Phylogenetic ubiquity and shuffling of the bacterial RecBCD and AddAB recombination complexes

Abstract

RecBCD and AddAB are bacterial enzymes that share similar helicase and nuclease activities and initiate repair of DNA double-strand breaks by homologous recombination. Examination of the phylogenetic distribution of AddAB and RecBCD revealed that one or the other complex is present in most sequenced bacteria. In addition, horizontal gene transfer (HGT) events involving addAB and recBCD appear to be common, with the genes encoding one complex frequently replacing those encoding the other. HGT may also explain the unexpected identification of archaeal addAB genes. More than 85% of addAB and recBCD genes are clustered on the genome, suggesting operon structures. A few organisms, including the Mycobacteria, encode multiple copies of these complexes of either the same or mixed classes. The possibility that the enzymatic activities of the AddAB and RecBCD enzymes promote their horizontal transfer is discussed, and the distribution of AddAB/RecBCD is compared to that of the RecU/RuvC resolvases. Finally, it appears that two sequence motifs, the Walker A box involved in ATP binding and an iron-sulfur-cysteine cluster, are present only in subsets of AddB proteins, suggesting the existence of mechanistically distinct classes of AddB.

FIG. 1.

Structure of the RecBCD and AddAB proteins. The RecB, RecD, and AddA proteins include a helicase domain (solid green) with the canonical six-helicase motifs of helicase superfamily I. The most N-terminal of these motifs is the Walker A box (orange). The RecC and AddB proteins include an inactivated helicase domain (striped green). The RecB, AddA, and AddB proteins additionally possess similar short nuclease domains toward their C termini (red). Some, but not all, AddB proteins possess an N-terminal Walker A and/or a mostly C-terminal iron-sulfur motif made up of four cysteines (blue).

FIG. 2.

Wide distribution of the AddAB and RecBCD proteins across 513 fully sequenced bacterial genomes. The 513 fully genome-sequenced bacteria considered here were split into 18 major taxa. Closely related strains or species have been grouped (yellow background), and species with multiple copies of addAB or recBCD genes are indicated by an additional line (pale blue background). The presence of genes encoding AddA, RecB, RecC, and RecD1 and identified by my methods is indicated by a black box. Boxes for AddB indicate the presence of both the Walker A and iron-sulfur motifs (green), neither motif (black), the Walker A box only (blue), or the iron-sulfur motif only (red). In all cases, a gray box indicates absence of the corresponding gene. Note that Magnetococcus species strain MC-1, here grouped with the Alphaproteobacteria, is now designated as an “unclassified proteobacterium.” Examples for which no protein meets the criteria used here but for which a potential homolog is identified by other means are shown as gray boxes marked “P” (see Fig. S1 in the supplemental material).

FIG. 3.

Distribution of AddAB and RecBCD across the bacterial tree argues for multiple displacements of one complex by the other via horizontal transfer of genes. A cladogram (from reference 13) of the bacterial taxa from Fig. 2 is color coded to indicate the occurrence of the AddAB and RecBCD complexes. When both complexes are found in a taxon (green boxes), they can occur within the same, or different, species. Branches supported by bootstrap values of less than 80% are indicated by dashed lines.

FIG. 4.

Distribution of AddAB in 28 sequenced archaea. The 28 fully genome-sequenced archaeal species considered here were split into three major taxa. The presence or absence of genes encoding AddA, AddB, RecB, RecC, and RecD1, determined by my methods, is indicated. Note that only AddB sequences with the iron-sulfur motif alone or sequences lacking both the Walker A and iron-sulfur motifs are shown.

FIG. 5.

Evidence for HGT in the RecBCD phylogenetic tree. An unrooted maximum likelihood tree was produced for the bacterial RecBCD complexes, with individual subunits concatenated to form single sequences for each complex. aLRT SH-like branch support values are shown. The scale bar indicates the number of amino acid substitutions per site.

FIG. 6.

Evidence for HGT in the AddAB phylogenetic tree. An unrooted maximum likelihood tree was produced for the bacterial and archaeal AddAB complexes, with individual subunits concatenated to form single sequences for each complex. aLRT SH-like branch support values are shown. The scale bar indicates the number of amino acid substitutions per site.