
Full text loading...
Category: Clinical Microbiology
Hairpin Telomere Resolvases, Page 1 of 2
< Previous page | Next page > /docserver/preview/fulltext/10.1128/9781555819217/9781555819200_Chap12-1.gif /docserver/preview/fulltext/10.1128/9781555819217/9781555819200_Chap12-2.gifAbstract:
Hairpin telomere resolvases (also known as protelomerases) have emerged as a unique solution to the end replication problem ( 1 , 2 ). These enzymes promote the formation of covalently closed hairpin ends on linear DNA molecules in some phage ( 3 , 4 , 5 ), bacterial plasmids and bacterial chromosomes ( 6 , 7 , 8 , 9 ). Telomere resolvases are mechanistically related to tyrosine recombinases and type IB topoisomerases and are also believed to play a role in the genome plasticity that characterizes Borrelia species. Fig. 1 shows the reaction pathway for replication of linear DNA molecules with covalently closed hairpin telomeres. Duplication of the DNA molecule results in replicated telomeres (rTel, also referred to as dimer junctions) that are recognized and processed in a DNA breakage and reunion reaction promoted by a hairpin telomere resolvase. The reaction products are covalently closed hairpin telomeres at both ends of linear monomeric DNA molecules. At this writing telomere resolvases have been purified from three phage and seven bacterial species: E. coli phage N15 ( 3 ), Klebsiella oxytoca phage ɸKO2, Yersinia enterocolitica phage PY54 ( 5 ), Agrobacterium tumefaciens ( 8 ), the Lyme spirochete Borrelia burgdorferi ( 6 ), the relapsing fever borreliae B. hermsii, B. parkeri, B. recurrentis, B. turicatae, and the avian spirochete B. anserina ( 7 ). The B. burgdorferi enzyme, ResT ( Res olvase of T elomeres) has been the most extensively studied at the biochemical level ( 6 , 7 , 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 ) and is the primary focus of this review, with properties of the other enzymes noted ( 3 , 4 , 5 , 8 , 24 ). Structural studies of the Klebsiella phage ɸKO2 ( 25 ) and the Agrobacterium ( 26 ) resolvases have been reported and have shed additional light on reaction mechanisms and on differences between the resolvases from different organisms.
Full text loading...
Hairpin telomere resolution as a solution to the end replication problem. Replication of a linear molecule results in the formation of dimer junctions or replicated telomeres (L′L, RR′) that are processed by telomere resolution, a unique type of DNA breakage and reunion reaction. Telomere resolution results in the formation of hairpin telomeres at the ends of the linear DNA molecule and separates the dimer replication intermediate into monomeric products of DNA replication. See text for further details. This figure is adapted from reference 6 and reprinted from reference 1 .
ResT promotes telomere resolution via a two-step transesterification. (1) The telomere resolvase, ResT, binds to a replicated telomere (rTel) substrate, which corresponds to the dimer junction L′L or RR′ shown in Fig. 1 ( 11 ). The black dots denote the scissile phosphates and the vertical bar in the center of rTel the axis of 180-degree rotational symmetry. (2) Positive supercoiling is believed to facilitate the cooperative formation of a cross-axis complex where communication between ResT protomers bound on both sides of the symmetry axis occurs ( 14 , 16 ). (3) The action of the “hairpin binding module,” a region with sequence similarity to a motif found in cut-and-paste transposases, induces a DNA distortion that facilitates (4) DNA cleavage, the first transesterification event ( 12 ). (5) A conformational change then occurs to juxtapose the free 5′-OH groups to the 3-phosphotyrosyl enzyme intermediates on the opposite strands. (6) Nucleophilic attack of the phosphotyrosyl linkage by the 5′-OH groups, the second transesterification step, results in phosphodiester bond formation to stabilize the hairpins generated during step 5. This figure and legend are adapted from reference 43 and reference 21 , and reprinted from reference 2 .
Sequence alignment of the domains carrying the active site residues of purified telomere resolvases. An alignment is shown for the telomere resolvases from the phages N15, TelN ( 3 ), φKO2, TelK ( 4 ), PY54, TelY ( 5 ) and the bacterial resolvases from Agrobacterium tumefaciens, TelA ( 8 ), the Lyme spirochete Borrelia burgdorferi, ResTBb ( 6 ) and the relapsing fever species Borrelia hermsii, ResTBh ( 7 ). Several other purified hairpin resolvases from other Borrelia species are not included in the lineup. The catalytic residues are indicated by asterisks and the active site tyrosine by a red asterisk. The double colon indicates the position of the proline in B. burgdorferi ResT that confers permissiveness for Type 2 telomeres ( 7 ). The corresponding active site residues for the tyrosine recombinase family are indicated above and below the alignment in red and those for type IB topoisomerases in blue.
B. burgdorferi telomere sequence alignment. Telomere sequences are arranged in descending order, according to the initial rate of telomere resolution. The initial rate, expressed in fmol/min is shown in the right-hand column and telomere sequences are aligned with the hairpins (or symmetry axis in the replicated telomeres) at the left end. The telomeres shown are half of the actual replicated telomere substrates used in the telomere resolution reactions. The colored boxes labeled 1 and 3 refer to previously identified regions of sequence homology ( 11 ), with some modifications. The original box 1 sequence, TATAAT is indicated by a light blue box, while the newly identified box 1 sequence, TATTAT is shown in dark blue. The homology box 3 region has been expanded from the five nucleotide sequence TAGTA to the eight nucleotide sequence TTAGTATA. The telomere sequences of lp17L, lp17R, lp21R, lp28-1R, lp56R ChromL and ChromR have been reported previously ( 45 , 47 , 48 , 70 , 71 , 72 ). Reprinted from reference 20 .
Telomere exchanges believed to be mediated by ResT promoted telomere fusions. The proposed mechanism for telomere exchange between linear plasmids and the right end of the B. burgdorferi chromosome is a two-step process. The first is a telomere fusion event and the second is a deletion or other type of mutation to inactivate or remove the newly fused telomere and prevent its resolution and to remove competing replication maintenance functions. The telomere fusion event is promoted by reversal of the ResT reaction such that two hairpin telomeres from different molecules are fused to generate a single DNA molecule carrying a replicated telomere. A deletion removing the telomere resolution site might be specifically targeted to the fused telomere by incomplete joining in the reverse reaction, to leave a ResT molecule covalently linked at a nick in the telomere; such covalent protein–DNA complexes are known to be foci for the formation of deletions and other chromosomal aberrations ( 73 ). Alternatively, a deletion could be derived from palindrome instability induced by passage of a replication fork through the inverted repeat of the fused telomere ( 74 , 75 ). B. burgdorferi chromosome extensions that may have arisen by ResT-mediated telomere fusions followed by deletion formation (see references 1 , 2 , 15 ). The B31 chromosome appears to have arisen from fusion of an lp28-1 plasmid with the N40 chromosome. Subsequently, a single fusion of the B31 chromosome with lp21, followed by sequence deletion would have generated the 297 chromosome. Similarly, two rounds of fusion/deletion of the B31 chromosome, first with an lp28-1 and subsequently with lp28-5, would have generated the JD1 chromosome. The sequence relatedness of the chromosomes and plasmids shown were reported by references 44 , 46 , 47 , 48 . This figure is slightly modified from reference 46 .
The domain structure of hairpin telomere resolvases. (A) The domain composition of the telomere resolvases from Klebsiella phage ɸKO2 (TelK), Agrobacterium (TelA) and Borrelia species (ResT) are shown (see also text). All telomere resolvases have a central catalytic domain with active site residues similar those of tyrosine recombinases and type IB topoisomerses (see Fig. 3 and text). In addition, hairpin resolvases carry divergent N-terminal domains; they may also carry a divergent C-terminal region or it may be absent as for TelA. Domains represented by the same shape and color scheme are structurally related. Domains delimited by brackets represent protein sequences dispensable for telomere resolution in vitro ( 8 , 25 ). Also shown for each hairpin telomere resolvase is the tyrosine nucleophile (Y). The numbers above the graphics refer to the amino acid numbers of the proteins. Precise domain boundaries have been adjusted from those previously reported based upon structural alignments. (B) and (C) A structural view of the domain organization of TelK and TelA, respectively. The domains are represented by the same color scheme used in (A) on one monomer in each dimer. The N-core domain, represented in blue, is composed of a helical bundle that forms the top of the C-clamp embrace the resolvases make with the substrate DNA and the long linker α-helix that connects the top and bottom of the C-clamp. TelK has a large insertion in the N-core domain called the muzzle; this is represented in yellow. The shared catalytic domain is represented in red. TelK has an additional C-terminal domain essential for telomere resolution called the stirrup; this domain is represented in grey. Beyond the shared catalytic domain TelA lacks an additional domain like the stirrup but instead has a short segment that contributes to dimerization contacts; this short C-terminal extension is represented in green. The arrows with residue numbers indicate the position in the structure of the first and last resolvable residues present in the structures. The structures presented were generated with The PyMOL Molecular Graphics System, Version 1.7 Schrödinger, LLC. (http://www.pymol.org/) using PDB ID 2v6e for TelK ( 25 ) and 4e0g for TelA ( 26 ).
TelK and TelA-DNA complexes showing displacement of the helical axes at the dimer interface by 7.5 and >10 angstroms, respectively. The structures presented were generated with The PyMOL Molecular Graphics System, Version 1.7 Schrödinger, LLC. (http://www.pymol.org/) using PDB ID 2v6e for TelK ( 25 ) and 4e0g for TelA ( 26 ).