The structure of the CRISPR-associated protein Csa3 provides insight into the regulation of the CRISPR/Cas system

Abstract

Adaptive immune systems have recently been recognized in prokaryotic organisms where, in response to viral infection, they incorporate short fragments of invader-derived DNA into loci called clustered regularly interspaced short palindromic repeats (CRISPRs). In subsequent infections, the CRISPR loci are transcribed and processed into guide sequences for the neutralization of the invading RNA or DNA. The CRISPR-associated protein machinery (Cas) lies at the heart of this process, yet many of the molecular details of the CRISPR/Cas system remain to be elucidated. Here, we report the first structure of Csa3, a CRISPR-associated protein from Sulfolobus solfataricus (Sso1445), which reveals a dimeric two-domain protein. The N-terminal domain is a unique variation on the dinucleotide binding domain that orchestrates dimer formation. In addition, it utilizes two conserved sequence motifs [Thr-h-Gly-Phe-(Asn/Asp)-Glu-X(4)-Arg and Leu-X(2)-Gly-h-Arg] to construct a 2-fold symmetric pocket on the dimer axis. This pocket is likely to represent a regulatory ligand-binding site. The N-terminal domain is fused to a C-terminal MarR-like winged helix-turn-helix domain that is expected to be involved in DNA recognition. Overall, the unique domain architecture of Csa3 suggests a transcriptional regulator under allosteric control of the N-terminal domain. Alternatively, Csa3 may function in a larger complex, with the conserved cleft participating in protein-protein or protein-nucleic acid interactions. A similar N-terminal domain is also identified in Csx1, a second CRISPR-associated protein family of unknown function.

Figure 1

Csa3 is a dimer with two domains per subunit (Embedded 3D content, see below for instructions). (A) Stereo ribbon diagram of the Csa3 protomer. The N-terminal domain is colored cyan, the linker in blue and the C-terminal domain in deep teal. Secondary structural elements are labeled by domain and in ascending order from the N- to C-termini. (B) Stereo ribbon diagram of the Csa3 dimer. The N-terminal domain, linker and C-terminal domains are colored cyan, light blue and deep teal in chain A and light orange, tan and orange in chain B. The recognition helices in both chains are colored pale blue. (C) Stereo view of the Csa3 dimer looking down upon the winged helix-turn helix domains. Relative to panel B, the dimer has been rotated 90° about a horizontal axis in the plane of the page. Embedded interactive content requires the free Adobe Reader software, version 9 or later and can be activated by clicking on any part of the figure. The model can be manipulated interactively using the mouse. Options for selecting, rotating, panning and zooming are available in the toolbar or contextual menu. Parts of the model can be individually accessed and toggled on or off using the model tree. Preset views can be accessed using the dropdown “views” menu. To end -D viewing, right-click on the model and select “disable content”;for MAC users, Ctrl+click.

Figure 2

Validation of the biological unit by SAXS. (A) The parabolic nature of the Kratky plot at low values of q indicates that Csa3 is well folded. (B) The solution dimer corresponds to the crystallographic dimer present in a single asymmetric unit. The observed scattering curve of Csa3 (black) was compared with hypothetical scattering curves for the asymmetric unit (red), a potential monomer (light blue), and four alternative dimers (blue, violet, orange and green). Only the hypothetical scattering curve for the dimer corresponding to the crystallographic asymmetric unit agrees well with the observed scattering.

Figure 3

Superpositional docking of DNA to Csa3. (A) The Csa3 dimer is docked to the ohrR promoter. Csa3 is colored as in Fig. 1B except that the recognition helix is colored light teal and the additional putative DNA-interacting regions are colored blue. (B) Close-up view of the chain A wHTH domain docked to DNA. Structural features predicted to interact with the DNA are labeled.

Figure 4

Two conserved sequence motifs are found in the Csa3 N-terminal domain. The genome properties tool and CRISPR database were used to identify csa3 genes that are near clusters of cas genes and at least one CRISPR locus. The Csa3 sequences were aligned using 3D-COFFEE . The two sequence motifs which are conserved among Csa3 orthologs are outlined with boxes. Csa3 residues contributing to the dimer interface are marked with asterisks’ (*). Structure-based pairwise alignments of the VC1899 and Sso1389 (Csx1) N-terminal domains and the OhrR wHTH domain with Sso1445 were calculated using Dali . Secondary structural elements of Sso1445 (Csa3), VC1899, Sso1389 (Csx1) and OhrR were identified using DSSP. The sequence motifs, which are conserved among Csa3 orthologs are not present in VC1899 or Csx1. Csx1 has a 16-residue insertion at the site corresponding to Motif 1 in Csa3. GI numbers associated with the listed Csa3 proteins are presented in Table S1. The figure was generated using ALINE .

Figure 5

Putative ligand-binding pocket on the Csa3 N-terminal domain. (A) Surface conservation of Csa3. Residues are colored based on their relative conservation among the Csa3 orthologs using Consurf . Blue residues are more conserved and yellow residues are less conserved. (B) Stereo image of the conserved pocket on the N-terminal domain. The most highly conserved residues are labeled. Because the pocket spans the dimer interface, there are two copies of each residue. A four-atom length of ordered PEG is shown in stick rendering. The oxygen atoms are colored red and the carbon atoms are colored green. (C) Close-up view of the putative ligand-binding pocket. Chain A is colored cyan and chain B is colored orange. Phe10, Glu122, and the Gly-h-Arg motif (Gly96, selenomethionine (Mse) 97, and Arg98) are shown in stick rendering with nitrogen atoms in blue and oxygen atoms in red. The four atoms of ordered PEG are shown in stick rendering with the carbon atoms colored green and oxygen atoms red. The yellow mesh depicts the 2Fo-Fc omit map for the PEG contoured at 1 σ. The black dotted lines depict inter-protomer hydrogen bonds between Arg98 and Gly96, and hydrogen bonds between Arg98 and the ordered PEG.

Figure 6

Csa3, VC1899 and Csx1 share similar N-terminal domains but differ in overall domain architecture. Csa3, VC1899, and Sso1389 (Csx1) are shown in equivalent orientations based on structural superposition of their N-terminal domains (cyan and orange). The structurally dissimilar C-terminal domains are shown in shades of grey. Note that the C-terminal domains in VC1899 are found on the opposite side of the Csa3-like N-terminal domain.