Evolution and horizontal transfer of dUTPase-encoding genes in viruses and their hosts

Abstract

dUTPase is a ubiquitous and essential enzyme responsible for regulating cellular levels of dUTP. The dut gene exists as single, tandemly duplicated, and tandemly triplicated copies. Crystallized single-copy dUTPases have been shown to assemble as homotrimers. dUTPase is encoded as an auxiliary gene in a number of virus genomes. The origin of viral dut genes has remained unresolved since their initial discovery. A comprehensive analysis of dUTPase amino acid sequence relationships was performed to explore the evolutionary dynamics of dut in viruses and their hosts. Our data set, comprised of 24 host and 51 viral sequences, includes representative sequences from available eukaryotes, archaea, eubacteria cells, and viruses, including herpesviruses. These amino acid sequences were aligned by using a hidden Markov model approach developed to align divergent data. Known secondary structures from single-copy crystals were mapped onto the aligned duplicate and triplicate sequences. We show how duplicated dUTPases might fold into a monomer, and we hypothesize that triplicated dUTPases also assemble as monomers. Phylogenetic analysis revealed at least five viral dUTPase sequence lineages in well-supported monophyletic clusters with eukaryotic, eubacterial, and archaeal hosts. We have identified all five as strong examples of horizontal transfer as well as additional potential transfer of dut genes among eubacteria, between eubacteria and viruses, and between retroviruses. The evidence for horizontal transfers is particularly interesting since eukaryotic dut genes have introns, while DNA virus dut genes do not. This implies that an intermediary retroid agent facilitated the horizontal transfer process between host mRNA and DNA viruses.

FIG. 1

Three types of motif arrangements. (A) Of the 75 dUTPases in this study, 62 consist of a single set of five conserved motifs (I to V). In addition to this common arrangement, two other variations are known. (B) All of the available alpha- and gammaherpesvirus sequences (six each) encode a dUTPase that has a single set of motifs but is twice as long as the common single arrangement. Motif III is conserved only in the amino portion of the protein, while motifs I, II, IV, and V are conserved in the carboxy portion. Possible remnants of motifs which imply a duplication of a single copy in the ancestor of these lineages and a subsequent loss of duplicate motifs are indicated by a question mark. The conservation of these remnants is variable within and between taxa (data not shown). Of the 12 available sequences, 10 have no amino acid residues in the position where the remnant of motif V would be; this is indicated by a dash (–). Sequences with this arrangement were analyzed in two sections: amino and carboxy. (C) The C. elegans sequence contains a triplicated dut colinearly transcribed and translated. Each copy of the triplicated protein has a complete set of the five motifs found in the common single copy arrangement of the gene. Each copy (amino, middle, and carboxy) was analyzed separately.

FIG. 2

The location of the dut gene is variable in the genomes of eubacteria, DNA viruses, and retroviruses. Closely related proteobacteria, E. coli, P. aeruginosa, and C. burnetii, all encode a dUTPase as do their more distant relatives. A eubacterial gene map comparison indicates that the type and number of genes surrounding dut varies. The DNA regions bounding the dut gene of P. aeruginosa were determined by conceptual translation. An additional potential ORF of unknown homology is indicated by a question mark. Alpha- and gammaherpesviruses both encode dUTPase, but the dut gene is not in the same position relative to the primase in these two lineages (24). Poxviruses have substituted genes upstream of dut (37). Between dut and the ribonucleotide reductase there is a single homologous but unidentified reading frame in both vaccinia virus and suid poxvirus. Between the transcription initiation factor gene and dut, however, neither of the two ORFs in suid poxvirus are homologous to any of the five ORFs in vaccinia virus. Two lineages of retroviruses (MMTV-like and nonprimate lentiviruses) transcribe dut as part of the gag and pol polycistrons, respectively. dUTPase is not encoded by close relatives of MMTV and nonprimate lentiviruses, neither human T-lymphotropic virus-like retroviruses, Rous sarcoma virus, nor HIV. Gene region and gene abbreviations are as follows. Single-copy dut genes are designated by an asterisk and duplicated dut genes by two asterisks. Symbols are as follows. Eubacteria: , the DNA synthesis flavoprotein; , unknown function; , the phosphomannomutase; , the outer membrane protein PIB. Herpesvirus genes: , the ribonucleotide reductase-related protein; , the primase. Poxviruses: , the transcription initiation factor; , the ribonucleotide reductase. Retroviruses: , the protease; , the reverse transcriptase; , the ribonuclease H; , the integrase. Unidentified homologous reading frames are indicated (). Unidentified nonhomologous reading frames are indicated with a question mark.

FIG. 3

Representative alignment of sequences analyzed in this study. OSM regions are indicated with bars and Roman numerals. MIRs include all residues between motifs I, II, III, IV, and V and those carboxyl to motif V (Arabic numbers in parentheses indicate residues removed for display). dUTPase residues involved in substrate binding as determined by X-ray crystallography are indicated above the alignment: E. coli (ECOL) by a plus symbol (+) (32) and H. sapiens (HSAP) by an asterisk (∗) (50). Residues implicated in dUTPase function by mutagenesis are also indicated above the alignment: FIV (1FIV) by a pound symbol (#) (74), E. coli also by a pound symbol (#) (73), and EIAV by a percent symbol (%) (59, 72). The carboxy copy of alpha- and gammaherpesvirus dUTPases (HH1C and HH4C) have no recognizable motif III. The amino copies of these dUTPases (HH1A and HH4A) are very divergent in motifs I, II, and IV. Motif V is truncated or divergent in HH1A, HH4A, and archaeal/archaeal virus dUTPases (MJAN/SIRV). There is a large insertion (21 to 38 residues) in the carboxy alpha- and gammaherpesvirus dUTPases (HH1C and HH4C) between motifs IV and V. Amino acid residues in a column are highlighted when more than half are identical or highly similar: (FY), (ILMV), (ST), (AG), (DE), (NQ), and (KR). The number of residues trimmed from the amino terminus of each sequence are indicated in parentheses at the beginning of the alignment. For a list of all sequence codes, see the Appendix.

FIG. 4

Shortest unrooted phylogenetic tree consistent with additive distances (Fitch-Margoliash algorithm). Poxvirus and avian adenovirus dUTPases are most similar to those of vertebrates. Chlorella virus (PBCV) dUTPase clusters with those of the eukaryotes. The bacteriophage r1t (BPRT) dUTPase clusters with those of the firmicutes. dUTPase from bacteriophage SPβ (BPSP) clusters with the sequence from host, B. subtilis (BSUB). Bacteriophage T5 (BPT5) also has a dUTPase similar to that of its proteobacterial host, E. coli (ECOL). Archaeal virus SIRV encodes a dUTPase similar to that of D. ambivalens (DAMB). Two proteobacterial dUTPases H. pylori (HPYL) and B. japonicum (BJAP) cluster with those of the groups Firmicutes, Chlamydiales, and Spirochaetales. MMTV relatives and nonprimate lentiviruses encode similar dUTPases. Strong cases for horizontal transfer supported by high bootstraps are indicated in this tree by an asterisk (∗). Other potential but less-well-supported cases are indicated with a pound symbol (#). Branch lengths are roughly proportional to distances, with the exception of very short branches, which are exaggerated for visibility. Groups of taxa corresponding to those listed in Table 1 are indicated with brackets and labels. Bootstrap values of 50% and higher are shown above supported branches. All taxon identifiers are listed in the Appendix.

FIG. 5

Assembly of dUTPase subunits. The four known dUTPase crystals (FIV, EIAV, E. coli, and H. sapiens) indicate a homotrimer folding to form three separate active sites in which each of the five conserved motifs is present. (A) H. sapiens (single). It was observed in the H. sapiens dUTPase that each subunit contributes at least one motif to each active site (50). In this diagram, the characteristic beta-barrel has been flattened to show the components of the trimer fold. Due to this flattening, the position of motif III is misleading. In the three-dimensional structure, motif III on the solid-gray chain associates with motifs I, II, and IV of the solid-black chain and motif V of the dashed chain. To help clarify these relationships, each active site is indicated by a shape (triangle, square, and circle) surrounding the motifs of which it is comprised (based on the published structure and assembly of H. sapiens dUTPase, [50]). (B) Herpesvirus (duplicated). Based on the H. sapiens structure and our alignment (Fig. 3), a possible structure for an alpha- and a gammaherpesvirus monomer is proposed. Motif III is present in the amino portion of the repeat instead of between motifs II and IV as in H. sapiens (Fig. 1). (C) C. elegans (triplicated). The dUTPase of C. elegans is the only dUTPase example to date of a tandem triplication comprising of three complete single copies (Fig. 1). This protein could potentially fold into a single monomeric dUTPase of the approximate size of the H. sapiens homotrimer. Straight lines indicate conserved structural regions inferred from the alignment. Curved lines indicate that the structure between motif V of one copy and motif I of the next is unknown. Each subunit is indicated by solid-black, dashed, or solid-gray lines. The amino (N) and carboxy (C) termini of each subunit are labeled. Motifs are labeled I to V in their approximate positions in the dUTPase alignment (Fig. 3).