
Full text loading...
Category: Clinical Microbiology
Phage-encoded Serine Integrases and Other Large Serine Recombinases, Page 1 of 2
< Previous page | Next page > /docserver/preview/fulltext/10.1128/9781555819217/9781555819200_Chap11-1.gif /docserver/preview/fulltext/10.1128/9781555819217/9781555819200_Chap11-2.gifAbstract:
Conservative site-specific recombination systems are ubiquitous in bacteria, where they play important roles in the horizontal transfer of genetic information, genome stability, and the control of gene expression. The outcomes of site-specific recombination are DNA integration, DNA excision (sometimes referred to as resolution), and DNA inversion. The systems comprise a recombinase, the sequence specific DNA substrates, and any accessory factors that are required for control. There are two evolutionarily and mechanistically different families of site-specific recombinases, the tyrosine and serine recombinases ( 1 ). All types of recombination outcomes are mediated by recombinases from both families. In the serine recombinase family there is a clear division between the resolvase/invertases and the enzymes that mediate integration/excision. The resolvase/invertases are approximately 180 to 200 amino acid proteins and are increasingly referred to as the small serine recombinases. The (pro)phage-encoded serine integrases, the transposases, such as those from clostridial ICE elements, Tn4451 and Tn5397, and the recombinases from the staphylococcal cassette chromosomes (SCC) elements are between 400 and 700 amino acids and are collectively known as the large serine recombinases (LSRs) ( 2 ). The first LSRs to be studied in in vitro recombination systems were the integrases from the Streptomyces phage, ɸC31 ( 3 ) and mycobacteriophage Bxb1 ( 4 ). Understanding the mechanism of the LSRs, however, was greatly hindered by the lack of structural information. In a breakthrough paper, Rutherford et al. published the structure of the large C-terminal domain (CTD) of a serine integrase bound to one half site of one substrate ( 5 ). This work has led to a step change in our understanding of the mechanism of the serine integrases ( 6 , 7 ).
Full text loading...
Role of large serine recombinases in (A) phage integration and excision and (B) movement by mobile elements. Black and blue double lines represent the host and phage/mobile element DNA, respectively. Triangles are the attP/attTn/attSCC sites (blue) or host attB sites (black). The hybrid attL and attR sites are mixed. Figure adapted from Smith ( 79 ) with permission from Elsevier Press.
The recombination pathway by large serine recombinases (LSRs). (i) Diagram of the domain structure of the LSRs; the N-terminal domain (NTD; green) is the catalytic domain; αE (green line) is the long alpha helix that connects the NTD to the recombinase domain (RD; red). The short linker between the RD and the zinc ribbon domain (ZD) is indicated by the arrow. The coiled-coil (CC) motif is depicted by rectangular projections. (ii) The LSRs in the unbound state are likely to be compact and globular ( 6 ). (iii) The LSRs binding to attP (gray) and attB (black). The relative positions of the ZD domains are different on attP and attB and these have consequences for the positions of the CC motifs. The CC motifs are either flared and dark, indicating they are projecting out of the paper or not flared and light, where they project into the paper. The scissile phosphates are located flanking the two nucleotide crossover sites shown as two white vertical lines. Bound to attP and attB, the CC to CC motifs can begin to interact and initiate synapsis. (iv) A synaptic complex is formed that requires interactions between the CC motifs as well as through the NTD tetrameric interface ( 41 ). Conformational changes occur in the NTDs to generate a flat interface for subunit rotation. (v) DNA cleavage occurs with concomitant formation of the phosphoseryl bonds. (vi) Subunit rotation swaps half sites, in this case B′ and P′. (vii) Joining of the recombinant products leads to a closed conformation of LSRs on the attL and attR sites and conformation changes in the NTDs. Figure adapted from Rutherford and van Duyne ( 6 ) with permission from Elsevier Press.
Structures of the Listeria innocua integrase C-terminal domain (CTD) bound to the A118 attP half site. Four views are shown to illustrate two different trajectories of the coiled-coil (CC) motif. (A) (i) A schematic of a dimer of a large serine recombinase bound to an attP site. The structures shown in (ii) and (iii) relate to the boxed area in (i) and display two different trajectories of the CC motif as described in Rutherford et al. ( 5 ). (B) (i) The schematic indicates that the views in (ii) and (iii), which show the same structures as in (A)(ii) and (A)(iii), are looking down through the DNA. Domains are color-coded: green is the C-terminal end of αE from the N-terminal domain (NTD), Red is the recombinases domain (RD), blue is the zinc ribbon domain (ZD) and the light blue region within ZD is the CC motif. The NTDs, which are absent in the structures, would connect to the αE helix (green) to bind to the opposite side of the DNA from the CTDs. Figures were constructed using the PDP file 4KIS.
Binding motifs in the attachment sites for Listeria innocua (LI) integrase, Bxb1, and ϕC31 integrases. Numbering of the bases is outwards from the crossover dinucleotide (00) to the left (minus) and the right (plus). The red and blue boxes indicate the recombinase domain (RD) and zinc ribbon domain (ZD) binding motifs, respectively. The three base pairs recognized by the RD-ZD linker in the attP sites are shown in pink. Highlighted orange are bases that are mutational sensitive and yellow is the discriminatory base described in Singh et al. ( 64 ).
Structural models of a large serine recombinase bound to a full attP site (A) and a full attB site (B) adapted from Rutherford et al. ( 5 ) with permission from Oxford University Press and redrawn by Dr. Greg van Duyne. The domains are colored using the same scheme as in Fig. 3 . ZD, zinc ribbon domain; RD, recombinases domain; NTD, N-terminal domain.
Site-selectivity explained by the geometry of the CC motifs bound to the attachment sites. (A) The three panels show the hypothetical assembly of synaptic complexes by integrase bound to two attP sites (left panel), two attB sites (middle panel), and an attP and attB site (right panel). The integrase subunits are colored red if bound to P or P′ and blue on B or B′. Line 1 in each panel shows an integrase dimer bound to an attachment site and the dimer is viewed from the perspective of zinc ribbon domain (ZD) and recombinase domain (RD) and looking down towards the N-terminal domain (NTD) bound to the opposite face of the DNA (black line). Line 2 is a second dimer bound to an attachment site but viewed the other way, that is, from the NTD and looking down towards the RD and ZD domains underneath. Line 3 shows what happens when the dimer from line 1 is superimposed on the dimer from line 2 to generate an integrase tetramer and line 4 is where this complex is rotated by 90°. In line 4, the CC motifs are flared and dark where they project out of the page and pale and thin where they project into the page. The CC motifs project in opposite directions from dimers bound to two attP sites or two attB sites but are proposed to be close enough to interact between integrase dimers bound to attP and attB site. (B) Possible pathway for assembly of the excision synapse with so-called complementary interactions between the integrase subunits is shown on the left ( 48 ). The attL and attR sites are proposed to be in a closed conformation with respect to the CC motif in the absence of the recombination directionality factor (RDF). Addition of the RDF might bind to the ZD domain to change the trajectory of the CC motifs so that they are in a more open conformation. As the CC motifs in the dimers bound to attL and attR project in the same direction (here projecting out of the page), there is an opportunity for them to interact, but only in the presence of the RDF. If one of the attL or attR sites aligns in the opposite orientation as shown by the tetramer synapse on the right, the CC motifs cannot interact as they are projecting in different directions, possibly explaining noncomplementary interaction and the bias against this type of synaptic complex in excision ( 48 ).
The role of the central dinucleotide in site polarity. The central dinucleotide determines the polarity of the attP and attB sites and the identity of the attL and attR sites. (A) The wild type recombination sites for ϕC31 integrase have a nonpalindromic 5′TT at the crossover dinucleotides that after cleavage and strand exchange by subunit rotation [indicated by arrow (i)] can base pair in the recombinants allowing joining of the DNA backbone [indicated by arrow (ii)]. (B) Any sites that have a mismatch at the dinucleotide cannot join after DNA cleavage and one round of strand exchange by subunit rotation. Strand exchange can either reverse or the rotation iterates to regenerate the substrates [indicated by arrows (iii)]. The recombination pathway can then begin again. (C) Fifty percent of synaptic complexes assemble with attP and attB in an antiparallel orientation such that cutting of the substrates yields dinucleotides at the crossover that cannot base pair in the recombinants leading to reversal or iteration of strand exchange (iii). In excision, attL and attR do not normally assemble an antiparallel synapse as this would require noncomplementary interactions between integrases bound to two P type (or two B type) half sites. However, attR × attR (or attL × attL) can form a synapse with the permitted complementary interactions by integrase subunits but joining of the products is prevented.
Location of attB sites