Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M308002200 on September 8, 2003

J. Biol. Chem., Vol. 278, Issue 47, 46219-46229, November 21, 2003
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
278/47/46219    most recent
M308002200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Görke, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Görke, B.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Regulation of the Escherichia coli Antiterminator Protein BglG by Phosphorylation at Multiple Sites and Evidence for Transfer of Phosphoryl Groups between Monomers*

Boris Görke{ddagger}

From the Institut für Biologie III, Universität Freiburg, Schänzlestrasse 1, D-79104 Freiburg, Germany

Received for publication, July 23, 2003 , and in revised form, August 26, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Activity of antiterminator protein BglG regulating the {beta}-glucoside operon in Escherichia coli is controlled by the phosphoenolpyruvate:carbohydrate phosphotransferase system (PTS) in a dual manner. It requires HPr phosphorylation to be active, whereas phosphorylation by the {beta}-glucoside-specific transport protein EIIBgl inhibits its activity. BglG and its relatives carry two PTS regulation domains (PRD1 and PRD2), each containing two conserved histidines. For BglG, histidine 208 in PRD2 was reported to be the negative phosphorylation site. In contrast, other antiterminators of this family are negatively regulated by phosphorylation of the first histidine in PRD1, and presumably activated by phosphorylation of the histidines in PRD2. In this work, a screen for mutant BglG proteins that escape repression by EIIBgl yielded exchanges of nine residues within PRD1, including conserved histidines His-101 and His-160, and C-terminally truncated proteins. Genetic and phosphorylation analyses indicate that His-101 in PRD1 is phosphorylated by EIIBgl and that His-160 contributes to negative regulation. His-208 in PRD2 is essential for BglG activity, suggesting that it is phosphorylated by HPr. Surprisingly, phosphorylation by HPr is not fully abolished by exchanges of His-208. However, phosphorylation by HPr is inhibited by exchanges in PRD1 and the phosphorylation of these mutants is restored in the presence of wild-type BglG. These results suggest that the activating phosphoryl group is transiently donated from HPr to PRD1 and subsequently transferred to His-208 of a second BglG monomer. The active His-208-phosphorylated BglG dimer can subsequently be inhibited in its activity by EIIBgl-catalyzed phosphorylation at His-101.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
The bacterial phosphoenolpyruvate:carbohydrate phosphotransferase system (PTS)1 is central in uptake and utilization of various carbohydrates (reviewed in Ref. 1). Primarily, it catalyzes a chain of phosphoryl transfer reactions toward the incoming carbohydrates. The first steps in this chain are the autophosphorylation of EI with PEP and subsequent transfer of the phosphoryl group from phospho-EI to HPr. Phospho-HPr is the general donor of phosphoryl groups to the various carbohydrate-specific transport proteins called enzymes II (EIIs). The EIIs consist of at least three domains, IIA, IIB, and IIC. HPr phosphorylates a histidine residue in domain IIA, from which the phosphoryl group is transferred to a residue in domain IIB, and from there to the sugar during its uptake by the integral membrane domain, IIC. Uptake of PTS substrates leads to dephosphorylation of the PTS proteins. The phosphorylation state of the PTS is measured to modulate the activity of a large number of proteins involved in carbohydrate uptake, in chemotaxis, and in gene regulation (2–4).

Transcriptional antiterminators of the BglG/SacY family are gene regulators directly controlled by PTS-catalyzed phosphorylations (reviewed in Refs. 5–8). They all control genes coding for EIIs and other functions involved in uptake and breakdown of PTS carbohydrates. All antiterminators carry a ~55 residue long RNA-binding domain called CAT at the N terminus (9), followed by two homologous domains, which putatively receive the signals from the PTS and transmit them by their dimerization, and which are therefore called PRDs (PTS regulation domains; Ref. 6). PRDs are also found in DNA binding activators, which, together with antiterminators, form a superfamily (10).

BglG/SacY-type antiterminators allow readthrough of transcription at specific Rho-independent terminators present within the transcription units they control. They bind as dimers to a conserved sequence in the RNA called RAT, thereby preventing formation of the overlapping terminator stem loop structure. Each antiterminator controls the expression of an assigned EII, which negatively regulates its antiterminator. This is achieved by phosphorylation of the antiterminator in the absence of substrate for the respective EII. The antiterminator becomes dephosphorylated by its EII and thereby reactivated when substrate becomes available, as has been demonstrated first for the BglG/EIIBgl pair in Escherichia coli (11, 12). In this case, a direct transfer of phosphoryl groups from the phosphorylation site cysteine 24 in IIBBgl to BglG has been proposed (13). BglG regulates the bgl operon, which codes for BglG itself, EIIBgl, and other functions for uptake and utilization of aryl-{beta}-glucosides (14, 15).

Recent work has elucidated that some of the antiterminators, including BglG in E. coli, SacT and LicT in Bacillus subtilis, and LacT in Lactobacillus casei, are in addition positively controlled by HPr. These antiterminators have to be phosphorylated by HPr at a second site to gain activity (16–20). This second phosphorylation event may provide a catabolite control mechanism to down-regulate activity of the antiterminators: when other, more favorable PTS-sugars, become available, the activating phosphoryl groups are withdrawn toward the preferred PTS-carbohydrate. The two antagonistic acting phosphorylations presumably modulate the transition from the inactive monomeric to the active dimeric form and vice versa. According to the current model these antiterminators exist in three different states, non-phosphorylated, phosphorylated by HPr, and multiple-phosphorylated by HPr and EII (reviewed for BglG in Refs. 3 and 4). At physiological concentrations, exclusively, the HPr phosphorylated form, which appears in the presence of the respective substrate but absence of preferred PTS substrates, is believed to dimerize and to be active in vivo.

Each PRD contains two highly conserved histidines, and all evidence suggests that they represent the phosphorylated sites (19–23). BglG is an exception in that it lacks the second histidine within PRD2. Where investigated, the first conserved histidine in PRD1 was found to be crucial for negative control. Exchange of this histidine leads to constitutively high antitermination activity in vivo even in the absence of the specific substrate (20, 24, 25). The second histidine in PRD1 participates in negative control of LicT and LacT (20, 25). In contrast, the histidine(s) in PRD2 were shown to be necessary for activity of the dually controlled proteins SacT, LicT, and LacT (6, 20, 25), and preferential phosphorylation of these residues by HPr has been demonstrated for LicT (19). Based on these observations, it has been proposed that these proteins are stimulated by HPr-dependent phosphorylation of the histidines in PRD2 and inhibited by EII-dependent phosphorylation of the first histidine in PRD1. However, quite different findings have been reported for BglG. In this case it was concluded that the conserved His-208 residue in PRD2 is the target of negatively acting phosphorylation by EIIBgl (26). Thus, no general rules appear to exist concerning the molecular mechanisms governing the regulation of this protein family (8, 20).

The work presented here was initiated to define residues in BglG important for negative regulation by EIIBgl. Using a genetic screen a number of BglG variants that escape negative regulation by EIIBgl were identified. In addition, BglG mutants with different substitutions for His-208 were constructed. Analysis of activity and regulation of the various mutants in vivo demonstrates a direct role of PRD1 rather than PRD2 in negative control. The role of the histidyl residues in BglG resembles their counterparts in the other dually PTS-controlled antiterminators; His-101 and His-160 in PRD1 both contribute to negative control, whereas His-208 in PRD2 appears to be a target of activating phosphorylation. Protein phosphorylation studies confirmed that indeed His-101 is phosphorylated by EIIBgl, but yielded surprising results for the phosphorylation event directly catalyzed by HPr. The data suggest that HPr primarily donates its phosphoryl group to PRD1, from where it is subsequently transferred to the His-208 residue of a second BglG monomer within a dimer in a trans intramolecular reaction.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Strains—The genotypes of E. coli strains are given in Table I. For phenotypic screening of mutant BglG proteins, strain R2243 was constructed as follows: First, the {lambda}PL-bglt2-lacZ cassette of plasmid pFDX3128 (see below) was integrated into the {lambda} attachment site (attB) of strain R1279 as recently described (27). From the resulting strain (R1735) a streptomycin-resistant mutant derivative was isolated, into which the F' lacIq lacZ{Delta}M15 lacY+ proA+ proB+ episome was introduced by conjugation with strain BMH71-18. Strain constructions followed the procedures given by Miller (28).


View this table:
[in this window]
[in a new window]
  		TABLE I
Strains and plasmids used in this study

 
Plasmids—The relevant structures of plasmids are given in Table I. Mutant plasmids isolated in the screen were sequenced using primers GK2 and GK3. These plasmids and their derivatives are listed in Table V, and oligonucleotides used in this study are given in Table VI under Supplemental Materials, respectively. Details on construction of recombinant plasmids are also described under Supplemental Materials.

Random Mutagenesis of bglG by Error-prone PCR and Isolation of Mutant bglG Alleles—Mutagenic PCRs were carried out according to Ref. 29. The 200-µl reactions were 10 mM Tris-HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, 50 µM of each dNTP, and contained 30 pmol of primers 619 and 620, 5 units of Taq polymerase and 2 fmol of the BspHI-SspI fragment of plasmid pFDX3736 (Fig. 2) as template. Primer 619 spans the AflII site in bglG and primer 620 spans the XbaI site between bglG and bglF and ends with the bglG stop codon at the 3'-end. Thirty cycles were carried out (1' 94 °C, 2' 59 °C, 3' 72 °C). The PCR fragments were digested with AflII and XbaI and ligated to the XbaI-AflII vector moiety of plasmid pFDX3736.



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 2.
Strategy for selection of mutant bglG genes encoding antiterminators that escape inhibition by EIIBgl. The DNA sequence encoding the PRDs in BglG was amplified from plasmid pFDX3736 by mutagenic PCRs with primers 619 and 620. The mutagenized products were used to substitute the corresponding wild-type fragment of pFDX3736. Recombinants were selected in strain R2243 on MacConkey indicator plates supplemented with lactose and tetracycline. Strain R2243 carries a {lambda}PL-bglt2-lacZ antitermination reporter cassette in the chromosome (27) and allows visual discrimination of active and inactive BglG species by monitoring the ability to utilize lactose as carbon source.

 
Recombinants were selected on MacConkey lactose plates containing tetracyclin (12.5 µg/ml). Plasmid DNAs of 100 Lac-positive colonies were isolated and subjected to restriction enzyme analysis. Plasmids with restriction patterns indistinguishable from that of the original plasmid pFDX3736 were reintroduced into strain R2243. All transformants retained the Lac+ phenotype. The transport function of EIIBgl encoded on the mutagenized plasmids was tested on MacConkey plates supplemented with arbutin and IPTG as inducer for expression of the bglG-bglF cassette. Arbutin is taken up by EIIBgl and is subsequently hydrolized by phospho-{beta}-glucosidase A which is encoded by the constitutively expressed bglA gene that is not part of the bgl operon/regulon (30, 31). All but two transformants produced Arb-positive phenotypes indistinguishable from transformants with the original plasmid pFDX3736. Sequencing of the plasmids isolated from the two Arb-negative colonies revealed in both cases a spontaneous frameshift mutation within the 5'-part of the bglF gene.

Determination of {beta}-Galactosidase Activities—Assays were according to Miller (28) as described (32). Enzyme activities are expressed in Miller units.

In Vivo Protein Phosphorylation—Cells were labeled with HIn Vivo Protein Phosphorylation[32P]O4 as described (18). For induction of tacOP-controlled gene expression, 0.1 mM IPTG was added. Proteins were separated on 13.5% SDS-polyacrylamide gels. The experiments were carried out at least twice, yielding reproducible results. It should be noted that due to a change in laboratory during this study different electrophoresis systems were used. In Fig. 5 proteins were separated by a workshop-made electrophoresis system whereas in the other cases a Hoefer system was applied, explaining apparent differences in separation.



View larger version (61K):
[in this window]
[in a new window]
  		FIG. 5.
EIIBgl-dependent phosphorylation of GalK{phi}BglG and its mutant derivatives carrying double exchanges of the conserved histidines. Strain R1279 was transformed with plasmids carrying galK{phi}bglG (lane 2; plasmid pFDX3226) or various alleles thereof carrying double exchanges (plasmids pFDX3268-3279) as indicated on top. The galK{phi}bglG alleles were expressed together with bglF as an operon under tacOP control. Lac repressor was delivered from plasmid pFDY226. Cells were labeled and analyzed as in Fig. 4. Lane 1, untransformed strain.

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
The PRD-containing Moiety of BglG Is Essential for All Signal Reception from the PTS—The first step was a verification that the putative PRDs in BglG are the domains that relay the signals from the PTS to the RNA-binding domain by their dimerization. A chimeric BglG derivative was constructed in which the PRDs were replaced by the dimerization domain of an unrelated protein. The first 51 codons of bglG encoding the RNA-binding domain (CAT) were fused in-frame to codons 96–237 of the phage {lambda} cI repressor gene encoding a flexible linker and its dimerization domain. To investigate whether the fusion protein is still subject to negative regulation, its gene was placed on a plasmid downstream of the IPTG-inducible tacOP either alone or together with the bglF gene (encoding EIIBgl) in tandem. For comparison, plasmids were analyzed that carried wild-type bglG alone or together with bglF or its mutant with a C24S exchange. Cys-24 represents the second phosphorylation site of EIIBgl and is essential for negative regulation of BglG (13). To decide about the role of the PRDs in dimerization, a plasmid, which encoded only the CAT domain was additionally employed. The various expression plasmids were introduced into strain R1279 ({Delta}bgl, {Delta}lac) as well as its derivative R1653 lacking the ptsHIcrr operon (which encodes HPr, EI and IIAGlc, respectively), in order to determine whether activity of the CAT{Phi}{lambda}CI fusion protein is still modulated by phospho-HPr. Antitermination activity was tested using reporter plasmid pFDY226 (Fig. 1), which allows quantitative monitoring of BglG activity (18). The {beta}-galactosidase enzyme activities were determined in the presence of different concentrations of IPTG (0, 0.05, 0.1, and 1 mM) to compare the antitermination activities at different expression levels (Fig. 1). This was important, because the intracellular concentration of BglG is crucial for its undisturbed regulation by the PTS (18). As expected, expression of the bglG-bglF cassette in the pts-positive strain led to only low enzyme activities (Fig. 1A, bar 1) close to the background level of this system (= 37 units, the activity detected with plasmid pFDY226 alone; Ref. 18). In contrast, activities up to ~1600 units at maximum expression levels (1 mM IPTG) were obtained in the absence of bglF or when the bglF-C24S allele was present (Fig. 1A, bars 2 and 3). These results reflect negative control of BglG by wild-type EIIBgl as well as its escape from this control in the presence of mutant EIIBgl-C24S (13, 18). Expression of the bglG as well as the bglG-bglF-C24S cassette in the pts-negative strain led to drastically reduced activities when compared with the pts-positive strain (bars 2 and 3) due to the second level of activity control of BglG, which requires phosphorylation by HPr to be active (18). However, as has been shown previously (18), in the pts-negative background a considerable level of BglG activity is detectable at higher expression levels (induction with 1 mM IPTG), suggesting that the unphosphorylated BglG forms active dimers at high protein concentrations. A similar result has also been obtained for LicT (33).



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 1.
The PRDs are indispensable for dual regulation of BglG as well as for dimerization. Strains R1279 (A, pts+) and R1653 (B, {Delta}pts) were co-transformed with antitermination reporter plasmid pFDY226 and one of the following expression plasmids, which carried the genes as indicated downstream of tacOP, respectively. Bar 1, pFDX3736; bar 2, pFDX3453; bar 3, pFDX4219; bar 4, pFDX3562; bar 5, pFDX3518; bar 6, pFDX4217. Reporter plasmid pFDY226 carries the lacZ reporter gene preceded by terminator t2 of the bgl operon, which in the absence of active BglG blocks transcription initiated at constitutive promoter P16 (32). In addition the lacIq gene is present on this plasmid, delivering Lac repressor for IPTG-inducible expression of the tacOP-controlled genes. {beta}-Galactosidase activities were determined in the presence of different concentrations of IPTG as indicated.

 
Expression of the CAT domain (BglG1–51) led to only background activities in both strains (bar 4), suggesting that the PRDs are indispensable for dimerization. However, upon a further increase of BglG1-51 expression using a more efficient ribosomal binding site, antitermination activity became again detectable (data not shown), suggesting that the CAT domain alone already has a weak potential to dimerize. Similar observations have been made for the CAT domains of SacY and LicT (9, 34). Antitermination activity was completely restored when the dimerization domain of {lambda}CI was fused to the CAT domain (Fig. 1, bar 5). Most importantly, activity of this fusion was constitutive and neither affected by the additional co-expression of bglF (bar 6) nor by the absence of the general pts genes (Fig. 1, bars 5 and 6) and is thus not subject to any PTS control. This demonstrates that the PRDs in BglG mediate both negative as well as positive control of activity by the PTS and that their dimerization is a prerequisite for antitermination.

Amino Acid Exchanges in PRD1 and Truncations of PRD2 Account for a Release of Negative Control of BglG—To get an insight into the differential roles of the two PRDs in BglG activity control and to possibly identify the histidine(s) phosphorylated by EIIBgl and other residues involved in negative control, the DNA encompassing the two PRDs were mutated and screened for mutant BglG variants that escape repression by EIIBgl. The experimental strategy is outlined in Fig. 2. For identification of the mutants, strain R2243 was constructed, which carries a chromosomal {lambda}PL-bglt2-lacZ antitermination reporter cassette and exhibits a Lac-positive phenotype exclusively in the presence of active BglG. It produces red colonies (Lac+) on MacConkey lactose indicator plates when transformed with plasmids carrying the bglG or the bglG-bglF-C24S cassette, respectively (Fig. 1), while colorless colonies (Lac–) are obtained with the untransformed strain or when plasmid pFDX3736 carrying the bglG-bglF cassette was present. The strategy for mutagenesis (Fig. 2) was to amplify by error-prone PCR a 636-bp fragment of bglG that encompasses both PRDs (codons 67–278) and to subsequently replace the corresponding wild-type fragment in plasmid pFDX3736 (carrying bglG, bglF). Recombinants were selected on MacConkey-lactose plates on which Lac-positive mutants appeared with a frequency of about 1%. Plasmid DNAs of such mutants were isolated and control experiments were carried out to verify that the Lac+ phenotype was conferred by the respective plasmid rather than by e.g. a spontaneous mutation in the chromosomal reporter cassette. A spontaneous mutation in bglF that results in activation of BglG, was ruled out by testing the transport function of EIIBgl (see "Experimental Procedures"). Finally, the mutated regions of 38 plasmids were sequenced.

The results of sequencing of the bglG mutants are summarized in Table II. 28 plasmids carried a single mutation in bglG, while 10 plasmids had 2 or more. Of the single mutations, 18 resulted in 10 amino acid exchanges in altogether 8 different positions (Table IIA). These were residues Cys-76, Gln-91, Ser-97, Asp-100, His-101, Phe-104, Glu-153, and His-160. For Asp-100 three different exchanges were found. The other 10 single mutations were frameshift or stop mutations leading to truncated BglG variants either with or without an additional short tail of heterologous codons (Table IIB). Of the 10 different amino acid exchanges 7 were found at least twice, of which 5 were derived from independent PCRs, supporting the importance of these exchanges for negative control by EIIBgl. Of the 10 bglG alleles carrying more than one mutation, 5 had a mutation that was also detected as a single mutation. Several of the double or multiple mutations were individually subcloned and the phenotypes of the single mutants were analyzed (Table IIC). This led to the identification of I80T as an additional exchange causing a Lac+ phenotype. As expected, the S97P and the L200-> amber mutations that were also isolated as single mutations, turned out to be responsible for the Lac+ phenotype of the respective multiple mutants. These experiments also revealed that the exchanges L70P, S97L, Q110H, N112D, P116S, and H219R conferred no Lac+ phenotype. The V147G exchange only conferred a rather weak Lac+ phenotype and was not further analyzed. Fig. 3 summarizes the data and depicts the distribution of all finally identified mutations. Intriguingly, all amino acid exchanges were located within PRD1 and none in PRD2. More strikingly, exchanges of the highly conserved histidines His-101 and His-160, candidates for the phosphorylation sites in BglG, were also found. The distribution of the stop and frameshift mutations also displayed an intriguing pattern. They all terminate bglG translation within a stretch of 58 codons encoding the N-terminal half of PRD2.


View this table:
[in this window]
[in a new window]
  		TABLE II
Mutations identified in the screen for BglG variants not subject to negative control by EIIBgl

 



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 3.
Location of amino acid exchanges and protein chain terminations in BglG that cause loss of repression by EIIBgl. The RNA-binding domain CAT and the two PRDs are represented as gray-shaded boxes. The positions of the PRDs and their predicted {alpha}-helices (depicted as black bars) are plotted according to Ref. 36. The three conserved putative histidine phosphorylation sites His-101, His-160, and His-208 are represented as black circles and the leucines at positions 179, 186, 193, 200, previously reported to be required for dimerization (38), are symbolized by Ls. The arrows indicate the positions of the amino acid exchanges (above) and of the stop and frameshift mutations (below), identified in the screen for derepressed BglG derivatives.

 
Activities and Regulatory Properties of the Derepressed BglG Proteins—The Lac+ phenotypes caused by the various mutant BglG proteins showed gradual differences that suggested corresponding differences in antitermination activities. In order to obtain a quantitative estimate of these differences, strain R1279 ({Delta}bgl {Delta}lac) was cotransformed with reporter plasmid pFDY226 (Fig. 1) and the various mutant plasmids. From these transformants, {beta}-galactosidase activities were determined in the absence (data not shown) and in the presence of 10–4 and 10–3 M IPTG as inducer for BglG and EIIBgl synthesis. As expected, expression of all mutant bglG genes led to significantly elevated enzyme activities as compared with expression of the wild-type bglG gene (Table III, column I). However, exclusively the mutants with exchanges of Asp-100 and His-101 showed the high level of activity seen with the wild-type BglG protein in the absence of a functional EIIBgl (Table III; compare values in column I with those of column II, line one). This result suggested that only these mutants completely escape negative control by EIIBgl, while all other mutants were either not fully released from negative control or were reduced in their potential to dimerize and thus less active. To discriminate between these possibilities, the various mutants were studied in the presence of the bglF-C24S allele, where negative control is not operative. In this case, the presence of bglG alleles with exchanges of I80T, Q91R, S97P, F104V, and H160R all led to further increases in activity to about the level seen with the wild-type bglG gene (Table III; compare columns I and II). It can thus be concluded that these mutants only partially escaped negative control by EIIBgl. In contrast, expression of the C76R and the E153V allele as well as all 3'-truncated derivatives led to unaltered activities. These mutants were thus simultaneously affected in negative control as well as in antitermination activity. The latter could either be the result of a direct interference with dimerization or less efficient activation by HPr.


View this table:
[in this window]
[in a new window]
  		TABLE III
Regulation of BglG and its various mutant derivatives by EIIBgl and HPr

 
The next question to resolve was to what extent activity of the different mutant BglG proteins still required a functional PTS, i.e. are still under HPr-exerted positive control. The experiments were therefore repeated in the isogenic {Delta}pts strain (Table III, columns III and IV). Since EIIBgl cannot be phosphorylated in this strain, antitermination activities were as expected roughly constitutive and unaffected by the presence of either wild-type EIIBgl or the non-functional EIIBgl-C24S protein (compare columns III and IV). The activities of the BglG proteins with exchanges C76R, I80T, S97P, H101L, F104V, E153V, and H160R, respectively, were significantly lower in the pts-negative strain and therefore still dependent on a functional PTS (Table III, compare columns III, IV, and II). In contrast, all Asp-100 and the Q91R mutant displayed high activities, suggesting that dimerization of these mutants is independent of activation by HPr. Similarly, activities of all truncated BglG derivatives were not reduced in the pts deletion strain. These proteins behaved like the BglG-(1–51)-{lambda} CI fusion protein (Fig. 1). They showed constitutive activity in all genetic backgrounds and became fully independent of both mechanisms of activity control of BglG.

His-101 in PRD1 Is the Major Site Involved in Negative Regulation Whereas His-208 in PRD2 Is Required for Activation of BglG—The results obtained above suggest that His101 represents the residue that is phosphorylated by EIIBgl. The second conserved histidine in PRD1, His-160, is likely to be mechanistically involved but not essential for negative regulation. Activity of the H160R mutant protein was still 4–7-fold repressed by EIIBgl, whereas the H101L mutant was highly active and completely unaffected by EIIBgl (Table III). To obtain further evidence for the importance of His101, a mutant bglG gene with a H101D exchange was constructed. An aspartate residue was found to mimic a phosphorylated site within HPr (35) and also in antiterminators LicT and LacT (20, 25). In addition, the possible role of His208, which represents the conserved histidine in PRD2, was addressed. For the latter, bglG alleles with exchanges of His-208 replaced by arginine, alanine, or aspartate were constructed. The H101D mutant protein was, like the H101L protein, not subject to repression by EIIBgl (compare columns I and II in Table IV). However, its activity was significantly lower than that of the H101L protein, particularly at lower expression levels. Depending on the concentration of IPTG present (10–3, 10–4, or 10–5 M), it had 43, 20, or only 15% of the wild-type activity (Table IV, column 2; data not shown). Thus replacement of His-101 by the negatively charged aspartate inhibits BglG activity, but this effect can partially be overcome by an increase in concentration of BglG.


View this table:
[in this window]
[in a new window]
  		TABLE IV
Effects of mutations of the conserved histidines His-101 and His-208 on activity and regulation of BglG by EIIBgl and HPr

 
All three His-208 mutant proteins displayed very low activities in the pts-positive strain, irrespective of the presence or absence of a functional bglF gene (Table IV, columns I and II). Interestingly activities increased slightly in the pts negative background. This effect was most pronounced with mutant protein H208D (Table IV, columns III and IV), indicating that a negative charge at the His-208 position is capable to partly restore the activity. The fact that this effect exclusively occurs in the pts-negative background suggests that one of the general components of the PTS may inhibit this mutant protein.

PRD1 Is Necessary and a Target for Phosphorylation by EIIBgl as Well as by HPr—The genetic data suggest that BglG may be activated by HPr catalyzed phosphorylation taking place at His-208 within PRD2, whereas it may be inactivated by EIIBgl-catalyzed phosphorylation at His-101 within PRD1. Provided that these reactions occurred independently of each other, one would expect that any exchange of His-208 should abolish phosphorylation of BglG by HPr, but not by EIIBgl, whereas the opposite should be true for BglG proteins with exchanges of His-101. The other exchanges identified in the mutagenesis screen, which all map within PRD1, should also lead to a reduction in phosphorylation by EIIBgl, but not by HPr, at least in those cases where the mutant proteins were still fully dependent on a functional PTS (Table III). To investigate these hypotheses, phosphorylation of BglG and its various mutant derivatives by HPr and EIIBgl in vivo was studied.

Since phosphorylated BglG co-migrates on SDS gels with a weakly phosphorylated, diffuse band of unknown origin (Ref. 18 and see also Fig. 6), impeding the analysis of weakly phosphorylated BglG species, I made use of a GalK-BglG fusion protein that shifts the position of the labeled protein by increasing the molecular weight. We have previously shown that this fusion protein becomes correctly phosphorylated by HPr as well as by EIIBgl (18). Plasmids carrying the various galK{phi}bglG alleles downstream of tacOP were introduced into the pts-positive strain. A second plasmid encoding Lac repressor with or without gene bglF under tacOP control, was additionally present. The cells were labeled with H3[32P]O4, and proteins were separated by SDS-PAGE and analyzed by autoradiography. Analysis of whole protein extracts by SDS-PAGE and coomassie staining confirmed that all fusion derivatives were properly expressed and protein amounts were roughly identical (data not shown). In the presence of the bglF expression plasmid, a strongly phosphorylated, extensive band at the position expected for EIIBgl (Mr 66.4 kDa, Fig. 4, lane 3) appeared, which was absent in the presence of the isogenic construct lacking bglF (lane 2) or in the untransformed strain (lane 1). As shown previously (18), the wild-type GalK-BglG fusion protein became moderately phosphorylated by HPr in the absence of bglF (Fig. 4, lane 4), and phosphorylation was stronger when EIIBgl in addition contributed to its phosphorylation (lane 5). The results with the mutant BglG proteins from the mutagenesis screen are shown in lanes 6–27 (Fig. 4). Phosphorylation of all of these proteins was reduced or even abolished. Surprisingly, this was not only the case in the presence of EIIBgl, but also in its absence (Fig. 4; compare lanes 6–27 –/+bglF, respectively, with lanes 4 and 5). No phosphorylation by HPr could be detected for the proteins with exchanges C76R, S97P, H101L, E153V, and H160R (lanes 6, 12, 20, 24, 26), and signals were extremely reduced with the mutants carrying exchanges of I80 and D100 (lanes 8, 14, 16, 18). Surprisingly, when cells expressed the Q91R protein, an additional phosphorylated band appeared approximately 3 kDa above the phosphorylated GalK-BglG fusion protein (lanes 10 and 11). The source of this species has still to be elucidated.



View larger version (112K):
[in this window]
[in a new window]
  		FIG. 6.
Presence of wild-type BglG restores HPr-dependent phosphorylation of GalK{phi}BglG-proteins carrying amino acid exchanges in PRD1. Strain R1279 was transformed with plasmid pFDY226 as source for Lac repressor and with one of different expression plasmids carrying bglG or galK{phi}bglG-alleles downstream of tacOP. In lanes 4, 6, 8, 10, the galK{phi}bglG alleles were expressed together with wild-type bglG as an operon. Cultures were labeled and analyzed as in Fig. 4. Lane 1, untransformed strain; lane 2, pFDX2942 (bglG); lane 3, pFD3225 (galk{phi}bglG); lane 4, pBGO36 (galK{phi}bglG + bglG); lane 5, pBGO17 (galk{phi}bglG-D100V); lane 6, pBGO37 (galK{phi}bglG-D100V + bglG); lane 7, pBGO18 (galk{phi}bglG-H101L); lane 8, pBGO38 (galK{phi}bglG-H101L + bglG); lane 9, pBGO21 (galk{phi}bglG-H160R); lane 10, pBGO39 (galK{phi}bglG-H160R + bglG).

 



View larger version (63K):
[in this window]
[in a new window]
  		FIG. 4.
Phosphorylation of GalK{phi}BglG and its various mutant derivatives by HPr and EIIBgl in vivo. Transformants of strain R1279 were grown in the presence of IPTG (0.1 mM) as inducer for tacOP driven expression of galK{phi}bglG and its various mutant alleles encoded on plasmids (plasmid pFDX3255 and its derivatives, listed in Table I and Table V in Supplemental Materials). Lac repressor was delivered from plasmid pFDY226 (lanes with even numbers) or plasmid pFDX3283, which additionally carries bglF under tacOP control (lanes with odd numbers starting with lane 3). As a control, data with the untransformed strain are presented in lane 1. Cultures were labeled with H3[32P]O4, proteins separated by SDS-PAGE, and gels analyzed by autoradiography.

 
In the presence of EIIBgl, the H101L-exchange was the only mutation which completely abolished any detectable phosphorylation (lane 21). In all other cases, phosphorylation remained detectable, but the signals were weaker as compared with the wild-type (Fig. 4; compare lanes 7, 9, 11, 13, 15, 17, 19, 23, 25, 27 with lane 5). Normalization of these data by subtraction of the corresponding signals in the absence of EIIBgl (Fig. 4; lanes with even numbers) confirmed that all these mutant proteins were indeed less efficiently phosphorylated by EIIBgl (data not shown). These data correlate well with the results of the genetic analysis of the mutants shown above (Table III). Those proteins that were not (mutant H101L; lane 21) or only weakly (mutants C76R, D100(G,N,V), E153V; lanes 7, 15, 17, 19, 25) phosphorylated by EIIBgl, were also completely independent of negative control, whereas those proteins that were more efficiently phosphorylated (mutants I80T, Q91R, S97P, F104V, and H160R; lanes 9, 11, 13, 23, 27), were still subject to residual repression by EIIBgl (Table III). This correlation suggests that the level of escape from repression seen for the individual mutant proteins may indeed be caused by a corresponding less efficient EIIBgl-catalyzed phosphorylation.

Inspection of the data also revealed that the individual mutations affected the two different phosphorylation events to different extents. For example, the three D100 mutant proteins were weakly phosphorylated in the absence as well as in the presence of EIIBgl (lanes 14–19). In contrast, phosphorylation of mutant proteins C76R, S97P, and H160R was undetectable in the absence of EIIBgl (lanes 6, 12, 26), whereas in its presence, phosphorylation of all these mutant proteins was more efficient than that of the D100 mutant proteins (Fig. 4, compare lanes 7, 13, 27 with lanes 15, 17, 19). These observations support the hypothesis that two mechanistically different phosphorylation events take place on BglG. However, these results also suggest that identical residues within PRD1 are involved in both phosphorylations.

The data obtained with the BglG proteins carrying different exchanges of His-208 are shown in Fig. 4, lanes 28–33. Phosphorylation of all three proteins was clearly detectable in the presence of EIIBgl, but the signals were reduced as compared with the wild-type protein (compare lanes 29, 31, and 33 with lane 5). The protein with the H208R exchange displayed the weakest signal (lane 29). Surprisingly, phosphorylation of the three mutant proteins did not only occur in the presence of EIIBgl, but very weakly also in its absence (lanes 28, 30, 32). This result demonstrates that HPr is able to phosphorylate BglG at a position different from His-208, most likely at a histidine within PRD1. The drastic reduction in the phosphorylation suggests, however, that His-208 is involved in the phosphorylations taking place in BglG.

EIIBgl Phosphorylates Histidine 101 in BglG—To obtain direct evidence that His-101 in PRD1 becomes phosphorylated by EIIBgl, double mutants simultaneously carrying two exchanges of the three conserved histidines were studied. This analysis was performed with all three possible combinations that leave only one of the conserved histidines as the potential target for phosphorylation. To explore the potential role of negative and positive charges, each histidine was replaced by arginine or glutamate, respectively, and all four combinations were tested with each of the three double mutants. In vivo phosphorylation of the various GalK-BglG fusion proteins carrying the double exchanges was analyzed in the absence and the additional presence of EIIBgl. None of these proteins became phosphorylated in the absence of EIIBgl (data not shown), supporting the observation (Fig. 4) that residues His101 as well as His-160 are both essential for the phosphorylation reaction directly catalyzed by HPr. However, in the presence of EIIBgl (Fig. 5), weak phosphorylation signals were present in the case of the His-160/His-208 double mutant proteins (Fig. 5, lanes 7–10) but not in that of any of the other tested combinations (Fig. 5, lanes 3–6 and lanes 11–14). Phosphorylation was most readily detectable with the H160R-H208E double mutant protein (lane 8), suggesting that a negative charge at residue 208 and simultaneously a positive charge at residue 160 enhances the phosphorylation, whereas positive charges at both positions (lane 7) inhibit the phosphorylation by EIIBgl, at His-101.

Evidence for a Transfer of the HPr-donated Phosphoryl Groups within the BglG Dimer—The data at first appeared enigmatic for the mechanism of activation of BglG by HPr-catalyzed phosphorylation. The genetic data suggested that His-208 in PRD2 is the target of this phosphorylation. However, the in vivo phosphorylation experiments indicate that (i) HPr phosphorylates a histidine in PRD1 and (ii) that, besides other residues within PRD1, both conserved histidines are essential for this phosphorylation to take place. This contradiction could be resolved, however, if one assumes that the entry point of the phosphoryl groups donated by HPr to BglG was different from the finally phosphorylated position, i.e. when HPr first donated its phosphoryl group to PRD1 from where it was subsequently transferred to His-208 in PRD2. Such an intramolecular phosphoryl group transfer could principally occur within a monomer or between the monomers of a BglG dimer.

In order to test the latter idea, phosphorylation of mutant BglG proteins that were not (exchanges H101L and H160R) or only very weakly (exchange D100V) phosphorylated by HPr (Fig. 4) were studied in the additional presence of wild-type BglG protein to allow the formation of heterodimers. To discriminate between phosphorylated mutant and wild-type protein species, the mutant bglG alleles were fused to the 5'-part of galK, whereas the wild-type gene had no tag. The galK-bglG fusion genes carrying the various exchanges were placed on plasmids together with the wild-type bglG gene as an operon under tacOP control for their simultaneous expression. As reference, an isogenic plasmid was employed that carries the wild-type galK-bglG fusion gene. In vivo phosphorylation experiments were carried out with transformants harboring these various expression plasmids together with a second plasmid as source for Lac repressor synthesis. For direct comparison, transformants expressing solely one of the various bglG alleles were also labeled. Expression of the wild-type bglG or the wild-type galK-bglG fusion gene, respectively, led to detection of the expected phosphorylated protein bands as compared with the empty strain (Fig. 6, compare lanes 2 and 3 with lane 1). When both genes were expressed together, simultaneous phosphorylation of both BglG species could be detected (Fig. 6, lane 4). As already demonstrated in Fig. 4, presence of the different mutations extremely reduced (exchange D100V; Fig. 6, lane 5) or completely abolished phosphorylation of the GalK-BglG fusion protein by HPr (exchanges H101L and H160R; lanes 7 and 9). However, in the additional presence of wild-type BglG protein, phosphorylation of the D100V mutant protein was clearly enhanced (compare lanes 6 and 5). Most importantly, phosphorylation of the otherwise unphosphorylatable mutant proteins H101L and H160R now became detectable (compare lane 8 with 7 and lane 10 with 9 in Fig. 6).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Data presented in this report demonstrate that the activity of the dually controlled antiterminator protein BglG is controlled by its phosphorylation at conserved histidine residues His-101 and His-208 located within PRD1 and PRD2, respectively. Negative regulation by EIIBgl is mediated by phosphorylation of BglG at His-101. Positive regulation by phosphorylation of His-208 within PRD2 is likely to occur indirectly. The data suggest that in this case a histidine in PRD1 is transiently phosphorylated by HPr, and that this phosphoryl group is subsequently transferred in a trans intramolecular reaction to the His-208 residue of the other monomer. These results and their interpretations are in agreement with and extent data on other antiterminator proteins of the BglG/SacY family (8). Thus, in this report a unifying view on the mechanism of regulation of the BglG/SacY family of conserved proteins can be presented.

Interaction Surface in BglG—Among the residues identified to participate in negative regulation, the two histidines within PRD1, His-101 and His-160, as well as Asp-100 and Glu-153 are fully conserved in all antiterminators of the BglG/SacY family. Residues Ile-80 and Phe-104 are also conserved, and for Gln-91 and Ser-97 equivalent amino acids are often found (10, 36). Based on the known structure of an active form of antiterminator LicT from B. subtilis and on sequence comparison, each PRD is predicted to consist of a bundle of five {alpha}-helices (36). With the exception of Gln-91, all other residues reside in putative {alpha}-helices in PRD1; Cys-76 and Ile-80 reside in helix {alpha}1, Ser-97, Asp-100, His-101, and Phe-104 in helix {alpha}2, and Glu-153 and His-160 in helix {alpha}5 (Fig. 3). The distance of the residues places them on the same side of the respective {alpha}-helix. In the structure of LicT, the two histidines in PRD1 are found close together, at 3 Å distance, and the glutamate corresponding to the here identified Glu-153 interacts with His-100 (His-101 in BglG). This putative active center is surrounded by a ring of 7 residues of which 3 correspond to the here identified Ser-97, Asp-100, and Phe-104. The hydrophobic half of this ring was proposed to provide the interaction surface for the phosphorylating protein (36). These coincidences suggest a very similar mechanism underlying the phosphoryl transfer reaction and the interaction of BglG and LicT with their negative regulators, respectively. However, the residues corresponding to Cys-76, Ile-80, and Gln-91 are not part of this surface and the respective residues of LicT are exposed to the environment. Perhaps they constitute a further region in BglG involved in interaction with EIIBgl and its sequestration to the membrane, which has recently been observed (27, 37).

A second class of mutants obtained in the mutagenesis screen comprises truncated BglG proteins completely or partly lacking PRD2 (Fig. 3). All these mutants exhibited constitutive activities, which were not subject to any regulation by the PTS (Table III). Thus the presence of PRD2 may not only be required for positive but also for negative regulation of BglG activity.

Role of the Conserved Histidines in BglG Regulation by Phosphorylation and Comparison to LicT—The genetic effects of the different exchanges in the three conserved histidines (Tables III and IV) were strikingly similar to results from a recent mutational analysis of LicT (20), which is BglGs closest relative in Gram-positive bacteria (10). In both cases, substitution of the first conserved histidine in PRD1 (His-101 in BglG) by an uncharged residue leads to full derepression, but does not affect dependence on HPr. Substitution by a negatively charged Asp residue, able to mimic the negative phosphorylation in LicT, also led to a less active BglG variant (Table IV). Mutation of the second histidine in PRD1 (His-160 in BglG) led to only partial derepression of LicT as well as BglG. Accordingly, EIIBgl-dependent phosphorylation of the BglG-H160 mutant protein was reduced but not abolished. Like its counterpart LicT, BglG became inactive when the conserved histidine in PRD2 (His-208 in BglG) was replaced by alanine. However, the replacement of His-208 by a negatively charged aspartate, which was reported to restore activity of LicT, had only a marginal effect on BglG as compared with the H208A(R) exchanges (Table IV). Activity of this mutant protein increased, however, in the pts-negative background, which can perhaps be explained by the fact that it is still weakly phosphorylated by HPr, which should occur at PRD1 and should therefore lead to its inactivation (see below). Conclusive evidence for H101 as the direct target for the inactivating phosphorylation catalyzed by EIIBgl can be drawn from analyses of mutant BglG proteins simultaneously carrying exchanges of two of the three histidines. Exclusively the His-160/His-208 double mutants were phosphorylated by EIIBgl but none of the other combinations, His-101/His-160 or His-101/His-208 (Fig. 5). In conclusion it appears that the histidines in BglG and LicT, respectively, serve quite similar functions, i.e. that the first histidine in PRD1 becomes phosphorylated by the respective EII and that the second histidine contributes to this reaction, whereas the activating phosphorylation by HPr should occur at the histidine(s) in PRD2.

Both PRDs Contribute to Dimerization of BglG—The truncated BglG proteins, completely or partly lacking PRD2, exhibited reduced but clearly detectable activities, whereas the CAT domain alone was inactive under the same conditions (Fig. 1). Truncations within PRD1 were not obtained in the screen, suggesting that integrity of this domain is essential for the residual potential to dimerize. In conclusion both PRDs may contribute to dimerization of BglG. In agreement, a crystal structure of a constitutively active LicT variant suggested that also in LicT both PRDs are involved in dimerization by making numerous protein contacts at the dimer interphase (36). The data presented here do not support a role of a leucine zipper motif (Leu-179, Leu-186, Leu-193, Leu-200; Fig. 3) present in the N-terminal-half of PRD2 of BglG which was reported to be absolutely required for dimerization (38). The respective variants containing or lacking these leucines did not significantly differ in activity (Table III). Interestingly, a leucine zipper is not apparent in any of the other PRD-containing proteins.

Comparison of the Results from the Present Study to Conclusions Drawn in a Previous Work—While the results presented here are in accordance with studies performed on other members of the BglG/SacY family of antiterminators (20, 25) and are particularly in agreement with results from in-depth studies performed on LicT (see above), they are in contradiction to conclusions previously drawn on BglG itself. In a different experimental setup, the H208R mutant protein shown here to be inactive was found to be active and to escape negative regulation (26). Moreover, in contrast to the in vivo studies presented here, no phosphorylation of this mutant protein could be detected in an in vitro system containing EI, HPr, and EIIBgl. The authors thus concluded that residue His-208 in BglG is the target of inhibitory phosphorylation. The different results concerning the phosphorylation of this mutant protein could be explained by differences in sensitivity of the assays. The discrepancies in the genetic data may be due to the different reporter systems used to determine antitermination activity. Chen et al. (26) used a transcriptional bgl'-lacZ fusion located downstream of bgl terminator t1 as the reporter gene. This fusion encompasses the first 199 codons of bglG (39), which are consequently co-expressed with the lacZ gene. In the mutagenesis screen performed here, a mutant bglG gene was isolated that carried a stop codon exactly behind codon 199. The resulting truncated protein showed considerable activity and was not subject to negative control any more (Table III). It may be relevant in this context that small changes in initial concentrations of active BglG species encoded downstream of bglt1 can result in high bglG expression levels due to auto-amplification of bglG expression when a certain threshold limit is exceeded (40). In conclusion, it appears possible that inactivity of the H208R mutant was masked by antitermination activity of the truncated BglG protein in the above study. In addition, heterodimers with yet unpredictable properties could have formed in this case.

Regulation of BglG by Multiple Phosphorylations—Genetic analyses had suggested that activation of BglG by HPr-catalyzed phosphorylation should occur at histidine 208 within PRD2. However, the phosphorylation experiments (Fig. 4) indicated that HPr phosphorylates a residue within PRD1. The apparent contradiction could be resolved by an experiment that demonstrated that HPr-mediated phosphorylation of the Asp-100 His-101, and His-160 mutant proteins is restored in the presence of wild-type BglG protein (Fig. 6), suggesting a phosphoryl transfer from wild-type protein molecules to histidine 208 of mutant molecules. The phosphorylation signals were, however, not restored to wild-type levels. This can at least in part be expected, because upon co-expression of wild-type and mutant proteins 50% heterodimers will theoretically form. Altogether the results of this study let me propose a mechanism of BglG regulation as it is depicted in Fig. 7. In a first step, HPr may transiently phosphorylate PRD1 (Fig. 7A). Next, the phosphoryl group is transferred to H208 of a second monomer in a trans intramolecular reaction. This reaction could lead to stabilization of the preformed dimer (Fig. 7A). When {beta}-glucoside transport slows down or stops, BglG becomes additionally phosphorylated at His-101, which occurs in an EIIBgl-dependent reaction and leads to inactivation of BglG (Fig. 7B).



View larger version (20K):
[in this window]
[in a new window]
  		FIG. 7.
Model for the regulation of BglG by PTS-catalyzed phosphorylations. A, under inducing conditions, i.e. in the presence of {beta}-glucosides but absence of other PTS substrates, HPr transiently phosphorylates BglG at PRD1. The presence of both conserved histidines and additional residues in PRD1 is essential for this reaction to take place. It is unknown whether this phosphorylation occurs in the monomer or requires a preformed dimer of the unphosphorylated protein. Next, the phosphoryl group is transferred to the His-208 residue in the PRD2 of a second BglG molecule, which may stabilize the dimer, thereby leading to activation of BglG. B, when {beta}-glucoside transport slows down, BglG becomes additionally phosphorylated. This second phosphorylation requires the presence of phospho-EIIBgl and takes place at His-101 within PRD1. The multiple phosphorylation shifts the monomer-dimer equilibrium toward the monomeric, inactive form.

 
The model (Fig. 7) does not explain why activities of the BglG proteins with exchanges of His-101 and His-160, respectively, were still fully dependent on a functional PTS, although they could not be phosphorylated by HPr. However, it could be envisioned that phospho-HPr still interacts with these proteins and induces the conformational changes necessary for dimerization. It is notable in this context that the corresponding exchanges in LacT render its activity thermosensitive (25).

In case of the dually controlled antiterminators LicT and SacT it was assumed that they are activated by HPr-catalyzed phosphorylation at their conserved histidines in PRD2 (6, 8), but whether this reaction, which has been demonstrated for LicT in vitro (19), is direct or indirect is unknown. Interestingly, a phosphoryl transfer between monomers involving the PRDs has also been proposed for MtlR, a DNA-binding PRD containing activator in Bacillus stearothermophilus (23).

Altogether, the data suggest that the PRDs in BglG represent a functional unit rather than independent modules in negative and positive regulation. Relevant in this context may be a recent in silico analysis, which indicates that the BglG/SacY family of antiterminators evolved from a single primordial antiterminator that had already both PRDs, and that the domains were not rearranged and shuffled during evolution (10).


   FOOTNOTES
 
* This work was supported by grants from the Landesschwerpunkt Baden-Württemberg and the Fonds der Chemischen Industrie to Bodo Rak. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

The on-line version of this article (available at http://www.jbc.org) contains Supplemental Data and Tables V and VI. Back

{ddagger} To whom correspondence should be addressed: Laboratoire de Chimie Bactérienne, CNRS, 31, chemin Joseph Aiguier, 13402 Marseille, France. Tel.: 33-491164571; Fax: 33-491718914; E-mail: bogoerke{at}ibsm.cnrs-mrs.fr.

1 The abbreviations used are: PTS, phosphoenolpyruvate (PEP) carbohydrate phosphotransferase system; EI, enzyme I of the PTS; HPr, histidine-containing protein of the PTS; EIIBgl, aryl-{beta}-glucoside-specific PTS transporter encoded by bglF; CAT, co-antiterminator; PRDs, PTS regulation domains; RAT, ribonucleic antiterminator; IPTG, isopropyl-{beta}-D-thiogalactopyranoside. Back


   ACKNOWLEDGMENTS
 
The author thanks Bodo Rak, who hosted this study and contributed with permanent support and discussion; Anne Galinier, LCB-CNRS Marseille, for the opportunity to finish the experiments in her laboratory; Anne Galinier, Bodo Rak, and Jörg Stülke for critical reading of the manuscript; and Elge Koalick for construction of plasmids pFDX3518 and pFDX3562 and for help with the {beta}-galactosidase assays. This study was carried out for the most part in the laboratory of Bodo Rak, Universität Freiburg.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Postma, P. W., Lengeler, J. W., and Jacobson, G. R. (1993) Microbiol. Rev. 57, 543–595[Abstract/Free Full Text]
  2. Lengeler, J. W., Bettenbrock, K., and Lux, R. (1994) in Phosphate in Microorganisms: Cellular and Molecular Biology (Torriani-Gorini, A., Yagil, E., and Silver, S., eds) pp. 182–188, Washington, D. C.
  3. Gunnewijk, M. G. W., van den Bogaard, P. T. C., Veenhoff, L. M., Heuberger, E. H., de Vos, W. M., Kleerebezem, M., Kuipers, O. P., and Poolman, B. (2001) J. Mol. Microbiol. Biotechnol. 3, 401–413[CrossRef][Medline] [Order article via Infotrieve]
  4. Plumbridge, J. (2002) Curr. Opin. Microbiol. 5, 187–193[CrossRef][Medline] [Order article via Infotrieve]
  5. Rutberg, B. (1997) Mol. Microbiol. 23, 413–421[CrossRef][Medline] [Order article via Infotrieve]
  6. Stülke, J., Arnaud, M., Rapoport, G., and Martin-Verstraete, I. (1998) Mol. Microbiol. 28, 865–874[CrossRef][Medline] [Order article via Infotrieve]
  7. Stülke, J. (2002) Arch. Microbiol. 177, 433–440[CrossRef][Medline] [Order article via Infotrieve]
  8. van Tilbeurgh, H., and Declerck, N. (2001) Curr. Opin. Struct. Biol. 11, 685–693[CrossRef][Medline] [Order article via Infotrieve]
  9. Manival, X., Yang, Y., Strub, M. P., Kochoyan, M., Steinmetz, M., and Aymerich, S. (1997) EMBO J. 16, 5019–5029[CrossRef][Medline] [Order article via Infotrieve]
  10. Greenberg, D. B., Stülke, J., and Saier, M. H., Jr. (2002) Res. Microbiol. 153, 519–526[Medline] [Order article via Infotrieve]
  11. Amster-Choder, O., Houman, F., and Wright, A. (1989) Cell 58, 847–855[CrossRef][Medline] [Order article via Infotrieve]
  12. Schnetz, K., and Rak, B. (1990) Proc. Natl. Acad. Sci. U. S. A. 87, 5074–5078[Abstract/Free Full Text]
  13. Chen, Q., Arents, J. C., Bader, R., Postma, P. W., and Amster-Choder, O. (1997) EMBO J. 16, 4617–4627[CrossRef][Medline] [Order article via Infotrieve]
  14. Schnetz, K., Toloczyki, C., and Rak, B. (1987) J. Bacteriol. 169, 2579–2590[Abstract/Free Full Text]
  15. Andersen, C., Rak, B., and Benz, R. (1999) Mol. Microbiol. 31, 499–510[CrossRef][Medline] [Order article via Infotrieve]
  16. Arnaud, M., Vary, P., Zagorec, M., Klier, A., Debarbouille, M., Postma, P. W., and Rapoport, G. (1992) J. Bacteriol. 174, 3161–3170[Abstract/Free Full Text]
  17. Gosalbes, M. J., Monedero, V., and Perez-Martinez, G. (1999) J. Bacteriol. 181, 3928–3934[Abstract/Free Full Text]
  18. Görke, B., and Rak, B. (1999) EMBO J. 18, 3370–3379[CrossRef][Medline] [Order article via Infotrieve]
  19. Lindner, C., Galinier, A., Hecker, M., and Deutscher, J. (1999) Mol. Microbiol. 31, 995–1006[CrossRef][Medline] [Order article via Infotrieve]
  20. Tortosa, P., Declerck, N., Dutartre, H., Lindner, C., Deutscher, J., and Le Coq, D. (2001) Mol. Microbiol. 41, 1381–1393[CrossRef][Medline] [Order article via Infotrieve]
  21. Amster-Choder, O., and Wright, A. (1997) J. Bacteriol. 179, 5621–5624[Abstract/Free Full Text]
  22. Tortosa, P., Aymerich, S., Lindner, C., Saier Jr., M. H., Reizer, J., and Le Coq, D. (1997) J. Biol. Chem. 272, 17230–17237[Abstract/Free Full Text]
  23. Henstra, S. A., Duurkens, R. H., and Robillard, G. T. (2000) J. Biol. Chem. 275, 7037–7044[Abstract/Free Full Text]
  24. Bachem, S., and Stülke, J. (1998) J. Bacteriol. 180, 5319–5326[Abstract/Free Full Text]
  25. Gosalbes, M. J., Esteban, C. D., and Perez-Martinez, G. (2002) Microbiology 148, 695–702[Abstract/Free Full Text]
  26. Chen, Q., Engelberg-Kulka, H., and Amster-Choder, O. (1997) J. Biol. Chem. 272, 17263–17268[Abstract/Free Full Text]
  27. Görke, B., and Rak, B. (2001) J. Mol. Biol. 308, 131–145[CrossRef][Medline] [Order article via Infotrieve]
  28. Miller, J. H. (1972) Experiments in Molecular Genetics, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York
  29. Zhou, Y., Zhang, X., and Ebright, R. (1991) Nucleic Acids Res. 19, 6052[Free Full Text]
  30. Prasad, I., Young, B., and Schaefler, S. (1973) J. Bacteriol. 114, 909–915[Abstract/Free Full Text]
  31. Junne, T., Schnetz, K., and Rak, B. (1990) J. Bacteriol. 172, 6615–6616[Free Full Text]
  32. Schnetz, K., Stülke, J., Gertz, S., Krüger, S., Krieg, M., Hecker, M., and Rak, B. (1996) J. Bacteriol. 178, 1971–1979[Abstract/Free Full Text]
  33. Declerck, N., Dutartre, H., Receveur, V., Dubois, V., Royer, C., Aymerich, S., and van Tilbeurgh, H. (2001) J. Mol. Biol. 314, 671–681[CrossRef][Medline] [Order article via Infotrieve]
  34. Declerck, N., Vincent, F., Hoh, F., Aymerich, S., and van Tilbeurgh, H. (1999) J. Mol. Biol. 294, 389–402[CrossRef][Medline] [Order article via Infotrieve]
  35. Wittekind, M., Reizer, J., Deutscher, J., Saier, M. H., Jr., and Klevit, R. E. (1989) Biochemistry 28, 9908–9912[CrossRef][Medline] [Order article via Infotrieve]
  36. van Tilbeurgh, H., Le Coq, D., and Declerck, N. (2001) EMBO J. 20, 3789–3799[CrossRef][Medline] [Order article via Infotrieve]
  37. Livnat, L., Nussbaum-Shochat, A., O'Day-Kerstein, K., Wright, A., and Amster-Choder, O. (2003) Proc. Natl. Acad. Sci. U. S. A. 100, 7099–7104[Abstract/Free Full Text]
  38. Boss, A., Nussbaum-Shochat, A., and Amster-Choder, O. (1999) J. Bacteriol. 181, 1755–1766[Abstract/Free Full Text]
  39. Mahadevan, S., Reynolds, A. E., and Wright, A. (1987) J. Bacteriol. 169, 2570–2578[Abstract/Free Full Text]
  40. Dole, S., Kühn, S., and Schnetz, K. (2002) Mol. Microbiol. 43, 217–226[CrossRef][Medline] [Order article via Infotrieve]
  41. Kramer, W., Drutsa, V., Jansen, H. W., Kramer, B., Pflugfelder, M., and Fritz, H. J. (1984) Nucleic Acids Res. 12, 9441–9456[Abstract/Free Full Text]
  42. Görke, B. (2000) Signal Transduction on the bgl-Operon from Escherichia coli. Ph.D. thesis, Universität Freiburg i. Br., Germany
  43. Spee, J. H., de Vos, W. M., and Kuipers, O. P. (1993) Nucleic Acids Res. 21, 777–778[Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Bacteriol.Home page
G. Monderer-Rothkoff and O. Amster-Choder
Genetic Dissection of the Divergent Activities of the Multifunctional Membrane Sensor BglF
J. Bacteriol., December 1, 2007; 189(23): 8601 - 8615.
[Abstract] [Full Text] [PDF]

Home page
 J. Bacteriol.Home page
B. Reichenbach, D. A. Breustedt, J. Stulke, B. Rak, and B. Gorke
Genetic Dissection of Specificity Determinants in the Interaction of HPr with Enzymes II of the Bacterial Phosphoenolpyruvate:Sugar Phosphotransferase System in Escherichia coli
J. Bacteriol., July 1, 2007; 189(13): 4603 - 4613.
[Abstract] [Full Text] [PDF]

Home page
 Microbiol. Mol. Biol. Rev.Home page
J. Deutscher, C. Francke, and P. W. Postma
How Phosphotransferase System-Related Protein Phosphorylation Regulates Carbohydrate Metabolism in Bacteria
Microbiol. Mol. Biol. Rev., December 1, 2006; 70(4): 939 - 1031.
[Abstract] [Full Text] [PDF]

Home page
 J. Bacteriol.Home page
T. Wolf, W. Janzen, C. Blum, and K. Schnetz
Differential Dependence of StpA on H-NS in Autoregulation of stpA and in Regulation of bgl.
J. Bacteriol., October 1, 2006; 188(19): 6728 - 6738.
[Abstract] [Full Text] [PDF]

Home page
 MicrobiologyHome page
S. Madhusudan, A. Paukner, Y. Klingen, and K. Schnetz
Independent regulation of H-NS-mediated silencing of the bgl operon at two levels: upstream by BglJ and LeuO and downstream by DnaKJ
Microbiology, October 1, 2005; 151(10): 3349 - 3359.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
M. Graille, C.-Z. Zhou, V. Receveur-Brechot, B. Collinet, N. Declerck, and H. van Tilbeurgh
Activation of the LicT Transcriptional Antiterminator Involves a Domain Swing/Lock Mechanism Provoking Massive Structural Changes
J. Biol. Chem., April 15, 2005; 280(15): 14780 - 14789.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplemental Data
Right arrow All Versions of this Article:
278/47/46219    most recent
M308002200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Görke, B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Görke, B.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2003 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement