Xinhua Ji, Ph.D.

Biomolecular Structure and Mechanism, Structure-Based Drug Design

Our research is focused on the structural biology of RNA biogenesis, with an emphasis on RNA-processing proteins and RNA polymerase-associated transcription factors, and structure-based development of therapeutic agents. The goal of structural analysis is to map the reaction trajectory or functional cycle of selected biological macromolecules, and that of drug discovery is to design, synthesize, and characterize novel anticancer and antimicrobial agents. To date, we have characterized the reaction trajectory and/or functional cycle of HPPK (a folate pathway enzyme essential for microorganisms but absent in mammals, View), Era (an essential GTPase that couples cell growth with cell division, View), RapA (a Swi2/Snf2 protein that recycles RNA polymerase, View), bacterial RNase III, and yeast RNase III (Rnt1p). Several biomolecules mentioned above are attractive molecular targets and structure-based drug development is an integral part of our research. See our Science Gallery and Drug Discovery Patents for the scope and depth of our science. Our contributions of RNase III research is summarized.

RNase III (ribonuclease III) enzymes, exemplified by prokaryotic RNase III and eukaryotic Rnt1p, Dcr1, Dicer, and Drosha, play important roles in RNA processing and maturation, post-transcriptional gene silencing, and defense against viral infection. For mechanistic studies, bacterial enzyme is a valuable model system for the entire family. We have shown how the dimerization of the  RNase III endonuclease domain (RIIID) creates a catalytic valley where  two cleavage sites are located, how the catalytic valley accommodates a dsRNA in a manner such that each of the two RNA strands is aligned with one of the two cleavage sites, how the  hydrolysis of each strand involves both RIIIDs, and how RNase  III uses the two cleavage sites to create the 2-nucleotide 3' overhangs in its products (View). We have also shown how magnesium is essential for the formation of a catalytically competent protein-RNA complex, how the use of two magnesium ions can drive the hydrolysis of each phosphodiester bond, and how conformational changes in both the substrate and the protein are critical elements for assembling the catalytic complex. Moreover, we have provided a stepwise mechanism for the enzyme to execute the phosphoryl transfer reaction (View). As informative as the bacterial enzyme for the mechanism of RNase III action, yeast Rnt1p is a valuable model system for eukaryotic RNase III enzymes. Unlike bacterial enzymes that use four catalytic side chains, eukaryotic RNase IIIs use six. It is also distinguished from bacterial enzymes that every eukaryotic RNase III has an N-terminal extension. What is more, Rnt1p exhibits a strict guanine nucleotide specificity, which is unique among RNase III enzymes. We have shown how the substrate-binding mode of Rnt1p is  distinct from that of bacterial RNase III (View), how all of the six catalytic side chains are engaged in the cleavage site (View), how a new RNA-binding motif of Rnt1p functions as a guanine-specific clamp (View), and how the double-stranded RNA-binding domain and N-terminal domain of Rnt1p  function as two rulers measuring the distance between the guanine nucleotide to the cleavage sites (View). This unusual mechanism of substrate selectivity represents an example of the evolution of substrate selectivity and provides a framework for understanding the mechanism of action of other eukaryotic RNase  III enzymes, including Dcr1, Dicer, and Drosha.

The worldwide effort in structural analysis of other eukaryotic RNase III enzymes resulted in several important structures, including the crystal structures of Dicer (View), Dcr1 (View), and Drosha (View). These structures, however, do not contain RNA and thus are not able to explain their mechanisms of action. Our structures of RNase III:dsRNA complexes greatly enhanced the significance of these important structures. Based on the protein-RNA interactions revealed by our structures of RNase III and Rnt1p, models with RNA can be reliably constructed for Dicer, Dcr1, and Drosha. A model complex of Dicer with RNA explains how Dicer enzymes recognize the 2-nucleotide 3' overhang of dsRNA substrate and measure 22 nucleotides up to position the scissile bond over the cleavage site. A model complex of Dcr1 with RNA explains how homodimers of non-canonical Dicer enzymes bind cooperatively along dsRNA substrate such that the distance between active centers in adjacent homodimers is the length of 22 nucleotides. A model complex of Drosha with RNA explains how Drosha enzymes recognize the last base pair in the basal junction of the primary microRNA substrate and measure 11 nucleotides up to position the scissile bond over the cleavage site.