The RhaS and RhaR proteins are transcription activators thatrespond to the availability of L-rhamnose and activate transcriptionof the operons in the Escherichia coli L-rhamnose catabolicregulon . RhaR activates transcription of rhaSR, and RhaS activatestranscription of the operon that encodes the L-rhamnose catabolic enzymes, rhaBAD, as well as the operon that encodes the L-rhamnosetransport protein, rhaT . RhaS is 30% identical to RhaR at theamino acid level, and both are members of the AraC/XylS familyof transcription activators . The RhaS and RhaR binding sitesoverlap the –35 hexamers of the promoters they regulate,suggesting they may contact the ⁷⁰ subunit of RNA polymeraseas part of their mechanisms of transcription activation . Insupport of this hypothesis, our lab previously identified aninteraction between RhaS residue D241 and ⁷⁰ residue R599 . Inthe present study, we first identified two positively chargedamino acids in ⁷⁰, K593 and R599, and three negatively chargedamino acids in RhaR, D276, E284, and D285, that were importantfor RhaR-mediated transcription activation of the rhaSR operon.Using a genetic loss-of-contact approach we have obtained evidencefor a specific contact between RhaR D276 and ⁷⁰ R599 . Finally,previous results from our lab separately showed that RhaS D250Aand ⁷⁰ K593A were defective at the rhaBAD promoter . Our genetic loss-of-contact analysis of these residues indicates that they identify a second site of contact between RhaS and ⁷⁰.

 
Transcription activation in Escherichia coli often involves the interaction of a DNA-binding activator protein with oneof the subunits of RNA polymerase [RNAP], most often the sigma[] or alpha [] subunit . Transcription activators that bind immediatelyupstream and adjacent to RNAP, in some cases overlapping the–35 promoter hexamer, may interact with the C-terminaldomain [domain 4] of the subunit of RNAP [8, 27] . The cI proteinof bacteriophage is required for the establishment and maintenanceof lysogeny and is perhaps the best-characterized example ofa transcription activator that contacts ⁷⁰ . The cI proteinactivates transcription of the P_RM promoter when bound at theO_R2 operator site, which overlaps the P_RM –35 hexamerby 2 bp [30] . Current evidence suggests that ⁷⁰ residues R588,K593, and R596 are required for activation by cI [23, 26, 35]. Genetic and molecular modeling studies, as well as the recent structure of a ternary cI- domain 4-DNA complex, indicate that cI D38 contacts both ⁷⁰ K593 [^A K418] and R596 [^A R421] and cI E34 contacts ⁷⁰ R588 [^A R413] [8, 19, 26, 35] . Prior tothe identification of the ternary complex structure, a molecularmodel of the interaction indicated that ⁷⁰ K593 [^A K418] contactsDNA but was not positioned to contact cI [6, 8, 35] . However,the ternary structure showed that the ^A residue that alignswith ⁷⁰ K593 has moved away from the DNA [relative to the modelof the interaction] and instead makes a protein-protein contactwith cI D38 [19].

There is also evidence that activation by several transcription activators in the AraC/XylS family involves ⁷⁰ domain 4 . Ourlab previously identified two ⁷⁰ residues, K593 and R599, whichare required for full activation by RhaS and further obtainedgenetic evidence that ⁷⁰ R599 is directly involved in a contactwith RhaS D241 [4] . These genetic results are also stronglysupported by molecular modeling of the RhaS-⁷⁰ complex on DNA[4] . Evidence for AraC interactions with ⁷⁰ come from early ⁷⁰ mutations [eventually identified at R596] that increasedaraBAD expression in the absence of activation by cyclic AMPreceptor protein [CRP] [18, 39, 44], as well as the findingthat araBAD expression in a cya strain was increased by ⁷⁰ E591Aand R596A and decreased by ⁷⁰ K593A [27] . In addition, witha DNA that mimicked an open complex, a small amount of DNA-bindingcooperativity could be detected between AraC and ⁷⁰ [7] . Atthe melAB promoter, genetic evidence indicated that ⁷⁰ R596interacts with MelR D261 and T265 while ⁷⁰ R599 also interactswith MelR D261, which aligns with RhaS D241 [16] . The Ada proteinhas two activation domains, one of which is an AraC/XylS familydomain which is required to activate transcription of the alkAoperon [33] . Alanine substitutions of ⁷⁰ residues K593, K597,and R603 each led to significant defects in Ada-dependent alkAtranscription in vivo and in vitro [24].

The transcription activator RhaS, and the closely related RhaR protein, activate transcription of the E . coli L-rhamnose catabolicoperons in the presence of the sugar L-rhamnose [10, 11, 42]. RhaS activates transcription of the rhaBAD and rhaT operons by binding as a dimer to sites that overlap the –35 hexamersof the promoters by 4 bp and extend upstream to –81 and–82, respectively [see Fig . 1 for the rhaBAD promoter] [10, 45] . Similarly, RhaR activates transcription of the rhaSRoperon by binding as a dimer to a site that overlaps the RNAPbinding site by 4 bp and extends upstream to –82 [Fig.1] [43] . The long binding sites for RhaS and RhaR each consistof two 17-bp imperfect inverted repeat half sites that are separatedby 16 or 17 bp of uncontacted DNA [10, 43, 45] . Each RhaS andRhaR monomer is predicted to contain two helix-turn-helix DNA-bindingmotifs and thereby contact two consecutive major grooves ofDNA [38] . CRP also activates transcription at all three of therha promoters . At the rhaBAD and rhaT promoters, the CRP bindingsite is located immediately upstream of the RhaS binding siteand is centered at –92.5 and –93.5, respectively[see Fig . 1 for the rhaBAD promoter] [11, 45] . The CRP siterequired for full activation at rhaSR is located upstream butnot adjacent to the RhaR binding site and is centered at –111.5[Fig . 1] [17].

 
RhaS and RhaR are members of a subset of the AraC/XylS familythat share amino acid sequence similarity with AraC over itsentire length [9, 13, 28] . Based on this similarity, RhaS andRhaR are predicted to consist of two domains connected by aflexible linker [5, 12, 29, 41] . The N-terminal domains arepredicted to be responsible for L-rhamnose binding and dimerization,while the C-terminal domains contain the 99-amino-acid regionthat classifies them as members of the AraC/XylS family . Inall studied cases of AraC/XylS family members, including RhaSand RhaR, the characteristic 99-amino-acid region constitutesa DNA-binding domain [3, 10, 43] . This DNA-binding domain has also been shown to be involved in transcription activation ina number of AraC/XylS family proteins including Ada, RhaS, AraC,MelR, MarA, SoxS, XylS, and UreR [1, 4, 5, 15, 16, 20-22, 37].

In this study, we further explored the mechanisms of transcription activation by RhaR and RhaS . We identified amino acid residuesin the C-terminal domain of ⁷⁰ and in RhaR that are importantfor RhaR-mediated transcription activation at the rhaSR promoter.We then used a genetic loss-of-contact approach to identifyan interaction between RhaR D276 and ⁷⁰ R599 that is requiredfor RhaR-mediated activation . We also extended the previousstudies by Bhende and Egan [4] of RhaS-mediated transcriptionactivation at rhaBAD . Here we identified a second interactionbetween RhaS and ⁷⁰, in this case, RhaS D250 and ⁷⁰ K593.

 
Culture media and conditions. E . coli cultures for ß-galactosidase assays were grownin morpholinepropanesulfonic acid-buffered medium [4, 34] . Tryptone-yeastextract liquid medium [0.8% tryptone, 0.05% yeast extract, 0.05%NaCl] was used to grow cells for most other experiments . SacBselection medium [1% tryptone, 0.5% yeast extract, 1.5% agar,5% sucrose [pH 7.8]] was used to select against sacB⁺ strains[14] . Antibiotics were used as indicated at the following concentrations:ampicillin [200 µg/ml], chloramphenicol [25 µg/ml],kanamycin [25 µg/ml], and tetracycline [20 µg/ml].

General methods. Standard methods were used for restriction endonuclease digestionand ligation using restriction endonucleases and T4 DNA ligasepurchased from New England Biolabs [Beverly, Mass.] . Transformationwas carried out using chemically induced competent cells ofE . coli, and plasmid DNA was purified by alkaline lysis . DNAsequencing reactions were carried out using custom-synthesizedIRD41 dye-labeled primers [Table 1] from LI-COR Inc . [Lincoln,Nebr.] and the Thermo Sequenase primer cycle sequencing kitfrom Amersham Life Sciences [Arlington Heights, Ill.] . DNA sequenceswere analyzed by automated dideoxy sequencing on a LI-COR 4000Lsequencer [University of Kansas Biochemical Research ServiceLaboratory] . The Expand High Fidelity PCR system [Roche, Indianapolis,Ind.] was used to amplify DNA fragments for cloning as wellas to generate templates for DNA sequencing from rhaS and rhaRalleles that were recombined into the chromosome . The DNA sequencesof both strands were determined for the entire cloned regionof all cloned, mutagenized, and recombined DNA fragments . The QIAquick PCR Purification kit [QIAGEN, Chatsworth, Calif.] was used to clean up PCR products.

 
Strains, plasmids, and phages. Table 2 lists the strains, phages, and plasmids used in thisstudy . All strains used in ß-galactosidase assayswere derived from ECL116 [2] and carried lacZ fusions in a singlecopy on phage integrated at att [40] . P1 phage-mediated generalizedtransduction was used to move [recC ptr recB recD]::Plac-betexo kan [from KM22] into SME1216 [selecting for kanamycin resistance]to make SME2417 . SME2495 was made using P1 transduction to move [rhaSR]::kan zih-35::Tn10 [from SME2800] into SME1217, selectingfor tetracycline resistance and then screening for a Rha^– phenotype . SME2496 was made from SME2416 by transformation witha PCR product containing [rhaSR]::cat-sac [amplified from pSE254,described below], which was recombined onto the chromosome byusing the recombination genes of bacteriophage [encoded by [recC ptr recB recD]::Plac-bet exo kan] [31] . The plasmid pSE250was made by restriction endonuclease digestion of pSE101 withBamHI at the rhaT' end of the clone and EcoRI [a natural sitewithin the rhaBAD promoter], creating the fragment rhaSRT',which was purified from an agarose gel by using QIAGEN's QIAEXGel Extraction kit . The rhaSRT' fragment was then ligated topUC18 [46], which had been digested with BamHI and EcoRI . Tomake pSE254, long-way-around PCR using pSE250 as the templateand primers 2297 and 2298 amplified all of the pSE250 sequenceexcept rhaSR and added a BglII site at each end . Then, PCR withprimers 2299 and 2300, using a PCR product containing the cat-saccassette [provided by Kenan Murphy] [32] as the template, generateda product that was ligated to the BglII sites of the long-way-aroundPCR product to create pSE254 . Plasmids pML148 to -169 [containingmutations in the rpoD gene] were obtained from the laboratoryof Carol Gross and were sequenced to ensure that they stillcarried the expected mutations . Several of the rpoD alleleswere initially found to be wild type . Assays involving thesealleles were repeated upon obtaining true mutants.

 
Mutagenesis of rhaS and rhaR. The mutant rhaS D250A allele was moved from pSE172 into thecontext of pSE101 by digesting pSE172 with BstEII and BglII[both sites are native to the wild-type rhaS gene] to createa fragment encoding the RhaS D250A substitution . This fragmentwas then ligated to similarly digested pSE101 to make pSE249.Genes encoding alanine substitution derivatives of RhaR D276Aand RhaR D285A were constructed by oligonucleotide-directedmutagenesis of rhaR in pGEM-11Zf[+] [Promega, Madison, Wis.],using the GeneEditor kit [Promega] and oligonucleotides 2208and 2210 . The mutant rhaR alleles were then subcloned into pSE250,using NheI and SmaI restriction endonuclease sites [both sitesoccur naturally within rhaR], to make pSE251 and pSE253, respectively.The rhaR E284A mutagenesis was performed using PCR to make oligonucleotide-directed mutations at the desired position with primer 2381, which also contained the recognition sequence for the EarI restriction endonuclease . Second, a nonmutagenized PCR fragment was made,also using a primer with EarI restriction sites . Finally, ligationof the mutant and wild-type DNA fragments allowed seamless reconstruction[25] of the full-length rhaR E284A gene in the context of rhaSRT' to make pSE252 . Oligos Etc [Wilsonville, Oreg.], IntegratedDNA Technologies [Coralville, Iowa], and MWG-Biotech [High Point,N.C.] synthesized oligonucleotide primers used in mutagenesis[Table 1] . Mutations were initially identified by diagnostic PCR using the following method . The very 3' nucleotides of the diagnostic oligonucleotide contained the desired substitution[s], such that amplification was possible only [in combination witha suitable downstream primer] when the template DNA carriedthe desired mutation . Putative mutants identified by this methodwere confirmed by DNA sequencing of both strands of the entirecloned region [see Table 1 for sequencing oligonucleotides].

Recombination of rhaS and rhaR alleles onto the chromosome. The mutant rhaS and rhaR alleles constructed as described above were recombined onto the E . coli chromosome such that they replaced the wild-type rhaS or rhaR allele by the following methods. Each rhaS or rhaR mutant was present on a plasmid in the contextof rhaSRT' [pSE249, pSE251, pSE252, or pSE253] . Oligonucleotides1170 and 2292 were used to amplify each mutant rhaSRT' regionby high-fidelity PCR . Approximately 500 ng of the rhaSRT' PCRproduct carrying a mutant allele was used to transform eitherSME2495 [ [rhaB-lacZ]110 [recC ptr recB recD]::Plac-bet exokan [rhaSR]::kan, zih-35::Tn10] or SME2496 [ [rhaB-lacZ]110 [recC ptr recB recD]::Plac-bet exo kan [rhaSR]::cat-sac, zih-35::Tn10].Since SME2495 and SME2496 both contained P_lac-bet exo, whichencodes the phage recombination proteins, the frequency ofhomologous recombination was much higher in these strains thanin wild-type E . coli strains [31] . The transformants were screened by spread plating on media containing 5-bromo-4-chloro-3-indolyl-ß-D-galactopyranoside [X-Gal] [40 µg/ml] and L-rhamnose [0.2%] to identify functional,or partially functional, rhaSR genes that replaced the rhaSR allele . There was no selection for successful recombinants in SME2495; rather we screened for blue colonies among a lawn ofwhite colonies . When transforming SME2496, which contains acat-sacB cassette [32] in place of rhaSR, we selected for the sucrose resistance of cells that had lost the sacB gene [which confers sucrose sensitivity] by homologous recombination . However, due to a significant background of spontaneous sucrose-resistant mutants, the transformants were also screened for at least a partially functional rhaS or rhaR gene by adding X-Gal [40 µg/ml] and L-rhamnose [0.2%] to the SacB selection plates . We foundthat sucrose inhibition of the sacB⁺ cells worked most reproduciblyat room temperature, although it took about 3 days for the cellsto grow . Phage P1-mediated generalized transduction was thenused to transfer the rhaS or rhaR allele of interest [linkedto zih-35::Tn10] to either SME1851 [ [rhaB-lacZ]84] or SME2515[ [rhaS-lacZ]92] by selecting for the tetracycline resistanceconferred by zih-35::Tn10 . Diagnostic PCR, as described above,was used to initially identify transductants that containedthe rhaS or rhaR mutation of interest . High-fidelity PCR wasthen used to amplify rhaSRT' from the chromosome, using oligonucleotides2097 and 1170, and the entire 3-kb PCR product was sequenced,as described above, to verify the presence of the desired mutationwith no additional mutations . Phage P1-mediated transductionwas then used to introduce recA::kan into each strain to makeSME2689, -2691, -2692, and -2933 . Finally, competent cells ofeach strain were made and transformed with plasmids containingeither the wild type-, K593A-, L595A-, R599A-, or R608A-encoding ⁷⁰ gene for ß-galactosidase assays.

ß-Galactosidase assay. ß-Galactosidase assays were performed as previouslydescribed [3] . In all cases, chromosomal rpoD was expressedfrom its own promoter, not the trp promoter described by Lonettoet al . [27], and the plasmid-encoded ⁷⁰ derivatives were expressedin the absence of isopropyl-ß-D-thiogalactopyranoside. Under these conditions, the ⁷⁰ derivatives are expected to accountfor approximately 50% of the total ⁷⁰ in the cells [27] . Specificactivities were averaged from at least three independent assayswith two replicates in each assay . The assays were performedon at least two different days, with independent cell growthsteps [starter tryptone-yeast extract culture, overnight culture,and final growth culture] for each assay.

 
⁷⁰ derivatives at the rhaSR promoter. We wished to determine whether any residues near the C-terminalend of ⁷⁰ were important for transcription activation by RhaR.Lonetto et al . [27] constructed a library of alanine substitutionsat 17 different positions near the C terminus of ⁷⁰ and foundthat some substitutions resulted in defects at class II activator-dependentpromoters . Previous work from our lab found that two of thealanine substitutions in this library were defective at a truncatedrhaBAD promoter where RhaS was the only transcription activator[4] . We assayed this library of alanine substitutions in ⁷⁰ at two RhaR-activated single-copy translational fusions, [rhaS-lacZ]216 and [rhaS-lacZ]92 . The [rhaS-lacZ]216 promoter contained theRhaR binding site as well as upstream CRP sites, while the [rhaS-lacZ]92 promoter contained only the RhaR binding site [Fig . 1] . Sincethe assays were carried out with a strain that also expressed wild-type ⁷⁰ from the chromosome [27], we considered values below 80% of wild-type activity to be significant defects . At [rhaS-lacZ]216, ⁷⁰ derivative L595A had 66% of the activityof wild-type ⁷⁰ [Fig . 2A] while the remaining ⁷⁰ derivativeswere not significantly defective . When the same sigma derivativeswere assayed at [rhaS-lacZ]92, L595A was still significantlydefective, with 53% activity compared to wild-type ⁷⁰ [Fig.2B] . In addition, three alanine substitutions of positivelycharged amino acid residues, K593A [79%], R599A [48%], and R608A[77%] were defective at [rhaS-lacZ]92 . These results suggestedthat ⁷⁰ residues K593, L595, R599, and R608 might make protein-protein contacts with RhaR that are required for transcription activation. The lack of defect from ⁷⁰ K593A, R599A, or R608A at the [rhaS-lacZ]216 promoter is similar to previous findings with RhaS and AraC that substitutions at some ⁷⁰ residues were defective only inthe absence of CRP activation [4, 27].

 
Based on the previous finding in our lab that a contact betweenRhaS residue D241 and ⁷⁰ residue R599 is required for full transcriptionactivation by RhaS [4], we predicted that RhaR D276, which alignswith RhaS D241 [Fig. 3], might contact ⁷⁰ R599 at the rhaSRpromoter . Molecular modeling of the RhaR-⁷⁰ interaction [Fig.4] indicates that the negatively charged RhaR residue D276 isvery close to the positively charged ⁷⁰ residue R599 . RhaR D276is also near ⁷⁰ R608, although in the model they do not appearclose enough to interact . The molecular model further showsthat two adjacent negatively charged RhaR residues, E284 andD285, are located near ⁷⁰ K593 . Based on these pieces of evidence,we hypothesized that contacts between some or all of the RhaRresidues D276, E284, and D285 and ⁷⁰ might be required for maximaltranscription activation by RhaR . We therefore tested alaninesubstitutions at these positions in RhaR for defects in transcriptionactivation.

 
RhaR residues D276, E284, and D285 are important for rhaSR transcription activation. In order to test whether the side chains of RhaR residues D276,E284, and D285 might play a role in transcription activationby RhaR, we constructed alanine substitutions at each of thesepositions . If the amino acid residues at these positions are required for transcription activation, the alanine substitution should result in a significant decrease in activation of rhaSR transcription . To assay the RhaR derivatives, the mutant rhaR alleles on plasmids were first recombined onto the chromosomesuch that they replaced the wild-type rhaR gene [see Materialsand Methods] . The wild-type and mutant rhaR alleles were then assayed for activation of [rhaS-lacZ]92 [Fig . 5] . The resultsshowed that all three of the alanine substitutions in RhaR weresignificantly defective, highlighting the importance of thewild-type residues at those positions . The especially largedefect of RhaR D285A may be partly due to a role in DNA bindingbased on its alignment with D250 in RhaS [Fig . 3], which makesbase-specific contacts with DNA [3] . However, a role in DNAbinding for RhaR D285 does not rule out interactions with ⁷⁰; therefore, all three of these RhaR residues are candidates for specific contacts with ⁷⁰.

 
Evidence for an interaction between RhaR D276 and ⁷⁰ R599. We used a genetic loss-of-contact approach to test potential interactions between RhaR D276 and ⁷⁰ K593 or R599 . Using thisapproach, we separately combined wild-type RhaR or the RhaRD276A derivative with each of three plasmids encoding either ⁷⁰ wild type, K593A, or R599A in a strain carrying a singlecopy of [rhaS-lacZ]92 . The results shown in Fig . 6A were plottedso that the activity with wild-type ⁷⁰ was set to 100% for eachRhaR derivative, thereby illustrating the relative effects ofeach ⁷⁰ derivative . On this graph, therefore, a ⁷⁰ derivativethat does not define a site of interaction with a given RhaRderivative is expected to have the same relative defect when combined with the indicated RhaR derivative as it does with wild-type RhaR, since the defects will be independent of eachother . On the other hand, if a ⁷⁰ derivative does define a siteof interaction with a given RhaR derivative, the ⁷⁰ derivativewill confer no further defect when combined with the indicatedRhaR derivative, since the interaction would already have beenlost with the RhaR derivative . Using this method to analyze the results in Fig . 6A, our first conclusion is that there isno interaction between RhaR D276 and ⁷⁰ K593 since the ⁷⁰ K593Aderivative had approximately the same relative defect in combinationwith either wild-type RhaR or RhaR D276A . Therefore, the defectsof ⁷⁰ K593A and RhaR D276A are independent . These results andthose in Fig. 2B also show that the ⁷⁰ R599A derivative by itself[in a wild-type rhaR strain] had approximately 50% activitycompared to wild-type ⁷⁰ . However, when ⁷⁰ R599A was combinedwith RhaR D276A, the ⁷⁰ R599A derivative conferred no furtherdefect upon RhaR D276A . In fact, the strain with the combinationof ⁷⁰ R599A and RhaR D276A had approximately 1.7-fold-higheractivity than the strain with wild-type ⁷⁰ and RhaR D276A . Theseresults fit the criteria for an allele-specific contact between ⁷⁰ R599 and RhaR D276 . As mentioned above, molecular modelingis consistent with this interaction since RhaR D276 is in close proximity to ⁷⁰ R599 in the model [Fig . 4] . We also tested the ⁷⁰ R608A derivative in combination with RhaR D276A and foundthat it had the same relative defective as it did with wild-typeRhaR [79% of wild-type ⁷⁰ activity in both cases [data not shown]];therefore, there was no indication of an interaction betweenthese two residues.

 
RhaR E284 and D285 and ⁷⁰. Using the same genetic loss-of-contact approach, we also testedfor potential interactions between ⁷⁰ and RhaR E284 and D285.The results in Fig . 6B show that the K593A ⁷⁰ derivative wasnot defective in combination with RhaR D285A [104% of wild-type ⁷⁰ activity] but the R599A ⁷⁰ derivative also became less defective[81% of wild-type ⁷⁰ activity] . In the absence of the resultsobtained for ⁷⁰ 599A, one might conclude that RhaR D285 contacts ⁷⁰ K593, since ⁷⁰ K593A had no significant defect when combinedRhaR D285A; however, the lack of strict allele specificity shedsdoubt on this conclusion . To further investigate the non-allele-specificdefects of ⁷⁰ substitutions in combination with RhaR D285A,we tested ⁷⁰ L595A and R608A derivatives, which were both defectivein a wild-type rhaR strain, as shown in Fig . 2B . When combined with RhaR D285A, the ⁷⁰ L595A and R608A derivatives were notsignificantly defective, with 86 and 87% of wild-type ⁷⁰ activity,respectively [data not shown] . These results suggest that RhaRD285A may reduce the ability of RhaR to interact with ⁷⁰ ina non-allele-specific manner; therefore, we can't conclude whether RhaR D285 contacts any of these ⁷⁰ residues.

Figure 6C shows the results of assays to identify potential interactions involving RhaR E284 . Our results showed that neither ⁷⁰ K593A nor R599A conferred a significant defect on RhaR E284A.In fact, the ⁷⁰ K593A-RhaR 284A combination gave much higheractivity [362%] than the RhaR 284A derivative with wild-type ⁷⁰ . Therefore, as described above, we tested the ⁷⁰ L595A andR608A derivatives in the rhaR E285A strain and found 139 and93% activity, respectively, compared to wild-type ⁷⁰ [data notshown] . Thus, we again found that all four of the ⁷⁰ derivativesthat were defective in the wild-type rhaR strain were no longersignificantly defective in the rhaR E284A strain . These resultssuggest that, similar to the RhaR D285A derivative, the RhaRE284A derivative may reduce the ability of RhaR to interactwith ⁷⁰ in a non-allele-specific manner . One explanation forthe very high relative activity of RhaR E284A in combinationwith ⁷⁰ K593A is that a new interaction may have been createdin this case.

Evidence for a specific interaction between RhaS D250 and ⁷⁰. Our lab previously identified an interaction between RhaS D241and ⁷⁰ R599, and we also found that ⁷⁰ K593A was defective at [rhaB-lacZ]84 but did not identify an amino acid in RhaS thatmight contact ⁷⁰ K593 [4] . The molecular model in Fig . 4 showsthat the only negatively charged RhaS residue that is in close proximity to the positively charged ⁷⁰ K593 is RhaS D250, suggestingthat these two residues might make a contact . Previous resultsfrom our lab showed that RhaS D250A was 12-fold defective for [rhaB-lacZ]84 activation; however, they also indicated thatthis residue participates in a base-specific DNA contact [3].In contrast to other approaches to identify positive-controlmutants, the genetic loss-of-contact approach does not requirethat the protein have wild-type DNA-binding capability; henceit has the potential to identify residues that have dual DNA-bindingand transcription activation functions . We therefore used thegenetic loss-of-contact approach to test whether RhaS D250 and ⁷⁰ K593 might be involved in an interaction . The results [Fig. 6D] support the hypothesis of an interaction between RhaS D250 and ⁷⁰ K593 since the K593A derivative was not significantlydefective when combined with RhaS D250A [87% activity comparedto wild-type ⁷⁰] . However, when ⁷⁰ R599A was combined with RhaSD250A, it maintained approximately the same relative defectas it had with wild-type RhaS . These results suggest that thereis an allele-specific interaction between RhaS D250 and ⁷⁰ K593.Molecular modeling is consistent with this interaction since, as mentioned above, RhaS D250 is in close proximity to ⁷⁰ K593in the model [Fig . 4].

 
The C terminus of ⁷⁰ is important for RhaS- and RhaR-mediated transcription activation. The binding sites for both RhaS and RhaR overlap the –35region of their respective core promoters by 4 bp, placing themin ideal positions to interact with the ⁷⁰ subunit of RNAP.Previous results reported by Bhende and Egan [4] identifiedtwo amino acid residues in ⁷⁰, K593 and R599, that were importantfor RhaS-mediated transcription activation at rhaBAD and rhaT.In the present study, we identified four amino acid residuesin ⁷⁰, K593, L595, R599, and R608, which were important forRhaR-mediated transcription activation of rhaSR [Fig . 2B] . Two of the alanine substitutions in ⁷⁰, K593A and R599A, were defectiveat all three of the rha promoters, suggesting similar mechanismsof activation by RhaS and RhaR.

The results reported in this paper [Fig . 2], as well as thosefrom a previous study [4], showed that ⁷⁰ K593A and R599A weredefective only at truncated rha promoters that did not includethe upstream CRP binding sites . This is similar to the findingsobtained for several other promoters that require multiple activators,such as araBAD, uhpT, and narG [18, 27, 36, 39] . Two possibleexplanations for this trend are that the second activator increasesthe total number of interactions such that the relative importanceof each individual interaction decreases or that the secondactivator creates redundancies in activation that mask the importanceof other interactions . A third possibility is that the secondactivator alters the orientation of the first activator relativeto ⁷⁰ such that the primary activator is no longer in an idealposition to interact with ⁷⁰ . In the first two models, the activatorinteraction with ⁷⁰ occurs in both the presence and absenceof the second activator but can be detected only in its absence,whereas in the third model, the interaction between the firstactivator and ⁷⁰ occurs only in the absence of the second activator.Further experiments will be needed to distinguish these models.At the rhaSR promoter, the ⁷⁰ L595A derivative was unique inthat it was defective in both the presence and absence of thesecond activator, CRP, but it's role in RhaR-mediated transcriptionactivation is not yet known.

Specific amino acid contacts between ⁷⁰ and RhaR. Previous results showing an interaction between RhaS D241 and ⁷⁰ R599 at rhaBAD [4] led us to investigate whether an interactionbetween RhaR D276 and ⁷⁰ R599 might be required for RhaR activationat rhaSR . We also used a molecular model of the RhaR-⁷⁰ domain4 interaction in which the structure of MarA [38] representedRhaR [Fig . 4] to identify the only two negatively charged RhaRresidues, E284 and D285, that were near ⁷⁰ K593 . RhaR E284 andD285 were therefore considered candidates for residues thatmight interact with ⁷⁰ K593 . After determining that alaninesubstitutions at RhaR residues D276, E284, and D285 were alldefective for rhaSR activation [Fig . 5], we used a genetic loss-of-contact approach to test for specific amino acid interactions between ⁷⁰ and RhaR.

To carry out a loss-of-contact analysis, one must first identify defective derivatives of each of the potentially interacting proteins . In the simplest case, the full defect of both of thetwo interacting residues is due to loss of the interaction—inother words, the only role of the two residues is the interaction.The rationale behind this approach in this simple case is thatmutation of one or the other of the interacting residues willeliminate the interaction; therefore, the phenotype of a straincarrying both mutations will be the same as the phenotype ofthe strains carrying the individual mutations . If one of theresidues has a second role in addition to the interaction, thenthe strain carrying both mutations will have a phenotype thatis no worse than the more defective of the strains carryingthe two individual mutations . This analysis does not provideconclusive results if both residues have roles in addition to the interaction . It is also expected that the predicted interactions will be allele specific . The majority of combinations of defective derivatives are not expected to identify interacting residues, and in these cases the defects resulting from each of the two mutations will at least be additive.

Using this rationale to interpret our genetic loss-of-contact assays, the results in Fig . 6A provide evidence for an interactionbetween ⁷⁰ R599 and RhaR D276 . This result is similar to previousresults from our lab [4] that indicate an interaction between ⁷⁰ R599 and RhaS D241 and evidence from Grainger et al . that ⁷⁰ R599 interacts with MelR D261 [16], which aligns with RhaS D241 and RhaR D276 . Molecular modeling of the RhaR-⁷⁰ complex[Fig . 4] shows that ⁷⁰ R599 and RhaR D276 are in close proximity,consistent with our interpretation that these two residues interact.Our genetic loss-of-contact results do not provide evidencefor an interaction between ⁷⁰ K593 and RhaR E284 or D285 [Fig.6C] . Instead, our results indicate that alanine substitutionsat RhaR E284 and D285 result in non-allele-specific decreasesin the defects of all of the ⁷⁰ alleles tested . One hypothesisis that RhaR E284A and D285A alter the details of the RhaR DNAinteraction such that RhaR is no longer in an ideal positionto interact with ⁷⁰ domain 4.

The role of ⁷⁰ K593 in transcription activation by RhaS. Residue K593 of ⁷⁰ has been found to be important for severaltranscription activators, including AraC, UhpA, cI, FNR, Ada,RhaR [this study], and RhaS [4, 24, 27, 35, 36] . With the exceptionof cI and RhaS [this study], evidence that ⁷⁰ K593 directlycontacts an activator has not been obtained . Our results indicatethat ⁷⁰ K593 contacts RhaS D250 as a part of the mechanism ofactivation by RhaS [Fig . 6D] . Our molecular model of the RhaS-⁷⁰ interaction shows that ⁷⁰ K593 and RhaS D250 are in close proximity,consistent with this result [Fig . 4] . While the binary complexof Taq ^A domain 4 and DNA shows that the residue that correspondsto ⁷⁰ K593 contacts DNA, in the cI- domain 4-DNA ternary complex,this residue participates in a protein-protein contact with cI instead [6, 19] . These findings indicate that ⁷⁰ K593 iscapable of interacting with an appropriately positioned transcriptionactivator and are consistent with our proposal that ⁷⁰ K593may contact RhaS D250.

Comparison of transcription activation by RhaS and RhaR. In this study we identified an interaction between RhaR D276and ⁷⁰ R599 that is equivalent to our previously identifiedinteraction of RhaS D241 and ⁷⁰ R599 . Further, although a RhaRequivalent of the RhaS D250 interaction with ⁷⁰ K593 was notidentified, our results do not rule out that such an interactionoccurs with RhaR . Therefore, our current evidence suggests thatthe RhaS-⁷⁰ interface is similar to the RhaR-⁷⁰ interface . Wecertainly expect, however, that not all aspects of RhaS activationand RhaR activation will be identical . For example, we knowthat the CRP site at rhaBAD is centered at position –92.5,whereas the most important CRP site at rhaSR is centered atposition –111.5 . It is not possible to draw conclusions about how or whether differences in the RhaS-⁷⁰ and RhaR-⁷⁰ interfaces might relate to this difference in CRP binding site position since all but one of the ⁷⁰ derivatives tested weredefective only in the absence of CRP . However, it is likelythat there is a difference in the mechanisms of RhaS and RhaRactivation that relates to this difference in the positionsof the CRP binding sites.

 
We thank Carol Gross for the library of alanine substitutionsin ⁷⁰, Richard Wolf for alerting us that some of the ⁷⁰ mutantshad reverted to the wild type, Kenan Murphy for providing strainKM22 and the cat-sacB cassette, Jeff Urbauer for assistancewith the modeling of ⁷⁰ domain 4 in the MarA-DNA structure,and Vydehi Rao for performing the assays of the ⁷⁰ library inthe strain containing [rhaS-lacZ]92.

This work was supported by Public Health Service grant GM55099 from the National Institute of General Medical Sciences andNIH Grant RR-P20 RR17708 from the Institutional DevelopmentAward Program of the National Center for Research Resources,both to S.M.E.

 
* Corresponding author . Mailing address: Department of Molecular Biosciences, University of Kansas, Lawrence, KS 66045 . Phone: [785] 864-4294 . Fax: [785] 864-5294 . E-mail: sme@ku.edu.

1. Akimaru, H., K . Sakumi, T . Yoshikai, M . Anai, and M . Sekiguchi. 1990 . Positive and negative regulation of transcription by a cleavage product of Ada protein . J . Mol . Biol . 216:261-273.
2. Backman, K., Y.-M . Chen, and B . Magasanik. 1981 . Physical and genetic characterization of the glnA-glnG region of the Escherichia coli chromosome . Proc . Natl . Acad . Sci . USA 78:3743-3747.
3. Bhende, P . M., and S . M . Egan. 1999 . Amino acid-DNA contacts by RhaS: an AraC family transcription activator . J . Bacteriol . 181:5185-5192 .
4. Bhende, P . M., and S . M . Egan. 2000 . Genetic evidence that transcription activation by RhaS involves specific amino acid contacts with sigma 70 . J . Bacteriol . 182:4959-4969 .
5. Bustos, S . A., and R . F . Schleif. 1993 . Functional domains of the AraC protein . Proc . Natl . Acad . Sci . USA 90:5638-5642.
6. Campbell, E . A., O . Muzzin, M . Chlenov, J . L . Sun, C . A . Olson, O . Weinman, M . L . Trester-Zedlitz, and S . A . Darst. 2002 . Structure of the bacterial RNA polymerase promoter specificity subunit . Mol . Cell 9:527-539.
7. Dhiman, A., and R . Schleif. 2000 . Recognition of overlapping nucleotides by AraC and the sigma subunit of RNA polymerase . J . Bacteriol . 182:5076-5081 .
8. Dove, S . L., S . A . Darst, and A . Hochschild. 2003 . Region 4 of sigma as a target for transcription regulation . Mol . Microbiol . 48:863-874.
9. Egan, S . M. 2002 . Growing repertoire of AraC/XylS activators . J . Bacteriol . 184:5529-5532.
10. Egan, S . M., and R . F . Schleif. 1994 . DNA-dependent renaturation of an insoluble DNA binding protein . Identification of the RhaS binding site at rhaBAD . J . Mol . Biol . 243:821-829.
11. Egan, S . M., and R . F . Schleif. 1993 . A regulatory cascade in the induction of rhaBAD . J . Mol . Biol . 234:87-98.
12. Eustance, R . J., S . A . Bustos, and R . E . Schleif. 1994 . Reaching out: locating and lengthening the interdomain linker in AraC protein . J . Mol . Biol . 242:330-338.
13. Gallegos, M.-T., R . Schleif, A . Bairoch, K . Hofmann, and J . L . Ramos. 1997 . AraC/XylS family of transcriptional regulators . Microbiol . Mol . Biol . Rev . 61:393-410.
14. Gay, P., D . Le Coq, M . Steinmetz, T . Berkelman, and C . I . Kado. 1985 . Positive selection procedure for entrapment of insertion sequence elements in gram-negative bacteria . J . Bacteriol . 164:918-921.
15. Grainger, D . C., T . A . Belyaeva, D . J . Lee, E . I . Hyde, and S . J . Busby. 2004 . Transcription activation at the Escherichia coli melAB promoter: interactions of MelR with the C-terminal domain of the RNA polymerase alpha subunit . Mol Microbiol . 51:1311-1320.
16. Grainger, D . C., C . L . Webster, T . A . Belyaeva, E . I . Hyde, and S . J . Busby. 2004 . Transcription activation at the Escherichia coli melAB promoter: interactions of MelR with its DNA target site and with domain 4 of the RNA polymerase sigma subunit . Mol Microbiol . 51:1297-1309.
17. Holcroft, C . C., and S . M . Egan. 2000 . Interdependence of activation at rhaSR by cyclic AMP receptor protein, the RNA polymerase alpha subunit C-terminal domain and RhaR . J . Bacteriol . 182:6774-6782 .
18. Hu, J . C., and C . A . Gross. 1985 . Mutations in the sigma subunit of E . coli RNA polymerase which affect positive control of transcription . Mol . Gen . Genet . 199:7-13.
19. Jain, D., B . E . Nickels, L . Sun, A . Hochschild, and S . A . Darst. 2004 . Structure of a ternary transcription activation complex . Mol . Cell 13:45-53.
20. Jair, K., R . G . Martin, J . L . Rosner, N . Fujita, A . Ishihama, and R . E . Wolf, Jr. 1995 . Purification and regulatory properties of MarA protein, a transcriptional activator of Escherichia coli multiple antibiotic and superoxide resistance promoters . J . Bacteriol . 177:7100-7104.
21. Jair, K.-W., W . P . Fawcett, N . Fujita, A . Ishihama, and R . E . Wolf, Jr. 1996 . Ambidextrous transcriptional activation by SoxS: requirement for the C-terminal domain of the RNA polymerase alpha subunit in a subset of Escherichia coli superoxide-inducible genes . Mol . Microbiol. 19:307-317.
22. Kaldalu, N., U . Toots, V . de Lorenzo, and M . Ustav. 2000 . Functional domains of the TOL plasmid transcription factor XylS . J . Bacteriol . 182:1118-1126 .
23. Kuldell, N., and A . Hochschild. 1994 . Amino acid substitutions in the –35 recognition motif of ⁷⁰ that result in defects in phage repressor-stimulated transcription . J . Bacteriol . 176:2991-2998.
24. Landini, P., and S . J . Busby. 1999 . The Escherichia coli Ada protein can interact with two distinct determinants in the ⁷⁰ subunit of RNA polymerase according to promoter architecture: identification of the target of Ada activation at the alkA promoter . J . Bacteriol . 181:1524-1529 .
25. LaPointe, C . F., and R . K . Taylor. 2000 . The type 4 prepilin peptidases comprise a novel family of aspartic acid proteases . J . Biol . Chem . 275:1502-1510 .
26. Li, M., H . Moyle, and M . M . Susskind. 1994 . Target of the trancriptional activation function of phage cI protein . Science 263:75-77.
27. Lonetto, M . A., V . Rhodius, K . Lamberg, P . Kiley, S . Busby, and C . Gross. 1998 . Identification of a contact site for different transcription activators in region 4 of the Escherichia coli RNA polymerase ⁷⁰ subunit . J . Mol . Biol . 284:1353-1365.
28. Martin, R . G., and J . L . Rosner. 2001 . The AraC transcriptional activators . Curr . Opin . Microbiol . 4:132-137.
29. Menon, K . P., and N . L . Lee. 1990 . Activation of ara operons by a truncated AraC protein does not require inducer . Proc . Natl . Acad . Sci . USA 87:3708-3712.
30. Meyer, B . J., R . Maurer, and M . Ptashne. 1980 . Gene regulation at the right operator [OR] of bacteriophage lambda . II . OR1, OR2, and OR3: their roles in mediating the effects of repressor and cro . J . Mol . Biol . 139:163-194.
31. Murphy, K . C. 1998 . Use of bacteriophage recombination functions to promote gene replacement in Escherichia coli . J . Bacteriol . 180:2063-2071 .
32. Murphy, K . C., K . G . Campellone, and A . R . Poteete. 2000 . PCR-mediated gene replacement in Escherichia coli . Gene 246:321-330.
33. Nakabeppu, Y., and M . Sekiguchi. 1986 . Regulatory mechanisms for induction of synthesis of repair enzymes in response to alkylating agents: ada protein acts as a transcriptional regulator . Proc . Natl . Acad . Sci . USA 83:6297-6301.
34. Neidhardt, F . C., P . L . Bloch, and D . F . Smith. 1974 . Culture medium for enterobacteria . J . Bacteriol . 119:736-747.
35. Nickels, B . E., S . L . Dove, K . S . Murakami, S . A . Darst, and A . Hochschild. 2002 . Protein-protein and protein-DNA interactions of sigma70 region 4 involved in transcription activation by lambdacI . J . Mol . Biol . 324:17-34.
36. Olekhnovich, I . N., and R . J . Kadner. 1999 . RNA polymerase and ⁷⁰ subunits participate in transcription of the Escherichia coli uhpT promoter . J . Bacteriol . 181:7266-7273 .
37. Poore, C . A., C . Coker, J . D . Dattelbaum, and H . L . Mobley. 2001 . Identification of the domains of UreR, an AraC-like transcriptional regulator of the urease gene cluster in Proteus mirabilis . J . Bacteriol . 183:4526-4535 .
38. Rhee, S., R . G . Martin, J . L . Rosner, and D . R . Davies. 1998 . A novel DNA-binding motif in MarA: the first structure for an AraC family transcriptional activator . Proc . Natl . Acad . Sci . USA 95:10413-10418 .
39. Silverstone, A . E., M . Goman, and J . G . Scaife. 1972 . ALT: a new factor involved in the synthesis of RNA by Escherichia coli . Mol . Gen . Genet . 118:223-234.
40. Simons, R . W., F . Houman, and N . Kleckner. 1987 . Improved single and multicopy lac-based cloning vectors for protein and operon fusions . Gene 53:85-96.
41. Soisson, S . M., B . MacDougall-Shackleton, R . Schleif, and C . Wolberger. 1997 . Structural basis for ligand-regulated oligomerization of AraC . Science 276:421-425 .
42. Tobin, J . F., and R . F . Schleif. 1987 . Positive regulation of the Escherichia coli L-rhamnose operon is mediated by the products of tandemly repeated regulatory genes . J . Mol . Biol . 196:789-799.
43. Tobin, J . F., and R . F . Schleif. 1990 . Purification and properties of RhaR, the positive regulator of the L-rhamnose operons of Escherichia coli . J . Mol . Biol . 211:75-89.
44. Travers, A. 1974 . RNA polymerase-promoter interactions: some general principles . Cell 3:97-104.
45. Via, P., J . Badia, L . Baldoma, N . Obradors, and J . Aguilar. 1996 . Transcriptional regulation of the Escherichia coli rhaT gene . Microbiology 142:1833-1840.
46. Yanisch-Perron, C., J . Vieira, and J . Messing. 1985 . Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mp18 and pUC19 vectors . Gene 33:103-119.

Free Online Full-text Article

What Is Salmonella?, What Is Bioassay?, What Is Activated Sludge?, What Is Pcr?, What Is Listeria Monocytogenes?, c, Bacterium, s, Microbes, s, Bacteria, r, Microorganism, s, Microbe, s, Bioreactors, n, Salmonella typhimurium, c, Acinetobacter, i, Escherichia coli, i, Bacteriophages, i, Staphylococcus, e, Pseudomonas aeruginosa, a, Biological reactors, e, Microorganism,