli tryptophan repressor binds multiple sites within the aroH and trp operators. Genes Dev. 1 556–564. [DOI] [PubMed] [Google Scholar]
  15. LeTilly, V. and Royer, C.A. 1993. Fluorescence anisotropy assays implicate protein–protein interactions in regulating trp repressor DNA binding. Biochemistry 32 7753–7758. [DOI] [PubMed] [Google Scholar]
  16. Liu, Y.-C. and Matthews, K.S. 1993. Dependence of trp repressor-operator affinity, stoichiometry, and apparent cooperativity on DNA sequence and size. J. Biol. Chem. 268 23239–23249. [PubMed] [Google Scholar]
  17. ———. 1994. trp repressor mutations alter DNA complex stoichiometry. J. Biol. Chem. 269 1692–1698. [PubMed] [Google Scholar]
  18. Marmorstein, R.Q., Sprinzl, M., and Sigler, P.B. 1991. An alkaline phosphatase protection assay to investigate trp repressor/operator interactions. Biochemistry 30 1141. [DOI] [PubMed] [Google Scholar]
  19. Morton, T.A. and Myszka, D.G. 1998. Kinetic analysis of macromolecular interactions using surface plasmon resonance biosensors. Methods Enzymol. 295 268–294. [DOI] [PubMed] [Google Scholar]
  20. Oda, M., Furukawa, K., Sarai, A., and Nakamura, H. 1999. Kinetic analysis of DNA binding by the c-Myb DNA-binding domain using surface plasmon resonance. FEBS Lett. 454 288–292. [DOI] [PubMed] [Google Scholar]
  21. Paluh, J.L. and Yanofsky, C. 1986. High-level production and rapid purification of the E. coli trp repressor. Nucleic Acids. Res. 14 7851–7860. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Parsons, I.D., Persson, B., Mekhalfia, A., Blackburn, G.M., and Stockley, P.G. 1995. Probing the molecular mechanism of action of co-repressor in the E. coli methionine repressor-operator complex using surface plasmon resonance (SPR). Nucleic Acids Res. 23: 211–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Reedstrom, R.J., Brown, M.P., Grillo, A., Roen, D., and Royer, C.A. 1997. Affinity and specificity of trp repressor-DNA interactions studied with fluorescent oligonucleotides. J. Mol. Biol. 273 572–585. [DOI] [PubMed] [Google Scholar]
  24. Shapiro, M., Arvidson, D.N., Pfau, J., and Youderian, P. 1993. The challenge-phage assay reveals differences in the binding equilibria of mutant Escherichia coli Trp super-repressors in vivo. Nucleic Acids Res. 21 5661–5666. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Tsoi, P.Y. and Yang, M. 2002. Kinetic study of various binding modes between human DNA polymerase β and different DNA substrates by surface-plasmon-resonance biosensor. Biochem. J. 361 317–325. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society

RESOURCES