Journal of Bacteriology, March 2004, p . 1381-1387, Vol . 186, No . 5

Studies of the Interaction of Escherichia coli YjeQ with the Ribosome In Vitro

Denis M . Daigle¹ and Eric D . Brown^1*

Antimicrobial Research Centre, Department of Biochemistry, McMaster University, Hamilton, Ontario, Canada L8N 3Z5¹

Received 20 September 2003/ Accepted 7 November 2003

 
Escherichia coli YjeQ represents a conserved group of bacteria-specific nucleotide-binding proteins of unknown physiological function that have been shown to be essential to the growth of E . coli and Bacillus subtilis . The protein has previously been characterized as possessing a slow steady-state GTP hydrolysis activity [8h^-1] [D . M . Daigle, L . Rossi, A . M . Berghuis, L . Aravind, E.V . Koonin, and E . D . Brown, Biochemistry 41: 11109-11117, 2002].In the work reported here, YjeQ from E . coli was found to copurify with ribosomes from cell extracts . The copy number of the proteinper cell was nevertheless low relative to the number of ribosomes[ratio of YjeQ copies to ribosomes, 1:200] . In vitro, recombinantYjeQ protein interacted strongly with the 30S ribosomal subunit,and the stringency of that interaction, revealed with salt washes,was highest in the presence of the nonhydrolyzable GTP analog 5'-guanylylimidodiphosphate [GMP-PNP] . Likewise, associationwith the 30S subunit resulted in a 160-fold stimulation of YjeQGTPase activity, which reached a maximum with stoichiometricamounts of ribosomes . N-terminal truncation variants of YjeQrevealed that the predicted OB-fold region was essential forribosome binding and GTPase stimulation, and they showed thatan N-terminal peptide [amino acids 1 to 20 in YjeQ] was necessaryfor the GMP-PNP-dependent interaction of YjeQ with the 30S subunit.Taken together, these data indicate that the YjeQ protein participatesin a guanine nucleotide-dependent interaction with the ribosomeand implicate this conserved, essential GTPase as a novel factorin ribosome function.

 
The YjeQ protein from Escherichia coli represents a family of orthologous proteins that are broadly conserved in bacteriaand absent in eukaryotes . YjeQ and its ortholog [YloQ] fromBacillus subtilis have been shown to be essential in their respective organisms [4] . Sequence analysis and homology modeling of YjeQ have revealed all diagnostic motifs of the P-loop GTPases, albeit in an unusual arrangement [9] . YjeQ and its orthologs exhibitan altered connectivity described by the G4-G1-G3 pattern of motifs as opposed to the consensus G1-G3-G4 pattern seen in most GTPases . By using purified, recombinant protein, it hasbeen shown that YjeQ is an unusual GTPase that catalyzes rapidhydrolysis of GTP [100 s^-1] despite having a low steady-state turnover of 8 h^-1 [9] . The low turnover is consistent with aslow, rate-limiting release of its products, GDP and phosphate.The kinetic disconnect between the chemical and product releasesteps of YjeQ is consistent with a role for the protein in transductionof the energy of hydrolysis of GTP into signal generation ormechanical work . Sequence analysis of YjeQ and its orthologshas also revealed an N-terminal, S1-like, OB-fold domain, foundin various RNA-binding proteins such as translation factorsand regulators of mRNA metabolism [3, 9] . Given the presenceof this domain and the fact that many GTPases, particularlythose that are highly conserved during evolution, function intranslation, we reasoned that YjeQ and its orthologs may befactors with a role in ribosome function [9].

Here we present the first experimental evidence that YjeQ associates with the ribosome . We report that YjeQ copurified with ribosomes from E . coli extracts . Using recombinant, purified protein, we have revealed a 160-fold stimulation of the GTPase activity of YjeQ through interaction with the 30S ribosomal subunit.Using N-terminal truncation variants, we have delineated a rolefor the OB-fold region of YjeQ and for amino acids 1 to 20 inthe interaction with the ribosome . The work is thus consistentwith the hypothesis that YjeQ has a role in ribosome function,and it provides a solid foundation for ongoing studies to furtherexplore this theory.

 
Materials. Tris, HEPES, and dithiothreitol [DTT] were from Bioshop CanadaInc . [Burlington, Ontario, Canada] . 2-Mercaptoethanol, 5'-guanylylimidodiphosphate[GMP-PNP], diethyl pyrocarbonate, Triton X-100, Malachite green,and ammonium molybdate were from Sigma-Aldrich [Oakville, Ontario,Canada] . RNase-free DNase I and Complete EDTA-free proteaseinhibitor cocktail were from Roche Diagnostics [Laval, Quebec,Canada] . Recombinant tobacco etch virus [TEV] protease was akind gift from Murray Junop [McMaster University] . Rabbit polyclonalanti-YjeQ[21-350] antibodies were produced at McMaster Universityand affinity purified by standard methods [5] . The secondaryantibody, horseradish perixidase [HRP]-conjugated donkey anti-rabbitimmunoglobulin G [IgG], was from Jackson ImmunoResearch, a branchof BIO/CAN [West Grove, Pa.] . GTP was from Amersham Biosciences[Baie d'Urfe, Quebec, Canada] . The Gateway Recombination Cloningand Expression kits were from Invitrogen-Life Technologies [Carlsbad,Calif.].

Construction of overexpression clones. Previous work characterizing recombinant YjeQ protein from E.coli showed that the expressed protein lacked the first 20 aminoacids [9] . Full-length YjeQ [YjeQ[1-350]] was produced withGateway Recombination technology using a TEV protease-cleavableN-terminal His₆ affinity purification tag . The gene was PCRamplified from E . coli MG1655 genomic DNA by using Vent DNApolymerase [New England Biolabs, Beverly, Mass.] and primersP1 [5'-G GGG ACA AGT TTG TAC AAA AAA GCA GGC TTA GAT TAC GATATC CCA ACG ACC GAA AAC CTG TAT TTT CAG * GGC AGT AAA AAT AAACTC TCC AAA GGC-3'] and P2 [5'-G GGG ACC ACT TTG TAC AAG AAAGCT GGG TCT CAG TCA TCC GTA TCA GAA AAG TTT TTA CGC G-3'] [theprotease cleavage site is marked with an asterisk; coding sequencesare underlined] . Cleavage of this protein with recombinant TEVprotease yielded full-length YjeQ[1-350] where the initiatorMet was replaced by Gly . YjeQ[21-350] and YjeQ[21-350] S221Awere constructed and purified as previously described [9] . YjeQ[114-350]was PCR amplified from plasmid pLR-1 [9] with primers P3 [5'-GGGG ACA AGT TTG TAC AAA AAA GCA GGC TTA GAA GGA GAT AGA ACCATG GAC GGC GTA AAA CCT ATT GCC GCC-3'] [coding sequences areunderlined] and P2 . The PCR-amplified products were cloned intoplasmid pDest14 [native] or pDest17 [His₆ tagged] by using Invitrogen-Life Technologies' Gateway Cloning and Expression kits . All plasmid constructs were confirmed by sequencing [MOBIX; McMaster University].

Purification of YjeQ[1-350] and variants. To purify YjeQ[1-350], 4 liters of E . coli BL21[DE3]/pDest17YjeQ-TEVwas grown at 37°C to an optical density at 600 nm [OD₆₀₀]of 0.3 and was induced with isopropyl-ß-D-thiogalactopyranoside [IPTG] to a final concentration of 1 mM for 3 h at 37°C[final OD₆₀₀, 1.2] . Cells were harvested by centrifugation at15,000 x g for 20 min . The cells were resuspended in bufferA [20 mM sodium phosphate [pH 7.2]-15 mM imidazole] containing1x Complete EDTA-free protease inhibitor cocktail [Roche Diagnostics]and lysed by three consecutive passes through a French pressurecell at 20,000 lb/in² . Cell debris was pelleted by centrifugationat 40,000 x g for 1 h at 4°C . The clarified lysate was loadedonto a HiTrap metal chelation column [bed volume, 5 ml] [AmershamBiosciences] charged with 5 ml of 100 mM nickel sulfate andpreequilibrated with buffer A . A linear gradient of buffer B[buffer A plus 350 mM imidazole] over 40 column volumes wasused to elute the protein . His₆-YjeQ[1-350] eluted between 200and 250 mM imidazole . Fractions containing His₆-YjeQ were pooledand concentrated to 2 ml by using Amicon Ultra [Fisher Scientific,Nepean, Ontario, Canada] centrifugal concentrators [molecularsize cutoff, 15 kDa], and buffer was gradually changed by bufferremoval and replacement with 100 mM Tris [pH 8] . The His₆ tagwas removed by digestion with purified recombinant TEV protease under the following conditions: 260 mg of YjeQ was digested with 6.5 mg of TEV protease for 2 h at 16°C in 50 mM Tris[pH 8]-0.5 mM EDTA-100 mM NaCl-0.25 mM DTT . The proteolyzedsample was dialyzed to remove DTT and EDTA and was rechromatographedover the HiTrap metal chelation column . YjeQ[1-350] lackingthe His₆ tag was located in the column flowthrough and concentratedto 2 ml . The sample was dialyzed in buffer C [50 mM HEPES [pH7.5], 1 mM DTT], loaded onto a Q Sepharose [Amersham Biosciences]anion-exchange column [2.6 by 20 cm], and eluted with a lineargradient of buffer D [buffer C plus 1 M NaCl] over 35 columnvolumes . Fractions containing YjeQ eluted between 300 and 350mM NaCl and were pooled and concentrated to 1.2 ml, followedby gel filtration chromatography on a Sephacryl S-200 column[1.6 by 70 cm] preequilibrated in buffer C . Fractions containingYjeQ were pooled, concentrated, and purified further by a secondanion-exchange step on a Mono Q [Amersham Biosciences] column[bed volume, 1 ml] preequilibrated in buffer C . By use of alinear gradient of buffer D, pure YjeQ[1-350] [eluting between320 and 340 mM NaCl and judged to be >99% pure by sodium dodecyl sulfate-polyacrylamide gel electrophoresis [SDS-PAGE] and staining with Coomassie brilliant blue R250]—was pooled, concentrated, dialyzed in storage buffer [50 mM HEPES [pH 7.5],0.5 mM DTT, 50 mM NaCl, 15% [vol/vol] glycerol], and storedat -80°C.

Untagged truncation variants of YjeQ, namely, YjeQ[21-350], YjeQ[21-350] S221A, and YjeQ[114-350], were purified as described previously [9] . Protein concentrations of all variants were determined by a Bradford assay [8] and verified by the methodof Gill and von Hippel [10].

Rapid isolation of ribosomes from E . coli cell extracts to determine the localization of YjeQ. Crude and rapid isolation of ribosomes from E . coli cell extractswas achieved by centrifugation at 40,000 x g for 1 h . The supernatantwas further clarified by ultracentrifugation at 150,000 x gfor 2 h . Stringency washes, first with 0.5% Triton, then with60 mM NH₄Cl, and finally with 1 M NH₄Cl, were performed for2 h at 4°C, according to standard methods [13] . Immunoblotting using SDS-15% polyacrylamide gels was performed as describedin the legend to Fig . 1.

 
Isolation of ribosomes and subunits depleted of YjeQ. Highly purified ribosomes and ribosomal subunits, containingno YjeQ detectable by Western blotting, were prepared usingmultiple, lengthy centrifugations over sucrose cushions andgradients as previously described [16] . Four liters of Luriabroth [LB] was inoculated with 40 ml of a saturated overnightculture of E . coli MG1655, grown to an OD₆₀₀ of 0.8, and slowly cooled to 15°C to produce runoff ribosomes, free of mRNA[16] . Cells were harvested by centrifugation at 8,500 x g for15 min, resuspended, and washed with buffer A [20 mM Tris-HCl[pH 7.5 at 4°C], 10.5 mM magnesium acetate, 100 mM NH₄Cl, 0.5 mM EDTA, 3 mM 2-mercaptoethanol] . All subsequent steps were performed at 4°C . The cell suspension was lysed by threeconsecutive passes through a French pressure cell at 10,000lb/in² followed by the addition of 500 U of RNase-free DNaseI [Roche Diagnostics] to the extract . An S30 fraction was generatedby centrifugation of the extract at 30,000 x g for 1 h . Thetop three-fourths of the S30 supernatant was recovered and overlaidonto an equal volume of 1.1 M sucrose cushions made up in bufferB [20 mM Tris-HCl [pH 7.5 at 4°C], 10.5 mM magnesium acetate,500 mM NH₄Cl, 0.5 mM EDTA, 3 mM 2-mercaptoethanol] . Sampleswere centrifuged at 100,000 x g for 15 h to produce sucrose- and salt-washed ribosomes [16] . The clear ribosomal pellet wasseparated from the flocculent brownish material above it, gentlywashed with buffer A, and then resuspended in buffer A by gentleagitation for 1 h . The sucrose-salt washing step described abovewas repeated . The clear ribosomal pellet was washed, and sucrosewas removed by two consecutive resuspensions in buffer C [10 mM Tris-HCl [pH 7.5 at 4°C], 10.5 mM magnesium acetate,500 mM NH₄Cl, 0.5 mM EDTA, 7 mM 2-mercaptoethanol] followedby centrifugation at 100,000 x g for 16 h.

To obtain 70S ribosomes, the ribosomal pellet was resuspendedin buffer D [10 mM Tris-HCl [pH 7.5 at 4°C], 5.25 mM magnesiumacetate, 60 mM NH₄Cl, 0.25 mM EDTA, 3 mM 2-mercaptoethanol]and separated by centrifugation on 10-to-30% [wt/vol] linearsucrose gradients made up in buffer D . The gradients were centrifugedat 48,000 x g for 15 h . The gradients were fractionated by upwarddisplacement using 60% [vol/vol] glycerol, and the fractions[200 µl] were analyzed by absorbance at 260 nm as wellas by SDS-15% PAGE to pool the appropriate subunits or 70S ribosomesbased on their protein complement . To ensure high purity ofthe ribosomal subunits, fractions overlapping the A₂₆₀ peakabsorbance for the subunits and those judged by SDS-PAGE and Coomassie brilliant blue R250 staining to be impure were eliminated as described previously [2] . Fractionation of the gradientswas followed by mixing of the 70S pool with an equal volume of buffer E [10 mM Tris-HCl [pH 7.5 at 4°C], 10 mM magnesium acetate, 60 mM NH₄Cl, 3 mM 2-mercaptoethanol] and centrifugation at 56,000 x g for 24 h . The 70S ribosomes were resuspended inbuffer E and stored at a concentration of 1,000 A₂₆₀ units/mlat -80°C [1 A₂₆₀ unit is equal to 23 pmol of 70S ribosomesin buffer E] . The A₂₆₀/A₂₈₀ ratio was determined to be 1.98.

Similarly, to obtain 30 and 50S subunits, the ribosomal pelletwas resuspended in buffer F [10 mM Tris-HCl [pH 7.5 at 4°C],1.1 mM magnesium acetate, 60 mM NH₄Cl, 0.1 mM EDTA, 2 mM 2-mercaptoethanol]. Fifty A₂₆₀ units of the subunit suspension was layered onto10-to-30% [wt/vol] sucrose gradients made up in buffer F, followedby centrifugation at 43,000 x g for 16 h . The gradients werefractionated as described above . The individual subunit poolswere recovered by pelleting at 200,000 x g for 12 h . The pellets for both 30 and 50S ribosomes were resuspended in buffer E, clarified by centrifugation at 15,000 x g for 15 min, and storedat -80°C as described above for 70S ribosomes . Quantitationof subunits was determined by absorbance at 260 nm [1 A₂₆₀ unitis equivalent to 69 or 34.5 pmol of 30 or 50S ribosomes, respectively].

 
YjeQ copurifies with ribosomes from cell extracts. Purification of ribosomes from wild-type E . coli and Westernblotting with anti-YjeQ antibodies revealed that nearly allof the YjeQ in the cell copurified with ribosomes [Fig . 1].In these experiments we employed simple detergent and salt washesfollowed by ultracentrifugation to simply isolate ribosomesfrom E . coli extracts and assess the localization of YjeQ . Indeed,the interaction was stable to detergent and high salt wash conditions,conventionally used to prepare ribosomes that are substantiallyfree from translation factors [13] . Furthermore, quantitative Western blotting revealed that YjeQ was of low abundance in E . coli, possessing a protein copy number of 100 copies/celland consequently in a substoichiometric association with ribosomes[YjeQ/ribosome ratio, 1:200] [data not shown].

Production of YjeQ[1-350] and its variants. Figure 2 shows a scaled diagram and SDS-PAGE analysis of full-length YjeQ and variants that were purified for this work . We produced full-length YjeQ[1-350] and N-terminal truncation variants lacking either the first 20 amino acids [YjeQ[21-350]] or the N-terminal OB-fold region [YjeQ[114-350]] . We also prepared an S221A variant previously characterized as having a significant impairmentof the chemical hydrolytic steps in catalysis with only a minorimpact on steady-state turnover [9] . Full-length YjeQ[1-350] was produced by engineering a TEV protease cleavable N-terminalHis₆ tag to prevent proteolysis of the N terminus . Overexpressionand purification of untagged YjeQ[1-350] from E . coli resultedin isolation of the truncated protein YjeQ[21-350] lacking thefirst 20 residues, as described previously [9] . All proteins were isolated to high purity [Fig . 2B], and all the truncation variants of YjeQ possessed similar steady-state GTPase activities [Table 1].

 
Binding of YjeQ[1-350] to 30, 50, and 70S ribosomes. To further characterize the interaction of YjeQ with ribosomes,it was necessary to produce ribosomes free of the YjeQ protein.We turned to a ribosome purification procedure involving multipleand lengthy sedimentations through sucrose cushions and gradientsfor the preparation of 70S ribosomes as well as 30 and 50S subunits[16] . Western blotting of these preparations revealed that allforms were free of YjeQ [data not shown] . Figure 3 shows the results of an in vitro pelleting assay in which the YjeQ-ribosome incubations are pelleted through sucrose cushions to test the interaction of full-length YjeQ[1-350] with the ribosome andits subunits under a variety of conditions [with or withoutGDP, GTP, or GMP-PNP] . YjeQ pelleted, to some extent, with allforms of the ribosome but showed the most extensive interactionwith the 30S subunit . In the presence or absence of GTP or GDP,YjeQ was distributed equally between the pellet and supernatantfractions when incubated with the 30S subunit . In the presenceof saturating levels of GMP-PNP, nearly all of the protein wasfound associated with the ribosomal pellet [Fig . 3A] . Whilenot tested explicitly here, this finding is consistent witha stronger association of YjeQ with the 30S subunit in the presenceof GMP-PNP, where a lower off-rate [i.e., rate of protein releasefrom the ribosome] for YjeQ would be manifest in more completepelleting over the 3-h ultracentrifugation run . This observationsuggests that the binding affinity of YjeQ for 30S ribosomalsubunits might be modulated by substrate hydrolysis and characterizedby maximal affinity for 30S subunits when GTP is found in theactive site.

 
Binding of YjeQ to 50 and 70S ribosomes was also observed, butto a lesser extent, and appeared to be independent of the presenceof GMP-PNP [Fig . 3B and C] . Interaction with the 50S subunit was slightly increased in the presence of saturating levelsof GDP and may have some functional significance . Western blottingin these experiments revealed a doublet of protein bands cross-reactive with anti-YjeQ polyclonal antibodies . We believe that this doubletis a result of proteolytic activity contaminating these ribosomal preparations . Indeed, the contaminating proteolytic activitymay be the source of the previously observed phenomenon wherethe overexpression of the recombinant untagged YjeQ in E . coli resulted in a cleavage product, lacking the first 20 amino acids[9] . In fact, the protein that copurified with ribosomes fromwild-type E . coli and was detected by Western blotting [Fig. 1] was found to comigrate with full-length YjeQ[1-350] [datanot shown] . Thus, the cleavage product seen in Fig . 3 upon incubationwith 50 and 70S ribosomes is presumed not to have a physiologicalfunction.

To further substantiate the binding of YjeQ to ribosomal subunits, and to confirm that results obtained by the pelleting assayreflected ribosomal interaction and not simply YjeQ precipitationin the presence of ribosomes, a complementary in vitro bindingexperiment was devised . YjeQ was incubated with GMP-PNP and70S ribosomes under conditions [lower magnesium acetate concentration,1.1 instead of 10.5 mM] that result in dissociation of 70S ribosomesto 50 and 30S ribosomal subunits [16] . The sample was subsequently separated on a 10-to-30% sucrose gradient instead of being pelleted through sucrose cushions . In this gradient system, all of the YjeQ was found to comigrate with the separated ribosomal subunits, with the majority [more than 75%] comigrating with the 30S material [Fig . 4].

 
Stringency of the interaction between YjeQ and 30S ribosomal subunits. The stringency of the association of YjeQ[1-350] with the 30S ribosomal subunit in the presence of GDP or GMP-PNP was evaluatedby using the pelleting assay with increasing concentrationsof either KCl or NH₄Cl [Fig . 5] as previously described [11]. At lower salt concentrations [100 and 250 mM KCl], the binding of YjeQ to the 30S subunit was essentially complete with GMP-PNPand about 50% complete with GDP, in agreement with the experimentsfor which results are presented in Fig . 3 . Addition of 500 mM NH₄Cl [and 250 mM KCl] reduced binding to about 50% in the presenceof GMP-PNP and completely abolished binding in the presence of GDP . One-third of the YjeQ protein remained associated with the 30S subunit at the highest salt concentration tested [1M NH₄Cl] . Thus, the stringency experiments support the conclusionthat the GMP-PNP-bound form of YjeQ has a stronger associationwith the 30S ribosomal subunit.

 
The GTPase activity of YjeQ is stimulated by the 30S ribosomal subunit. The intrinsic GTPase levels observed with YjeQ were strongly stimulated by the presence of ribosomes, particularly the 30S ribosomal subunit, with which we observed a 160-fold increasein k_cat [Table 1] . We further confirmed, by using the catalytically impaired [Ser221Ala] variant of YjeQ, that the GTPase activity upon stimulation originated from YjeQ . We previously showedthat the Ser221Ala variant had an impairment in the chemicalsteps of GTP hydrolysis, and we show here that no stimulationof GTPase activity by the 30S subunit was observed with thisYjeQ variant [Table 1] . Figure 6 shows the dependence of the GTPase stimulation of YjeQ on the amount of the 30S subunit present and reveals a plateau corresponding to a stoichiometryof 1 YjeQ copy to 1 30S subunit . As was the case for binding,the level of stimulation of GTPase activity by 70S ribosomeswas lower [96-fold] than that with 30S subunits [Table 1] . The50S subunit showed considerably less stimulation [13-fold] ofintrinsic GTPase levels of YjeQ [Table 1].

 
Impact of N-terminal truncations on ribosome binding and ribosome-stimulated GTPase activity. Ribosome binding and GTPase stimulation were measured in experimentsconducted to assess the contribution of the N-terminal regionof YjeQ to the ribosome interaction . Figure 7 shows the resultsof ribosome pelleting assays with N-terminal truncation variantsperformed in the presence of GMP-PNP . YjeQ[21-350], which lacksthe N-terminal 20 amino acids, demonstrated an affinity forboth 50 and 70S ribosomes that was similar to that seen in Fig.3 for full-length YjeQ[1-350] . In contrast, where nearly allof the full-length YjeQ[1-350] protein was associated with the30S subunit in the presence of GMP-PNP, the 20-amino-acid truncationvariant [YjeQ[21-350]] was found equally distributed betweenthe pellet and supernatant fractions, suggesting that this varianthad lost significant GMP-PNP-dependent binding affinity forthe 30S subunit . YjeQ[114-350], lacking the first 113 aminoacids, which encompass the OB-fold domain of the protein, wasunable to bind to any ribosome form [Fig . 7] . Similar experimentsyield identical findings when incubations are carried out withGDP or in the absence of nucleotide [data not shown].

 
Table 1 details the ribosome-stimulated GTPase activities of the N-terminal truncation variants . Stimulation of the GTPase of the 20-amino-acid truncation variant [YjeQ[21-350]] was slightly less than that of full-length YjeQ [e.g., 100- versus 160-fold stimulation by the 30S subunit for the 20-amino-acid truncation variant and the full-length protein, respectively] . Loss ofthe OB-fold of YjeQ[114-350] resulted in the near-eliminationof stimulation of GTPase by the ribosome and its subunits [e.g.,only 2.8-fold stimulation by the 30S subunit].

Taken together, the effects on ribosome binding and GTPase stimulation observed with the N-terminal truncation variants suggest that the OB-fold is critical for ribosome association and associated GTPase stimulation, while the N-terminal sequence [residues1 to 20] of YjeQ appears to be essential in imparting GMP-PNP-dependent binding affinity for the 30S ribosomal subunit.

 
Daigle et al . previously presented steady-state and pre-steady-state kinetic characterizations of recombinant YjeQ to show that the protein was an unusual GTPase enzyme where the chemical stepsof catalysis were 45,000-fold faster than those of product release[9] . They likewise presented sequence analysis of YjeQ and itsorthologs to reveal an unusual circular permutation in the GTPasedomain of the protein and the presence of an N-terminal S1-likeOB-fold domain [9], typical of proteins that interact with RNAs[3] . The unusual kinetics associated with the GTPase functionis consistent with a role for YjeQ in signal or energy transduction.In the work presented here, we characterize a selective interaction between YjeQ and the ribosome, mediated by the OB-fold domain, modulated by the nonhydrolyzable GTP analog GMP-PNP, and witha stimulatory impact on the GTPase activity of YjeQ . The resultsare intriguing and consistent with a role for YjeQ in ribosomefunction.

Our work began with the routine isolation of ribosomes from wild-type E . coli . To our surprise, isolation of 70S ribosomes from E . coli and Western blotting with anti-YjeQ antibodies revealed that nearly all of the YjeQ in the cell was associatedwith ribosomes . Indeed, the interaction was stable to wash conditionsthat are typically used to remove translation factors from ribosome preparations [13] . Quantitative Western blotting put the copy number of YjeQ at about 100 copies per cell, in a stoichiometry of about 1 for every 200 ribosomes, consistent with the factthat this protein has not been reported in ribosome preparations previously . In speculating on a role for this protein in ribosome function, the low copy number of the protein is noteworthy.The celebrated translational GTPases, EF-Tu, EF-G, and IF2,for example, are abundant and in near-stoichoimetry with ribosomes.Elongation factor P, on the other hand, has a relatively lowcopy number [EF-P/ribosome ratio, 1:30] [1] and functions in stimulating the peptidyltransferase to enhance peptide bondformation only in certain dipeptides [2] . It is conceivable that YjeQ also has a critical but narrow role in a subset of translating ribosomes.

Some technical hurdles are noteworthy in the work presentedhere . Meticulous sucrose gradient density sedimentation procedures[16] were required to prepare highly purified ribosomes andsubunits depleted of YjeQ . That material was critical to ourexaminations of the YjeQ-ribosome interaction in vitro by useof a ribosome pelleting assay . We also elected to troubleshootthe purification of full-length recombinant YjeQ for these experiments.It was noted previously that an N-terminal truncation variant[YjeQ[21-350]] was consistently generated upon overexpressionand purification of the untagged protein [9] . Here, full-lengthYjeQ[1-350] was produced by engineering a TEV protease cleavagesite to remove an N-terminal polyhistidine tag from affinity-purifiedYjeQ protein . Apparently, the N-terminal tag protected YjeQfrom the proteolysis that beset the native protein upon overexpression.

While YjeQ bound to all forms of the ribosome in our pelleting assay, the extent of binding, judged by the fraction of YjeQthat pelleted with a stoichiometric amount of ribosomes, variedand was the highest with the 30S ribosomal subunit, where aboutone-half of the YjeQ protein copelleted . Addition of GDP orGTP to the pelleting assay had no impact on the extent of binding,while the nonhydrolyzable GTP analog GMP-PNP resulted in completecopelleting of full-length YjeQ[1-350] with the ribosomal subunit.The identical outcome with GTP, GDP, and no added nucleotideis consistent with the fact that YjeQ is purified in a formbound by 0.6 equivalent of GDP and rapidly hydrolyzes GTP toGDP with a rate constant of 100 s^-1 [9] . The fact that the GMP-PNP-boundform of YjeQ had a higher affinity for the 30S subunit thanthe GDP-bound form was also evident in the stringencies of therespective interactions to increasing salt concentrations.

We speculate that we have probed, in these experiments, a physiologically relevant modulation of the 30S subunit binding activity of YjeQ in its GTP- and GDP-bound states, where the GTP analog GMP-PNP facilitates the production of a static mimic of the GTP-boundform . Such modulations are paradigmatic of signal and energytransducing G-proteins and are frequently associated with animpact on GTPase function . Thus, it follows in this work thatwe have also noted a significant stimulation of the steady-stateGTPase activity of YjeQ by the ribosome, in particular by the30S subunit . Also remarkable is the fact that maximum stimulationof GTPase activity occurs at a 1:1 stoichiometry of YjeQ withthe ribosome . This implies that despite the low copy numberof YjeQ, it is capable of a functionally significant and stoichiometricinteraction with the 30S subunit.

Low intrinsic GTPase activity is not uncommon among prokaryotic translational GTPases, including EF-Tu and EF-G, which require interaction with the ribosome for maximal activity [6, 7] . Interestingly,Era, another prokaryotic GTPase possessing low intrinsic activity,is stimulated by 16S rRNA and has recently been discovered tobe a factor involved in the maturation of 16S rRNA [12] . Full-lengthYjeQ's steady-state GTPase activity of 3.1 h^-1 is comparableto intrinsic GTPase levels observed with EF-Tu [1.8 h^-1] [15, 17] . Stimulation of EF-Tu by unprogrammed ribosomes [lackingan mRNA template and associated translation factors] is aboutto 2- to 20-fold, while the binding of EF-Tu to programmed ribosomesresults in 100,000-fold stimulation of the GTPase [to 500 s^-1]. Thus, the 160-fold stimulation seen here with YjeQ and unprogrammed ribosomes has a precedent . In the case of YjeQ, the pre-steady-state kinetic analysis described a rapid GTP hydrolysis step [100s^-1] followed by a much slower steady-state turnover, apparentlylimited by-product release [9] . It seems likely that the interaction of YjeQ with the ribosome impacts primarily on product release kinetics . We will test this hypothesis in due course.

The ribosome binding experiments performed with the N-terminal truncation variants of recombinant YjeQ, presented here, have revealed the importance of the first 20 amino acids and theOB-fold for YjeQ function on the ribosome . Typically composedof a five-stranded closed ß-barrel structure and oftencapped by an alpha helix, OB-folds form a binding surface employedfor binding oligosaccharides, proteins, and most often oligonucleotides[3] . We found low but significant sequence similarity betweenthe OB-fold in YjeQ and its orthologs and the OB-fold of theprotein translation factor eIF-1A [9] . Truncation of the OB-foldin the YjeQ[114-350] variant abolished binding and GTPase stimulationby the ribosome or its individual subunits . Truncation of thefirst 20 amino acids in YjeQ[21-350], on the other hand, showedlittle effect on ribosome binding or on ribosome stimulationof the GTPase activity of YjeQ . This variant exhibited behaviorcomparable to that of the full-length enzyme, except that its30S binding activity was no longer modulated by the nonhydrolyzableGTP analog GMP-PNP . Given the critical role of the N-terminal20 amino acids in the latter phenomenon, it is worth emphasizingour previous finding that this peptide is proteolytically sensitiveto removal and is absent when the untagged protein is overexpressedand purified from E . coli [9] . Thus, its role is likely criticalto a fully functional YjeQ protein.

Information gathered in this study provides further supportfor the hypothesis that the YjeQ protein from E . coli and its orthologs are bacteria-specific factors with a role in ribosome function . We have concentrated in this work on the interactionof YjeQ with the ribosome and the impact of that interactionon YjeQ function . Work to address the impact of YjeQ on thefunction of the ribosome is ongoing.

 
* Corresponding author . Mailing address: Antimicrobial Research Centre, Department of Biochemistry, McMaster University, 1200 Main St . West, Hamilton, Ontario, Canada L8N 3Z5 . Phone: [905] 525-9140, ext . 22392 . Fax: [905] 522-9033 . E-mail: ebrown@mcmaster.ca.

1. An, G., B . R . Glick, J . D . Friesen, and M . C . Ganoza. 1980 . Identification and quantitation of elongation factor EF-P in Escherichia coli cell-free extracts . Can . J . Biochem . 97:23-28.
2. Aoki, H., K . Dekany, S.-L . Adams, and M . C . Ganoza. 1997 . The gene encoding the elongation factor P protein is essential for viability and is required for protein synthesis . J . Biol . Chem . 272:32254-32259 .
3. Arcus, V. 2002 . OB-fold domains: a snapshot of the evolution of sequence, structure and function . Curr . Opin . Struct . Biol . 12:794-801.
4. Arigoni, F., F . Talabot, M . Peitsch, M . D . Edgerton, E . Meldrum, E . Allet, R . Fish, T . Jamotte, M . L . Curchod, and H . Loferer. 1998 . A genome-based approach for the identification of essential bacterial genes . Nat . Biotechnol . 16:851-856.
5. Ausubel, F . M., R . Brent, R . E . Kingston, D . D . Moore, J . G . Seidman, J . A . Smith, and K . Struhl. 1994 . Current protocols in molecular biology, vol . 2, chapter 11, section 11.12-11.13 . John Wiley & Sons Inc., Boston, Mass.
6. Bourne, H . R., D . A . Sanders, and F . McCormick. 1990 . The GTPase superfamily: a conserved switch for diverse cell functions . Nature 348:125-132.
7. Bourne, H . R., D . A . Sanders, and F . McCormick. 1991 . The GTPase superfamily: conserved structure and molecular mechanism . Nature 349:117-127.
8. Bradford, M . M. 1976 . A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding . Anal . Biochem . 34:248-254.
9. Daigle, D . M., L . Rossi, A . M . Berghuis, L . Aravind, E . V . Koonin, and E . D . Brown. 2002 . YjeQ, an essential, conserved, uncharacterized protein from Escherichia coli, is an unusual GTPase with circularly permuted G-motifs and marked burst kinetics . Biochemistry 41:11109-11117.
10. Gill, S . C., and P . H . von Hippel. 1989 . Calculation of protein extinction coefficients from amino acid sequence data . Anal . Biochem. 182:319-326.
11. Hesterkamp, T., E . Deuerling, and B . Bukau. 1997 . The amino-terminal 118 amino acids of Escherichia coli trigger factor constitute a domain that is necessary and sufficient for binding to ribosomes . J . Biol . Chem . 272:21865-21871 .
12. Inoue, K., J . Alsina, J . Chen, and M . Inouye. 2003 . Suppression of defective ribosome assembly in a rbfA deletion mutant by overexpression of Era, an essential GTPase in Escherichia coli. Mol . Microbiol . 48:1005-1016.
13. Kung, H . F., C . Spears, T . Schulz, and H . Weissbach. 1974 . Studies on the in vitro synthesis of beta-galactosidase: necessary components in the ribosomal wash . Arch . Biochem . Biophys . 162:578-584.
14. Laemmli, U . K. 1970 . Cleavage of structural proteins during the assembly of the head of bacteriophage T4 . Nature 227:680-685.
15. Rodnina, M . V., H . Stark, A . Savelsbergh, H.-J . Wieden, N . B . Matassova, F . Peske, T . Daviter, C . O . Gualerzi, and W . Wintermeyer. 2000 . GTPases: mechanisms and functions of translation factors on the ribosome . Biol . Chem . 381:377-387.
16. Spedding, G. 1990 . Isolation and analysis of ribosomes from prokaryotes, eukaryotes and organelles, p . 5-7 . In G . Spedding [ed.], Ribosomes and protein synthesis: a practical approach . IRL Press, New York, N.Y.
17. Sprinzl, M., S . Brock, Y . Huang, P . Milovnik, M . Nanninga, M . Nesper-Brock, H . Rutthard, and K . Szkaradkiewicz. 2000 . Regulation of GTPases in the bacterial translation machinery . Biol . Chem . 381:367-375.

Free Online Full-text Article

What Is Nitrification?, What Is Salmonella?, What Is Activated Sludge?, What Is Molecular Biology?, What Is Water Purification?, i, Microbes, r, Microorganism, r, Microbiology, n, Microorganisms, r, Bacterium, r, Microbiological, e, S. cerevisiae, e, Escherichia coli, o, Klebsiella, s, Alcaligenes, s, Pseudomonas aeruginosa, n, Escherichia coli, c, Cell suspensions, i, Salmonella typhimurium, i, Bacteria, s, Bacillus, c, Gram negative, c, Bacteria, n, Bacteriological, e, Bacteriological, n, Candida albicans, i, Agrobacterium, n, Microbial, n, Proteus, e, Escherichia coli, n, Escherichia coli




 

© 2005 Transgalactic Ltd (manufacturer of Bioscreen C software) | Privacy Statement | P.O. Box 1393, 00101 Helsinki, Finland, phone: +358 9 85172920, fax: +358 9 8749481, e-mail: microbiology@bionewsonline.com
 

Last modified: May 25, 2005