Atomic force microscopy measurements and model of DNA bending caused by binding of AraC protein.

Atomic force microscopy (AFM) was used to conduct single-molecule imaging of protein/DNA complexes involved in the regulation of the arabinose operon of Escherichia coli. In the presence of arabinose, the transcription regulatory protein AraC binds to a 38 bp region consisting of the araI1 and araI2 half-sites. The domain positioning of full-length AraC, when bound to DNA, was not previously known. In this study, AraC was combined with 302 and 560 bp DNA and arabinose, deposited on a mica substrate, and imaged with AFM in air. High resolution images of 560 bp DNA, where bound protein was visible, showed that AraC induces a bend in the DNA with an angle 60° ± 12° with a median of 55°. These results are consistent with earlier gel electrophoresis measurements that measured the DNA bend angle based on migration rates. By using known domain structures of AraC, geometric constraints, and contacts determined from biochemical experiments, we developed a model of the tertiary and quaternary structure of DNA-bound AraC in the presence of arabinose. The DNA bend angle predicted by the model is in agreement with the measurement values. We discuss the results in view of other regulatory proteins that cause DNA bending and formation of the open complex to initiate transcription.

Allosteric regulation within the highly interconnected structural scaffold of AraC/XylS homologs tolerates a wide range of amino acid changes.

To create bacterial transcription "circuits" for biotechnology, one approach is to recombine natural transcription factors, promoters, and operators. Additional novel functions can be engineered from existing transcription factors such as the E. coli AraC transcriptional activator, for which binding to DNA is modulated by binding L-arabinose. Here, we engineered chimeric AraC/XylS transcription activators that recognized ara DNA binding sites and responded to varied effector ligands. The first step, identifying domain boundaries in the natural homologs, was challenging because (i) no full-length, dimeric structures were available and (ii) extremely low sequence identities (≤10%) among homologs precluded traditional assemblies of sequence alignments. Thus, to identify domains, we built and aligned structural models of the natural proteins. The designed chimeric activators were assessed for function, which was then further improved by random mutagenesis. Several mutational variants were identified for an XylS•AraC chimera that responded to benzoate; two enhanced activation to near that of wild-type AraC. For an RhaR•AraC chimera, a variant with five additional substitutions enabled transcriptional activation in response to rhamnose. These five changes were dispersed across the protein structure, and combinatorial experiments testing subsets of substitutions showed significant non-additivity. Combined, the structure modeling and epistasis suggest that the common AraC/XylS structural scaffold is highly interconnected, with complex intra-protein and inter-domain communication pathways enabling allosteric regulation. At the same time, the observed epistasis and the low sequence identities of the natural homologs suggest that the structural scaffold and function of transcriptional regulation are nevertheless highly accommodating of amino acid changes.

Engineering sensitivity and specificity of AraC-based biosensors responsive to triacetic acid lactone and orsellinic acid.

We previously described the design of triacetic acid lactone (TAL) biosensor 'AraC-TAL1', based on the AraC regulatory protein. Although useful as a tool to screen for enhanced TAL biosynthesis, this variant shows elevated background (leaky) expression, poor sensitivity and relaxed inducer specificity, including responsiveness to orsellinic acid (OA). More sensitive biosensors specific to either TAL or OA can aid in the study and engineering of polyketide synthases that produce these and similar compounds. In this work, we employed a TetA-based dual-selection to isolate new TAL-responsive AraC variants showing reduced background expression and improved TAL sensitivity. To improve TAL specificity, OA was included as a 'decoy' ligand during negative selection, resulting in the isolation of a TAL biosensor that is inhibited by OA. Finally, to engineer OA-specific AraC variants, the iterative protein redesign and optimization computational framework was employed, followed by 2 rounds of directed evolution, resulting in a biosensor with 24-fold improved OA/TAL specificity, relative to AraC-TAL1.

Characterization of a novel AraC/XylS-regulated family of N-acyltransferases in pathogens of the order Enterobacterales.

Enteroaggregative Escherichia coli (EAEC) is a diarrheagenic pathotype associated with traveler's diarrhea, foodborne outbreaks and sporadic diarrhea in industrialized and developing countries. Regulation of virulence in EAEC is mediated by AggR and its negative regulator Aar. Together, they control the expression of at least 210 genes. On the other hand, we observed that about one third of Aar-regulated genes are related to metabolism and transport. In this study we show the AggR/Aar duo controls the metabolism of lipids. Accordingly, we show that AatD, encoded in the AggR-regulated aat operon (aatPABCD) is an N-acyltransferase structurally similar to the essential Apolipoprotein N-acyltransferase Lnt and is required for the acylation of Aap (anti-aggregation protein). Deletion of aatD impairs post-translational modification of Aap and causes its accumulation in the bacterial periplasm. trans-complementation of 042aatD mutant with the AatD homolog of ETEC or with the N-acyltransferase Lnt reestablished translocation of Aap. Site-directed mutagenesis of the E207 residue in the putative acyltransferase catalytic triad disrupted the activity of AatD and caused accumulation of Aap in the periplasm due to reduced translocation of Aap at the bacterial surface. Furthermore, Mass spectroscopy revealed that Aap is acylated in a putative lipobox at the N-terminal of the mature protein, implying that Aap is a lipoprotein. Lastly, deletion of aatD impairs bacterial colonization of the streptomycin-treated mouse model. Our findings unveiled a novel N-acyltransferase family associated with bacterial virulence, and that is tightly regulated by AraC/XylS regulators in the order Enterobacterales.

Structures of the transcriptional regulator BgaR, a lactose sensor.

The structure of BgaR, a transcriptional regulator of the lactose operon in Clostridium perfringens, has been solved by SAD phasing using a mercury derivative. BgaR is an exquisite sensor of lactose, with a binding affinity in the low-micromolar range. This sensor and regulator has been captured bound to lactose and to lactulose as well as in a nominal apo form, and was compared with AraC, another saccharide-binding transcriptional regulator. It is shown that the saccharides bind in the N-terminal region of a jelly-roll fold, but that part of the saccharide is exposed to bulk solvent. This differs from the classical AraC saccharide-binding site, which is mostly sequestered from the bulk solvent. The structures of BgaR bound to lactose and to lactulose highlight how specific and nonspecific interactions lead to a higher binding affinity of BgaR for lactose compared with lactulose. Moreover, solving multiple structures of BgaR in different space groups, both bound to saccharides and unbound, verified that the dimer interface along a C-terminal helix is similar to the dimer interface observed in AraC.

Helical Behavior of the Interdomain Linker of the Escherichia coli AraC Protein.

In Escherichia coli, the dimeric AraC protein actively represses transcription from the l-arabinose araBAD operon in the absence of arabinose but induces transcription in its presence. Here we provide evidence that, in shifting from the repressing to the inducing state, the behavior of the interdomain linker shifts from that of an α helix to that of a more flexible form. In vivo and in vitro experiments show that AraC with a linker sequence that favors helix formation is shifted toward the repressing state in the absence and presence of arabinose. Conversely, AraC containing a linker sequence that is unfavorable for helix formation is shifted toward the inducing state. Experiments in which the presumed helical linker is shortened or lengthened, protein helical twist experiments, are also consistent with a helix transition mechanism. Previous experiments have shown that, upon the binding of arabinose, the apparent rigidity with which the DNA binding domains of AraC are held in space decreases. Thus, arabinose likely controls the stability or rigidity of the interdomain linker. Circular dichroism experiments with peptides show that the helicity of the linker sequence can be controlled by the helicity of residues preceding the linker, providing a plausible mechanism for arabinose to control the repressing-inducing state of AraC protein.

Arabinose Alters Both Local and Distal H-D Exchange Rates in the Escherichia coli AraC Transcriptional Regulator.

In the absence of arabinose, the dimeric Escherichia coli regulatory protein of the l-arabinose operon, AraC, represses expression by looping the DNA between distant half-sites. Binding of arabinose to the dimerization domains forces AraC to preferentially bind two adjacent DNA half-sites, which stimulates RNA polymerase transcription of the araBAD catabolism genes. Prior genetic and biochemical studies hypothesized that arabinose allosterically induces a helix-coil transition of a linker between the dimerization and DNA binding domains that switches the AraC conformation to an inducing state [Brown, M. J., and Schleif, R. F. (2019) Biochemistry, preceding paper in this issue (DOI: 10.1021/acs.biochem.9b00234)]. To test this hypothesis, hydrogen-deuterium exchange mass spectrometry was utilized to identify structural regions involved in the conformational activation of AraC by arabinose. Comparison of the hydrogen-deuterium exchange kinetics of individual dimeric dimerization domains and the full-length dimeric AraC protein in the presence and absence of arabinose reveals a prominent arabinose-induced destabilization of the amide hydrogen-bonded structure of linker residues (I167 and N168). This destabilization is demonstrated to result from an increased probability to form a helix capping motif at the C-terminal end of the dimerizing α-helix of the dimerization domain that preceeds the interdomain linker. These conformational changes could allow for quaternary repositioning of the DNA binding domains required for induction of the araBAD promoter through rotation of peptide backbone dihedral angles of just a couple of residues. Subtle changes in exchange rates are also visible around the arabinose binding pocket and in the DNA binding domain.

AraC-like transcriptional activator CuxR binds c-di-GMP by a PilZ-like mechanism to regulate extracellular polysaccharide production.

Cyclic dimeric GMP (c-di-GMP) has emerged as a key regulatory player in the transition between planktonic and sedentary biofilm-associated bacterial lifestyles. It controls a multitude of processes including production of extracellular polysaccharides (EPSs). The PilZ domain, consisting of an N-terminal "RxxxR" motif and a ß-barrel domain, represents a prototype c-di-GMP receptor. We identified a class of c-di-GMP-responsive proteins, represented by the AraC-like transcription factor CuxR in plant symbiotic α-proteobacteria. In Sinorhizobium meliloti, CuxR stimulates transcription of an EPS biosynthesis gene cluster at elevated c-di-GMP levels. CuxR consists of a Cupin domain, a helical hairpin, and bipartite helix-turn-helix motif. Although unrelated in sequence, the mode of c-di-GMP binding to CuxR is highly reminiscent to that of PilZ domains. c-di-GMP interacts with a conserved N-terminal RxxxR motif and the Cupin domain, thereby promoting CuxR dimerization and DNA binding. We unravel structure and mechanism of a previously unrecognized c-di-GMP-responsive transcription factor and provide insights into the molecular evolution of c-di-GMP binding to proteins.

A genetic and physical study of the interdomain linker of E. Coli AraC protein--a trans-subunit communication pathway.

Genetic experiments with full length AraC and biophysical experiments with its dimerization domain plus linker suggest that arabinose binding to the dimerization domain changes the properties of the inter-domain linker which connects the dimerization domain to the DNA binding domain via interactions that do not depend on the DNA binding domain. Normal AraC function was found to tolerate considerable linker sequence alteration excepting proline substitutions. The proline substitutions partially activate transcription even in the absence of arabinose and hint that a structural shift between helix and coil may be involved. To permit fluorescence anisotropy measurements that could detect arabinose-dependent dynamic differences in the linkers, IAEDANS was conjugated to a cysteine residue substituted at the end of the linker of dimerization domain. Arabinose, but not other sugars, decreased the steady-state anisotropy, indicating either an increase in mobility and/or an increase in the fluorescence lifetime of the IAEDANS. Time-resolved fluorescence measurements showed that the arabinose-induced anisotropy decrease did not result from an increase in the excited-state lifetime. Hence arabinose-induced decreases in anisotropy appear to result from increased tumbling of the fluorophore. Arabinose did not decrease the anisotropy in mutants incapable of binding arabinose nor did it alter the anisotropy when IAEDANS was conjugated elsewhere in the dimerization domain. Experiments with heterodimers of the dimerization domain showed that the binding of arabinose to one subunit of the dimer decreases the fluorescence anisotropy of only a fluorophore on the linker of the other subunit.

Analysis of amino acid substitutions in AraC variants that respond to triacetic acid lactone.

The Escherichia coli regulatory protein AraC regulates expression of ara genes in response to l-arabinose. In efforts to develop genetically encoded molecular reporters, we previously engineered an AraC variant that responds to the compound triacetic acid lactone (TAL). This variant (named "AraC-TAL1") was isolated by screening a library of AraC variants, in which five amino acid positions in the ligand-binding pocket were simultaneously randomized. Screening was carried out through multiple rounds of alternating positive and negative fluorescence-activated cell sorting. Here we show that changing the screening protocol results in the identification of different TAL-responsive variants (nine new variants). Individual substituted residues within these variants were found to primarily act cooperatively toward the gene expression response. Finally, X-ray diffraction was used to solve the crystal structure of the apo AraC-TAL1 ligand-binding domain. The resolved crystal structure confirms that this variant takes on a structure nearly identical to the apo wild-type AraC ligand-binding domain (root-mean-square deviation 0.93 Å), suggesting that AraC-TAL1 behaves similar to wild-type with regard to ligand recognition and gene regulation. Our results provide amino acid sequence-function data sets for training and validating AraC modeling studies, and contribute to our understanding of how to design new biosensors based on AraC.

Identification and characterization of Aeromonas hydrophila genes encoding the outer membrane receptor of ferrioxamine B and an AraC-type transcriptional regulator.

We found that, under iron-limiting conditions, Aeromonas hydrophila ATCC 7966(T) could utilize the xenosiderophore desferrioxamine B (DFOB) for growth by inducing the expression of its own outer membrane receptor. Two consecutive genes, desR and desA, were selected as candidates involved in DFOB utilization. The presence of the ferric-uptake regulator boxes in their promoters suggested that these genes are under iron-dependent regulation. Mutation of desA, a gene that encodes the outer membrane receptor of ferrioxamine B, disrupted the growth of the amonabactin-deficient mutant in the presence of DFOB. ß-Galactosidase reporter assays and reverse transcriptase-quantitative PCR demonstrated that desR, a gene that encodes an AraC-like regulator homolog is required for the induction of desA transcription in the presence of DFOB and under iron-limiting conditions. The functions of desA and desR were analyzed using complementation experiments. Our data provided evidence that DesA is powered primarily by the TonB2 system.

Computational and experimental investigation of constitutive behavior in AraC.

Many mutations in the N-terminal arm of AraC result in constitutive behavior in which transcription of the araBAD genes occurs even in the absence of arabinose. To begin to understand the mechanism underlying this class of mutations, we used molecular dynamics with self-guided Langevin dynamics to simulate (1) wild-type (WT) AraC, (2) known constitutive mutants resulting from alterations in the regulatory arm, particularly alanine and glycine substitutions at residue 8 because P8G is constitutive, whereas P8A behaves like wild type, and (3) selected variant AraC proteins containing alterations in the dimerization core. In all of the constitutive arm mutants, but not the WT protein, residues 37-42, which are located in the core of the dimerization domain, became restructured. This raised the question of whether or not these structural changes are an obligatory component of constitutivity. Using molecular dynamics, we identified alterations in the core that produced a similar restructuring. The corresponding mutants were constructed and their ara constitutivity status was determined experimentally. Because the core mutants were not found to be constitutive, we conclude that restructuring of core residues 37-42 does not, itself, lead to constitutivity of AraC. The available data lead to the hypothesis that the interaction of the N-terminal arm with something other than the front lip is the primary determinant of the inducing versus repressing state of AraC.

A novel assay for assessing juxtamembrane and transmembrane domain interactions important for receptor heterodimerization.

Understanding the basis of specificity in receptor homodimerization versus heterodimerization is essential in determining the role receptor plays in signal transduction. Specificity in each of the interfaces formed during signal transduction involves cooperative interactions between receptor extracellular, transmembrane (TM), and cytoplasmic domains. While methods exist for studying receptor heterodimerization in cell membranes, they are limited to either TM domains expressed in an inverted orientation or capture only heterodimerization in a single assay. To address this limitation, we have developed an assay (DN-AraTM) that enables simultaneous measurement of homodimerization and heterodimerization of type I receptor domains in their native orientation, including both soluble and TM domains. Using integrin αIIb and RAGE (receptor for advanced glycation end products) as model type I receptor systems, we demonstrate both specificity and sensitivity of our approach, which will provide a novel tool to identify specific domain interactions that are important in regulating signal transduction.

Understanding the basis of a class of paradoxical mutations in AraC through simulations.

Most mutations at position 15 in the N-terminal arm of the regulatory protein AraC leave the protein incapable of responding to arabinose and inducing the proteins required for arabinose catabolism. Mutations at other positions of the arm do not have this behavior. Simple energetic analysis of the interactions between the arm and bound arabinose do not explain the uninducibility of AraC with mutations at position 15. Extensive molecular dynamics (MD) simulations, carried out largely on the Open Science Grid, were done of the wild-type protein with and without bound arabinose and of all possible mutations at position 15, many of which were constructed and measured for this work. Good correlation was found for deviation of arm position during the simulations and inducibility as measured in vivo of the same mutant proteins. Analysis of the MD trajectories revealed that preservation of the shape of the arm is critical to inducibility. To maintain the correct shape of the arm, the strengths of three interactions observed to be strong in simulations of the wild-type AraC protein need to be preserved. These interactions are between arabinose and residue 15, arabinose and residues 8-9, and residue 13 and residue 15. The latter interaction is notable because residues L9, Y13, F15, W95, and Y97 form a hydrophobic cluster which needs to be preserved for retention of the correct shape.

Snap denaturation reveals dimerization by AraC-like protein Rns.

Here we show that the Rns regulator of Escherichia coli dimerises in vivo and in vitro. Furthermore, we demonstrate that Rns forms aggregates in vitro and describe a methodology to ameliorate aggregation thus permitting the analysis of Rns by cross-linking.

Computer-based annotation of putative AraC/XylS-family transcription factors of known structure but unknown function.

Currently, about 20 crystal structures per day are released and deposited in the Protein Data Bank. A significant fraction of these structures is produced by research groups associated with the structural genomics consortium. The biological function of many of these proteins is generally unknown or not validated by experiment. Therefore, a growing need for functional prediction of protein structures has emerged. Here we present an integrated bioinformatics method that combines sequence-based relationships and three-dimensional (3D) structural similarity of transcriptional regulators with computer prediction of their cognate DNA binding sequences. We applied this method to the AraC/XylS family of transcription factors, which is a large family of transcriptional regulators found in many bacteria controlling the expression of genes involved in diverse biological functions. Three putative new members of this family with known 3D structure but unknown function were identified for which a probable functional classification is provided. Our bioinformatics analyses suggest that they could be involved in plant cell wall degradation (Lin2118 protein from Listeria innocua, PDB code 3oou), symbiotic nitrogen fixation (protein from Chromobacterium violaceum, PDB code 3oio), and either metabolism of plant-derived biomass or nitrogen fixation (protein from Rhodopseudomonas palustris, PDB code 3mn2).

A new and unexpected domain-domain interaction in the AraC protein.

An interaction between the dimerization domains and DNA binding domains of the dimeric AraC protein has previously been shown to facilitate repression of the Escherichia coli araBAD operon by AraC in the absence of arabinose. A new interaction between the domains of AraC in the presence of arabinose is reported here, the regulatory consequences of which are unknown. Evidence for the interaction is the following: the dissociation rate of arabinose-bound AraC from half-site DNA is considerably faster than that of free DNA binding domain, and the affinity of the dimerization domains for arabinose is increased when half-site DNA is bound. In addition, an increase in the fluorescence intensity of tryptophan residues located in the arabinose-bound dimerization domain is observed upon binding of half-site DNA to the DNA binding domains. Direct physical evidence of the new domain-domain interaction is demonstrated by chemical crosslinking and NMR experiments.

Extranucleosomal DNA binding directs nucleosome sliding by Chd1.

Chd1- and ISWI-type chromatin remodelers can sense extranucleosomal DNA and preferentially shift nucleosomes toward longer stretches of available DNA. The DNA-binding domains of these chromatin remodelers are believed to be responsible for sensing extranucleosomal DNA and are needed for robust sliding, but it is unclear how these domains contribute to directional movement of nucleosomes. Here, we show that the DNA-binding domain of Chd1 is not essential for nucleosome sliding but is critical for centering mononucleosomes on short DNA fragments. Remarkably, nucleosome centering was achieved by replacing the native DNA-binding domain of Chd1 with foreign DNA-binding domains of Escherichia coli AraC or Drosophila melanogaster engrailed. Introducing target DNA sequences recognized by the foreign domains enabled the remodelers to rapidly shift nucleosomes toward these binding sites, demonstrating that these foreign DNA-binding domains dictated the direction of sliding. Sequence-directed sliding occluded the target DNA sequences on the nucleosome enough to promote release of the remodeler. Target DNA sequences were highly stimulatory at multiple positions flanking the nucleosome and had the strongest influence when separated from the nucleosome by 23 or fewer base pairs. These results suggest that the DNA-binding domain's affinity for extranucleosomal DNA is the key determinant for the direction that Chd1 shifts the nucleosome.

Active role of the interdomain linker of AraC.

Mutations in the interdomain linker of the gene for the AraC regulatory protein of Escherichia coli that severely interfere with the protein's ability either to repress or to activate transcription have been found. These mutations have relatively small effects on the dimerization domain's ability to bind arabinose or to dimerize the protein or on the DNA-binding domain's affinity for a single DNA half-site. The linker mutations, however, dramatically change the affinity of AraC for binding to two direct-repeat DNA half-sites. Less dramatically, the induction-deficient linker variants also display altered DNA sequence selectivity. These results show that changing the sequence of the interdomain linker can profoundly affect the dimerization domain-DNA-binding domain interactions in AraC. The smaller effects on the functions of the individual domains could be the direct result of the linker alterations but more likely are the indirect result of the altered dimerization domain-DNA-binding domain interactions. In summary, the linker does not simply function as a passive and flexible connector between the domains of AraC but, instead, is more directly involved in the protein's dimerization domain-DNA-binding domain interactions.

Computational predictions of the mutant behavior of AraC.

An algorithm implemented in Rosetta correctly predicts the folding capabilities of the 17-residue N-terminal arm of the AraC gene regulatory protein when arabinose is bound to the protein and the dramatically different structure of this arm when arabinose is absent. The transcriptional activity of 43 mutant AraC proteins with alterations in the arm sequences was measured in vivo and compared with their predicted folding properties. Seventeen of the mutants possessed regulatory properties that could be directly compared with folding predictions. Sixteen of the 17 mutants were correctly predicted. The algorithm predicts that the N-terminal arm sequences of AraC homologs fold to the Escherichia coli AraC arm structure. In contrast, it predicts that random sequences of the same length and many partially randomized E. coli arm sequences do not fold to the E. coli arm structure. The high level of success shows that relatively "simple" computational methods can in some cases predict the behavior of mutant proteins with good reliability.