comment

Description

TECHNICALCOMMENT– transition statelikely to contain an MgF3Commenton“ThePentacovalent analog, rather than to represent the high-energy phosphorane intermediate suggestedin (1). This change in interpretation is crucial,PhosphorusIntermediateofabecause the structural analog can be inter-preted with equal validity as a mimic of thePhosphorylTransferReaction”transition state for a concerted reaction, as aphosphorane intermediate, or as a metaphos-The claim by Lahiri et al.( 1) to have iden- some of the attributes of various Fourier phate species in a dissociative process. Thetified the pentacovalent phosphorus interme- maps described by Lahiri et al.( 1) did not difference between these systems is simply an– moiety.diate of a phosphoryl transfer reaction ad- seem consistent with a PO few tenths of an angstrom in the apical bond3dresses an important, long-disputed problem: Is there an alternative, better interpreta- distances and cannot be resolved solely onwhether the enzyme-catalyzed transfer of tion of these data? The crystallization con- the basis of a model structure. Therefore, wephosphate is a dissociative or associative pro- ditions used by Lahiri et al.( 1) contained believe that the catalytic mechanism of PGM2 and 100 mM NH F. Ourcess. The apparent observation of a pen- 10 mM Mg remains unidentified.4taoxy-phosphorane suggests that the transfer published results (5) have shown that Seeing may be believing (8)—but only– is a good mimic of the ...

Informations

• Publié par: Onsa
• Nombre de lectures: 16
• Langue: English

Extrait

T ECHNICAL C OMMENT – likely to contain an MgFtransition state 3 Comment on “The Pentacovalent analog, rather than to represent the high energy phosphorane intermediate suggested Phosphorus Intermediate of a in (1). This change in interpretation is crucial, because the structural analog can be inter preted with equal validity as a mimic of the Phosphoryl Transfer Reaction” transition state for a concerted reaction, as a phosphorane intermediate, or as a metaphos T he claim by Lahiriet al. (1phate species in a dissociative process. The) to have idensome of the attributes of various Fourier tified the pentacovalent phosphorus intermemaps described by Lahiriet al. (1difference between these systems is simply a) did not n– diate of a phosphoryl transfer reaction adseem consistent with a PO moiety. few tenths of an angstrom in the apical bond 3 dresses an important, longdisputed problem:Is there an alternative, better interpretadistances and cannot be resolved solely on whether the enzymecatalyzed transfer oftion of these data? The crystallization conthe basis of a model structure. Therefore, we phosphate is a dissociative or associative proditions used by Lahiriet al. (1) containedbelieve that the catalytic mechanism of PGM 2 cess. The apparent observation of a pen10 mM Mg and 100 mM NHF. Our remains unidentified. 4 taoxyphosphorane suggests that the transferpublished results (5) have shown thatSeeing may be believing (8only)— but – occurs through an associative, addition–MgF isa good mimic of the transitionwhen you truly apprehend the object itself! 3 elimination process and contrasts with thestate of phosphoryl transfer reactions and G. Michael Blackburn solution mechanism, which is dissociative incan be formed under conditions such as Nicholas H. Williams character and has little bond order to thethose employed by Lahiriet al. (1). Our Krebs Institute nucleophile and leaving group in the transicrystal structure of a small GTPase com University of Sheffield tion state (2with its GTPaseactivating protein). plexed Sheffield, S3 7HF, UK – However, the apical P–O bond lengths for(GAP) shows that MgF sits in the active Email: [e-mail]; 3 this trigonal bipyramidal intermediate (2.0site and adopts a trigonal bipyramidal ar [e-mail] and 2.1 Å) are significantly longer than thoserangement with similar apical bond lengths Steven J. Gamblin in smallmolecule phosphorane structures(2.0 and 2.3 Å) to the nucleophile and Stephen J. Smerdon [e.g., 1.65 Å and 1.67 Å for pentacyclohexyleaving group as those assigned to the National Institute for Medical Research loxyphosphorane (3)]. These distances are“phosphorane” above. It therefore seems Mill Hill not unreasonable for a transition state strucplausible that the PGM structure contains – n– London, NW7 1AA, UK ture with partial apical bonds (4), but byan MgFrather than a POspecies. We 3 3 Email: [e-mail]; definition this is not a stable species. Even atfurther note that fluoride catalyses the [e-mail] 93 K, the halflife for decomposition of andephosphorylation of a phosphoenzyme 11 activated complex iswithout affecting activity (so this cannot10 s,6), furnishing a References and Notes be a kinetically trapped transition state. To bekinetic pathway for loss of phosphate from 1. S. D. Lahiri, G. Zhang, D. Dunaway-Mariano, K. N. thermodynamically viable, such a TS comthe PGM complex. Allen,Science299, 2067 (2003). 2. A.C. Hengge, inComprehensive Biological Catalysis, plex would need to be comparable in stabilityWe have previously confirmed that our G M. Sinnott, Ed. (Academic Press, San Diego, CA, – with the starting and product complexes,protein complex really contained the MgF 3 1997), vol. 1, p. 517. which implies that the kinetic barrier to reacmoiety by protoninduced xray emission spec E. Davies, A. J. Kirby, C. D. Roussev,3. J.Acta Crystallogr. tion has disappeared (and so the transfertroscopy (PIXE) (5). This technique is capable E57(no. 11), o994 (2001). 4. A.L. Mildvan,Proteins29, 401 (1997). reaction should become ultrafast at low temof establishing directly the ratio of P/Mg in the 5. D. L. Grahamet al.,Chem. Biol.9, 375 (2002); perature) and that the thermodynamic propcrystals of the PGM complex in question. If its coordinates deposited with PDB code 1OW3. erties of the transition state are very unusualstructure is indeed a phosphorane intermediate, P. Layne, V. A. Najjar,6. P.J. Biol. Chem.250, 966 (1975). – 7. The proximate amino acids (V9, D10, A115, K145) (because the change in temperature from 291the P/Mg ratio will be 2:1; if it is an MgF 3 make shorter hydrogen bonds to the equatorial fluo- to 93 K would have to lead to a decrease incomplex, the ratio will be 1:2. rines than those described in (1). ‡1 The highresolution crystallographic datafor the reactionsome 12 kcal molG of R. Knowles,8. J.Science299, 2002 (2003). within the ES complex). From a structuralof Lahiriet al. (1) are fully compatible (7) point of view, we were also concerned thatwith our analysis that the PGM structure is 10 April 2003; accepted 3 July 2003

www.sciencemag.org SCIENCE VOL301 29AUGUST 2003

1184c