Technical Comments

Comment on “The Mechanism for Activation of GTP Hydrolysis on the Ribosome”

See allHide authors and affiliations

Science  01 Jul 2011:
Vol. 333, Issue 6038, pp. 37
DOI: 10.1126/science.1202472


Voorhees et al. (Reports, 5 November 2010, p. 835) determined the structure of elongation factor Tu (EF-Tu) and aminoacyl–transfer RNA bound to the ribosome with a guanosine triphosphate (GTP) analog. However, their identification of histidine-84 of EF-Tu as deprotonating the catalytic water molecule is problematic in relation to their atomic structure; the terminal phosphate of GTP is more likely to be the proper proton acceptor.

Voorhees et al. (1) determined the structure of the 70S ribosome bound to elongation factor Tu (EF-Tu) in a ternary complex with the guanosine triphosphate (GTP) analog β-γ-methyleneguanosine 5′-triphosphate (GDPCP) and aminoacyl–transfer RNA (tRNA). Their study has greatly increased our understanding of GTP hydrolysis by the translational guanosine triphosphatases (trGTPases). EF-Tu is a very poor GTPase off the ribosome, but if the anticodon of the tRNA matches the codon of the mRNA, GTP hydrolysis is rapidly induced, leading to dissociation of EF-Tu and guanosine diphosphate (GDP) from the ribosome and A-site entry of the tRNA (2).

In most members of the large family of GTPases, a glutamine in a specific loop of the protein (switch II) is essential for induction of GTP hydrolysis (3). By interaction with a protein, called GTPase activating protein (GAP), the Gln residue places a water molecule at the γ-phosphate of the GTP molecule. The mechanism of activation of the water molecule for hydrolysis has been intensely debated. Now there seems to be a consensus that the γ-phosphate of the GTP molecule acts as a general base by removing a proton from the water molecule, enabling it to attack the γ-phosphate and hydrolyze the phosphate ester (46).

For the trGTPases, the part of the system that corresponds to the GAP has not previously been identified. Several alternatives have been discussed. Voorhees et al. (1) demonstrate that the phosphate of A2662 of the 23S RNA is the component that corresponds to the GAP. A2662 is part of a universally conserved sequence, the sarcin-ricin loop (SRL). This was first identified in eukaryotes as a functional region of the rRNA. When covalently modified by the ribotoxins α-sarcin and ricin, the SRL cannot induce GTP hydrolysis in translation factors EF-1 and EF-2 or their bacterial counterparts, EF-Tu and EF-G, respectively. A2662 is at the tip of the SRL in a GAGA tetraloop sequence (7). In trGTPases, the glutamine of switch II is replaced by a histidine (in EF-Tu, His84). Voorhees et al. (1) reported that the role of A2662 is to interact with His84 and move it into position to place the water molecule close to the γ-phosphate of the GTP molecule. In numerous previously published structures, this histidine is seen in many different positions but rarely, if ever, at the water molecule near to the γ-phosphate. Two residues (Ile60 and Val21) have been implicated as a “hydrophobic gate,” preventing the histidine from accessing the water molecule. But Voorhees et al. (1) argue that it is the A2662-dependent positioning of His84 and the water molecule, rather than the opening of the hydrophobic gate, that allows for rapid hydrolysis of GTP. The authors suggest that His84 acts as a general base and activates the water molecule by abstracting one of its protons. The water molecule can thereby attack the γ-phosphate and hydrolyze GTP to GDP and inorganic phosphate, Pi. This scenario seems unlikely for several reasons.

To illustrate, consider the charges and hydrogen bonds in the GTPase hydrolysis center. The resolution of the current structure does not allow any detailed discussion of hydrogen bonds, but, because all the bonds of interest are on the short side, even errors of several tenths of an Ångström still do not make them unlikely as hydrogen bonds. Initially, the γ-phosphate is not protonated. The γ- and β-phosphates interact with a magnesium ion and a lysine from the switch I loop (Fig. 1). The water molecule can donate two hydrogen bonds: one to the carbonyl oxygen of Thr61 and one to the oxygen of the γ-phosphate. In the hydrogen bond to His84, the water molecule must be an acceptor. His84, positioned between two negatively charged phosphate groups, must have an elevated pKa (where Ka is the acid dissociation constant) value and is thus likely to be protonated and positively charged. His84 donates one hydrogen bond through its Nε to the GAP function, the phosphate of A2662, and another through its protonated Nδ to the water molecule. When His84 is fully protonated, it cannot abstract a water proton, suggesting another GTPase mechanism than the one proposed by Voorhees et al. (1). The reaction, we suggest, instead proceeds by γ-phosphate removal of a proton from the water molecule like in the part of the GTP-binding protein (G protein) family with a glutamine instead of a histidine. The substrate-generated hydroxide ion attacks the γ-phosphate, either concerted with the proton transfer or in a stepwise mechanism leading to GTP hydrolysis. This agrees with a general mechanism for GTPases proposed by Schweins et al. (4) and is in line with a previous suggestion for GTP hydrolysis on EF-Tu in the ribosome-bound ternary complex by (8).

Fig. 1

The atomic structure around the phosphates of the GTP molecule in EF-Tu (1). The histidine is firmly bound to a unique place because of its interaction with the phosphate of A2662 in the sarcin-ricin loop. The water molecule is moved into contact with the γ-phosphate and donates hydrogen bonds to the carbonyl oxygen of Thr61 and the γ-phosphate. The water molecule is induced to donate a proton to the γ-phosphate and attack it, leading to the hydrolysis of the GTP molecule. [Credit: Saraboji Kadhirvel]

Guanosine 5′-O-(γ-thio)triphosphate (GTP-γ-S) hydrolysis in the ribosome-bound ternary complex is not affected by a pH variation in the 6.5 to 8.5 range (8), in line with a substrate-activated water molecule but not with His84 acting as a general base. Furthermore, mutation of His84 to Ala84 reduces the rate of GTP hydrolysis in ribosome-bound ternary complex by six orders of magnitude, whereas a mutation to Gln84, the classical GTPase residue in this position, only moderately reduces the rate of GTP hydrolysis (8). The latter finding is important, because Gln84 cannot function as a general base. It is therefore likely that trGTPases hydrolyze GTP by the same general mechanism as all GTPases.

For the trGTPases, the search for an arginine finger has been unsuccessful (9). It seems possible that the positively charged histidine could also take the role of this arginine and stabilize the transition state.

In conclusion, we consider the Voorhees et al. (1) study a milestone in the understanding of GTP hydrolysis by the trGTPases but suggest that the detailed mechanism proposed be revised in line with the general GTPase proposal by Schweins et al. (4).

References and Notes

  1. Acknowledgments: We greatly appreciate the kind assistance of S. Kadhirvel.
View Abstract

Navigate This Article