The structural and functional basis for the kirromycin resistance of mutant EF-Tu species in Escherichia coli.

A structural and functional understanding of resistance to the antibiotic kirromycin in Escherichia coli has been sought in order to shed new light on the functioning of the bacterial elongation factor Tu (EF-Tu), in particular its ability to act as a molecular switch. The mutant EF-Tu species G316D, A375T, A375V and Q124K, isolated by M13mp phage-mediated targeted mutagenesis, were studied. In this order the mutant EF-Tu species showed increasing resistance to the antibiotic as measured by poly(U)-directed poly(Phe) synthesis and intrinsic GTPase activities. The K'd values for kirromycin binding to mutant EF-Tu.GTP and EF-Tu.GDP increased in the same order. All mutation sites cluster in the interface of domains 1 and 3 of EF-Tu.GTP, not in that of EF-Tu.GDP. Evidence is presented that kirromycin binds to this interface of wild-type EF-Tu.GTP, thereby jamming the conformational switch of EF-Tu upon GTP hydrolysis. We conclude that the mutations result in two separate mechanisms of resistance to kirromycin. The first inhibits access of the antibiotic to its binding site on EF-Tu.GTP. A second mechanism exists on the ribosome, when mutant EF-Tu species release kirromycin and polypeptide chain elongation continues.     

Codon-dependent conformational change of elongation factor Tu preceding GTP hydrolysis on the ribosome.

The mechanisms by which elongation factor Tu (EF-Tu) promotes the binding of aminoacyl-tRNA to the A site of the ribosome and, in particular, how GTP hydrolysis by EF-Tu is triggered on the ribosome, are not understood. We report steady-state and time-resolved fluorescence measurements, performed in the Escherichia coli system, in which the interaction of the complex EF-Tu.GTP.Phe-tRNAPhe with the ribosomal A site is monitored by the fluorescence changes of either mant-dGTP [3'-O-(N-methylanthraniloyl)-2-deoxyguanosine triphosphate], replacing GTP in the complex, or of wybutine in the anticodon loop of the tRNA. Additionally, GTP hydrolysis is measured by the quench-flow technique. We find that codon-anticodon interaction induces a rapid rearrangement within the G domain of EF-Tu around the bound nucleotide, which is followed by GTP hydrolysis at an approximately 1.5-fold lower rate. In the presence of kirromycin, the activated conformation of EF-Tu appears to be frozen. The steps following GTP hydrolysis--the switch of EF-Tu to the GDP-bound conformation, the release of aminoacyl-tRNA from EF-Tu to the A site, and the dissociation of EF-Tu-GDP from the ribosome--which are altogether suppressed by kirromycin, are not distinguished kinetically. The results suggest that codon recognition by the ternary complex on the ribosome initiates a series of structural rearrangements resulting in a conformational change of EF-Tu, possibly involving the effector region, which, in turn, triggers GTP hydrolysis.     

Delayed release of inorganic phosphate from elongation factor Tu following GTP hydrolysis on the ribosome

The dissociation of inorganic phosphate (P(i)) following GTP hydrolysis is a key step determining the functional state of many GTPases. Here, the timing of P(i) release from elongation factor Tu (EF-Tu) and its implications for the function of EF-Tu on the ribosome were studied by rapid kinetic techniques. It was found that P(i) release from EF-Tu is >20-fold slower than GTP cleavage and limits the rate of the conformational switch of EF-Tu from the GTP- to the GDP-bound form. The point mutation Gly94Ala in the switch 2 region of EF-Tu abolished the delay in P(i) release, suggesting that P(i) release is controlled by the mobility of the switch 2 region with Gly94 acting as a pivot. The rate of P(i) release or the conformational switch of EF-Tu does not affect the selection of aminoacyl-tRNA on the ribosome. Rather, the slow P(i) release may be a consequence of the tight interaction of the switch regions of EF-Tu with the gamma-phosphate and the ribosome in the GTPase activated state of the factor.     

Intact aminoacyl-tRNA is required to trigger GTP hydrolysis by elongation factor Tu on the ribosome

GTP hydrolysis by elongation factor Tu (EF-Tu) on the ribosome is induced by codon recognition. The mechanism by which a signal is transmitted from the site of codon-anticodon interaction in the decoding center of the 30S ribosomal subunit to the site of EF-Tu binding on the 50S subunit is not known. Here we examine the role of the tRNA in this process. We have used two RNA fragments, one which contains the anticodon and D hairpin domains (ACD oligomer) derived from tRNA(Phe) and the second which comprises the acceptor stem and T hairpin domains derived from tRNA(Ala) (AST oligomer) that aminoacylates with alanine and forms a ternary complex with EF-Tu. GTP. While the ACD oligomer and the ternary complex containing the Ala-AST oligomer interact with the 30S and 50S A site, respectively, no rapid GTP hydrolysis was observed when both were bound simultaneously. The presence of paromomycin, an aminoglycoside antibiotic that binds to the decoding site and stabilizes codon-anticodon interaction in unfavorable coding situations, did not increase the rate of GTP hydrolysis. These results suggest that codon recognition as such is not sufficient for GTPase activation and that an intact tRNA molecule is required for transmitting the signal created by codon recognition to EF-Tu.     

EF-Tu dynamics during pre-translocation complex formation: EF-Tu·GDP exits the ribosome via two different pathways

The G-protein EF-Tu, which undergoes a major conformational change when EF-Tu·GTP is converted to EF-Tu·GDP, forms part of an aminoacyl(aa)-tRNA·EF-Tu·GTP ternary complex (TC) that accelerates the binding of aa-tRNA to the ribosome during peptide elongation. Such binding, placing a portion of EF-Tu in contact with the GTPase Associated Center (GAC), is followed by GTP hydrolysis and P_i release, and results in formation of a pretranslocation (PRE) complex. Although tRNA movement through the ribosome during PRE complex formation has been extensively studied, comparatively little is known about the dynamics of EF-Tu interaction with either the ribosome or aa-tRNA. Here we examine these dynamics, utilizing ensemble and single molecule assays employing fluorescent labeled derivatives of EF-Tu, tRNA, and the ribosome to measure changes in either FRET efficiency or fluorescence intensity during PRE complex formation. Our results indicate that ribosome-bound EF-Tu separates from the GAC prior to its full separation from aa-tRNA, and suggest that EF-Tu·GDP dissociates from the ribosome by two different pathways. These pathways correspond to either reversible EF-Tu·GDP dissociation from the ribosome prior to the major conformational change in EF-Tu that follows GTP hydrolysis, or irreversible dissociation after or concomitant with this conformational change.     

Molecular dynamics of ribosomal elongation factors G and Tu

Translation on the ribosome is controlled by external factors. During polypeptide lengthening, elongation factors EF-Tu and EF-G consecutively interact with the bacterial ribosome. EF-Tu binds and delivers an aminoacyl-tRNA to the ribosomal A site and EF-G helps translocate the tRNAs between their binding sites after the peptide bond is formed. These processes occur at the expense of GTP. EF-Tu:tRNA and EF-G are of similar shape, share a common binding site, and undergo large conformational changes on interaction with the ribosome. To characterize the internal motion of these two elongation factors, we used 25 ns long all-atom molecular dynamics simulations. We observed enhanced mobility of EF-G domains III, IV, and V and of tRNA in the EF-Tu:tRNA complex. EF-Tu:GDP complex acquired a configuration different from that found in the crystal structure of EF-Tu with a GTP analogue, showing conformational changes in the switch I and II regions. The calculated electrostatic properties of elongation factors showed no global similarity even though matching electrostatic surface patches were found around the domain I that contacts the ribosome, and in the GDP/GTP binding region.     

Insights into the Structural Dynamics of Nucleocytoplasmic Transport of tRNA by Exportin-t

Exportin-t (Xpot) transports mature 5′- and 3′-end processed tRNA from the nucleus to the cytoplasm by associating with a small G-protein Ran (RAs-related nuclear protein), in the nucleus. The release of tRNA in cytoplasm involves RanGTP hydrolysis. Despite the availability of crystal structures of nuclear and cytosolic forms of Xpot, the molecular details regarding the sequential events leading to tRNA release and subsequent conformational changes occurring in Xpot remain unknown. We have performed a combination of classical all-atom and accelerated molecular dynamics simulations on a set of complexes involving Xpot to study a range of features including conformational flexibility of free and cargo-bound Xpot and functionally critical contacts between Xpot and its cargo. The systems investigated include free Xpot and its different complexes, bound either to Ran (GTP/GDP) or tRNA or both. This approach provided a statistically reliable estimate of structural dynamics of Xpot after cargo release. The mechanistic basis for Xpot opening after cargo release has been explained in terms of dynamic structural hinges, about which neighboring region could be displaced to facilitate the nuclear to cytosolic state transition. Post-RanGTP hydrolysis, a cascade of events including local conformational change in RanGTP and loss of critical contacts at Xpot/tRNA interface suggest factors responsible for eventual release of tRNA. The level of flexibility in different Xpot complexes varied depending on the arrangement of individual HEAT repeats. Current study provides one of the most comprehensive and robust analysis carried out on this protein using molecular dynamics schemes.     

Distortion of tRNA upon near-cognate codon recognition on the ribosome

The accurate decoding of the genetic information by the ribosome relies on the communication between the decoding center of the ribosome, where the tRNA anticodon interacts with the codon, and the GTPase center of EF-Tu, where GTP hydrolysis takes place. In the A/T state of decoding, the tRNA undergoes a large conformational change that results in a more open, distorted tRNA structure. Here we use a real-time transient fluorescence quenching approach to monitor the timing and the extent of the tRNA distortion upon reading cognate or near-cognate codons. The tRNA is distorted upon codon recognition and remains in that conformation until the tRNA is released from EF-Tu, although the extent of distortion gradually changes upon transition from the pre- to the post-hydrolysis steps of decoding. The timing and extent of the rearrangement is similar on cognate and near-cognate codons, suggesting that the tRNA distortion alone does not provide a specific switch for the preferential activation of GTP hydrolysis on the cognate codon. Thus, although the tRNA plays an active role in signal transmission between the decoding and GTPase centers, other regulators of signaling must be involved.     

Structural basis for interaction of the ribosome with the switch regions of GTP-bound elongation factors

Elongation factor G (EF-G) catalyzes tRNA translocation on the ribosome. Here a cryo-EM reconstruction of the 70S*EF-G ribosomal complex at 7.3 A resolution and the crystal structure of EF-G-2*GTP, an EF-G homolog, at 2.2 A resolution are presented. EF-G-2*GTP is structurally distinct from previous EF-G structures, and in the context of the cryo-EM structure, the conformational changes are associated with ribosome binding and activation of the GTP binding pocket. The P loop and switch II approach A2660-A2662 in helix 95 of the 23S rRNA, indicating an important role for these conserved bases. Furthermore, the ordering of the functionally important switch I and II regions, which interact with the bound GTP, is dependent on interactions with the ribosome in the ratcheted conformation. Therefore, a network of interaction with the ribosome establishes the active GTP conformation of EF-G and thus facilitates GTP hydrolysis and tRNA translocation.     

Timing of GTP binding and hydrolysis by translation termination factor RF3

Protein synthesis in bacteria is terminated by release factors 1 or 2 (RF1/2), which, on recognition of a stop codon in the decoding site on the ribosome, promote the hydrolytic release of the polypeptide from the transfer RNA (tRNA). Subsequently, the dissociation of RF1/2 is accelerated by RF3, a guanosine triphosphatase (GTPase) that hydrolyzes GTP during the process. Here we show that—in contrast to a previous report—RF3 binds GTP and guanosine diphosphate (GDP) with comparable affinities. Furthermore, we find that RF3–GTP binds to the ribosome and hydrolyzes GTP independent of whether the P site contains peptidyl-tRNA (pre-termination state) or deacylated tRNA (post-termination state). RF3–GDP in either pre- or post-termination complexes readily exchanges GDP for GTP, and the exchange is accelerated when RF2 is present on the ribosome. Peptide release results in the stabilization of the RF3–GTP–ribosome complex, presumably due to the formation of the hybrid/rotated state of the ribosome, thereby promoting the dissociation of RF1/2. GTP hydrolysis by RF3 is virtually independent of the functional state of the ribosome and the presence of RF2, suggesting that RF3 acts as an unregulated ribosome-activated switch governed by its internal GTPase clock.     

A unique tRNA recognition mechanism of Caenorhabditis elegans mitochondrial EF-Tu2

Nematode mitochondria expresses two types of extremely truncated tRNAs that are specifically recognized by two distinct elongation factor Tu (EF-Tu) species named EF-Tu1 and EF-Tu2. This is unlike the canonical EF-Tu molecule that participates in the standard protein biosynthesis systems, which basically recognizes all elongator tRNAs. EF-Tu2 specifically recognizes Ser-tRNA^Ser that lacks a D arm but has a short T arm. Our previous study led us to speculate the lack of the D arm may be essential for the tRNA recognition of EF-Tu2. However, here, we showed that the EF-Tu2 can bind to D arm-bearing Ser-tRNAs, in which the D–T arm interaction was weakened by the mutations. The ethylnitrosourea-modification interference assay showed that EF-Tu2 is unique, in that it interacts with the phosphate groups on the T stem on the side that is opposite to where canonical EF-Tu binds. The hydrolysis protection assay using several EF-Tu2 mutants then strongly suggests that seven C-terminal amino acid residues of EF-Tu2 are essential for its aminoacyl-tRNA-binding activity. Our results indicate that the formation of the nematode mitochondrial (mt) EF-Tu2/GTP/aminoacyl-tRNA ternary complex is probably supported by a unique interaction between the C-terminal extension of EF-Tu2 and the tRNA.     

Domain III of elongation factor G from Thermus thermophilus is essential for induction of GTP hydrolysis on the ribosome

Two elongation factors (EF) EF-Tu and EF-G participate in the elongation phase during protein biosynthesis on the ribosome. Their functional cycles depend on GTP binding and its hydrolysis. The EF-Tu complexed with GTP and aminoacyl-tRNA delivers tRNA to the ribosome, whereas EF-G stimulates translocation, a process in which tRNA and mRNA movements occur in the ribosome. In the present paper we report that: (a) intrinsic GTPase activity of EF-G is influenced by excision of its domain III; (b) the EF-G lacking domain III has a 10(3)-fold decreased GTPase activity on the ribosome, whereas its affinity for GTP is slightly decreased; and (c) the truncated EF-G does not stimulate translocation despite the physical presence of domain IV, which is also very important for translocation. By contrast, the interactions of the truncated factor with GDP and fusidic acid-dependent binding of EF-G.GDP complex to the ribosome are not influenced. These findings indicate an essential contribution of domain III to activation of GTP hydrolysis. These results also suggest conformational changes of the EF-G molecule in the course of its interaction with the ribosome that might be induced by GTP binding and hydrolysis.     

Alignment/misalignment hypothesis for tRNA selection by the ribosome

Transfer RNAs (tRNAs) are the adaptor molecules that allow the ribosome to decode genetic information during protein synthesis. During decoding, the ribosome must chose the tRNA whose anticodon corresponds to the codon inscribed in the messenger RNA to incorporate the correct amino acid into the growing polypeptide chain. Fidelity is improved dramatically by a GTP hydrolysis event. Information about the correctness of the anticodon must be sent from the decoding center to the elongation factor, EF-Tu, where the GTP hydrolysis takes place. A second discrimination event entails the accommodation of the aminoacyl-tRNA into its fully bound A/A state inside the ribosome. Here, we present a hypothesis for a specific mechanism of signal transduction through the tRNA, which operates during GTPase activation and accommodation. We propose that the rigidity of the tRNA plays an important role in the transmission of the decoding signal. While the tRNA must flex during binding and accommodation, its anisotropic stiffness enables precise positioning of the acceptor arm in the A/T state, the A/A state and the accommodation corridor. Correct alignment will result in optimal GTPase activation and accommodation rates. Incorrect tRNAs, however, whose anticodons are misaligned, will also have acceptor arms that are misaligned, resulting in sub-optimal GTPase activation and accommodation rates. In the case of GTPase activation, it is possible that the misalignment of the acceptor arm affects the rate directly, by altering the conformational change of the switch region of EF-Tu, or indirectly, by changing the alignment of EF-Tu with respect to the sarcin-ricin loop (SRL) of the large ribosomal subunit.     

Streptomycin interferes with conformational coupling between codon recognition and GTPase activation on the ribosome

Aminoacyl-tRNAs (aa-tRNAs) are selected by the ribosome through a kinetically controlled induced fit mechanism. Cognate codon recognition induces a conformational change in the decoding center and a domain closure of the 30S subunit. We studied how these global structural rearrangements are related to tRNA discrimination by using streptomycin to restrict the conformational flexibility of the 30S subunit. The antibiotic stabilized aa-tRNA on the ribosome both with a cognate and with a near-cognate codon in the A site. Streptomycin altered the rates of GTP hydrolysis by elongation factor Tu (EF-Tu) on cognate and near-cognate codons, resulting in almost identical rates of GTP hydrolysis and virtually complete loss of selectivity. These results indicate that movements within the 30S subunit at the streptomycin-binding site are essential for the coupling between base pair recognition and GTP hydrolysis, thus modulating the fidelity of aa-tRNA selection.     

Conformational changes in ribosomes upon interaction with elongation factors

Conformational changes in the ribosomes upon interaction with EF-Tu were studied by limited proteolysis with a set of proteases. The main results are: (1) The cleavage rate of S1 protein strongly depends on the cooperative effect of poly(U) and tRNA: (2) The conformation of L7/L12 proteins is modulated by interaction of elongation factors with the ribosome and depends on hydrolysis of GTP; (3) The sensitivity of some ribosomal proteins (S6, S7, S18, S19, L9, L16, L19, and L27) to proteases changes upon binding of EF-Tu and depends on the ribosome functional state in accordance with GTP hydrolysis. Most of these proteins are located far from the factor-binding center of the ribosome. The possible mechanism of conformational changes is discussed.     

Elongation factor Tu ternary complex binds to small ribosomal subunits in a functionally active state

A complex between elongation factor Tu (EF-Tu), GTP, phenylalanyl-tRNA (Phe-tRNA), oligo(uridylic acid) [oligo(U)], and the 30S ribosomal subunit of Escherichia coli has been formed and isolated. Binding of the EF-Tu complex appears to be at the functionally active 30S site, by all biochemical criteria that were examined. The complex can be isolated with 0.25-0.5 copy of EF-Tu bound per ribosome. The binding is dependent upon the presence of both the aminoacyl-tRNA and the cognate messenger RNA. Addition of 50S subunits to the preformed 30S-EF-Tu-GTP-Phe-tRNA-oligo(U) complex ("30S-EF-Tu complex") causes a rapid hydrolysis of GTP. This hydrolysis is coordinated with the formation of 70S ribosomes and the release of EF-Tu. Both the release of EF-Tu and the hydrolysis of GTP are stoichiometric with the amount of added 50S subunits. 70S ribosomes, in contrast to 50S subunits, neither release EF-Tu nor rapidly hydrolyze GTP when added to the 30S-EF-Tu complexes. The inability of 70S ribosomes to react with the 30S-EF-Tu complex argues that the 30S-EF-Tu complex does not dissociate prior to reaction with the 50S subunit. The requirements of the 30S reaction for Phe-tRNA and oligo(U) and the consequences of the addition of 50S subunits resemble the reaction of EF-Tu with 70S ribosomes, although EF-Tu binding to isolated 30S subunits does not occur during the elongation microcycle. This suggests that the EF-Tu ternary complex binds to isolated 30S subunits at the same 30S site that is occupied during ternary complex interaction with the 70S ribosome.(ABSTRACT TRUNCATED AT 250 WORDS)     

Structural and functional properties of thesaurin a (42Sp50), the major protein of the 42 S particles present in Xenopus laevis previtellogenic oocytes

Thesaurin a is one of two protein components of a 42 S ribonucleoprotein particle that is very abundant in previtellogenic oocytes of Xenopus laevis. The primary function of the 42 S particle is the long-term storage of 5 S RNA and aminoacyl-tRNA. Thesaurin a is homologous to eukaryotic elongation factor 1 alpha (EF-1 alpha) and to prokaryotic elongation factor Tu (EF-Tu). Sequence comparison with EF-1 alpha and EF-Tu of different species indicates that thesaurin a is rather distantly related to all eukaryotic elongation factors. In spite of this, the secondary structure of thesaurin a, deduced from hydrophobic cluster analysis, is remarkably similar to that of EF-1 alpha and EF-Tu. The binding and catalytic properties of thesaurin a are also similar but not identical to those of EF-1 alpha. Like EF-1 alpha, purified thesaurin a binds tRNA, GDP, and GTP. Unlike EF-1 alpha, thesaurin a binds discharged tRNA more tightly than charged tRNA, and GTP more tightly than GDP. Thesaurin a also hydrolyzes GTP and catalyzes the mRNA-dependent binding of aminoacyl-tRNA to 80 S ribosomes. The functional properties of the 42 S particle are in general agreement with those of purified thesaurin a. In particular, the 42 S particle contains GTP and efficiently transfers aminoacyl-tRNA to 80 S ribosomes without addition of exogenous elongation factor.     

Fluorescence characterization of the interaction of various transfer RNA species with elongation factor Tu.GTP: evidence for a new functional role for elongation factor Tu in protein biosynthesis

The ubiquity of elongation factor Tu (EF-Tu)-dependent conformational changes in amino-acyl-tRNA (aa-tRNA) and the origin of the binding energy associated with aa-tRNA.EF-Tu.GTP ternary complex formation have been examined spectroscopically. Fluorescein was attached covalently to the 4-thiouridine base at position 8 (s4U-8) in each of four elongator tRNAs (Ala, Met-m, Phe, and Val). Although the probes were chemically identical, their emission intensities in the free aa-tRNAs differed by nearly 3-fold, indicating that the dyes were in different environments and hence that the aa-tRNAs had different tertiary structures near s4U-8. Upon association with EF-Tu.GTP, the emission intensities increased by 244%, 57%, or 15% for three aa-tRNAs due to a change in tRNA conformation; the fourth aa-tRNA exhibited no fluorescence change upon binding to EF-Tu.GTP. Despite the great differences in the emission intensities of the free aa-tRNAs and in the magnitudes of their EF-Tu-dependent intensity increases, the emission intensity per aa-tRNA molecule was nearly the same (within 9% of the average) for the four aa-tRNAs when bound to EF-Tu-GTP. Thus, the binding of EF-Tu.GTP induced or selected a tRNA conformation near s4U-8 that was very similar, and possibly the same, for each aa-tRNA species. It therefore appears that EF-Tu functions, at least in part, by minimizing the conformational diversity in aa-tRNAs prior to their beginning the recognition and binding process at the single decoding site on the ribosome. Since an EF-Tu-dependent fluorescence change was also observed with fluorescein-labeled tRNA(Phe), the protein-dependent structural change is effected by direct interactions between EF-Tu and the tRNA and does not require the aminoacyl group. The Kd of the tRNA(Phe).EF-Tu.GTP ternary complex was determined, at equilibrium, to be 2.6 microM by the ability of the unacylated tRNA to compete with fluorescent Phe-tRNA for binding to the protein. Comparison of this Kd with that of the Phe-tRNA ternary complex showed that in this case the aminoacyl moiety contributed 4.3 kcal/mol toward ternary complex formation at 6 degrees C but that the bulk of the binding energy in the ternary complex was derived from direct protein-tRNA interactions.(ABSTRACT TRUNCATED AT 400 WORDS)     

The solution structure of the Escherichia coli initiator tRNA and its interactions with initiation factor 2 and the ribosomal 30 S subunit

The conformation of the Escherichia coli initiator tRNA has been investigated using enzymatic and chemical probes. This study was conducted on the naked tRNA and on the tRNA involved in the various steps leading to the formation of the 30 S.IF-2.GTP.fMet-tRNA.AUG complex. A three-dimensional model of the initiator tRNA is presented, which displays several differences with yeast tRNAPhe: (i) the anticodon arm is more rigid; (ii) the presence of an additional nucleotide in the D loop results in specific features in both T and D loops; (iii) C1 and A72 might form a noncanonical base pair. Aminoacylation and formylation induce subtle conformational adjustments near the 3' end, the T arm and the D loop. Initiation factor (IF) 2 interacts with a rather limited portion of the tRNA, covering the T loop and the minor groove of the T stem, and induces an increased flexibility in the anticodon arm. The specific structural features observed in the T loop are probably recognized by IF-2. In the 30 S.IF-2.GTP.fMet-tRNA.AUG complex, additional protections are observed in the acceptor stem and in the anticodon arm, resulting from a strong steric hindrance and from the codon-anticodon interaction within the subunit decoding site.     

Conformational selection through electrostatics: Free energy simulations of GTP and GDP binding to archaeal initiation factor 2

Archaeal Initiation Factor 2 is a GTPase involved in protein biosynthesis. In its GTP-bound, "ON" conformation, it binds an initiator tRNA and carries it to the ribosome. In its GDP-bound, "OFF" conformation, it dissociates from tRNA. To understand the specific binding of GTP and GDP and their dependence on the conformational state, molecular dynamics free energy simulations were performed. The ON state specificity was predicted to be weak, with a GTP/GDP binding free energy difference of -1 kcal/mol, favoring GTP. The OFF state specificity is larger, 4 kcal/mol, favoring GDP. The overall effects result from a competition among many interactions in several complexes. To interpret them, we use a simpler, dielectric continuum model. Several effects are robust with respect to the model details. Both nucleotides have a net negative charge, so that removing them from solvent into the binding pocket carries a desolvation penalty, which is large for the ON state, and strongly disfavors GTP binding compared to GDP. Short-range interactions between the additional GTP phosphate group and ionized sidechains in the binding pocket offset most, but not all of the desolvation penalty; more distant groups also contribute significantly, and the switch 1 loop only slightly. The desolvation penalty is lower for the more open, wetter OFF state, and the GTP/GDP difference much smaller. Short-range interactions in the binding pocket and with more distant groups again make a significant contribution. Overall, the simulations help explain how conformational selection is achieved with a single phosphate group.