RNase AS versus RNase T: Similar yet Different

RNase AS versus RNase T: Similar yet Different

Structure Previews RNase AS versus RNase T: Similar yet Different Xuhua Tang,1 Siew Choo Lim,1 and Haiwei Song1,* 1Institute of Molecular and Cell Bi...

244KB Sizes 36 Downloads 107 Views

Structure

Previews RNase AS versus RNase T: Similar yet Different Xuhua Tang,1 Siew Choo Lim,1 and Haiwei Song1,* 1Institute of Molecular and Cell Biology, 61 Biopolis Drive, Proteos, Singapore 138673, Singapore *Correspondence: [email protected] http://dx.doi.org/10.1016/j.str.2014.04.006

In this issue of Structure, Romano and colleagues show that RNase AS specifically hydrolyses adenylatecontaining RNA and affects mycobacterial virulence. This study reveals the structural basis underlying the substrate specificity of this enzyme. Tuberculosis (TB) is a disease caused by a 30 to 50 exoribonuclease with strong insertion of Pro105 to contribute the forMycobacterium tuberculosis. Infections specificity for adenylate-containing RNA. mation of the cleft hosting AMP(+1) of M. tuberculosis are often asymptomStructurally, RNase AS resembles adenine base (Romano et al., 2014). Secatic, chronic in a clinically symptomatic RNase T, another DEDDh exonuclease ond, RNase AS and RNase T have state, and highly recalcitrant to antibi- from E. coli, despite their low sequence different electrostatic potential surfaces otics. Because latent TB infection affects identity. Although both RNase AS and (Romano et al., 2014), although they both about a third of the world population, RNase T function as a dimer and share a adopt an opposing dimeric arrangement treatment of both chronic and latent similar dimeric architecture, they have (Zuo et al., 2007). RNase T has a characforms of TB are essential in controll- distinct substrate recognition mecha- teristic positively charged patch (NSB ing and eliminating this infectious and nisms (Figure 1). RNase AS exhibits patch) from one monomer juxtaposed often fatal disease. The cell-envelope of a strong substrate preference of short with the catalytic DEDD residues from M. tuberculosis plays a key role in bacte- adenylate-containing RNA over DNA, the other monomer, which is involved in rial virulence and antibiotic resistance while RNase T is a more efficient DNase the binding of long nucleotides (Hsiao (Brennan and Nikaido, 1995); hence, than RNase, which specifically trims the et al., 2011). In contrast, RNase AS has a targeting the cell-envelope biosynthesis 30 end of structured DNA (Hsiao et al., positively charged patch located just 2011; Hsiao et al., 2014). Moreover, below the catalytic cleft from the same is a promising strategy for TB therapy. In this issue of Structure, Romano et al. RNase T has an unusual substrate speci- monomer (Romano et al., 2014). These (2014) identified RNase AS, an exonu- ficity in that its exonucleotic activity is structural differences likely account for clease important in mycobacterial viru- blocked by 30 -terminal C residues and their distinct substrate preferences. lence. The gene (rv2179c) encoding double-stranded structures (Hsiao et al., Based on the crystal structures of RNase AS was identified through trans- 2011). Structural comparison between RNase AS with bound AMP or UMP, poson mutagenesis in the orthologs of RNase AS and RNase T reveals two Romano et al. (2014) proposed that the M. marinum and M. smegmatis. Mutants significant differences in the structures. specificity for adenine moiety at the lacking this gene display different myco- First, a long a-helix in RNase T is broken nucleotide (+1) position is determined by bacterial cell surface properties compared into two helices in RNase AS by the a single hydrogen bond between the to the wild-type in a double filter NH2 group of the adenine base and the carbonyl oxygen of assay that measures the amount Met106. To confirm this, the authors of capsular a-glucan. Through experimentally proved that RNase knockout and rescue experiments AS is unable to degrade poly inoin zebra fish embryos, the authors sine, a polynucleotide in which the showed that this gene is important adenine NH2 group is replaced for mycobacterial virulence in vivo. by carbonyl oxygen, because the RNase AS belongs to the DEDDh latter is unable to act as a donor subfamily of the DEDD 30 -to-50 exonuclease superfamily. This large in a hydrogen bond interaction. family includes many essential proThis novel and simple substrate karyotic and eukaryotic exonucleselection mechanism is drastically ases, and all bear four conserved different from the one proposed metal-binding acidic residues and for RNase T. Structural analyses one general base residue in the of RNase T-DNA complexes show active site and share a common that RNase T dimer has an ideal two-metal ion mechanism of hydroarchitecture to specifically trim Figure 1. Schematic Diagrams of the Substrate Binding Modes for RNase AS and RNase T lysis (Zuo and Deutscher, 2001). the 30 end of structured DNAs The left panel shows the RNase AS dimer bound to a singlethrough an elegant mechanism to Through oligonucleotides degradastranded adenylate-containing RNA, and the right panel screen out 30 -terminal cytosines for tion experiments, Romano et al. shows the RNase T dimer bound to a double-stranded DNA cleavage by inducing an inactive duplex with a short 30 overhang. (2014) confirmed that RNase AS is Structure 22, May 6, 2014 ª2014 Elsevier Ltd All rights reserved 663

Structure

Previews conformation in the active site (Hsiao et al., 2011). Because only singlestranded RNA was used for this study (Romano et al., 2014), it will be interesting to know if RNase AS can also degrade double-stranded RNA with 30 overhangs and whether its activity will be blocked by double-stranded structures. In prokaryotes, the poly(A) tail functions in the regulation of RNA stability and quality control (Mohanty and Kushner, 2011). Given that RNase AS and PARN [a poly(A)-specific 30 exoribonuclease of the DEDDh subfamily; Wu et al., 2005] share a similar substrate preference, and the latter is a key deadenylase in eukaryotic mRNA turnover, RNase AS is likely involved in mRNA decay through

shortening of the 30 adenylate-containing mRNA. However, the transcripts targeted by RNase AS are not known. One of the future goals is to identify the genes responsible for the mycobacterial virulence whose stability is regulated by RNase AS through deadenylation. Identification of the physiological substrates of RNase AS and elucidation of the catalytic mechanism of this important RNase would greatly aid the drug design efforts toward more efficient TB therapy.

Hsiao, Y.Y., Yang, C.C., Lin, C.L., Lin, J.L., Duh, Y., and Yuan, H.S. (2011). Nat. Chem. Biol. 7, 236–243. Hsiao, Y.Y., Fang, W.H., Lee, C.C., Chen, Y.P., and Yuan, H.S. (2014). PLoS Biol. 12, e1001803. Mohanty, B.K., and Kushner, S.R. (2011). Wiley Interdiscip Rev RNA 2, 256–276. Romano, M., van de Weerd, R., Brouwer, F.C., Roviello, G.N., Lacroix, R., Sparrius, M., van den Brink-van Stempvoort, G., Maaskant, J.J., van der Sar, A.M., Appelmelk, B.J., et al. (2014). Structure 22, this issue, 719–730. Wu, M., Reuter, M., Lilie, H., Liu, Y., Wahle, E., and Song, H. (2005). EMBO J. 24, 4082–4093.

REFERENCES

Zuo, Y., and Deutscher, M.P. (2001). Nucleic Acids Res. 29, 1017–1026.

Brennan, P.J., and Nikaido, H. (1995). Annu. Rev. Biochem. 64, 29–63.

Zuo, Y., Zheng, H., Wang, Y., Chruszcz, M., Cymborowski, M., Skarina, T., Savchenko, A., Malhotra, A., and Minor, W. (2007). Structure 15, 417–428.

Membrane Interaction and Functional Plasticity of Inositol Polyphosphate 5-Phosphatases Werner Braun1,2,* and Catherine H. Schein3 1Sealy

Center for Structural Biology and Molecular Biophysics, University of Texas Medical Branch, Galveston, TX 77555, USA of Biochemistry and Molecular Biology, University of Texas Medical Branch, Galveston, TX 77555, USA 3Foundation for Applied Molecular Evolution, 720 SW 2nd Avenue Suite 201, Gainesville, FL 32601, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.str.2014.04.008 2Department

In this issue of Structure, Tre´saugues and colleagues determined the interaction of membrane-bound phosphoinositides with three clinically significant human inositol polyphosphate 5-phosphatases (I5Ps). A comparison to the structures determined with soluble substrates revealed differences in the binding mode and suggested how the I5Ps and apurinic endonuclease (APE1) activities evolved from the same metal-binding active center. A complex family of membrane-bound inositol-lipid conjugates plays a central role in cell proliferation, synaptic vesicle recycling, and actin polymerization. Control of these pathways depends on a family of enzymes that control the phosphorylation of insoluble phosphoinositides (PtdIns), i.e., glycerolphospholipids bound to the 1-hydroxyl of myo-inositol. Specific kinases add phosphate groups at the 3, 4, and 5 hydroxyls of the inositol ring, leading to seven different isoforms. These, in turn, are removed by a family of phosphatases. In this issue of Structure, Tre´saugues et al. (2014) show the binding site for three human phosphatases that

remove the 5-phosphate from PtdIns with two or three phosphates bound to PtdIns. Mutations in the similar 5-phophatase active site of these enzymes are related to Lowes syndrome and Dent disease (5-phosphatase OCRL), characterized by renal failure, as well as defects in insulin signaling and obesity (5-phophatase SHIP2). They also determined the structure of INPP5B, whose similarity in substrate specificity to OCRL suggests it may have overlapping function. The details of these new structures could help in the design of treatments for renal syndromes. Inhibitors of SHIP2 might even be novel weight-loss medications,

664 Structure 22, May 6, 2014 ª2014 Elsevier Ltd All rights reserved

because mice deficient in SHIP2 were resistant to diet-induced obesity. These structures are most remarkable for the details they can shed on how insoluble inositides bind, because the only structure of a 5-phosphatase catalytic domain previously reported is that of a complex of Schizosaccharomyces pombe synaptojanin (SPsynaptojanin) with a soluble ligand (Tsujishita et al., 2001). They are also important for the details they provide on a unifying mechanism for dephosphorylation of PtdIns with other types of phospho-transfer reactions. The inositol position is different in the new structures from models of the