Substantial Sequence Divergence Research Articles

The sequence of a 200 kb portion of a Plasmodium vivax chromosome reveals a high degree of conservation with Plasmodium falciparum chromosome 3

Within a 199 866 base pair (bp) portion of a Plasmodium vivax chromosome we identified a conserved linkage group consisting of at least 41 genes homologous to Plasmodium falciparum genes located on chromosome 3. There were no P. vivax homologues of the P. falciparum cytoadherence-linked asexual genes clag 3.2, clag 3.1 and a var C pseudogene found on the P. vivax chromosome. Within the conserved linkage group, the gene order and structure are identical to those of P. falciparum chromosome 3. This conserved linkage group may extend to as many as 190 genes. The subtelomeric regions are different in size and the P. vivax segment contains genes for which no P. falciparum homologues have been identified to date. The size difference of at least 900 kb between the homologous P. vivax chromosome and P. falciparum chromosome 3 is presumably due to a translocation. There is substantial sequence divergence with a much higher guanine+cytosine (G+C) content in the DNA and a preference for amino acids using GC-rich codons in the deduced proteins of P. vivax. This structural conservation of homologous genes and their products combined with sequence divergence at the nucleotide level makes the P. vivax genome a powerful tool for comparative analyses of Plasmodium genomes.

Interactions of human NKG2D with its ligands MICA, MICB, and homologs of the mouse RAE-1 protein family.

NKG2D is an activating receptor that is expressed on most natural killer (NK) cells, CD8 alphabeta T cells, and gammadelta T cells. Among its ligands is the distant major histocompatibility complex class I homolog MICA, which has no function in antigen presentation but is induced by cellular stress. To extend previous functional evidence, the NKG2D-MICA interaction was studied in isolation. NKG2D homodimers formed stable complexes with monomeric MICA in solution, demonstrating that no other components were required to facilitate this interaction. MICA glycosylation was not essential but enhanced complex formation. Soluble NKG2D also bound to cell surface MICB, which has structural and functional properties similar to those of MICA. Moreover, NKG2D stably interacted with surface molecules encoded by three newly identified cDNA sequences (N2DL-1, -2, and -3), which are identical to the human ULBP proteins and may represent homologs of the mouse retinoic acid-early inducible family of NKG2D ligands. Because of the substantial sequence divergence among these molecules, these results indicated promiscuous modes of receptor binding. Comparison of allelic variants of MICA revealed large differences in NKG2D binding that were associated with a single amino acid substitution at position 129 in the alpha2 domain. Varying affinities of MICA alleles for NKG2D may affect thresholds of NK-cell triggering and T-cell modulation.

A SINGLE ANCIENT ORIGIN OF BROOD PARASITISM IN AFRICAN FINCHES: IMPLICATIONS FOR HOST-PARASITE COEVOLUTION

Robust phylogenies for brood-parasitic birds, their hosts, and nearest nesting relatives provide the framework to address historical questions about host-parasite coevolution and the origins of parasitic behavior. We tested phylogenetic hypotheses for the two genera of African brood-parasitic finches, Anomalospiza and Vidua, using mitochondrial DNA sequence data from 43 passeriform species. Our analyses strongly support a sister relationship between Vidua and Anomalospiza, leading to the conclusion that obligate brood parasitism evolved only once in African finches rather than twice, as has been the conventional view. In addition, the parasitic finches (Viduidae) are not recently derived from either weavers (Ploceidae) or grassfinches (Estrildidae), but represent a third distinct lineage. Among these three groups, the parasitic finches and estrildids, which includes the hosts of all 19 Vidua species, are sister taxa in all analyses of our full dataset. Many characters shared by Vidua and estrildids, including elaborate mouth markings in nestlings, unusual begging behavior, and immaculate white eggs, can therefore be attributed to common ancestry rather than convergent evolution. The host-specificity of mouth mimicry in Vidua species, however, is clearly the product of subsequent host-parasite coevolution. The lineage leading to Anomalospiza switched to parasitizing more distantly related Old World warblers (Sylviidae) and subsequently lost these characteristics. Substantial sequence divergence between Vidua and Anomalospiza indicates that the origin of parasitic behavior in this clade is ancient (∼20 million years ago), a striking contrast to the recent radiation of extant Vidua. We suggest that the parasitic finch lineage has experienced repeated cycles of host colonization, speciation, and extinction through their long history as brood parasites and that extant Vidua species represent only the latest iterations of this process. This dynamic process may account for a significantly faster rate of DNA sequence evolution in parasitic finches as compared to estrildids and other passerines. Our study reduces by one the tally of avian lineages in which obligate brood parasitism has evolved and suggests an origin of parasitism that involved relatively closely related species likely to accept and provide appropriate care to parasitic young. Given the ancient origin of parasitism in African finches, ancestral estrildids must have been parasitized well before the diversification of extant Vidua, suggesting a long history of coevolution between these lineages preceding more recent interactions between specific hosts and parasites.Corresponding Editor: B. Crespi

Disruption of phacA, an Aspergillus nidulans Gene Encoding a Novel Cytochrome P450 Monooxygenase Catalyzing Phenylacetate 2-Hydroxylation, Results in Penicillin Overproduction

Aspergillus nidulans utilizes phenylacetate as a carbon source via homogentisate, which is degraded to fumarate and acetoacetate. Mutational evidence strongly suggested that phenylacetate is converted to homogentisate through two sequential hydroxylating reactions in positions 2 and 5 of the aromatic ring. Using cDNA substraction techniques, we have characterized a gene, denoted phacA, whose transcription is strongly induced by phenylacetate and which putatively encodes a cytochrome P450 protein. A disrupted phacA strain does not grow on phenylacetate but grows on 2-hydroxy- or 2, 5-dihydroxyphenylacetate. Microsomal extracts of the disrupted strain are deficient in the NADPH-dependent conversion of phenylacetate to 2-hydroxyphenylacetate. We conclude that PhacA catalyzes the ortho-hydroxylation of phenylacetate, the first step of A. nidulans phenylacetate catabolism. The involvement of a P450 enzyme in the ortho-hydroxylation of a monoaromatic compound has no precedent. In addition, PhacA shows substantial sequence divergence with known cytochromes P450 and defines a new family of these enzymes, suggesting that saprophytic fungi may represent a source of novel cytochromes P450. Phenylacetate is a precursor for benzylpenicillin production. phacA disruption increases penicillin production 3-5-fold, indicating that catabolism competes with antibiotic biosynthesis for phenylacetate and strongly suggesting strategies for Penicillium chrysogenum strain improvement by reverse genetics.

Sequence analysis of the oxidase/reductase genes upstream of the Rhodococcus erythropolis aldehyde dehydrogenase gene thcA reveals a gene organisation different from Mycobacterium tuberculosis.

The sequence of the DNA region upstream of the thiocarbamate-inducible aldehyde dehydrogenase gene thcA of Rhodococcus erythropolis NI86/21 was determined. Most of the predicted ORFs are related to various oxidases/reductases, including short-chain oxidases/reductases, GMC oxidoreductases, alpha-hydroxy acid oxidases (subfamily 1 flavin oxidases/dehydrogenases), and subfamily 2 flavin oxidases/dehydrogenases. One ORF is related to enzymes involved in biosynthesis of PQQ or molybdopterin cofactors. In addition, a putative member of the TetR family of regulatory proteins was identified. The substantial sequence divergence from functionally characterized enzymes precludes a reliable prediction about the probable function of these proteins at this stage. In Mycobacterium tuberculosis H37Rv, most of these ORFs have homologs that are also clustered in the genome, but some striking differences in gene organization were observed between Rhodococcus and Mycobacterium.

FAST-2 is a mammalian winged-helix protein which mediates transforming growth factor beta signals.

The mechanisms by which transforming growth factor beta (TGF-beta) and related ligands regulate transcription remain poorly understood. The winged-helix (WH) transcription factor fork head activin signal transducer 1 (FAST-1) was identified as a mediator of activin signaling in Xenopus embryos (X. Chen, M. J. Rubock, and M. Whitman, Nature 383:691-696, 1996). We have cloned a novel WH gene from the mouse which shares many properties with FAST-1. We find that this gene, which we call FAST-2, is able to mediate transcriptional activation by TGF-beta. FAST-2 also interacts directly with Smad2, a cytoplasmic protein which is translocated to the nucleus in response to TGF-beta, and forms a multimeric complex with Smad2 and Smad4 on the activin response element, a high-affinity binding site for FAST-1. Analysis of the sequences of FAST-1 and FAST-2 reveals substantial protein sequence divergence compared to known vertebrate orthologs in the WH family. This suggests that FAST-2 represents a new WH gene related to FAST-1, which functions to mediate TGF-beta signals in mammals. We have also examined the structure of the FAST-2 gene and find that it overlaps with a kinesin motor protein gene. The genes are transcribed in opposite orientations, and their transcripts overlap in the 3' untranslated region.

Molecular characterization of an exceptionally acidic lysozyme-like protein from the protozoon Entamoeba histolytica

The protozoan parasite Entamoeba histolytica contains a second antibacterial protein with lysozyme-like properties. The newly recognized bacteriolytic protein was purified from extracts of amoebic trophozoites to allow amino-terminal sequencing. Subsequent molecular cloning revealed that it is an isoform of the amoeba lysozyme described previously but also demonstrated a substantial sequence divergence of the two forms. As lysozymes typically are basic proteins, the novel amoebic protein differs markedly in having a p I of 4.5. There is no significant similarity of both amoeba lysozymes with any bacteriolytic protein of other organisms reported so far; however, striking sequence identity is found with predicted gene products of unknown function derived from the bacteria-feeding nematode Caenorhabditis elegans.

Genetic diversity in the Mycobacterium tuberculosis complex based on variable numbers of tandem DNA repeats.

Genetic loci containing variable numbers of tandem repeats (VNTR loci) form the basis for human gene mapping and identification, forensic analysis and paternity testing. The variability of bacterial tandem repeats has not been systematically studied. Eleven tandem repeat loci in the M. tuberculosis genome were analysed. Five major polymorphic tandem repeat (MPTR) loci contained 15-bp repeats with substantial sequence variation in adjacent copies. Six exact tandem repeat (ETR) loci contained large DNA repeats with identical sequences in adjacent repeats. These 11 loci were amplified in 48 strains to determine the number of tandem repeats at each locus. The strains analysed included 25 wild-type strains of M. tuberculosis, M. bovis, M. africanum and M. microti and 23 substrains of the attenuated M. bovis BCG vaccine. One of the five MPTR loci and all six ETR loci had length polymorphisms corresponding to insertions or deletions of tandem repeats. Most ETR loci were located in intergenic regions where copy number may influence expression of downstream genes. Each ETR locus had multiple alleles in the panel. Combined analysis identified 22 distinct allele profiles in 25 wild-type strains of the M. tuberculosis complex and five allele profiles in 23 M. bovis BCG substrains. Allele profiles were reproducible and stable, as demonstrated by analyses of multiple isolates of particular reference strains obtained from different laboratories. VNTR typing may be generally useful for strain differentiation and evolutionary studies in bacteria.

The autosomal chorion locus of the medfly Ceratitis capitata. I. Conserved synteny, amplification and tissue specificity but sequence divergence and altered temporal regulation.

We report the isolation, full sequence characterization, amplification and expression properties of medfly chorion genes corresponding to the autosomal chorion locus of Drosophila. These genes are found adjacent to the paramyosin gene and are organized in the same order and tandem orientation as their Drosophila homologues, although they are spaced further apart. They show substantial sequence divergence from their Drosophila homologues, including novel peptide repeats and a new spacing of the tyrosines, which are known to be cross-linked in Dipteran chorion. The genes are amplified and expressed during oogenesis, as in Drosophila. Three of them are expressed in the same relative temporal order as in Drosophila but the fourth gene, the homologue of s15, shows a clear shift to an earlier expression period. This is the first known instance of changed temporal regulation in dipteran chorion genes.

Specificity of PS integrin function during embryogenesis resides in the alpha subunit extracellular domain.

We tested the ability of different integrin alpha subunits to substitute for each other during embryonic development. Two alpha subunits, which form heterodimers with the same betaPS subunit, are expressed in complementary tissues in the Drosophila embryo, with alphaPS1 expressed in the epidermis and endoderm, and alphaPS2 expressed in the mesoderm. As a result the two integrin heterodimers are present on opposite surfaces at sites of interaction between the mesoderm and the other cell layers where they are required for normal development. Using the GAL4 system, we are able to rescue fully the embryonic lethality of an alphaPS2 null mutation with a UAS-alphaPS2 transgene, but only partially with a UAS-alphaPS1 gene, due to partial rescue of both muscle and midgut phenotypes. Similarly we are able to rescue the embryonic/first instar larval lethality of an alphaPS1 null mutation gene with UAS-alphaPS1, but only partially with UAS-alphaPS2. Each UAS-alpha gene, when it contains the cytoplasmic domain from the other alpha subunit, maintains an equivalent ability to rescue its own mutation and cannot fully rescue a mutation in the other alpha. We conclude that the two alpha subunits are not equivalent and have distinct functions which reside in the extracellular domains.

Independent genetic control of maize starch-branching enzymes IIa and IIb. Isolation and characterization of a Sbe2a cDNA.

In maize (Zea mays L.) three isoforms of starch-branching enzyme (SBEI, SBEIIa, and SBEIIb) are involved in the synthesis of amylopectin, the branched component of starch. To isolate a cDNA encoding SBEIIa, degenerate oligonucleotides based on domains highly conserved in Sbe2 family members were used to amplify Sbe2-family cDNA from tissues lacking SBEIIb activity. The predicted amino acid sequence of Sbe2a cDNA matches the N-terminal sequence of SBEIIa protein purified from maize endosperm. The size of the mature protein deduced from the cDNA also matches that of SBEIIa. Features of the predicted protein are most similar to members of the SBEII family; however, it differs from maize SBEIIb in having a 49-amino acid N-terminal extension and a region of substantial sequence divergence. Sbe2a mRNA levels are 10-fold higher in embryonic than in endosperm tissue, and are much lower than Sbe2b in both tissues. Unlike Sbe2b, Sbe2a-hybridizing mRNA accumulates in leaf and other vegetative tissues, consistent with the known distribution of SBEIIa and SBEIIb activities.

Analysis of protein binding to the Sma/Cla DNA repeat in pathogenic Neisseriae.

Antigenic variation of the pilus is an essential component of Neisseria gonorrhoeae pathogenesis. Unidirectional recombination of silent pilin DNA into an expressed pilin gene allows for substantial sequence variation of this highly immunogenic surface structure. While the RecA protein is required for pilin gene recombination, the factors which maintain the silent reservoir of pilin sequences and/or allow unidirectional recombination from silent to expression loci remain undefined. We have previously shown that a conserved sequence at the 3'end of all pilin loci (the Sma/Cla repeat) is required to be present at the expression locus for efficient recombination from the silent loci. In this study, the binding of gonococcal proteins to this DNA sequence was investigated. Gel mobility shift assays and competition experiments using deletion derivatives of the repeat, show that multiple activities bind to different regions of the Sma/Cla repeat and define the boundaries of the binding sequences. Moreover, only the pathogenic Neisseria harbor proteins which specifically bind to this repeat, suggesting a correlation between the expression of these DNA binding proteins and the potential to cause disease.

The genome of the pseudo T-even bacteriophages, a diverse group that resembles T4

Polymerase chain reaction analysis of a large collection of bacteriophages with T-even morphology revealed four phages that are distantly related to all the others. The genomes of these pseudo T-even phages hybridized under stringent conditions to only a limited portion of the T4 genome that encodes virus head, head-to-tail joining and contractile tail genes. Except for this region, no extensive hybridization was detected between most pairs of the different pseudo T-even genomes. Sequencing of this conserved region of the pseudo T-even phage RB49 revealed substantial nucleotide sequence divergence from T4 (∼30% to 40%), and random genomic sequencing of this phage indicated that more than a third of its sequences had no detectable homology to T4. Among those sequences related to the T-even genes were virion structural components including the constituents of the phage base plate. Only a few sequences had homology to T4 early functions; these included ribonucleotide diphosphatase reductase, DNA ligase and the large subunit of DNA topoisomerase. The genomes of the pseudo T-even phage were digested by restriction enzymes that are unable to digest the T-even DNAs which contain glucosylated hydroxymethyl-cytosine residues. This suggests that only limited nucleotide modifications must be present in the pseudo T-even genomes. Conservation of much of the morphogenetic region of these diverse phage genomes may reflect particularly strong sequence constraints on these gene products. However, other explanations are considered, including the possibility that the various morphogenetic segments were acquired by the pseudo T-even genomes by modular evolution. These results support the notion that phage evolution may proceed within a network of both closely and distantly related genomes.

Descriptive and functional characterization of variation in theFundulus heteroclitus Ldh-B proximal promoter

Variation in enzyme expression may be an important mechanism for physiological and evolutionary adaptation. The Ldh-B locus in the teleost fish Fundulus heteroclitus is one of a very few loci for which an evolutionary difference in transcription rate between populations has been demonstrated. To begin to understand the molecular modifications that are responsible for altering transcription, we have characterized the Ldh-B proximal promoter using a combination of sequence analysis, transient transfection, and in vivo footprinting. The Ldh-B gene has several transcription start sites and a TATA-less, Inr (initiator of transcription motif) containing promoter with multiple Sp1-like motifs. Transfection experiments reveal that Sp1 sites, TCC repeats, and Inrs are functional components of the proximal promoter. We find substantial sequence variation between populations within the proximal promoter (250 bp from the transcription start sites) and footprinting analysis indicates that some of this sequence variation is associated with differential protein binding to the apparent TFIID binding site and Sp1 sites. Together, these data suggest that variation in the Ldh-B proximal promoter may play a role in the observed difference in transcription rates between northern and southern populations of F. heteroclitus.

IS1236, a newly discovered member of the IS3 family, exhibits varied patterns of insertion into the Acinetobacter calcoaceticus chromosome.

Analysis of spontaneous mutations in Acinetobacter calcoaceticus revealed a 1237 bp insertion sequence named IS1236 and possessing a nucleotide sequence resembling those of members of the lS3 family. The chromosome of A. calcoaceticus strain ADP1 contains seven copies of IS1236 which appears to insert preferentially into pobR, the transcriptional activator of the structural gene for p-hydroxybenzoate hydroxylase. IS1236 creates tandem 3 bp DNA duplications flanking the sites of its insertion in pobR. Different duplication patterns are found following insertion of IS1236 into pcaH, a structural gene for protocatechuate 3,4-dioxygenase. Therefore the insertion of properties of IS1236 appear to be influenced by its DNA target. Amino acid sequences associated with the apparent transposase function have been conserved in ORFB of IS1236 whereas the presumed DNA-binding helix-turn-helix region of IS1236 ORFA exhibits substantial amino acid sequence divergence from its IS3 counterparts. IS1236 ORFA and ORFB coding sequences overlap considerably, and sequence evidence indicates mechanisms for ORFB expression in IS1236 may resemble those employed by other members of the IS3 family. Portions of the IS1236 terminal repeats exhibit substantial sequence divergence from other members of the IS3 family, but evolution appears to have conserved a mechanism preventing expression of the insertion sequence genes as a consequence of transcriptional readthrough.

Genotype determination of hepatitis C virus from northern India: identification of a new subtype.

Hepatitis C virus (HCV) shows substantial nucleotide sequence diversity distributed throughout the viral genome, with many variants showing only 68-79% overall sequence homology. This has led to problems in diagnosis of HCV using commercial immunoassays. Based on clustering of homologous sequences, various genotypes and subtypes of HCV have been described from different geographical regions. In the present study, 11 isolates from India were genotyped using sequence comparison for part of the non-structural (NS5) and structural (core) regions. Parts of the genome covering 451 bp (nt 9-459) of the core gene and a 249 bp fragment (nt 7959-8207) of the NS5 gene were reverse transcribed and amplified using nested polymerase chain reaction (RT-PCR). The amplified fragments were cloned and sequenced. The classification into genotypes was done on the basis of phylogenetic analysis. Four isolates showed sequence homology to type 1b. Two of the isolates were classified as type 3a. One isolate was classified as type 3b and the remaining four isolates were found to be variants of type 3 but did not belong to any designated subtype. On the basis of phylogenetic analysis two of the unclassified isolates were put into a new subtype of 3 named as 3g. In one of these variants, parts of a 5'-noncoding (5' NCR; 204 bp), envelope-E1 (435 bp), and NS3 (502 bp) regions were also amplified, cloned, and sequenced. This study demonstrates the type 3 variants including a new subtype (3g) to be the major cause of HCV infection in India.

Characterization of the gene encoding the polymorphic immunodominant molecule, a neutralizing antigen of Theileria parva.

Theileria parva, a tick-transmitted protozoan parasite related to Plasmodium spp., causes the disease East Coast fever, an acute and usually fatal lymphoproliferative disorder of cattle in Africa. Previous studies using sera from cattle that have survived infection identified a polymorphic immunodominant molecule (PIM) that is expressed by both the infective sporozoite stage of the parasite and the intracellular schizont. Here we show that mAb specific for the PIM Ag can inhibit sporozoite invasion of lymphocytes in vitro. A cDNA clone encoding the PIM Ag of the T. parva (Muguga) stock was obtained by using these mAb in a novel eukaryotic expression cloning system that allows isolation of cDNA encoding cytoplasmic or surface Ags. To establish the molecular basis of the polymorphism of PIM, the cDNA of the PIM Ag from a buffalo-derived T. parva stock was isolated and its sequence was compared with that of the cattle-derived Muguga PIM. The two cDNAs showed considerable identity in both the 5' and 3' regions, but there was substantial sequence divergence in the central regions. Several types of repeated sequences were identified in the variant regions. In the Muguga form of the molecule, there were five tandem repeats of the tetrapeptide, QPEP, that were shown, by transfection of a deleted version of the PIM gene, not to react with several anti-PIM mAbs. By isolating and sequencing the genomic version of the gene, we identified two small introns in the 3' region of the gene. Finally, we showed that polyclonal rat Abs against recombinant PIM neutralize sporozoite infectivity in vitro, suggesting that the PIM Ag should be evaluated for its capacity to immunize cattle against East Coast Fever.

A consensus prediction of the secondary structure for the 6‐phospho‐β‐D‐galactosidase superfamily

Two separate unrefined models for the secondary structure of two subfamilies of the 6-phospho-beta-D-galactosidase superfamily were independently constructed by examining patterns of variation and conservation within homologous protein sequences, assigning surface, interior, parsing, and active site residues to positions in the alignment, and identifying periodicities in these. A consensus model for the secondary structure of the entire superfamily was then built. The prediction tests the limits of an unrefined prediction made using this approach in a large protein with substantial functional and sequence divergence within the family. The protein belongs to the (alpha-beta class), with the core beta strands aligned parallel. The supersecondary structural elements that are readily identified in this model is a parallel beta sheet built by strands C, D, and E, with helices 2 and 3 connecting strands (C+D) and (D+E), respectively, and an analogous beta-alpha unit (strand G and helix 7) toward the end of the sequence. The resemblance of the supersecondary model to the tertiary structure formed by 8-fold alpha-beta barrel proteins is almost certainly not coincidental.

Precore and contiguous regions of hepatitis B virus in liver transplantation for end-stage hepatitis B

Background/Aims: Recurrent hepatitis B virus (HBV) infection is the leading cause of mortality and morbidity after orthotopic liver transplantation (OLT) for HBV-related liver disease, but the extent of viral genetic variation in this setting remains unknown. Methods: Eight patients who underwent OLT for HBV-related liver disease were studied; 7 had cirrhosis and 1 had fulminant hepatitis. Four patients received long-term hepatitis B immunoglobulin prophylaxis. A 240-base pair fragment (1742–1981) comprising the precore region of HBV was amplified by polymerase chain reaction from sera drawn before OLT and 6,12, and 24 months after OLT and analyzed. Results: All sera were positive by polymerase chain reaction. Nucleotide sequence variations were congruent within most patients before and after OLT; however, in one patient, substantial sequence variation was observed, suggesting infection with a new HBV strain. No sequence variation associated with a particular outcome could be identified. Two patients harbored HBV variants with a deletion or insertion upstream of the precore messenger RNA initiation site. Conclusions: Reinfection after OLT can occasionally be caused by HBV strains different from the one present before OLT. Changes within the sequenced region are not predictive of the outcome of reinfection.

Cloning, sequencing, and expression of RecA proteins from three distantly related thermophilic eubacteria.

Sequences of the recA genes of the highly divergent thermopholic eubacteria Thermus aquaticus (and Thermus thermophilus), Thermotoga maritima, and Aquifex pyrophilus were determined from fragments derived by polymerase chain reaction (PCR) with degenerate primers and from inverse PCR products obtained using unique primers based on the fragment sequences. The source of the PCR products was verified by Southern hybridization. Complete PCR-derived recA genes were cloned into an expression vector regulated by a temperature-sensitive lambda-repressor, and independently derived clones expressing thermostable recA were selected. DNA sequences were verified to be authentic by direct cycle-sequencing of PCR products and/or sequencing of several clones. In contrast to Escherichia coli RecA protein, all the purified thermophilic RecA proteins exhibited single-stranded DNA-dependent ATPase activity optima above 70 degrees C. Phylogenetic analysis of RecA sequences suggested that the thermophilic RecA proteins were at least as different from one another as were Gram-positive organisms, mesophilic Gram-negative organisms, and cyanobacteria. In spite of substantial sequence divergence, interesting characteristics of the thermostable RecA proteins included increased valine content, common amino acid replacements at two highly conserved sites, and an increase in the calculated isoelectric point of approximately a full pH unit.