Comprehensive analysis of DNA polymerase III α subunits and their homologs in bacterial genomes

Published online 7 October 2013 Nucleic Acids Research, 2014, Vol. 42, No. 3 1393–1413 doi:10.1093/nar/gkt900 SURVEY AND SUMMARY Comprehensive analysis of DNA polymerase III a subunits and their homologs in bacterial genomes Ke˛stutis Timinskas, Monika Balvočiūtė, Albertas Timinskas and Česlovas Venclovas* Institute of Biotechnology, Vilnius University, Graičiūno 8, Vilnius LT-02241, Lithuania Received July 31, 2013; Revised September 12, 2013; Accepted September 13, 2013 ABSTRACT INTRODUCTION DNA polymerase III is a tripartite protein machine responsible for replication of bacterial genome (1–5). *To whom correspondence should be addressed. Tel: +370 5 269 1881; Fax: +370 5 260 2116; Email: Present address:  _ , Institut für Mathematik und Informatik, Ernst Moritz Arndt Universität Greifswald, Germany. Monika Balvočiute The authors wish it to be known that, in their opinion, the first two authors should be regarded as Joint First Authors. ß The Author(s) 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. The analysis of 2000 bacterial genomes revealed that they all, without a single exception, encode one or more DNA polymerase III a-subunit (PolIIIa) homologs. Classified into C-family of DNA polymerases they come in two major forms, PolC and DnaE, related by ancient duplication. While PolC represents an evolutionary compact group, DnaE can be further subdivided into at least three groups (DnaE13). We performed an extensive analysis of various sequence, structure and surface properties of all four polymerase groups. Our analysis suggests a specific evolutionary pathway leading to PolC and DnaE from the last common ancestor and reveals important differences between extant polymerase groups. Among them, DnaE1 and PolC show the highest conservation of the analyzed properties. DnaE3 polymerases apparently represent an ‘impaired’ version of DnaE1. Nonessential DnaE2 polymerases, typical for oxygen-using bacteria with large GC-rich genomes, have a number of features in common with DnaE3 polymerases. The analysis of polymerase distribution in genomes revealed three major combinations: DnaE1 either alone or accompanied by one or more DnaE2s, PolC+DnaE3 and PolC+DnaE1. The first two combinations are present in Escherichia coli and Bacillus subtilis, respectively. The third one (PolC+DnaE1), found in Clostridia, represents a novel, so far experimentally uncharacterized, set. It consists of a DNA polymerase, its processivity factor b-clamp and a clamp loader complex. The actual DNA synthesis is performed by the polymerase III a-subunit (PolIIIa), classified into the C-family of DNA polymerases (6). Surprisingly, bacterial PolIIIa subunits are both structurally and evolutionary distinct from eukaryotic and archaeal replicative DNA polymerases (7,8) that belong to the B-family. Instead, the PolIIIa catalytic domain is distantly related to the X-family of DNA polymerases (7,8), exemplified by eukaryotic Polb, a polymerase acting in DNA excision repair (9,10). It should be noted that this unexpected relationship could not be detected by protein sequence comparison and only became apparent in the context of 3D structures (7,8). Although polymerases of C and X families are not globally similar, a strong case for their common evolutionary origin could be made based on the observation that they share a common fold of corresponding ‘palm’ domains and bind DNA in the same manner (11). In contrast, ‘palm’ domains of DNA polymerases belonging to A, B and Y families have entirely different fold. Taken together, these findings lend additional support for the hypothesis that bacterial replicative polymerases (C-family) on one hand and archaeal/eukaryotic replicative polymerases (B-family) on the other hand have evolved as components of two independent DNA replication systems (12). Another interesting observation is that C-family polymerases are essentially confined to the bacteria kingdom. Only a handful of PolIIIa homologs have been detected in bacteriophages, which predominantly use B-family (and to lesser extent A-family) DNA polymerases (13,14). One of the explanations for the scarcity of PolIIIa homologs even in bacteria-infecting viruses is that the C-family is evolutionary ‘young’ compared with the B-family (13). Owing to their relatively late emergence, C-family DNA polymerases might have failed to make a significant imprint in the B-family– dominated viral landscape (13), and a few instances of 1394 Nucleic Acids Research, 2014, Vol. 42, No. 3 If the number of distinct PolIIIa subunits and their role in a bacterial cell are considered, there also are notable differences. The widely studied E. coli encodes a sole DnaE-type PolIIIa subunit, which performs DNA synthesis of both leading and lagging strands (1,19,20). However, this is not a universal situation in the bacterial world. For example, low-GC Gram-positive bacteria were found to have both PolC and DnaE (17). Experiments with B. subtilis and some other Gram-positive bacteria showed that both types of PolIIIa subunits are essential (21–23). Initially, it was thought that PolC and DnaE are leading and lagging strand polymerases, respectively (21). However, more recently, in vitro experiments with the reconstituted B. subtilis replisome (24) revealed a different picture of their division of labor. It turned out that DnaE makes an initial extension of the RNA primer on both strands and then PolC takes over for rapid synthesis of long stretches of DNA (24). In this regard, B. subtilis DnaE is reminiscent of eukaryotic Pol a, which extends the RNA primer and then makes way for a processive replicase (25). Some bacteria have a second copy of DnaE, usually referred to as DnaE2. So far, genetic studies targeting dnaE2, all without a single exception, identified it as a nonessential gene (26–32), indicating that DnaE2 is not required for chromosomal DNA replication. Instead, DnaE2 has been associated with DNA damage-inducible error-prone translesion DNA synthesis (TLS) (26–28,31,32). In genomes, dnaE2 is typically found as part of LexA-regulated contiguous or split multigene cassette, which includes two other genes, imuA/imuA’ and imuB (27,33,34). The two genes encode catalytically Figure 1. Structural organization of DnaE and PolC forms of C-family DNA polymerases. Crystal structures of T. aquaticus DnaE (left, PDB ID: 3E0D) and G. kaustophilus PolC (right, PDB ID: 3F2B) complexes with the DNA displayed in same orientation. Protein structures are shown as solvent accessible surfaces with different structural modules shown in different colors. The missing NTD and the exonuclease domain (Exo) in PolC structure are represented correspondingly by a pair of elli (...truncated)