Identification of novel catalytic strategies of small ribozymes

Restricted (Penn State Only)
Author:
Bingaman, Jamie Lee
Graduate Program:
Chemistry
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
August 17, 2017
Committee Members:
  • Philip C. Bevilacqua, Dissertation Advisor
  • Philip C. Bevilacqua, Committee Chair
  • Squire J Booker, Committee Member
  • David D Boehr, Committee Member
  • Song Tan, Outside Member
Keywords:
  • ribozyme
  • glmS
  • RNA
  • catalysis
Abstract:
Since their discovery in 1981, RNA enzymes, or ribozymes, have been of interest for their potential applicability for site-specific catalysis and pathogenic targeting. The class of small ribozymes, which are generally less than 100 nucleotides in the catalytic core, utilize an internal RNA 2′OH group as the nucleophile in their self-cleavage mechanism, which results in 2′,3′-cyclic phosphate and 5′OH termini. There have been nine classes of small ribozymes identified thus far, but the recent increase in discovery rate due to bioinformatic searches suggests that more will be found in the near future. The glmS ribozyme is one class of small ribozymes found in the untranslated region of Gram-positive bacteria. The ribozyme is unique in that it requires binding of a small molecule cofactor, glucosamine-6-phosphate (GlcN6P), for activity, making it a riboswitch-ribozyme. Previous biochemical studies have suggested a role for GlcN6P as a general acid-base catalyst in the glmS ribozyme self-cleavage mechanism. Structural studies support this finding and show that the GlcN6P binds in such a way that it is positioned to protonate the 5′O leaving group and act as the general acid. In addition, structural and biochemical studies support a role for an active-site guanine, G33, in the self-cleavage mechanism. Though G33 is found positioned in crystal structures to act as the general base, there is still uncertainty as to whether G33 acts directly as the general base or as a proton donor to the 2′OH while some other moiety acts as the base. Studies from our lab in collaboration with Sharon Hammes-Schiffer’s lab suggest that the pKa of G33 is shifted away from neutrality, supporting its role as a proton donor to the 2′OH until either another moiety abstracts the proton from the nucleophile or nucleophilic attack by the 2′OH proceeds enough for the pKa of the 2′OH to be lowered and the proton to transfer to G33 (Zhang, S.; Ganguly, A.; Goyal, P.; Bingaman, J.L.; Bevilacqua, P.C.; Hammes-Schiffer, S. J. Am. Chem. Soc. 2015, 137, 784). We also probed the effect of a Mg2+ ion at the active-site of the glmS ribozyme in collaboration with the Hammes-Schiffer lab and found that although direct coordination with a non-bridging oxygen atom is detrimental due to electrostatic repulsion of the cofactor, obstruction of nucleophilic attack, and disruption of hydrogen bonding interactions, Mg2+ binding to the Hoogsteen face of could be beneficial to catalysis due to lowering of the pKa of G33 (Zhang, S.; Stevens, D. R.; Goyal, P.; Bingaman, J.L.; Bevilacqua, P.C.; Hammes-Schiffer, S. J. Phys. Chem. Lett. 2016, 7, 3984). The work in this thesis is geared towards uncovering new strategies that the glmS ribozyme, and small ribozymes in general, use in their catalytic mechanisms. Chapter 2 is focused on understanding the multifunctionality of the GlcN6P cofactor in the glmS ribozyme using scissile phosphate non-bridging oxygen (NBO) atom (pro-RP and pro-SP) thio substitutions, which have traditionally been used to test for metal ion binding. Previous studies have suggested a role for the cofactor as the general acid in the glmS ribozyme self-cleavage mechanism. Large, normal thio effects in the holoribozyme (in the presence of GlcN6P) that are not rescuable by thiophilic metal ions suggest that the GlcN6P cofactor and other nucleobases help to align the active site for catalysis. A large, stereospecific inverse thio effect in the aporibozyme (in the absence of GlcN6P) suggests that the 1-hydroxyl group of GlcN6P prevents an inhibitory interaction between the 2′OH and the pro-RP NBO atom. Additionally, thiophilic metal ion rescue in the aporibozyme with the RP and dithio substrates, the latter of which contains a sulfur substitution at both NBO atoms of the scissile phosphate, suggests the amine group of GlcN6P provides charge stabilization at the pro-RP NBO atom during self-cleavage. The experimental results presented in this chapter are supported by computations performed by Sharon Hammes-Schiffer’s lab. This work demonstrates that RNA can compensate for its lack of structural diversity by recruiting exogenous organic cofactors, which contribute in myriad ways to ribozyme catalysis. The focus of Chapter 3 is the mechanism behind activation of the glmS ribozyme through prevention of the 2′OH to NBO inhibitory interaction observed in Chapter 2. A series of ribozyme variants are prepared where certain active-site guanines that contact the NBO atoms are changed to adenines and a deoxy GlcN6P analog is synthesized to probe the effect of deleting various hydrogen bond donors to the NBO atoms of the scissile phosphate. Thio substitutions are used in an unconventional way to probe hydrogen bonding within the active site. Thio substitutions in the background of single G32A, G57A, and 1-deoxy-GlcN6P variants show diminished normal thio effects for the thio substrates whose mutated atom contacts the varied moiety in the wild-type (WT) holoribozyme. However, upon double variation in a G57A+1-deoxy-GlcN6P ribozyme and a G57A aporibozyme, the RP substrate, whose mutated atom contacts G57 and GlcN6P in the WT holoriboyzme, shows a large inverse thio effect, similar to the inverse thio effect observed for the WT aporibozyme in Chapter 2. These results suggest that the glmS ribozyme has evolved an overdetermined set of hydrogen bond donors within the active-site to compete with the 2′OH nucleophile. This in turn activates the nucleophile, which could be a general strategy for ribozymes. Chapter 4 aims to unify our understanding of small ribozyme catalysis through the evaluation of catalytic strategy usage based on published crystal structures. Custom Python plugins are used to probe small ribozyme crystal structures for their use of the in-line nucleophilic attack (α), charge stabilization at the NBO atoms (β), deprotonation of the 2′OH nucleophile (γ), and protonation of the 5′O leaving group (δ) strategies enumerated by Breaker. Results presented in this chapter reveal that evidence for these strategies is found primarily at the scissile phosphate and not at other phosphates throughout the ribozymes. Furthermore, this work suggests two additional strategies used by the small ribozymes for activation of the 2′OH nucleophile: (1) acidification of the nucleophile through hydrogen bond donation to the 2′OH (γ′) and (2) release of the 2′OH from inhibitory interactions through hydrogen bond donation to the NBO atoms (γ′′), the latter of which was investigated in Chapter 3. These findings reveal the conservation of catalytic strategies across small ribozymes and show how ribozymes selectively activate the 2′OH nucleophile at the scissile phosphate. The focus of Chapter 5 is the contribution of exogenous species on the self-cleavage of the glmS ribozyme. The glmS ribozyme is known to misfold and dimerize from work in our lab and other labs. In this chapter, Mg2+ ions, buffer molecules, and macromolecular crowders are found to enhance glmS ribozyme folding, catalysis, or both. These results suggest that the ribozyme might avoid misfolding and react efficiently in the cell by binding species in the cellular milieu. Additionally, a new method for detecting ionized guanines is presented in this chapter. This method involves chemical modification of the Watson-Crick face of guanines with glyoxal, a reagent that is sensitive to the ionization state of the guanine at the N1 position. Chemical probing with glyoxal in the glmS ribozyme reveals that the putative general base, G33, does not show enhanced reactivity, and in fact shows less reactivity, compared to other guanines in the ribozyme. This suggests that G33 does not have a pKa shifted towards neutrality and may in fact have one shifted away from neutrality, which is consistent with previous pH-dependent studies involving a guanine analog. This method could be applied to other small ribozymes to test for the presence of guanines with both down-shifted and up-shifted pKas. Overall, the work presented in this thesis provides mechanistic insight into novel catalytic strategies used by ribozymes and suggests that though these ribozymes may be limited structurally, they have developed ways of attaining catalytic proficiency.