Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Dirla, S. Articles by Schleif, R. Search for Related Content PubMed PubMed Citation Articles by Dirla, S. Articles by Schleif, R.

 Previous Article  |  Next Article 

Journal of Bacteriology, April 2009, p. 2668-2674, Vol. 191, No. 8
0021-9193/09/$08.00+0     doi:10.1128/JB.01529-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Constitutive Mutations in the Escherichia coli AraC Protein

Stephanie Dirla, John Yeh-Heng Chien, and Robert Schleif^*

Biology Department, Johns Hopkins University, 3400 N. Charles St., Baltimore, Maryland 21218

Received 29 October 2008/ Accepted 31 January 2009

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The Escherichia coli AraC protein represses and induces the araBAD operon in response to the absence or presence of L-arabinose. Constitutive mutations in the AraC gene no longer require the presence of L-arabinose to convert AraC from its repressing to its inducing state. Such mutations were isolated directly by virtue of their constitutivity or by their resistance to the nonmetabolizable arabinose analog, D-fucose. The majority of the constitutive mutations lie within the same residues of the N-terminal regulatory arm of AraC. Two, however, were found in the core of the dimerization domain. As predicted by the light switch mechanism of AraC, constitutive mutations increase the susceptibility of the N-terminal arms to digestion by trypsin or chymotrypsin, suggesting that these mutations weaken or disrupt the arm structure required for repression by AraC. Fluorescence, circular dichroism, and cysteine reactivity measurements show that the constitutive mutations in the core of the dimerization domain lead to a weakening of the support for the arms and reduce the stability of the minus-arabinose arm structure. These mutations also weaken the interaction between the two-helix bundle and the β-barrel subdomains of the dimerization domain and reduce the structural stability of the β-barrels.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In the absence of L-arabinose, AraC protein, the regulator of the arabinose operon in Escherichia coli, represses expression of the araBAD operon by binding to two well-separated DNA half-sites and forming a DNA loop (3, 11, 14). The loop itself helps repress expression by hindering the access of RNA polymerase to the p_BAD promoter (1). It also indirectly decreases expression by engaging one of the DNA binding domains of dimeric AraC in binding to the distal araO₂ half-site instead of allowing it to bind to the araI₂ half-site whose occupancy activates transcription (28). AraC protein loops the DNA in the absence of arabinose because in this condition its two DNA binding domains are held in orientations and positions such that DNA looping is energetically favored. Upon binding arabinose, the DNA binding domains are less rigidly held or not held at all, and the DNA binding domains are then free to bind to the adjacent I₁ and I₂ half-sites (2), a state from which transcription is highly activated (28) (Fig. 1).

View larger version (17K): [in this window] [in a new window]

FIG. 1. Depiction of AraC and pBAD in the presence or absence of arabinose. In the absence of arabinose, the arms of AraC rigidly hold the DNA binding domains in a conformation that makes it energetically more favorable for the protein to bind to the I1 and O2 half-sites. This creates a loop in the DNA that represses transcription. When arabinose is present, the arms release the DNA binding domains and AraC binds to I1 and I2. The opening of the loop gives RNA polymerase more free access to the promoter, and the positioning of a DNA binding domain of AraC at I2 stimulates transcription initiation by RNA polymerase. For simplicity in representation, the hinge of the arm and the dimerization interface are represented on opposite faces of the dimerization domain; in reality these parts lie on the same side of the dimerization domain.

The N-terminal arms of AraC play a critical role in controlling the protein's two regulatory states (21). In the absence of arabinose, the arms are structured by their interaction with the dimerization domains (18). The DNA binding domains then bind to the arms plus the dimerization domains, and the protein is in its rigid repressing state (9, 19, 26). When arabinose binds to the dimerization domains of the protein, the N-terminal arms adopt a totally different structure (5, 16, 18), and the DNA binding domains are no longer rigidly held (27). Hence, the protein is free to bind to adjacent DNA half-sites and activate transcription (11). This model has been termed the light switch mechanism because the structure of the arms turns transcription on or off.

Constitutive mutants of AraC induce even though arabinose is absent (4). Within the framework described above, two types of defective interaction could lead to constitutivity. First, mutations could interfere with the ability of the arms to form the structures on the dimerization domains that are required for repression. As a result, in the absence of arabinose, the DNA binding domains would not be held, and constitutivity would result. A second possibility is that while the arms could form the structures necessary for immobilization of the DNA binding domains, important contacts to the DNA binding domains may be missing, for example, by the alteration of side chains required in the interaction. Previously, we found that most substitutions in the N-terminal arm of AraC and some substitutions on the surface of the DNA binding domain resulted in constitutivity (17, 19, 26), but these results do not exclude the possibility that alterations elsewhere in the protein could also generate constitutivity. The locations and/or character of additional constitutive mutations could support, extend, or refute our current understanding AraC protein's mechanism of action as represented in the light switch model. Therefore, here we describe the isolation of mutations on the basis of their constitutivity, their mapping, and an investigation of their mechanisms of action.

Englesberg et al. (4) discovered that a sizeable fraction of mutants resistant to the growth-inhibitory effects of the nonmetabolizable arabinose analog, D-fucose, are constitutive (see also references 1, 13, 15, and 22). This provides a simple isolation method. Constitutive mutations may also be isolated directly by their ability to induce expression of the arabinose operon in the absence of L-arabinose (15). We therefore examined constitutive mutations isolated by these two methods. We also verified the prediction that constitutive mutations reduce the stability of the arm structure required for repression.

The vast majority of the constitutive mutations that we isolated were in the N-terminal arm. Two of the constitutive mutations found were, however, not in the arm. They were in the core of the dimerization domain, lying in the interface between the β-barrel structure in which arabinose binds and the pair of -helices that dimerize the protein. We examined these two mutations to determine whether the mechanism by which they generate constitutivity might be different from that of the arm mutations.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Plasmids and mutagenesis. A plasmid encoding AraC protein and placing green fluorescent protein (GFP) expression under the control of p_BAD, pBAD-GFPuv (Clontech), was used for the isolation of constitutive mutations. Fucose-resistant constitutive mutants were isolated after using the GeneMorph II EZClone domain mutagenesis procedure (Stratagene) to mutate the full-length AraC gene on the pBAD-GFPuv plasmid, transformation into SH321 cells, and selecting colonies from minimal plates containing 0.5% L-arabinose and 1.0% D-fucose. Constitutive mutants not isolated via fucose resistance were found by passing pBAD-GFPuv through the E. coli mutator strain XL1-Red (Stratagene) and then transforming the plasmid into SH321 cells (8), spreading them on YT plates (20), and picking green fluorescent colonies. Plasmid DNA was isolated from cultures grown from single colonies with the Wizard Plus Miniprep DNA purification system (Promega), and mutations were identified by sequencing.

AraCTF (12), a pET21b plasmid, was previously constructed to express the arm plus the dimerization domain of the AraC protein. It codes for residues 1 to 182, followed by a short C-terminal Leu-Glu linker and a His₆ tag. Mutations were introduced into the arm plus the dimerization domain on this plasmid by using QuikChange site-directed mutagenesis (Stratagene). Plasmid DNA was isolated and sequenced as described above. In addition to the indicated mutations, all arm-plus-dimerization domain proteins also contained the mutation Y31V, which increases the solubility of the protein in the absence of arabinose (25).

Dimerization domain purification. Wild-type and mutant arm plus dimerization domain were expressed and purified from cultures of E. coli BL21(DE3) cells (Stratagene) essentially as described previously (12). Cells were grown in 2-liter baffle flasks with 500 ml of YT medium plus 100 µg of ampicillin/ml, with continuous shaking at 37°C. At an optical density at 600 nm of ca. 0.5 to 1.0, expression was induced with 1 mM IPTG (isopropyl-β-D-thiogalactopyranoside) and the addition of 0.2% L-arabinose to increase dimerization domain solubility. After 3 h, the cells were harvested by sedimentation at 6,000 x g for 15 min. Cells were lysed by using an Avestin EmulsiFlex C3 homogenizer in lysis buffer consisting of 15 mM Tris-Cl (pH 8.0), 0.1 M NaCl, 5% glycerol, 10 mM MgCl₂, 50 mM L-arabinose, and 10 µg each of DNase I and RNase A/ml. Insoluble cellular debris was removed from the lysate by centrifugation at 10,000 x g for 15 min. The supernatant was incubated with 4 to 5 ml of Ni-NTA agarose beads (Qiagen), followed by gentle rocking at 4°C for at least 3 h. Next, the slurry was centrifuged at 1,000 x g for 5 min to settle the beads, and the supernatant was discarded. The beads were washed with 15 mM Tris-Cl (pH 8.0), 0.1 M NaCl, 50 mM L-arabinose, and 10 mM imidazole until the optical density at 280 nm of the wash was <0.05. Bound protein was eluted with 2 column volumes of elution buffer consisting of 15 mM Tris-Cl, 10 mM NaCl, 50 mM L-arabinose, and 2 M imidazole in which the pH was adjusted to 8.0 after all components were present. For protein used in urea melting experiments, the His₆ tag of 1.0 mg of arm plus dimerization domain/ml was cleaved by digestion with 5 µg of trypsin/ml at 4°C overnight. Trypsin was then inhibited with 5 µg of soybean trypsin inhibitor/ml. Anion-exchange chromatography using a Pharmacia Mono-Q HR 5/5 1-ml column on a Pharmacia fast-performance liquid chromatography system was performed in 15 mM Tris-Cl (pH 8.0)-50 mM L-arabinose buffer with an elution gradient of 10 mM to 1 M NaCl over 40 ml. Arm plus dimerization domain elutes at roughly 250 mM NaCl, with a protein concentration of 5 mg/ml, with purity of >95%. Fractions were analyzed by sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis and stained with GelCode blue stain reagent (Pierce). The protein concentration was determined based on the absorbance at 280 nm, assuming a molar extinction coefficient of 37,410 M^–1 cm^–1 estimated from the sequence composition (6).

Protease sensitivity experiments. Arm plus dimerization domain still containing the His₆ tag was digested in 100 µl of 0.06 mg of protein/ml with 0.01 mg of trypsin/ml at room temperature in 0.01 M Tris-Cl (pH 7.5), 0.1 M NaCl, 2 mM dithiothreitol, and 0.1 mM EDTA, with or without 10 mM L-arabinose. At the times indicated, 8 µl from the digestion was added to 2 µl of 5x sample buffer (0.3 M Trizma base, 10% SDS, 3.5 M 2-mercaptoethanol, 50% glycerol, 0.2% bromophenol blue) containing 0.25 µl of 6.9 mg of phenylmethylsulfonyl fluoride/ml freshly dissolved in ethanol. The reaction samples were analyzed on SDS-14% polyacrylamide gels.

Arm plus dimerization domain sensitivity to chymotrypsin cleavage was assessed by incubating 60 µl of 2.0 mg of dimerization domain/ml with 0.02 mg of bovine -chymotrypsin/ml in 15 mM Tris-Cl (pH 8.0)-50 mM NaCl with or without 50 mM L-arabinose at room temperature. Digestion was stopped at various times by combining 6 µl of reaction mixture with 1 µl of 1.0 mg of phenylmethylsulfonyl fluoride/ml freshly dissolved in ethanol. The reaction was analyzed as described above.

Mass spectrometry. Arm plus dimerization domain digested with trypsin or chymotrypsin, as described above, was exchanged into 0.1% trifluoroacetic acid by passage through a 250-µl Sephadex G-10 spin column. Ten to one-hundred picomoles of protein was spotted onto an Applied Biosystems Voyager sample plate, with sinapinic acid in 50% acetonitrile and 0.05% trifluoroacetic acid used as the matrix. An Applied Biosystems Voyager DE-STR matrix-assisted laser desorption ionization-time of flight instrument at the mass spectrometry facility of the Johns Hopkins School of Medicine was used for the mass analysis.

Urea denaturation measurements. (i) Tryptophan fluorescence. Stock solutions of 0 and 9 M urea solutions were prepared by dissolving urea in 15 mM Tris-Cl (pH 8.0)-50 mM NaCl, with or without 50 mM L-arabinose. Arm plus dimerization domain was added to each stock solution so that the protein concentration was 1 µM, and samples of intermediate urea concentrations were prepared by mixing various amounts of the two stock solutions. After mixing, samples were incubated for at least 1 h at room temperature. Tryptophan fluorescence was monitored at 24°C in 1-cm path length quartz cuvettes by exciting the samples at 295 nm with a 75-W xenon light source and monitoring the integrated emission from 310 to 400 nm. The validity of this protocol requires that the dimerization domain refold from 9 M urea. We verified that this is the case by repeating several of the denaturation experiments by directly adding urea to the folded protein instead of using the two stocks of protein plus urea.

(ii) Circular dichroism. Samples were prepared and equilibrated with urea as described above without L-arabinose at a protein concentration of 0.2 mg/ml. Circular dichroism was monitored from 240 to 215 nm in a 0.1-cm path length quartz cuvette using a Jasco J-710 spectropolarimeter. Scans were recorded with a scan rate of 20 nm/min, a bandwidth of 1 nm, and a response time of 2 s per point. Four scans per sample were performed to improve the signal-to-noise ratio.

(iii) DTNB assay. Reaction buffer containing 50 mM NaCl and 15 mM Tris-Cl (pH 8.0) was combined, in 1-cm path length quartz cuvettes, with various concentrations of urea dissolved in reaction buffer and 1 mM 5, 5'dithiobis-(2-nitrobenzoic acid (DTNB) dissolved in 50 mM phosphate buffer (pH 7.1) and 50 mM NaCl. A 25 Lambda UV/VIS (Perkin-Elmer) was blanked, and then arm plus dimerization domain was added so that the final protein concentration 10 µM. The absorbance at 412 nm was measured immediately after the addition of protein every second for 15 min.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Mutation spectra of two methods of constitutive isolation. Our first objective was to compare the fucose-resistant mutations and constitutive mutations isolated without a fucose resistance selection. If the mutational spectra of the two methods strikingly differ, then the two classes of mutants could be different and act through different mechanisms. For example, suppose that the only way to become fucose resistant were to alter the arm, and an incidental consequence of arm mutations is constitutivity, but that constitutivity could also result from mutations in the DNA binding domain that strengthen its interaction with RNA polymerase. In such a case, constitutive mutations isolated by their fucose resistance would be restricted to the arm, but constitutive mutations isolated merely by their expression of ara enzymes in the absence of arabinose would lie both in the arm and in the DNA binding domain.

The experiments used a plasmid in which expression of GFP has been placed under the control of the AraC gene and the araBAD promoter (24). Mutations were introduced into the AraC gene using GeneMorph II EZClone domain mutagenesis (Stratagene) or by passing the plasmid through XL1-Red cells (Stratagene). Constitutive mutations created using GeneMorph II EZClone domain mutagenesis were identified by selecting for growth on L-arabinose in the presence of the inhibitor, D-fucose. Those created by passage through XL1-Red cells were selected visually by identifying constitutive mutants as GFP positive in populations grown in the absence of arabinose and fucose. Table 1 shows that the two selection methods yield similar sets of residues whose mutation leads to constitutivity. This suggests that the underlying mechanism of constitutivity in the two groups is the same. The restricted number of different mutant residues found at each site is likely to be due to the structure of the genetic code, the fact that only single base changes were made, and that the mutagenesis methods that we used preferentially produce a restricted set of base changes, often being A-TG-C and G-CA-T, (16). Most of the mutations lie within residues 8 to 22 of the N-terminal arms of AraC. This emphasizes the critical importance of the arms in directing the protein either to repress or activate transcription. The unexpected positions of two constitutive mutations at residues 149 and 152 suggest that these two mutations may interfere with AraC repression by a mechanism different from that which functions when AraC responds to arabinose.

View this table: [in this window] [in a new window]

TABLE 1. Constitutive mutants and their locations, identity, and frequency of isolation

Structure of constitutive AraC arms. The results presented in the previous section found that the great majority of mutations generating constitutivity lie in the N-terminal regulatory arms. This shows that the arm residues play a major role in the immobilization of the DNA binding domains that is required for repression. According to the light switch mechanism, when the domains cannot be immobilized, AraC by default activates transcription, and the result is constitutivity. One reason that mutations in the arms could prevent immobilization of the DNA binding domains and, hence, fail to repress, is that in the absence of arabinose the arms do not form the structure on the dimerization domains necessary for immobilizing the DNA binding domains (18). Another reason is that the arms could form the proper structure, but the presumed contacts between the arms and the DNA binding domains that are necessary for immobilization may not be formed. Another prediction of the light switch mechanism involves low-level constitutive mutants that are further inducible by the addition of arabinose (17). The arms in such mutants should be less stably structured than wild-type arms but should become more structured upon arabinose addition.

To test the various predictions, we used a protease sensitivity assay. Structured proteins, or peptides in stable contact with a protein, usually are resistant to proteases (10). Unstructured proteins or peptides or quasi-stable regions are more sensitive to proteases. Wild-type AraC arm plus dimerization domain, residues 1 to 177, is highly resistant to trypsin digestion, but since the arms lack trypsin cleavage sites, this resistance says nothing about the arms. Through extensive screening (16, 18), we found that the function of AraC is largely unchanged by converting asparagine 16 to arginine in the N-terminal arms. Therefore, we introduced arginine at this position to provide a new potential trypsin cleavage site without changing the phenotype of AraC. Previously, we observed that wild type-like arms containing the N16R mutation are very slightly sensitive to trypsin cleavage in the presence or absence of arabinose and are only slightly more sensitive in the absence of arabinose due to a restructuring of the arm (18).

Figure 2 shows the trypsin sensitivity of the arms of wild type-like N16R dimerization domain and that of four constitutive mutants that were chosen for this experiment because they are further inducible by the addition of arabinose. These mutants came from the large set of previously characterized arm mutants (17). As anticipated, the rate of trypsin cleavage at R16 as observed on SDS gels and verified by mass spectrometry in constitutive mutants was found to be substantially increased compared to the wild type-like N16R background. The presence of arabinose reduced the trypsin sensitivity of G12T and H18L. This result indicates that the presence of arabinose alters the trypsin susceptibility of these arms, likely by stabilizing their structure. Presumably, the arabinose-stabilized arm structure is similar to the wild type in the presence of arabinose. These experiments were done using arm plus dimerization domain with the His₆ tail still attached to the protein. The unstructured His₆ tail was cleaved from the protein very rapidly, whereas it took considerably longer for the mutant arm to be removed. This suggests that either the protease sensitive mutant arms are structured, or that they spend a substantial portion of the time in a structured state.

View larger version (33K): [in this window] [in a new window]

FIG. 2. Trypsin digestion of the N-terminal regulatory arm in N16R and constitutive N16R mutants in the absence or presence of arabinose for the times indicated. Lysozyme (molecular weight, 14,000) and carbonic anhydrase (molecular weight, 30,000) size standards were included in the last two lanes of each gel. Arrows indicate the positions of the bands corresponding to the arm-dimerization domain-His6 and the digestion products. Lane identities are the same in all of the gels. The induced and uninduced expression levels of the constitutive mutants as a percentage of the fully induced wild-type expression levels are as follows: P8H, 30 and 200%; G12T, 50 and 400%; and H18L, 100 and 100% (17).

Examination of the unusual constitutive mutants. Most of the constitutive mutations that we found lie between residues 8 and 22. We did, however, find two constitutive mutations elsewhere, at positions 149 and 152 (Table 1). How do they generate constitutivity? The structures of the arm plus dimerization domain in the presence or absence of arabinose, Protein Data Bank entries 2ARC (23) and 1XJA (25), show that the arm plus dimerization domain consists of two subdomains, a sugar-binding β-barrel and a two-helix dimerization interface. Residues 149 and 152 lie within one of the -helices of the helical subdomain, with side chains pointing into the interface between the -helical and β-barrel subdomains and almost contacting residues 18 and 19 of the arm (Fig. 3A). Thus, there are two possibilities for the mechanism by which mutations E149F and A152V generate constitutivity. The mutations could directly alter the structural support of the arm's base by reducing the interactions between residues 18 and 19 of the arm with residues 149 and 152 of the -helical subdomain and, if so, the structure of the arms may be destabilized. Because the mutations introduce bulkier side chains into the interface between the β-sheet structure and the two -helices, it is also possible that they weaken the interaction between the two subdomains. This could reduce the structural stability of either or both of the subdomains or it could slightly open the subdomain interface (Fig. 3B). Either possibility would reduce the structural support for the arms in the minus arabinose condition and, as a result, not hold the DNA binding domains such that repression is favored.

View larger version (20K): [in this window] [in a new window]

FIG. 3. Apo dimerization domain of AraC from PDB 1XJA showing the dimerization subdomain on the left and the β-barrel subdomain on the right. The atoms of residues E149 and A152 are represented as gray spheres. The atoms of the closest arm residues, H18 and L19, are represented as black spheres. The subdomain interface, residues 98 to 111 and residues 121 to 123, is black in the diagrams. (A) Normal closed structure. (B) Opened structure.

To determine the structural basis of the mutations, we subjected purified wild-type and mutant dimerization domain to four assays that are sensitive to different aspects of the domain's structure: chymotrypsin cleavage, tryptophan fluorescence as a function of urea concentration, circular dichroism as a function of urea concentration, and solvent accessibility of a single cysteine residue as measured by DTNB sensitivity during urea denaturation.

Natural cleavage sites for chymotrypsin exist both in the arms and in the region between the β-barrel and -helical subdomains, residues 98 to 125. We found that the E149F and A152V mutations increase the sensitivity of the arms to chymotrypsin digestion, but no detectable cleavage of wild-type or mutant dimerization domain was observed within the region between the subdomains as demonstrated by the size of the digestion products measured on SDS gels or mass spectrometry (data not shown). Thus, the mutations destabilize the arm and may open the subdomains, but if they do, the opening likely is small.

Three of the five tryptophan residues present in the arm plus dimerization domain lie in the region between the subdomains where their exposure to water, and hence their fluorescence, would be significantly altered by exposure of the subdomain interface to the solvent. Therefore, if the mutations weaken, open, or eliminate the subdomain interface, then we would predict significant fluorescence differences between the mutant and wild-type proteins during urea-induced unfolding. Indeed, mutants E149F and A152V show strikingly different urea-induced denaturation curves compared to wild-type protein (Fig. 4). In the absence of arabinose, the mutant arm plus dimerization domains display midpoints in the transition region of the curve around 2 to 3 M urea, whereas the wild-type protein has a midpoint around 4.5 M urea.

View larger version (11K): [in this window] [in a new window]

FIG. 4. Urea denaturation profiles of wild type, E149F, and A152V dimerization domain monitored by tryptophan fluorescence in the presence (A) or absence (B) of arabinose.

The chymotrypsin and fluorescence data together indicate that under natural conditions, that is, in the absence of urea, the E149F and A152V mutations do not cause the dimerization domain of AraC to be significantly unfolded or to produce a substantial opening between the subdomains of the dimerization domain. Hence, the mechanism of action of the mutations appears to be a reduction in the support of the arm. The reduced support could be direct, by elimination of residue-residue interactions, or indirect, as a result of weaker interactions between the two subdomains involved. That there is a materially reduced interaction between the β-barrel and -helical subdomains of the dimerization domain is indicated by the strikingly different urea denaturation profiles of wild-type and the two mutant proteins above 2 M urea.

We performed additional experiments to determine the consequences of the weaker interaction during urea denaturation. The mutations could lead to subdomain opening or denaturation of either subdomain at low urea concentrations. Circular dichroism measurements of the dimerization domain during urea denaturation performed at a wavelength that is sensitive almost entirely to -helical structure (Fig. 5) showed that E149F does not significantly affect unfolding of the -helical portion of the dimerization domain. Consequently, whatever structural change reported on by tryptophan fluorescence of E149F protein beginning above 2 M urea is either subdomain opening or substantial unfolding of the β-barrel subdomain. The single cysteine residue in the dimerization domain allowed resolution of these two possibilities. The cysteine is in the β-barrel subdomain is appreciably buried and inaccessible to solvent and does not face the -helices. Therefore, its solvent accessibility as measured by DTNB reactivity provides a measure of unfolding of the subdomain and not opening of the subdomain interface. As shown in Fig. 6, its reactivity in the E149F mutant in 4 and 6 M urea is much greater than that of wild-type protein. Therefore, we conclude that E149F weakens the interaction between subdomains, which weakens the structural stability of the β-barrel in the presence of urea. In the absence of urea, the β-barrel structure remains intact, and the subdomain interface does not open significantly, as suggested by the DTNB reactivity of a cysteine residue in the and β-barrel region and the chymotrypsin sensitivity of the region between the subdomains. Thus, the slight weakening of the subdomain interface gives rise to a structural destabilization of the arms, and this results in the constitutive phenotype in vivo.

View larger version (7K): [in this window] [in a new window]

FIG. 5. Urea denaturation profiles of wild-type and E149F dimerization domains in the absence of arabinose monitored by the mean residue ellipticity of the circular dichroism signal at 222 nm.

View larger version (11K): [in this window] [in a new window]

FIG. 6. Cysteine exposure, as monitored by TNB production over time, for wild-type (A) and E149F (B) dimerization domains at different concentrations of urea and in the absence of arabinose.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
We have investigated the mechanistic basis of constitutive mutations in the AraC regulatory protein. Two different methods for isolating such mutations displayed overlapping mutational spectra, with the majority of mutations being found within residues 8 to 22 of the arms. The overlap suggests that neither of the two isolation methods, fucose resistance or direct identification of constitutives, selected a particular subset of constitutive mutations. This fact and the biochemical properties of the constitutive mutants are consistent with the light switch mechanism of the protein. A number of experiments (5, 7, 16, 18, 21, 27) have shown that the arms play an important role in regulating the state of the AraC protein. In the absence of arabinose, the arms interact with the dimerization domain. Then, the DNA binding domains bind to the arm plus dimerization domains in a way that favors binding of the protein to two well-separated half-sites and the formation of a repressive loop in the DNA. Mutations in the arms could alter their structure in a way that interferes with their interacting with the DNA binding domains. As a result, the DNA binding domains would not be held rigidly, and AraC by default would then bind to the araI₁ and araI₂ half-sites and activate transcription regardless of the presence or absence of arabinose.

We found that the trypsin susceptibility of the arms as measured on arm plus dimerization domain increased for most of the constitutive arm mutations that we tested. This implies that most of the mutations act by decreasing the stability of the structure of the arms. These experiments also showed that the protease sensitivity of the arms in some of the constitutive mutants is reduced by the presence of arabinose, indicating that in these cases, arabinose stabilizes the plus arabinose structure of the arm. That is, the mutations disrupt the minus arabinose structure of the arms more strongly than they disrupt the plus arabinose structure of the arms.

Two of the constitutive mutations that we found were not in the arm and are instead located in the core of the dimerization domain. Although the mutations are not found within the arm, the arms of these proteins are more susceptible than the wild type to chymotrypsin cleavage. The crystal structures of arm plus dimerization domain (23, 25) show that the side chains of these residues, E149 and A152, project into the interface between the -helical and β-barrel subdomains of the dimerization domain and also lie near residues 18 and 19 of the arm. Thus, the protease results can be explained by a decrease in the support of the structure of the arms. Fluorescence, circular dichroism, and DTNB sensitivity experiments indicate that the mutations reduce support for the arm. The experiments also indicate that the mutations weaken the interaction between the β-barrel and -helical subdomains of the dimerization domain with the consequence that in the E149F mutant, the β-barrel unfolds at a substantially lower urea concentration than in the wild-type protein.

In summary, constitutive mutations of AraC interfere with the protein's ability to form the repressive conformation. Consistent with our current understanding of the basis for AraC action, most constitutive mutations are found within the arm of AraC and act to reduce the stability with which the arm folds against the dimerization domain. This highlights the importance of the arm in regulating the state of the protein. Mutations not found within the arm can also act to destabilize its structure and likely reduce its interaction with the DNA binding domain.

 
This study was supported by National Institutes of Health grant GM18277 to R.S.

We thank Michael Rodgers, Katie Frato, and Jennifer Seedorf for assistance, discussions, and helpful comments on the manuscript.

 
* Corresponding author. Mailing address: Biology Department, Johns Hopkins University, 3400 North Charles St., Baltimore, MD 21218. Phone: (410) 516-5206. Fax: (410) 516-5213. E-mail: schleif{at}jhu.edu

Published ahead of print on 13 February 2009.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Beverin, S., D. E. Sheppard, and S. S. Park. 1971. D-Fucose as a gratuitous inducer of the arabinose operon in strains of Escherichia coli B/r mutant in gene araC. J. Bacteriol. 107:79-86.[Abstract/Free Full Text]
2
2. Carra, J., and R. Schleif. 1993. Variation of half-site organization and DNA looping by AraC protein. EMBO J. 12:35-44.[Medline]
3
3. Dunn, T., S. Hahn, S. Ogden, and R. Schleif. 1984. An operator at –280 base pairs that is required for repression of araBAD operon promoter: addition of DNA helical turns between the operator and promoter cyclically hinders repression. Proc. Natl. Acad. Sci. USA 81:5017-5020.[Abstract/Free Full Text]
4
4. Englesberg, E., J. Irr, J. Power, and N. Lee. 1965. Positive control of enzyme synthesis by gene C in the arabinose system. J. Bacteriol. 90:946-957.[Abstract/Free Full Text]
5
5. Ghosh, M., and R. R. Schleif. 2001. Biophysical evidence of arm-domain interactions in AraC. Anal. Biochem. 295:107-112.[CrossRef][Medline]
6
6. Gill, S. C., and P. H. von Hippel. 1989. Calculation of protein extinction coefficients from amino acid sequence data. Anal. Biochem. 189:319-326.
7
7. Gryczynski, U., and R. Schleif. 2004. A portable allosteric mechanism. Proteins 57:9-11.[CrossRef][Medline]
8
8. Hahn, S., T. Dunn, and R. F. Schleif. 1984. Upstream repression and CRP stimulation of the Escherichia coli arabinose operon. J. Mol. Biol. 180:61-72.[CrossRef][Medline]
9
9. Harmer, T., M. Wu, and R. Schleif. 2001. The role of rigidity in DNA looping-unlooping by AraC. Proc. Natl. Acad. Sci. USA 98:427-431.[Abstract/Free Full Text]
10
10. Hubbard, S. J., F. Eisenmenger, and J. M. Thornton. 1994. Modeling studies of the change in conformation required for cleavage of limited proteolytic sites. Protein Sci. 3:757-768.[Medline]
11
11. Lobell, R., and R. Schleif. 1990. DNA looping and unlooping by AraC protein. Science 250:528-532.[Abstract/Free Full Text]
12
12. LaRonde-LeBlanc, N., and C. Wolberger. 2000. Characterization of the oligomeric states of wild and mutant AraC. Biochemistry 39:11593-11601.[CrossRef][Medline]
13
13. MacInnes, K. R., D. E. Sheppard, and B. Falgout. 1978. Regulatory properties of araC (c) mutants in the arabinose operon of Escherichia coli B/r. J. Bacteriol. 133:178-184.[Abstract/Free Full Text]
14
14. Martin, K., L. Huo, and R. Schleif. 1986. The DNA loop model for ara repression: AraC protein occupies the proposed loop sites in vivo and repression-negative mutations lie in these same sites. Proc. Natl. Acad. Sci. USA 83:3654-3658.[Abstract/Free Full Text]
15
15. Nathanson, N. M., and R. Schleif. 1975. Paucity of sites mutable to constitutivity in the araC activator gene of the arabinose operon of Escherichia coli. J. Mol. Biol. 25:185-199.
16
16. Reed, W., and R. Schleif. 1999. Hemiplegic mutations in AraC protein. J. Mol. Biol. 294:417-425.[CrossRef][Medline]
17
17. Ross, J. J., U. Gryczynski, and R. Schleif. 2003. Mutational analysis of residue roles in AraC function. J. Mol. Biol. 328:85-93.[CrossRef][Medline]
18
18. Rodgers, M. E., N. D. Holder, S. Dirla, and R. Schleif. 2009. Functional modes of the regulatory arm of AraC. Proteins 74:81-91.[CrossRef][Medline]
19
19. Saviola, B., R. Seabold, and R. F. Schleif. 1998. Arm-domain interactions in AraC. J. Mol. Biol. 289:539-548.
20
20. Schleif, R. F., and P. C. Wensink. 1981. Practical methods in molecular biology. Springer-Verlag, New York, NY.
21
21. Seabold, R., and R. Schleif. 1998. Apo-AraC actively seeks to loop. J. Mol. Biol. 278:529-538.[CrossRef][Medline]
22
22. Sheppard, D. E. 1986. Dominance relationships among mutant alleles of regulatory gene araC in the Escherichia coli B/R arabinose operon. J. Bacteriol. 168:999-1001.[Abstract/Free Full Text]
23
23. Soisson, S., B. MacDougal-Shackleton, R. Schleif, and C. Wolberger. 1997. Structural basis for ligand-regulated oligomerization of AraC. Science 276:421-425.[Abstract/Free Full Text]
24
24. Timmes, A., M. Rodgers, and R. Schleif. 2004. Biochemical and physiological properties of the DNA binding domain of AraC protein. J. Mol. Biol. 340:731-738.[CrossRef][Medline]
25
25. Weldon, J. E., M. Rodgers, C. Larkin, and R. Schleif. 2007. Structure and properties of a truely apo form of AraC dimerization domain. Proteins 66:646-654.[CrossRef][Medline]
26
26. Wu, M., and R. Schleif. 2001. Mapping arm-DNA-binding domain interactions in AraC. J. Mol. Biol. 307:1001-1009.[CrossRef][Medline]
27
27. Wu, M., and R. Schleif. 2001. Strengthened arm-dimerization domain interactions in AraC. J. Biol. Chem. 276:2562-2564.[Abstract/Free Full Text]
28
28. Zhang, X., T. Reeder, and R. Schleif. 1996. Transcription activation parameters at ara pBAD. J. Mol. Biol. 258:14-24.[CrossRef][Medline]

---

Journal of Bacteriology, April 2009, p. 2668-2674, Vol. 191, No. 8
0021-9193/09/$08.00+0     doi:10.1128/JB.01529-08
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Dirla, S. Articles by Schleif, R. Search for Related Content PubMed PubMed Citation Articles by Dirla, S. Articles by Schleif, R.