Figure 1.
A. Unrooted bootstrapped phylogenetic tree showing a subset of Cas4 family proteins from archaea. Each protein is represented by a three-letter species code followed by the gene number from the respective genome sequences (Sso, S. solfataricus; Pfu, P. furiosus; Ttx, T. tenax). This neighbour-joining tree was generated from a T-coffee alignment of the proteins using MacVector, with pairwise distances between sequences uncorrected. The bootstrap values shown at each node represent the percentage of all trees (10000 total) agreeing with this topology. B. Cartoon showing the arrangement of the four cysteines acting as FeS ligands and the conserved residues contributing to the active site of the RecB nuclease, for the Cas4 family and related proteins. The blue shading denotes the central portion of the RecB-like domain in each protein.
Figure 2.
Cas4 family enzymes are Iron-Sulfur proteins.
A. SDS-PAGE gel lanes showing purified Sso0001 and Sso1391 proteins. Concentrated samples of Sso0001 and Sso1391 had an olive-green appearance. B. The Sso0001 protein eluted from a calibrated superpose 12 column in one major symmetrical peak with a size corresponding to a molecular weight of 175 kDa.
Figure 3.
Cas4 family proteins are metal dependent nucleases.
A. Sso0001 (0.6 µM) cleaves the ssDNA oligonucleotide 15T (1 µM) in the presence of either Mg2+ or Mn2+ but the ssRNA oligonucleotide 20U (1 µM) only in the presence of Mn2+. Both oligonucleotides were 5′-end labelled with a 5′-fluorescein moiety. Reactions were carried out at 75°C for 10 min. B. The Sso0001 D99A variant does not cleave ssDNA oligonucleotide 15T (1 µM) in the presence of Mg2+ ions. Assays were carried out as in panel A with 0.6 or 1.2 µM of the relevant protein. Reactions were carried out at 75°C for 10 min. C. Circular phiX174 virion ssDNA (60 nM) was incubated with 1.2 µM wild-type or 3 µM D99A Sso0001 for 10, 20 and 40 min at 55°C in the presence of MgCl2. No degradation of the DNA was observed, suggesting that Sso0001 requires a ssDNA end for activity. The control lane (C) was incubated in the same conditions in the absence of enzyme. dsDNA size markers (m) are indicated.
Figure 4.
Cas4 family proteins are 5′-3′ exonucleases.
A. Sso0001 (0.6 µM) cleaved the 5′ end labelled DNA 29mer (1 µM) in the presence of 10 mM Mg2+ into products with a size around 4 nt without any observable intermediates. In contrast, the 3′ end labelled DNA 29mer (1 µM) was cleaved to generate a ladder of progressively smaller products over the course of the reaction time in the presence of 10 mM Mg2+. Time points for both substrates were 0, 0.5, 1, 2, 3, 5, 10 and 15 min. The reaction temperature was 55°C. B. Sso0001 (0.3 µM) was incubated with 3′ end labelled DNA 50mer (500 nM) in reaction buffer at 55°C and products were analysed at 1, 2, 3, 5 and 10 min. In a parallel reaction (lanes 6–10), a 20 fold excess (10 µM) of unlabeled oligonucleotide of the same sequence was added to the reaction at the 1 min time point. A marked decrease in cleavage of the labelled oligonucleotide was observed, suggesting that Sso0001 is not highly processive, but can dissociate from substrates following each round of catalysis.
Figure 5.
Cas4 family proteins are specific for single-stranded DNA and stalled by an internal extrahelical fluorescein.
The left side of the gel shows degradation of a 31nt oligonucleotide with an internal fluorescein-dT at position 20. The right hand side shows this oligonucleotide in a 31bp DNA duplex (both at 1 µM final concentration). The assay was carried out with 0.6 µM Sso0001 at 55°C for time points of 0, 1, 2, 3, 5, 10, 20, 30 and 40 min in 10 mM Mg2+. Control lanes (c) lacked added Sso0001 enzyme.
Figure 6.
Structural similarities between AddB, Cas4 and a putative exonuclease DUF3799.
All three structures show the same arrangement of four cysteine residues forming a metal ligand, coupled with a RecB nuclease domain. However, unlike AddB (PDB: 3U44) and Cas4, DUF3799 (PDB: 3L0A) has a zinc ion (green sphere) coordinated with the cysteines instead of iron. The cartoons on the left show the conserved residues implicated in nuclease activity and FeS cluster binding.
Figure 7.
Possible role of Cas4 in the spacer acquisition pathway.
In the spacer acquisition stage of the CRISPR system, the duplex viral DNA containing the protospacer may be unwound by the Cas3 helicase, providing substrates for Cas4. Processing by Cas4 would generate 3′ overhangs suitable for strand invasion at the CRISPR locus on the host chromosome.