RNases H: Structure and mechanism

Abstract

RNases H are a family of endonucleases that hydrolyze RNA residues in various nucleic acids. These enzymes are present in all branches of life, and their counterpart domains are also found in reverse transcriptases (RTs) from retroviruses and retroelements. RNases H are divided into two main classes (RNases H1 and H2 or type 1 and type 2 enzymes) with common structural features of the catalytic domain but different range of substrates for enzymatic cleavage. Additionally, a third class is found in some Archaea and bacteria. Besides distinct cellular functions specific for each type of RNases H, this family of proteins is generally involved in the maintenance of genome stability with overlapping and cooperative role in removal of R-loops thus preventing their accumulation. Extensive biochemical and structural studies of RNases H provided not only a comprehensive and complete picture of their mechanism but also revealed key basic principles of nucleic acid recognition and processing. RNase H1 is present in prokaryotes and eukaryotes and cleaves RNA in RNA/DNA hybrids. Its main function is hybrid removal, notably in the context of R-loops. RNase H2, which is also present in all branches of life, can play a similar role but it also has a specialized function in the cleavage of single ribonucleotides embedded in the DNA. RNase H3 is present in Archaea and bacteria and is closely related to RNase H2 in sequence and structure but has RNase H1-like biochemical properties.

This review summarizes the mechanisms of substrate recognition and enzymatic cleavage by different classes of RNases H with particular insights into structural features of nucleic acid binding, specificity towards RNA and/or DNA strands and catalysis.

Introduction

RNases H are metal-dependent endonucleases that hydrolyze RNA residues in double-stranded nucleic acids with some sequence preference but not strict sequence specificity. They are present in all domains of life and belong to the retroviral integrase superfamily (RISF), comprising various enzymes that are involved in the metabolism of nucleic acids [1]. This large and evolutionarily ancient group includes both exo- and endonucleases, retroviral integrases, DNA transposases, Holliday junction resolvases, Argonaute, and spliceosomal protein Prp8 [2]. Moreover, structural similarities with RISF proteins were found in prokaryotic CRISPR-Cas proteins [3,4]. All of these enzymes possess catalytic domains that adopt the so-called RNase H fold. The central element of this fold is a mixed five-stranded β-sheet, with strands arranged 32145 and strand 2 running antiparallel to the others (Fig. 1). Strands 4 and 5 are usually shorter than strands 1–3 [1].

RNases H are divided into two main classes: RNases H1 and H2 (or type 1 and type 2 enzymes) (Fig. 2). These classes have common structural features of the catalytic domain but different ranges of substrates that they cleave. A third class, RNases H3, is found in some Archaea and bacteria (Fig. 2). RNases H3 have close structural similarity to RNases H2. In terms of substrate preference, however, they are similar to type 1 RNases H. Additionally, RNase H1 domains are found in multidomain reverse transcriptases (RTs) from retroviruses and retroelements [5].

Hydrolysis of the phosphodiester bond by RNases H results in the formation of 3′ hydroxyl and 5′ phosphate groups (Fig. 3). The active site of RNases H is composed of four highly conserved carboxylates that form a motif called DEDD (Fig. 1, Fig. 2). The first three residues form a core of the active site that is conserved in all RNases H. The negatively charged active site binds divalent metal ions that are essential for catalysis which occurs through a two-metal-ion mechanism. Mg²⁺ is the preferred metal ion, but other ions, such as Mn²⁺, can also support hydrolysis.