Ribozyme-Catalyzed and Nonenzymatic Reactions of Phosphate Effects upon Substitution of Sulfur for a Nonbridging Phosphoryl

The L-2 1 ScaI ribozyme derived from the intervening sequence of Tetrahymena thermophila pre-rRNA catalyzes a guanosine-dependent endonuclease reaction that is analogous to the first step in self-splicing of this intervening sequence. We now describe pre-steady-state kinetic experiments, with sulfur substituting for the pro-Rp (nonbridging) phosphoryl oxygen atom at the site of cleavage, that test aspects of a kinetic model proposed for the ribozyme reaction (Herschlag, D., & Cech, T. R. (1 990) Biochemistry 29, 10159-10171). Thio substitution does not affect the reaction with subsaturating oligonucleotide substrate and saturating guanosine ((kat/K,,JS), consistent with the previous finding that binding of the oligonucleotide substrate limits this rate constant. In contrast, there is a significant decrease in the rate of single-turnover reactions of ribozyme-bound (Le., saturating) oligonucleotide substrate upon thio substitution, with decreases of 2.3-fold for the reaction with guanosine ((k,t/Km)G) and 7-fold for hydrolysis [i.e., with solvent replacing guanosine; k,(-G)]. These “thio effects” are consistent with rate-limiting chemistry, as shown by comparison with model reactions. Nonenzymatic nucleophilic substitution reactions of the phosphate diester, methyl 2,4-dinitrophenyl phosphate monoanion, are slowed 4-1 1 -fold by thio substitution for reactions with hydroxide ion, formate ion, fluoride ion, pyridine, and nicotinamide. In addition, we have confirmed that thio substitution has no effect on the nonenzymatic alkaline cleavage of R N A (Burgers, P. M. J., & Eckstein, F. (1979) Biochemistry 18, 592-596). Considering the strong preference of Mg2+ for binding to oxygen rather than sulfur, the modest thio effect on the chemical step of the ribozyme-catalyzed reaction and the absence of a thio effect on the equilibrium constant for binding of the oligonucleotide substrate suggest that the pro-Rp oxygen atom is not coordinated to Mg2+ in the E4 complex or in the transition state. General implications of thio effects in enzymatic reactions of phosphate diesters are discussed. A sulfur atom has been substituted for a nonbridging oxygen atom of phosphate diester substrates in many enzymatic reactions, generally to determine the stereochemical course of these reactions (for review see Knowles (1980), Eckstein (1985), and Frey (1989)). In addition, “thio substitution” has been used to probe the rate-limiting step (Bryant et al., 1983; Mizrahi et al., 1985; McSwiggen & Cech, 1989; Griep et al., 1990; Pate1 et al., 1991; Wong et al., 1991). The latter approach relies on the expectation that substitution of sulfur will slow the nonenzymatic or “intrinsic” chemical reactivity of the phosphate diester; this expectation is derived from studies with phosphate triesters (Benkovic & Schray, 1973, 1978). The absence of a rate decrease in an enzymatic reaction could then be considered evidence for a rate-limiting step other than chemistry, such as substrate binding, product release, or a conformational change. The rate effect upon substitution of sulfur is commonly referred to as a “thio effect” (Le., thio effect A ribozyme, or RNA enzyme, derived from the intervening sequence of Tetrahymena thermophila pre-rRNA catalyzes the transesterification reaction shown in Figure 1 (Zaug et al., 1986). The substrate G,CCCUCUAGU forms a perfect “matched” duplex with the 5’ exon binding site. This reaction exhibits a small thio effect of 52-fold on the steady-state rate = kphapphateIkphcsphorothioate). ‘Supported in part by NIH Grant GM 28039 to T.R.C. and a grant *This paper is dedicated to Bob Abeles, in honor of his 65th birthday. * Author to whom correspondence should be addressed. 8 D.H. is a Lucille P. Markey Scholar in Biomedical Science and was supported in part by a postdoctoral fellowship from the Helen Hay Whitney Foundation. from the Lucille P. Markey Charitable Trust to D.H. 0006-296019 1 10430-4844$02.50/0 parameters when a sulfur atom is substituted for the pro-Rp oxygen at the cleavage site (Figure 1B; McSwiggen & Cech, 1989). A detailed kinetic analysis has shown that, with subsaturating concentrations of the oligonucleotide substrate and saturating guanosine (i.e., (k,t/Km)s), binding of the oligonucleotide substrate is rate-limiting; with saturating oligonucleotide substrate (i.e., kcat), release of the product G,CCCUCU is rate-limiting (Herschlag & Cech, 1990a).’ Furthermore, the chemical step2 is calculated to be very fast (-350 m i d ) relative to substrate and product dissociation (0.2 and 0.1 m i d , respectively), so that even if the chemical step were limited by the chemical reaction and not a conformational change, a large decrease in the rate of the chemical reaction upon substitution by sulfur would have been invisible to the steady-state analysis with saturating G.3 Thus, no significant thio effect would be expected, consistent with the I The kinetic analysis was performed with a substrate with AS rather than AGU at its 3’ end. However, as shown herein, the endonuclease reactions of these substrates have the same kinetic parameters. This is reasonable because the L-21 ScaI ribozyme doesn’t base pair with the 3’ of the cleavage site (see Figure 1) and doesn’t appear to have any other net interaction with them as judged by equilibrium binding measurements (Herschlag 8c Cech, 1990a,b; Pyle et al., 1990). The term ‘chemical step” is used to refer to all steps that involve the ternary complex, i.e, the chemical reaction and any associated conformational steps. Abbreviations: G, guanosine; S*, 5’-3zPP-end-labeled substrate (Le., p*G,CCCUCUAGU and the analogous specifically thio-substituted oligonucleotide); MES, 2-(N-morpholino)ethanesulfonic acid; CHES, (cyclohexylamino)ethanesulfonic acid; PAGE, polyacrylamide gel electrophoresis; EDTA, (ethylenedinitri1o)tetraacetic acid; MDP, methyl 2,4-dinitrophenyl phosphate; MDPS, methyl 2,4-dinitrophenyl phosphorothioate; DNP, 2,4-dinitrophenol.