Characterisation of XlCdc1, a Xenopus homologue of the small (PolD2) subunit of DNA polymerase δ; identification of ten conserved regions I–X based on protein sequence comparisons across ten eukaryotic species

Abstract

DNA polymerase δ (Pol δ), which plays keys roles in DNA replication, repair and recombination in eukaryotic cells, comprises at least two essential subunits — a large catalytic subunit (PolD1) possessing both DNA polymerase and 3′–5′ exonuclease activities, and a smaller subunit (PolD2) whose function is not yet clear. Here we describe the cloning and sequencing of a Xenopus cDNA encoding a homologue of the PolD2 subunit. This protein (designated XlCdc1) is 69% identical to the human PolD2 protein and 34% identical to fission yeast Cdc1. Alignment of PolD2 protein sequences across ten eukaryotic species identifies 36 invariant amino-acid positions. These 36 residues are located within ten conserved regions (designated I–X) likely to have key functional roles. Consistent with this, the mutations in six previously identified yeast mutant PolD2 proteins map within conserved regions III, VI, VII and VIII. Several of the invariant amino acids are also conserved across the archaeal DNA polymerase II DP1 protein family.

Introduction

Chromosomal DNA replication in eukaryotic cells requires the activity of a number of distinct multi-subunit DNA polymerases. DNA polymerase δ (Pol δ) is one such enzyme complex that is also involved in DNA repair and recombination (reviewed by Hindges and Hübscher, 1997b). Pol δ comprises a large catalytic subunit, possessing both polymerase and 3′–5′ exonuclease activities, and a variable number of smaller subunits depending on the organism from which the enzyme is purified and most likely also on the purification conditions employed. In mammalian cells, highly purified Pol δ is a heterodimeric enzyme, consisting of a 125 kDa catalytic subunit (hereafter referred to as the PolD1 protein) and an associated 55 kDa subunit (PolD2, reviewed by Downey and So, 1995).

Both the PolD1 and PolD2 proteins are well conserved across evolution. In addition to mammals, PolD1 proteins have been identified in a wide variety of other eukaryotic species, including the yeasts Saccharomyces cerevisiae, Schizosaccharomyces pombe and Candida albicans (Morrison et al., 1990, Nolan and Rosamond, 1996, Park et al., 1993, Pignede et al., 1991), the fruit fly Drosophila melanogaster (Chiang and Lehman, 1995) and the malarial parasite Plasmodium falciparum (Ridley et al., 1991). Multiple sequence alignments of these proteins have been used to identify amino acids that are conserved across evolution and which are therefore likely to be important for the function of the Pol δ enzyme complex (Cullmann et al., 1993, Hindges and Hübscher, 1997b). These conserved residues are also excellent targets for mutational analysis, in particular for reverse genetic studies in the yeasts (see for example Giot et al., 1997, Simon et al., 1991).

Homologues of the mammalian PolD2 proteins, approx. 35% identical at the primary sequence level to their mammalian counterparts, have been identified in S. cerevisiae and S. pombe (Burgers and Gerik, 1998, Eissenberg et al., 1997, Gerik et al., 1998, Giot et al., 1997, Hashimoto et al., 1998, MacNeill et al., 1996, Sugimoto et al., 1995). In both cases genetic analysis has demonstrated that the PolD2 function is an essential one. In S. cerevisiae, highly purified Pol δ is trimeric complex comprising homologues of PolD1 and PolD2 (termed Pol3 and Pol31, respectively; note that Pol31 is also known as Hys2 and Sdp5) together with an additional non-essential subunit (Pol32) that interacts directly with Pol31 (Burgers and Gerik, 1998, Eissenberg et al., 1997, Gerik et al., 1998, Giot et al., 1997, Hashimoto et al., 1998). In S. pombe Pol δ is tetrameric, comprising PolD1, PolD2 and Pol32 homologue proteins (Pol3, Cdc1 and Cdc27, respectively), along with a non-essential subunit Cdm1 (MacNeill et al., 1996, Reynolds et al., 1998, Zuo et al., 1997). In addition to Pol3 and Cdc1, the PolD2-interacting subunit Cdc27 is also an essential protein in this organism.

The processivity of Pol δ is dependent upon its interaction with its trimeric accessory factor PCNA. In vitro polymerase processivity assays carried out using recombinant human PolD1 and PolD2 proteins have shown that PolD2 is required for stimulation of the dimeric enzyme by PCNA (Zhou et al., 1997). How the PolD2 protein acts to facilitate the action of PCNA is unclear, as PolD2 does not appear to bind PCNA directly, nor is PolD1 able to bind PCNA in the absence of PolD2, at least under normal conditions. Instead, PolD2 may act to alter the conformation of the PolD1 protein to render it capable of interacting with PCNA or, alternatively, a stable interaction between the heterodimeric form of the enzyme and PCNA may require PCNA to bind simultaneously to both PolD1 and PolD2. These issues are currently being addressed in a number of laboratories. In the yeasts the situation is further complicated by the fact that the Pol32 and Cdc27 proteins, of S. cerevisiae and S. pombe respectively, also contact PCNA directly and in the case of S. pombe, interaction between Cdc27 and PCNA is essential for Pol δ function (Burgers and Gerik, 1998; Gerik et al., 1998; N. Reynolds and S.A. MacNeill, unpublished results).

Following on from our earlier work identifying the fission yeast Cdc1 protein as a PolD2 homologue (MacNeill et al., 1996), we recently embarked upon a more detailed mutational analysis of the Cdc1 protein using both random and directed mutagenesis techniques (J. Sanchez Garcia, E. Knatko and S.A. MacNeill, unpublished results). In order to focus our directed mutagenesis studies on those residues likely to be essential for PolD2 function, we wished to identify additional PolD2 homologues from other species, incorporate their protein sequences into a multiple sequence alignment and use this information to identify suitable targets for mutational analysis. In this paper we describe the cloning and sequencing of a cDNA encoding a new member of the PolD2 family, from the African clawed toad Xenopus laevis. The XlCdc1 protein, which is 463 amino acids in length, is 69% identical to human PolD2 and 34% identical to fission yeast Cdc1. Through database searching we have obtained the sequences of a total of ten members of the PolD2 family, including four unpublished sequences derived from genome sequencing projects. By aligning the PolD2 protein sequences from these ten eukaryotic species, we have identified ten conserved regions (designated I–X) likely to have important, if not essential, roles in PolD2 function. Consistent with this, the mutations in six previously identified yeast mutant PolD2 proteins map close to the centre of conserved regions III, VI, VII and VIII. In addition, a number of those amino acids found to be conserved across all ten eukaryotic sequences are also conserved in members of the recently described family of DNA polymerase II small subunits (DP1 proteins) from archaeal species (Cann et al., 1998, Ishino et al., 1998, Uemori et al., 1997), once again underlining their likely importance for PolD2 protein function.