Bacterial DNA uptake sequences can accumulate by molecular drive alone.

Uptake signal sequences are DNA motifs that promote DNA uptake by competent bacteria in the family Pasteurellaceae and the genus Neisseria. The genomes of these bacteria contain many copies of their canonical uptake sequence (often >100-fold overrepresentation), so the bias of the uptake machinery causes cells to prefer DNA derived from close relatives over DNA from other sources. However, the molecular and evolutionary forces responsible for the abundance of uptake sequences in these genomes are not well understood, and their presence is not easily explained by any of the current models of the evolution of competence. Here we describe use of a computer simulation model to thoroughly evaluate the simplest explanation for uptake sequences, that they accumulate in genomes by a form of molecular drive generated by biased DNA uptake and evolutionarily neutral (i.e., unselected) recombination. In parallel we used an unbiased search algorithm to characterize genomic uptake sequences and DNA uptake assays to refine the Haemophilus influenzae uptake specificity. These analyses showed that biased uptake and neutral recombination are sufficient to drive uptake sequences to high densities, with the spacings, stabilities, and strong consensus typical of uptake sequences in real genomes. This result greatly simplifies testing of hypotheses about the benefits of DNA uptake, because it explains how genomes could have passively accumulated sequences matching the bias of their uptake machineries.

Coevolution of DNA uptake sequences and bacterial proteomes.

Dramatic examples of repeated sequences occur in the genomes of some naturally competent bacteria, which contain hundreds or thousands of copies of short motifs called DNA uptake signal sequences. Here, we analyze the evolutionary interactions between coding-region uptake sequences and the proteomes of Haemophilus influenzae, Actinobacillus pleuropneumoniae, and Neisseria meningitidis. In all three genomes, uptake sequence accumulation in coding sequences has approximately doubled the frequencies of those tripeptides specified by each species' uptake sequence. The presence of uptake sequences in particular reading frames correlated most strongly with the use of preferred codons at degenerately coded positions, but the density of uptake sequences correlated only poorly with protein functional category. Genes lacking homologs in related genomes also lacked uptake sequences, strengthening the evidence that uptake sequences do not drive lateral gene transfer between distant relatives but instead accumulate after genes have been transferred. Comparison of the uptake sequence-encoded peptides of H. influenzae and N. meningitidis proteins with their homologs from related bacteria without uptake sequences indicated that uptake sequences were also preferentially located in poorly conserved genes and at poorly conserved amino acids. With few exceptions, amino acids at positions encoded by uptake sequences were as well conserved as other amino acids, suggesting that extant uptake sequences impose little or no constraint on coding for protein function. However, this state is likely to be achieved at a substantial cost because of the selective deaths required to eliminate maladaptive mutations that improve uptake sequences.

Evolutionary stability of DNA uptake signal sequences in the Pasteurellaceae.

The DNA-uptake signal sequence (USS) of the bacterium Haemophilus influenzae is highly over-represented in its genome (1,471 copies of the core sequence AAGTGCGGT), and DNA fragments containing USS are preferentially taken up by competent cells. Because this bias favors uptake of conspecific DNA, USSs are often considered a kind of mate recognition system in bacteria, acting as species-specific barriers against uptake of unrelated DNA. However, the H. influenzae USS is highly over-represented in the genomes of three otherwise-divergent Pasteurellaceae species (Pasteurella multocida, Haemophilus somnus, and Actinobacillus actinomycetemcomitans, 927, 1,205, and 1,760 copies, respectively), suggesting that USSs do not always limit exchange. USSs in all these genomes are mainly in coding regions and show no orientation bias around the chromosome, weakening proposed USS functions in transcription termination and chromosome replication. Alignment of homologous genes was used to determine evolutionary relationships between individual USSs. Most H. influenzae USSs were found to have perfect or imperfect homologs (USS at the same location) in at least one other species, and most USSs in the other species had perfect or imperfect homologs in H. influenzae. These homologies suggest that the use of a common USS is due to inheritance of the USS-based uptake system from a common ancestor of the Pasteurellaceae, and it indicates that individual USSs can be evolutionarily stable elements of their genomes. The pattern is consistent with a molecular drive model of USS evolution, with new USSs arising by mutation and preferentially spread to new genomes by the biased DNA-uptake system.

Do bacteria have sex?

Do bacteria have genes for genetic exchange? The idea that the bacterial processes that cause genetic exchange exist because of natural selection for this process is shared by almost all microbiologists and population geneticists. However, this assumption has been perpetuated by generations of biology, microbiology and genetics textbooks without ever being critically examined.

Competence development by Haemophilus influenzae is regulated by the availability of nucleic acid precursors.

DNA uptake by naturally competent bacteria provides cells with both genetic information and nucleotides. In Haemophilus influenzae, competence development requires both cAMP and an unidentified signal arising under starvation conditions. To investigate this signal, competence induction was examined in media supplemented with nucleic acid precursors. The addition of physiological levels of AMP and GMP reduced competence 200-fold and prevented the normal competence-induced transcription of the essential competence genes comA and rec-2. The rich medium normally used for growth allows only limited competence. Capillary electrophoresis revealed only a subinhibitory amount of AMP and no detectable GMP, and the addition of AMP or GMP to this medium also reduced competence 20- to 100-fold. Neither a functional stringent response system nor a functional phosphoenolpyruvate:glycose phosphotransferase system (PTS) was found to be required for purine-mediated repression. Added cAMP partially restored both transcription of competence genes and competence development, suggesting that purines may reduce the response to cAMP. Potential binding sites for the PurR repressor were identified in several competence genes, suggesting that competence is part of the PUR regulon. These observations are consistent with models of competence regulation, in which depleted purine pools signal the need for nucleotides, and support the hypothesis that competence evolved primarily for nucleotide acquisition.

Oxymonads are closely related to the excavate taxon Trimastix.

Despite intensive study in recent years, large-scale eukaryote phylogeny remains poorly resolved. This is particularly problematic among the groups considered to be potential early branches. In many recent systematic schemes for early eukaryotic evolution, the amitochondriate protists oxymonads and Trimastix have figured prominently, having been suggested as members of many of the putative deep-branching higher taxa. However, they have never before been proposed as close relatives of each other. We amplified, cloned, and sequenced small-subunit ribosomal RNA genes from the oxymonad Pyrsonympha and from several Trimastix isolates. Rigorous phylogenetic analyses indicate that these two protist groups are sister taxa and are not clearly related to any currently established eukaryotic lineages. This surprising result has important implications for our understanding of cellular evolution and high-level eukaryotic phylogeny. Given that Trimastix contains small, electron-dense bodies strongly suspected to be derived mitochondria, this study constitutes the best evidence to date that oxymonads are not primitively amitochondriate. Instead, Trimastix and oxymonads may be useful organisms for investigations into the evolution of the secondary amitochondriate condition. All higher taxa involving either oxymonads or Trimastix may require modification or abandonment. Affected groups include four contemporary taxa given the rank of phylum (Metamonada, Loukozoa, Trichozoa, Percolozoa), and the informal excavate taxa. A new "phylum-level" taxon may be warranted for oxymonads and Trimastix.

Point mutations in a peptidoglycan biosynthesis gene cause competence induction in Haemophilus influenzae.

We have identified three new Haemophilus influenzae mutations causing cells to exhibit extreme hypercompetence at all stages of growth. The mutations are in murE, which encodes the meso-diaminopimelate-adding enzyme of peptidoglycan synthesis. All are point mutations causing nonconservative amino acid substitutions, two at a poorly conserved residue (G(435)-->R and G(435)-->W) and the third at a highly conserved leucine (L(361)-->S). The mutant strains have very similar phenotypes and do not exhibit any defects in cell growth, permeability, or sensitivity to peptidoglycan antibiotics. Cells retain the normal specificity of DNA uptake for the H. influenzae uptake signal sequence. The mutations do not bypass genes known to be needed for competence induction but do dramatically increase expression of genes required for the normal pathway of DNA uptake. We conclude that the mutations do not act by increasing cell permeability but by causing induction of the normal competence pathway via a previously unsuspected signal.

Phylogenetic placement of Trichonympha.

Flagellated protists of the Class Hypermastigida have previously been classified on morphology alone, since no molecular sequences have been available. We have isolated DNA from 350 cells of the hypermastigote Trichonympha, manually collected from the hindgut of Zootermopsis angusticollis, and used this DNA as template for polymerase chain reaction amplification of the small-subunit ribosomal RNA gene. The DNA sequence of the amplified fragment is closely related to that of a previously-unidentified gut symbiont from the termite Reticulitermes flavipes, and phylogenetic analysis places both sequences as a sister group to the known trichomonads, in agreement with the morphological classification.

A 3',5' cyclic AMP (cAMP) phosphodiesterase modulates cAMP levels and optimizes competence in Haemophilus influenzae Rd.

Changes in intracellular 3',5' cyclic AMP (cAMP) concentration regulate the development of natural competence in Haemophilus influenzae. In Escherichia coli, cAMP levels are modulated by a cAMP phosphodiesterase encoded by the cpdA gene. We have used several approaches to demonstrate that the homologous icc gene of H. influenzae encodes a functional cAMP phosphodiesterase and that this gene limits intracellular cAMP and thereby influences competence and other cAMP-dependent processes. In E. coli, expression of cloned icc reduced both cAMP-dependent sugar fermentation and beta-galactosidase expression, as has been shown for cpdA. In H. influenzae, an icc null mutation increased cAMP-dependent sugar fermentation and competence development in strains where these processes are limited by mutations reducing cAMP synthesis. When endogenous production of cAMP was eliminated by a cya mutation, an icc strain was 10,000-fold more sensitive to exogenous cAMP than an icc+ strain. The icc strain showed moderately elevated competence under noninducing conditions, as expected, but had subnormal competence increases at onset of stationary phase in rich medium, and on transfer to a nutrient-limited medium, suggesting that excessive cAMP may interfere with induction. Consistent with this finding, a cya strain cultured in 1 mM cAMP failed to develop maximal competence on transfer to inducing conditions. Thus, by limiting cAMP levels, the H. influenzae cAMP phosphodiesterase may coordinate its responses to nutritional stress, ensuring optimal competence development.

The hotspot conversion paradox and the evolution of meiotic recombination.

Studies of meiotic recombination have revealed an evolutionary paradox. Molecular and genetic analysis has shown that crossing over initiates at specific sites called hotspots, by a recombinational-repair mechanism in which the initiating hotspot is replaced by a copy of its homolog. We have used computer simulations of large populations to show that this mechanism causes active hotspot alleles to be rapidly replaced by inactive alleles, which arise by rare mutation and increase by recombination-associated conversion. Additional simulations solidified the paradox by showing that the known benefits of recombination appear inadequate to maintain its mechanism. Neither the benefits of accurate segregation nor those of recombining flanking genes were sufficient to preserve active alleles in the face of conversion. A partial resolution to this paradox was obtained by introducing into the model an additional, nonmeiotic function for the sites that initiate recombination, consistent with the observed association of hotspots with functional sites in chromatin. Provided selection for this function was sufficiently strong, active hotspots were able to persist in spite of frequent conversion to inactive alleles. However, this explanation is unsatisfactory for two reasons. First, it is unlikely to apply to obligately sexual species, because observed crossover frequencies imply maintenance of many hotspots per genome, and the viability selection needed to preserve these would drive the species to extinction. Second, it fails to explain why such a genetically costly mechanism of recombination has been maintained over evolutionary time. Thus the paradox persists and is likely to be resolved only by significant changes to the commonly accepted mechanism of crossing over.

The evolution of bacterial transformation: sex with poor relations.

Bacteria are the only organisms known to actively take up DNA and recombine it into their genomes. While such natural transformation systems may provide many of the same benefits that sexual reproduction provides eukaryotes, there are important differences that critically alter the consequences, especially when recombination's main benefit is reducing the mutation load. Here, analytical and numerical methods are used to study the selection of transformation genes in populations undergoing deleterious mutation. Selection for transformability depends on the shape of the fitness function against mutation. If the fitness function is linear, then transformation would be selectively neutral were it not for the possibility that transforming cells may take up DNA that converts them into nontransformable cells. If the selection includes strong positive (synergistic) epistasis, then transformation can be advantageous in spite of this risk. The effect of low quality DNA (from selectively killed cells) on selection is then studied analytically and found to impose an additional cost. The limited data available for real bacterial populations suggest that the conditions necessary for the evolution of transformation are unlikely to be met, and thus that DNA uptake may have some function other than recombination of deleterious mutations.

Regulation of competence development and sugar utilization in Haemophilus influenzae Rd by a phosphoenolpyruvate:fructose phosphotransferase system.

Changes in intracellular cAMP concentration play important roles in Haemophilus influenzae, regulating both sugar utilization and competence for natural transformation. In enteric bacteria, cAMP levels are controlled by the phosphoenolpyruvate:glycose phosphotransferase system (PTS) in response to changes in availability of the preferred sugars it transports. We have demonstrated the existence of a simple PTS in H. influenzae by several methods. We have cloned the H. influenzae ptsI gene, encoding PTS Enzyme I; genome analysis locates it in a pts operon structurally homologous to those of enteric bacteria. In vitro phosphorylation assays confirmed the presence of functional PTS components. A ptsI null mutation reduced fructose uptake to 1% of the wild-type rate, and abolished fructose fermentation even when exogenous cAMP was provided. The ptsI mutation also prevented fermentation of ribose and galactose, but utilization of these cAMP-dependent sugars was restored by addition of cAMP. In wild-type cells the non-metabolizable fructose analogue xylitol prevented fermentation of these sugars, confirming that the fructose PTS regulates cAMP levels. Development of competence under standard inducing conditions was reduced 250-fold by the ptsI mutation, unless cells were provided with exogenous cAMP. Competence is thus shown to be under direct nutritional control by a fructose-specific PTS.

Cell transfer by filtration: evaluation of protocols for transformational competence.

Collection and washing of cells by centrifugation is a time consuming and inconvenient component of many microbiological protocols, especially when sterility must be maintained. Filtration onto disposable membrane filters efficiently replaced the centrifugation steps in standard transformational competence protocols for Escherichia coli, Haemophilus influenzae, and Saccharomyces cerevisiae. With all three protocols filtered cells were competent sooner and showed transformation efficiencies comparable to or higher than those of cells prepared by centrifugation. The procedure can be easily adapted to collect cells for many other purposes, and should be especially useful in metabolic studies and in preparing cells for electroporation, as it facilitates both thorough washing and rapid but gentle resuspension of cells without compromising sterility.

The Haemophilus influenzae sxy-1 mutation is in a newly identified gene essential for competence.

The sxy-1 mutation of Haemophilus influenzae causes a 100- to 1,000-fold increase in spontaneous natural competence. We have used mapping and sequencing to identify this mutation as a G-to-A transition in an open reading frame adjacent to the rec-1 locus. This mutation substitutes valine for isoleucine at amino acid 19 of the protein specified by this gene (now named sxy). A multicopy plasmid containing the wild-type sxy gene confers constitutive competence on wild-type cells. Cells carrying this plasmid exhibit, in all stages of growth, DNA uptake levels and transformation frequencies as high those normally seen only after full induction of competence by starvation; deletion of part of the sxy gene from the plasmid abolishes this effect. In contrast, a transposon insertion in sxy entirely prevents both DNA uptake and transformation, indicating that sxy encodes a function essential for competence. These findings suggest that sxy may act as a positive regulator of competence. However, because cells carrying the transposon-inactivated sxy::Tn allele grow slowly under conditions that do not induce competence, sxy may also have a role in noncompetent cells.

Male mutation rates and the cost of sex for females.

Although we do not know why sex evolved, the twofold cost of meiosis for females provides a standard against which postulated benefits of sex can be evaluated. The most reliable benefit is sex's ability to reduce the impact of deleterious mutations. But deleterious mutations may themselves generate a large and previously overlooked female-specific cost of sex. DNA sequence comparisons have confirmed Haldane's suggestion that most mutations arise in the male germ line; recent estimates of a, the ratio of male to female mutation rates, are ten, six and two in humans, primates and rodents, respectively. Consequently, male gametes may give progeny more mutations than the associated sexual recombination eliminates. Here I describe computer stimulations showing that the cost of male mutations can easily exceed the benefits of recombination, causing females to produce fitter progeny by parthenogenesis than by mating. The persistence of sexual reproduction by females thus becomes even more problematic.

The Haemophilus influenzae adenylate cyclase gene: cloning, sequence, and essential role in competence.

Competence for transformation in Haemophilus influenzae is stimulated by cyclic AMP (cAMP) and requires the cAMP-dependent catabolite regulatory protein CRP. Thus, understanding the control of competence will require understanding how cAMP levels are regulated. As a first step, we have cloned the H. influenzae adenylate cyclase gene (cya) by complementing the Lac- phenotype of delta cya Escherichia coli. Its sequence specifies an 843-amino-acid protein which has significant identity to other known bacterial adenylate cyclases (41 to 43% and 61% identical to the cya genes of enteric bacteria and of Pasteurella multocida, respectively). As seen in other bacterial cya genes, there is evidence for regulation similar to that demonstrated for E. coli: the presence of a strong consensus CRP binding site within the promoter of the gene may provide feedback control of cAMP levels by repressing cya transcription, and translation may be limited by the weak ribosome binding site and by initiation of protein synthesis with GUG rather than AUG or the UUG used in other bacterial cya genes. We confirmed the essential role of cAMP in competence by constructing and characterizing H. influenzae cya mutants. This strain failed to develop competence either spontaneously or after transfer to a competence-inducing medium. However, it became as competent as its wild-type parent in the presence of exogenous cAMP. This result suggests that the failure of exogenously added cAMP to induce optimum competence in wild-type cells is not due to a limitation to the entry of cAMP into the cells. Rather, it strongly favors models in which competence induction requires both an increase in intracellular cAMP and a second as yet unidentified regulatory event. H. influenzae strains mutant in cya or crp were unable to ferment xylose or ribose. This confirms that influenzae, like E. coli, uses cAMP and CRP to regulate nutrient uptake and utilization and lends increasing support to the hypothesis that DNA uptake is mechanism of nutrient acquisition.

Genes for breakfast: the have-your-cake-and-eat-it-too of bacterial transformation.

Bacterial transformation, in which cells take up and recombine free strands of DNA, is the simplest process thought to have evolved for genetic exchange, i.e., because of the potential benefits of producing progeny with recombinant genotypes. However, two other functions are equally plausible: acquisition of intact DNA strands to use for recombinational repair of DNA damage, and acquisition of the nutrients contained in DNA molecules. Although the recombinant progeny produced by transformation can be beneficial, the success of genes causing transformation is limited by other factors, especially by the genetic quality of DNA derived from dead cells. Our recent experiments in the naturally transformable bacteria Haemophilus influenzae and Bacillus subtilis suggest that the DNA-repair hypothesis is unlikely to be correct. In H. influenzae, transformation does not detectably increase the cells' ability to survive DNA damage. More importantly, we have found that, although competence (the ability to take up DNA) is induced by nutritional limitation in both H. influenzae and B. subtilis, it is not induced by DNA damage in either. Thus we favor the hypothesis that transformation evolved as a nutrient-uptake system, especially because unrelated DNA is abundant in the environments of many naturally transformable bacteria.