Crystal structure of a PIWI protein suggests mechanisms for siRNA recognition and slicer activity

Abstract

RNA silencing regulates gene expression through mRNA degradation, translation repression and chromatin remodelling. The fundamental engines of RNA silencing are RISC and RITS complexes, whose common components are 21-25 nt RNA and an Argonaute protein containing a PIWI domain of unknown function. The crystal structure of an archaeal Piwi protein (AfPiwi) is organised into two domains, one resembling the sugar-binding portion of the lac repressor and another with similarity to RNase H. Invariant residues and a coordinated metal ion lie in a pocket that surrounds the conserved C-terminus of the protein, defining a key functional region in the PIWI domain. Furthermore, two Asp residues, conserved in the majority of Argonaute sequences, align spatially with the catalytic Asp residues of RNase H-like catalytic sites, suggesting that in eukaryotic Argonaute proteins the RNase H-like domain may possess nuclease activity. The conserved region around the C-terminus of the PIWI domain, which is required for small interfering RNA (siRNA) binding to AfPiwi, may function as the receptor site for the obligatory 5' phosphate of siRNAs, thereby specifying the cleavage position of the target mRNA.

Figure 1

Structure of A. fulgidus Piwi (AfPiwi). (A) Ribbon representation of AfPiwi coloured salmon (domain A), cyan (domain B) and orange (N-domain). Positions of invariant residues are coloured red (group I, invariant in all Argonaute proteins), green (group II, Piwi and Ago subfamily switch residues), blue (group III, invariant to eukaryotic Argonaute) and yellow (group IV, highly conserved in all Argonaute proteins). (B) Surface representation of AfPiwi coloured coded as in (A). The three conserved regions are indicated CRI, CRII and CRIII, as is the cadmium ion bound to the C-terminus. The interdomain crevice and domain B channel are labelled. (C) Stereoview showing a superimposition of AfPiwi and the PIWI domain of P. furiosus Argonaute PfAgo (pink) (PDB code: 1U04). The two proteins share identical architectures, although display small differences in relative orientations of domains A and B. The figure was produced using PYMOL (http://www.pymol.org).

Figure 2

Sequence alignment of AfPiwi with PIWI domains of eukaryotic Argonaute proteins from the Piwi and Ago subfamilies together with a structure-based alignment of AfPiwi and PfAgo. Colour coding of conserved and invariant residues as in Figure 1. Residues coordinating the Cd²⁺ ion are indicated by red arrows, and the RNase H-like catalytic Asp and Glu residues are indicated by blue arrows (AfPiwi residues 159, 254, 264 and 427). Argonaute proteins demonstrated to slice mRNA (HsAgo2, DmAgo2 and SpAgo) are bracketed in red. For a more extensive multiple sequence alignment, see Supplementary Figure 1. The figure was produced using ALSCRIPT (Barton, 1993).

Figure 3

Structural similarities between domain B of AfPiwi and the RNase H-type fold. (A) Structure of RNase HII from Methanococcus jannaschii (Lai et al, 2000; PDB code: 1EKE). (B) Domain B of AfPiwi. In both proteins, the RNase H-like fold is coloured salmon. In (A), the catalytic Asp residues are displayed and the equivalent positions on the β5 and β8 strands of AfPiwi are indicated in red in (B).

Figure 4

Metal-binding site at the C-terminus of AfPiwi involving Gln159 and Leu427. (A) Stereoview of a 2F_o–F_c electron density map centred on the metal ion. (B) Details of the hexa-coordination of the Cd²⁺ ion and surrounding conserved residues (colour-coded as for Figures 1 and 2).

Figure 5

Model for an siRNA duplex associated with AfPiwi. (A) The molecular surface representation is coloured according to electrostatic potential, ramping from blue to red for positive to negative electrostatic potential. A 5′ phosphate of the guide strand RNA (yellow) is docked into CRI, placing the scissile phosphate of the target RNA strand (green) adjacent to the RNase H-like catalytic site. (B) Ribbon diagram showing the RNA duplex docked into the groove of AfPiwi. CRI and CRII are indicated. Invariant basic residues are displayed in CRI. The distance between the 5′ phosphate of the guide RNA and the scissile phosphate of the target RNA is 18 Å, matching the distance between the metal ion bound to Leu427 and the side chain of the RNase H-like catalytic Asp on β5. The catalytic Asp residues were modelled from the structural superimposition of M. jannaschii RNase HII onto AfPiwi domain B.

Figure 6

AfPiwi forms a distinct complex with an siRNA-like RNA duplex. (A) Sequence and structure of the self-complementary RNA oligonucleotide used in this study. The RNA was labelled with ³²P at the 5′ end (^*) using T4 PNK. (B) EMSA assessing complex formation between the end-labelled RNA (<0.5 nM) and increasing concentrations of AfPiwi (0, 0.2, 0.7, 2, 7, 20 and 60 μM; lanes 2–8). Binding reactions were analysed by nondenaturing PAGE. Lane 1 is a control demonstrating the absence of an interaction with a control protein (55 μM human protein kinase B). (C) UV crosslinking assay showing covalent complex formation between the RNA and AfPiwi upon UV irradiation. Samples were analysed by SDS–PAGE. Lanes 3–9 are equivalent to lanes 2–8 in (B). Lanes 1 and 2 are controls confirming the absence of covalent complex formation without UV irradiation (20 μM AfPiwi) or with a control protein (55 μM human protein kinase B), respectively.

Figure 7

Mutation of CRI reduces the affinity of AfPiwi for RNA. (A) UV crosslinking assay assessing complex formation between the RNA duplex (<1 nM) and AfPiwi and AfPiwi^MUT (each 7 μM). The autoradiograph and Coomassie-stained gel are shown. (B) EMSA comparing complex formation by AfPiwi and AfPiwi^MUT (7, 2 and 0.7 μM in lanes 1–3 or 4–6).