How hydrolytic exoribonucleases impact human disease: Two sides of the same story

Abstract

RNAs are extremely important molecules inside the cell, which perform many different functions. For example, messenger RNAs, transfer RNAs and ribosomal RNAs are involved in protein synthesis, whereas noncoding RNAs have numerous regulatory roles. Ribonucleases (RNases) are the enzymes responsible for the processing and degradation of all types of RNAs, having multiple roles in every aspect of RNA metabolism. However, the involvement of RNases in disease is still not well understood. This review focuses on the involvement of the RNase II/RNB family of 3'-5' exoribonucleases in human disease. This can be attributed to direct effects, whereby mutations in the eukaryotic enzymes of this family [defective in sister chromatid joining (Dis3; or Rrp44), Dis3-like exonuclease 1 (Dis3L1; or Dis3L) and Dis3-like exonuclease 2 (Dis3L2)] are associated with a disease, or indirect effects, whereby mutations in the prokaryotic counterparts of RNase II/RNB family (RNase II and/or RNase R) affect the physiology and virulence of several human pathogens. In this review, we compare the structural and biochemical characteristics of the members of the RNase II/RNB family of enzymes. The outcomes of mutations impacting enzymatic function are revisited, in terms of both the direct and indirect effects on disease. Furthermore, we also describe the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) viral exoribonuclease and its importance to combat the COVID-19 pandemic. As a result, RNases may be a good therapeutic target to reduce bacterial and viral pathogenicity. These are the two perspectives on RNase II/RNB family enzymes that are presented in this review.

Keywords: Dis3L2; RNA decay; RNase II/RNB family; human disease; pathogens; viral exoribonuclease.

Fig. 1

3′–5′ RNA decay (via hydrolysis) in prokaryotic and eukaryotic human cells. Left panel: In bacteria, the decay of the majority of RNA transcripts starts with an endoribonucleolytic cleavage (e.g. RNase E; faded image). After this initial step, the breakdown products can be hydrolytically degraded by 3′–5′ exoribonucleases from the RNB family of enzymes (RNase II and/or RNase R). This exoribonucleolytic activity can be promoted by the 3′‐polyadenylation of RNAs. RNase II accounts for 90% of RNA degradation in Gram‐negative bacteria, and it is sensitive to secondary structures. RNase R is a cold‐shock protein able to degrade structured RNA transcripts. Both enzymes generate end products of 4 and 2 nts, respectively. 3′–5′ exoribonucleolytic RNA degradation can be phosphorolytic (faded image). In that case, PNPase acts alone or together with other enzymes in a degradation complex called the degradosome. Right panel: In eukaryotic cells, mature cytoplasmic mRNAs are usually stabilised by a 5′ cap and a long 3′ poly(A) tail, to which poly(A) binding proteins (PABPs) may bind, thereby protecting that extremity from degradation. If mRNAs possess premature termination codons, they will probably be cleaved by an endoribonuclease (endoRNase, represented in the pathway on the left of the figure), with the resulting fragments being further digested by exoribonucleases (exoRNases) starting on the extremities generated by the internal disruption of the transcript. However, mRNA decay starts more frequently through partial or complete deadenylation (at the top centre of the figure), which is carried out by deadenylases (e.g. Ccr4‐Not and Pan‐2/Pan‐3 complexes). Subsequently, deadenylated mRNAs can be further decapped (represented at the bottom centre of the figure) or be readily degraded in the 3′–5′ direction by the RNA exosome. Decapping is stimulated by the Lsm1‐7/Pat‐1 complex, which recruits the decapping complex Dcp‐1/Dcp‐2. This step enables degradation in the 5′–3′ direction by enzymes of the 5PX superfamily: Xrn1 in the cytoplasm or Xrn2 in the nucleus. ‘*’ indicates that, despite Xrn1 being able to unwind highly structured RNA molecules, some of them block Xrn1 progression. The RNA exosome has distinct isoforms, which exhibit differences in cellular localization: (a) in the nucleolus, the Exo‐9 inert core binds to Rrp6 (an exoRNase from Escherichia coli RNase D family); (b) in the nucleoplasm, the major catalytic subunit is Dis3 (an exoRNase, which also possesses endoRNase activity and belongs to the RNB/RNase II family); (c) in the cytoplasm, the major subunit is Dis3L1 (an exoRNase of the RNB/RNase II family), but the nucleoplasmic isoform of the RNA exosome (with Dis3 instead of Dis3L1 enzyme associated with Exo‐9) may also be present in smaller amounts (dashed grey line in the box). Of note, Rrp6 is known to be present in all human isoforms of the RNA exosome, but it is absent in the cytoplasmic isoform of yeast. ‘?’ means there is still no evidence that Dis3L1 degrades secondary structures. Apart from the abovementioned pathways, transcripts can be targeted for uridylation by terminal uridylyltransferases (TUTases) or poly(U) polymerases (PUPs; on the right). This 3′‐end modification specifically recruits the Dis3L2 enzyme (an exoRNase from the RNB/RNase II family) that further degrades the molecule in the 3′–5′ direction. Small, capped degradation products are fully hydrolysed by DcpS. The faded parts of the image represent mechanisms of RNA decay not explored in this review. Created with BioRender.com.

Fig. 2

Protein domain organization of enzymes from the RNase II/RNB family. Representation of protein domains present in representative members of the RNase II/RNB family: bacterial RNase II and RNase R, and eukaryotic Dis3 (or Rrp44), Dis3L1 (or Dis3L) and Dis3L2. While the PIN domain confers an additional endonucleolytic activity to Dis3, it is inactive in Dis3L1 (depicted by a cross on the domain). The position of the amino acid substitutions in the hDIS3L2 protein that have been linked with human diseases (as detailed in Table 1) is indicated. The amino acid substitutions associated with PRLMNS/WT are denoted in red; the amino acid substitution associated with ATC is denoted in black. C, carboxyl terminus; N, amino terminus. The dimensions of the domains are approximately to scale.