Structural insights into DNA sequence recognition by Type ISP restriction-modification enzymes

Abstract

Engineering restriction enzymes with new sequence specificity has been an unaccomplished challenge, presumably because of the complexity of target recognition. Here we report detailed analyses of target recognition by Type ISP restriction-modification enzymes. We determined the structure of the Type ISP enzyme LlaGI bound to its target and compared it with the previously reported structure of a close homologue that binds to a distinct target, LlaBIII. The comparison revealed that, although the two enzymes use almost a similar set of structural elements for target recognition, the residues that read the bases vary. Change in specificity resulted not only from appropriate substitution of amino acids that contacted the bases but also from new contacts made by positionally distinct residues directly or through a water bridge. Sequence analyses of 552 Type ISP enzymes showed that the structural elements involved in target recognition of LlaGI and LlaBIII were structurally well-conserved but sequentially less-conserved. In addition, the residue positions within these structural elements were under strong evolutionary constraint, highlighting the functional importance of these regions. The comparative study helped decipher a partial consensus code for target recognition by Type ISP enzymes.

CgII cleaves DNA using a mechanism distinct from other ATP-dependent restriction endonucleases

“…The cleavage data show that linear DNA with a pair of sites in direct repeat (HtT1 or HtT2) is the most efficiently cut, where cleavage dominates at the second site (downstream to the first as drawn in Figure 4 ), with a rate that is similar to that of the one-site circular DNA and which is independent of local sequence (∼10 −3 s −1 , Supplementary Table S3 ). This would be consistent with a rear-end collision mechanism for activation of cleavage that is different from the Type I, ISP and III RM enzymes ( 5 , 16 , 20 , 22 , 40 , 41 ). However, we also observed cleavage of both DNA with sites in inverted repeat (HtH and TtT) with a similar rate (∼10 −4 s −1 ) that is 10-fold slower than the direct repeats (Figure 4B and Supplementary Table S3 ).…”

The H-subunit of the restriction endonuclease CglI contains a prototype DEAD-Z1 helicase-like motor

“… Comparison of the helicase-like domains of Types I, ISP and III restriction enzymes. Helicase consensus motifs are shown based on the work here and ( 61 , 78 , 80 , 81 ). The Type IIIA consensus sequences are presented.…”

The third restriction–modification system from Thermus aquaticus YT-1: solving the riddle of two TaqII specificities

“…For the TaqII branch of the Thermus -family enzymes we currently have just three proteins for which the recognition motif is known. However the TRDs of TaqII and TaqIII have significant similarity to the TRDs of the Type ISP enzymes LlaGI and LlaBIII, for which crystal structures have recently been determined and residues making direct DNA contacts have been identified ( 42 ) ( Supplementary data, Figure S6 ). We therefore aligned TaqII, TaqIII and TspGWI with the Type ISP enzymes described by Kulkarni, et al ( 42 ), and find that the TaqII and TaqIII proteins match well with the observed and co-variation-predicted DNA recognition elements in the Type ISP enzymes (Figure 8 ; Supplementary data, Figure S6 ).…”

“…However the TRDs of TaqII and TaqIII have significant similarity to the TRDs of the Type ISP enzymes LlaGI and LlaBIII, for which crystal structures have recently been determined and residues making direct DNA contacts have been identified ( 42 ) ( Supplementary data, Figure S6 ). We therefore aligned TaqII, TaqIII and TspGWI with the Type ISP enzymes described by Kulkarni, et al ( 42 ), and find that the TaqII and TaqIII proteins match well with the observed and co-variation-predicted DNA recognition elements in the Type ISP enzymes (Figure 8 ; Supplementary data, Figure S6 ). Note that in this discussion of recognition elements, we present the TaqII/TaqIII strand that contains the adenine which the enzymes methylate, and refer to the base positions as 0, 1, 2, 3, 4, 5 (and 6), where the fifth position is the methylated adenine, i.e.…”

“…Interestingly, these enzymes share a conserved DNA recognition domain with the Type ISP enzymes, such as LlaGI and LlaBIII, even though these enzymes employ a completely different mechanism for DNA cutting. The excellent characterization of Type ISP DNA recognition carried out by the Szczelkun and Saikrishnan labs ( 42 ) can be applied to the TaqII family enzymes to help identify putative positions and aa that specify recognition, particularly those responsible for the differing specific recognition at two positions within the TaqII and TaqIII recognition motifs (Figure 8 ). We have used the analysis of recognition elements to predict recognition motifs for 6 uncharacterized Thermus -family enzymes ( Supplementary data, Figures S6 and S7 ), which predictions can be readily tested.…”