Biochemical detection of adenosine and cytidine ionization within RNA by interference analysis.

Perturbation of active site functional group pIs 7). Deletion or modification of the base significantly reduces the catalytic rate (>900 fold), though removal of the 2'-OH group from A756 reduces the rate of cleavage only 10 fold (8). While mutation of A756 reduces the reaction rate, it does not appear to affect folding or substrate binding (7). These results suggest an important catalytic role for A756 in VS RNA activity. Other catalytic RNAs, which perform phosphotransfer reactions equivalent to that of the VS ribozyme, are proposed to utilize a general acid base mechanism involving ionization of active site residues (911). It is possible that the VS ribozyme may also employ an ionized base to facilitate catalysis, but there is little or no evidence in support of this hypothesis. To explore this possibility, we employed a series of adenosine and cytidine analogs with N1 and N3 perturbed pKa values, respectively, in nucleotide analog interference mapping (NAIM). These analogs make it possible to simultaneously, yet individually, assay for functionally important base ionization at every A or C residue in an RNA. Each of the analogs is prepared as a triphosphate for transcriptional incorporation into an RNA and tagged with an a phosphorothioate linkage, a bond that can be cleaved with iodine to reveal the position of analog incorporation within the RNA polymer. These analogs were instrumental in the identification of a protonated C (C300) important for folding of group I introns, ionization of a C (C75) in the active site of the genomic HDV ribozyme, and the ionization of an A (A10) in the active site of the hairpin ribozyme (9; 12; 13). They are ideally suited to assay for base ionization within the VS ribozyme. RESULTS AND DISCUSSION Seven nucleotide analogues were used to explore base ionization within the ribozyme. This included four adenosine and three cytidine analogues, specifically: 8-azaadenosine (n8A, pK,=2.2), formycin A (FormA, pK,=4.4), purine riboside (Pur, pK,=2. l ) , and 7-deaza-adenosine (7dA, pKa=5.2) for adenosine; 6-azacytidine (n6C, pK,=2.6), 5-fluorocytidine ( P C , pK,=2.3) and 230 Nucleic Acids Research Supplement N o . 3 pseudoisocytidine (YiC, pK,=9.4) for cytidine. We employed a ligation based NAIM assay using the selfligation construct VSE, to probe important functional groups throughout the VS RNA. Of the 171 nucleotides in this construct, 142 nucleotides (G640-A78 1) were informative for the assay. Each of the As and Cs within the VS sequence were assayed for ionization using this series of nucleotide analogues. None of the Cs within the VS sequence displayed an interference pattern consistent with ionization. However, the interference pattern using the adenosine analog series indicated that A756 undergoes ionization during the ligation reaction. The data to support this were as follows. At pH 7, nsA and Pur incorporation caused interference at A756. Both of these nucleotides have acidic pKas compared to adenosine. By contrast, both nucleotides with elevated pKas, FormA and 7dA, caused enhanced ligation activity when incorporated at A756. To further test if the interferences at A756 was due to base ionization, we performed the NAIM assays at pHs ranging from 8.0 to 5.4. nsA interference at A756 persisted at pH 7.0 and 6.0, however at pH 5.4 the interference was essentially eliminated. Similarly, the A756 Pur interference observed at pH 7.0 was rescued at pH 5.4. Thus, both the pK, perturbed adenosine analogs demonstrated a pH dependent interference pattern at A756. The specificity of the effect was tested using dA, an analog that does not have a reduced pK, but causes interference at A756. dA interference at A756 was unaffected across the pH range, including pH 5.4. These, and previously published data, provide compelling evidence that the chemical identity of A756 is important for VS ribozyme catalysis, with regard both to the array of functional groups that it presents in the active site and to its ability to be ionized at some point in the reaction pathway. If A756 is protonated, then is the pKa of A756 by necessity perturbed? RNA nucleosides do not have functional groups with near neutral pK,s. The N1 and N3 groups of adenosine and cytidine have potential for protonation, though their pK,s are too acidic to be significantly protonated at physiological pH (14). The pKas increase by about half a pH unit when the 5' and 3'hydroxyls are phosphorylated, as is the case in an RNA polymer, but the pKas remain too low to have much effect on a reaction proceeding efficiently at neutral pH (15). However, RNAs, like proteins, can fold into tertiary structures and create microenvironments that lead to local pK, perturbations of the bases. We cannot ascertain from the NAIM data how much the A756 pK, is perturbed, but the interference rescue at pH 5.4 appear to place 5.4 as a lower limit. Based upon the principle of microscopic reversibility, if A756 plays a role as a general base (for example) in this concerted ligation reaction, then it must also function as a general acid during cleavage. Unless the pK, of the base were shifted toward neutrality it could not make a meaningful contribution to the reaction, and its ionization would have gone undetected in our assay. If A756 is ionized, as suggested by these data, then how might protonation of this adenosine promote the reaction of the VS ribozyme? There are several lines of evidence that suggest that A756 protonation plays a catalytic, rather than a structural role in the VS reaction. First, mutation of A756 to any other base results in drastic reductions in self-cleavage activity (6; 7), but structural comparison of wild type and mutant VS constructs by FRET indicate that mutation of A756 does not affect the global folding of the VS RNA. (16) Second, A756 variants bind substrate efficiently (16). Third, 4-thioU cross-linking experiments pinpoint A756 as being in close proximity to the cleavage site (17). Fourth, a FRET-derived structural model of the VS ribozyme generated by Lilley and coworkers places A756 near the substrate helix, where it can make significant tertiary interactions with nucleotides at the cleavage site (16). At the very least, these studies place A756 in a position for it to serve a catalytic function. Evidence of base ionization makes A756 an ideal candidate for a role in charge neutralization of the transition state and/or a role in proton transfer. REFERENCES I . 2. 3. Griffiths, A. J. F. (1995) Microbiol. Reviews 59, 673- 685. Saville, B. J. & Collins, R. A. (1990) Cell 61, 685-696. Guo, H. C., De Abreu, D. M., Tillier, E. R., Saville, B. J., Olive, J . E. & Collins, R. A. (1993) J . Molec. Biol. 232, 351-61. Jones, F. D., Ryder, S. P. & Strobel, S. A. (2001) Nucleic Acids Res. 29,5115-5120. Beattie, T. L. & Collins, R. A. (1997) J . Mofec. Biol. 267, 830-840. Sood, V. D. & Collins, R. A. (2002) J . Molec. B id . 320,443-454. Lafontaine, D. A., Wilson, T. J., Norman, D. G. & Lilley, D. M. J . (2001) J . Molec. Biol. 312,663-674. Lafontaine, D. A,, Wilson, T. J . , Zhao, Z. & Lilley, D. M. J. (2002) J . Molec. Biol. 323,23-34. Ryder, S . P., Oyelere, A. K., Padilla, J . L., Klostermeier, D., Millar, D. P. & Strobel, S. A. (2001) 4.