Bliss
- 1. Do bacteria have sex? Rosie Redfield University of British Columbia
- 2. Do bacteria have any processes that evolved to make new combinations of chromosomal genes? Why is this an important question? Why is natural competence the only process we need to investigate? Why do bacteria take up DNA? How does Haemophilus influenzae decide when to take up DNA? How does H. influenzae decide which DNAs to take up? What kinds of evidence are most useful?
- 3. Cycles of meiosis and gamete fusion that randomly recombine alleles from two parents into offspring. Sexual reproduction (in eukaryotes): Do bacteria have any processes that evolved to make new combinations of chromosomal genes? Why is this an important question?
- 4. Why does sexual reproduction exist? Genes that cause sexual reproduction must end up in more fit genomes than they would have without sex.
- 5. How should we decide? (What kind of evidence is best?) Do bacteria have any processes that evolved to make new combinations of chromosomal genes? What are the candidates? Parasexual processes of bacteria:
- 6. How should we decide? (What kind of evidence will best show how natural selection has acted?) Have ‘parasexual’ processes evolved for genetic exchange? Make mathematical models or computer simulations? Run long-term evolution experiments (like Lenski)? Use the ‘comparative method’?
- 7. How should we decide? Have ‘parasexual’ processes evolved for genetic exchange? What has real-world selection produced? What genes are responsible for the process? What genes does the process benefit? Do these genes affect other processes? What goes wrong when a gene is defective? Is the process regulated? If so, how? First, characterize these processes in more detail, paying attention to evolutionary issues.
- 8. How should we decide? Have ‘parasexual’ processes evolved for genetic exchange? Make mathematical models and computer simulations. How can selection act (under very simplified conditions.)? How does selection act (under very unnatural conditions)? Run long-term evolution experiments. Use the ‘comparative method’. If selection is still unclear: What has selection produced (under other conditions)?
- 9. Case study #1: RecA Discovery: Needed for homologous recombination in bacteria. Phenotype of mutant: Recombination down >1000x. Mechanism: More mutant phenotypes: Viability down ~10%. Death from DNA damage up >100x. RecA senses DNA damage and induces the ‘SOS response’. RecA carries out ‘recombination repair’ of otherwise-lethal DNA damage.
- 10. Roles of RecA in DNA repair: RecA binds to damaged DNA strands This induces other DNA repair proteins RecA pairs damaged strands with complementary intact strands. The intact strand provides a template for repair of the damage. The same steps cause recombination with incoming DNA:
- 11. Recombination: Do ‘parasexual’ processes exist for genetic exchange? All of these genes have important 'housekeeping' functions in DNA replication and repair.
- 12. Transduction and conjugation are caused by genetic parasites. Do ‘parasexual’ processes exist for genetic exchange? Conjugation: Usually plasmid transfer Rarely host DNA transfer Transduction: Recipient Phage-infected source
- 13. Do ‘parasexual’ processes exist for genetic exchange? Why do bacteria take up DNA? Templates for DNA repair? Nucleotides and other nutrients (food)? Sources of environmental DNA Recipient cell x Could be for New combinations of genes (sex)?
- 14. DNA is a good source of nutrients: Nucleotide synthesis is very expensive. DNA is widely available in the environment: biofilms host mucosa dead eukaryotes 10-100 µg/ml in Serving size = 1 genome (3 pg) Fragments per serving = 12 Calories Carbon (as ribose) Phosphate Nitrogen Fats Amino acids Amounts Per Serving 0.000005 1 fg 1 fg 1 fg 0 fg 0 fg Percent daily values are based on a 3 mb genome. Not a significant source of calories from fat, saturated fat, cholesterol, dietary fiber, vitamin A, vitamin C, calcium, sulfur or iron.
- 15. Animation by Matt Baumann Competence in Haemophilus influenzae
- 16. Why do bacteria take up DNA? How does H. influenzae decide when to take up DNA? How does H. influenzae decide which DNAs to take up? If DNA serves mainly as food, we predict regulation by nutritional signals
- 17. Why do bacteria take up DNA? How does H. influenzae decide when to take up DNA? Competence genes are induced when both sugars and nucleotides are scarce.
- 18. CRP-S sequence
- 19. 1. The sugar signal: First, CRP activates energy-supply genes (and sxy ): Then, CRP and Sxy activate DNA uptake genes: CRP is activated (by cAMP) when preferred sugars are depleted. If…
- 20. Purine nucleotides and nucleosides prevent competence by inhibiting Sxy translation purines pyrim-idines purine bases
- 21. 10 -5 10 -6 10 -7 purR + purR - Constitutive production of purines prevents competence induction The purR gene encodes the purine repressor. In purR + cells the enzymes that synthesize purines are only made when the cell runs out of purines. In purR - cells the proteins are always made, so the cell never runs out of purines. Transformation frequency
- 22. 10 -4 10 -5 10 -6 10 -7 sxy + Hypercompetent sxy mutants Cells with hypercompetence mutations in sxy don’t care about purine levels. purR + purR -
- 23. Combining sxy-1 and sxy 3 restores base pairing and eliminates hypercompetence A A U U U U C U C A G U C A U C U U U A G A A G U U A G U A G G U C A A A U G U U U C . . . . . . . . 5’-G A A G U A C U 1 4 5 3 2 RBS A U G The 5 hypercompetence mutations all disrupt sxy mRNA base pairing sxy+ C : G sxy-3 T x G C x A sxy-1 sxy-1+3 T : A
- 24. Mutations in sxy stem 1 do alter mRNA secondary structure 7 1
- 25. Why do bacteria take up DNA? How does H. influenzae decide when to take up DNA? How does H. influenzae decide which DNAs to take up? Regulation is by nutritional signals.
- 26. H. influenzae cells (and their relatives) preferentially take up DNA containing a short sequence called the uptake signal sequence (USS). Their genomes have these sequences at thousands of sites. USS locations in the genome
- 27. Molecular drive explanation: Sequences preferred by the uptake machinery accumulate in the genome by recombination, replacing allelic versions that are not as easily taken up. Computer simulation modeling confirms that this drive can explain the observed properties of uptake sequences. Sex explanation: Species with uptake sequences in their genomes and a corresponding uptake bias benefit more from recombination. Why are uptake sequences so abundant in the genome?
- 28. Molecular drive explanation: Sequences preferred by the uptake machinery accumulate in the genome by recombination, replacing allelic versions that are not as easily taken up. But why are the uptake sequences preferred by the uptake machinery? Hypothesis: Uptake sequences are easier to kink Sex explanation: Species with uptake sequences in their genomes and a corresponding uptake bias benefit more from recombination. DNA must kink to enter the pore. outer membrane inner membrane
- 29. USS Randomized 200bp 300bp core s-2 s-3 Uptake sequences are bent. Core Seg 2 Seg 3 Core Seg 2 Seg 3 Neighbors are correlated at bend positions Ethylation of some positions enhances uptake
- 30. New research direction: What role do uptake sequences play in the mechanism of DNA uptake? Want to see it? Our lab web pages include our grant proposals.
- 31. If uptake sequences didn’t evolve for sex, bacteria don’t have ANY processes that evolved to generate recombination. Bacteria apparently get whatever recombination they might ‘need’ by accident. So why do eukaryotes need so much more?
- 32. With funding from CIHR & NIH
- 34. Uptake sequences are not species-specific. Redfield et al. 2006 80 96 100 80 A. actinomycetemcomitans P. multocida H. somnus M. succiniciproducens H. influenzae H. ducreyi M. haemolytica A. pleuropneumoniae 100 100 1465 1760 927 1216 1485 742 199 973 Consensus ‘cores’ Hin-type USS Apl -type USS AAGTGCGGT ACAAGCGGT
- 35. 136 260 459 581 648 640 384 180 436 583 628 445 174 43 Percent identities to homologs Mean uptake sequences per gene Uptake sequences are rare in strongly conserved genes. Findlay and Redfield 2009 H. influenzae N. meningitidis
- 36. Laterally transferred genes have few uptake sequences: Mean number of uptake sequences per gene N. meningitidis H. influenzae Number of standard genomes with homologs Number of standard genomes with homologs Mean uptake sequences per gene Uptake sequences don’t promote LGT; they accumulate after LGT. Findlay and Redfield 2009
- 37. Uptake sequences are less variable than other sequences in the genome. BLAST comparison of homologous USS in H. influenzae strains
- 38. Molecular drive explanation: Sequences preferred by the uptake machinery accumulate in the genome by recombination, replacing allelic versions that are not as easily taken up. Both predict that the USS motif found in the genome should reflect the uptake bias, but… Sex explanation: Species with uptake sequences in their genomes and a corresponding uptake bias benefit more from recombination.
- 39. Core Seg 2 Seg 3 Genomic USS motif DNA uptake of fragments with mutant USS Why doesn’t the uptake bias more closely match the sequences in the genome? 3 0 1 2 n n n n A A A G T G C G G T n A A A T T T n n n n n n A T T T T T n n n n n % variation Genetic variation at USS positions 0 10 20 30 A A G T G C G G T n AA n TT % uptake C C C G C A C C G . . . . . . . . . TT . G . . . T GG
- 40. Tripeptide frequency (per 1000 amino acids) Peptides encodable by Hin uptake sequences are enriched only in the Hin proteome Biased DNA uptake modifies tripeptide frequencies: Findlay and Redfield 2009 H. influenzae proteome ( Hin ) A. pleuropneumoniaeae proteome ( Apl ) N. meningitidis proteome ( Nme) E. coli proteome (control) Reversed peptides show no enrichment Peptides encodable by Nme uptake sequences are enriched only in the Nme proteome Peptides encodable by Apl uptake sequences are enriched only in the Apl proteome
- 41. Competence genes are regulated similarly in many bacteria Gene present
- 42. Artificially inducing E. coli Sxy turns on many E. coli genes:
- 43. Does E. coli have natural competence? Same competence genes: Same induction by Sxy and CRP acting at CRP-S sites: So why won’t E. coli become competent? Perhaps because we don’t know how to turn Sxy on...
- 45. CAGTTAGTA -10 -35 A 0 50 100 150 200 250 300 10 -2 10 -3 10 -4 10 -5 10 -6 10 -7 10 -8 10 -9 Transformation frequency Time in rich medium (min) No colonies Wildtype cells sxy-1 mutant WT sxy-1 sxy-2 sxy-3 sxy-4 sxy-5 Transformation frequency 10 -3 10 -4 10 -5 10 -6 10 -7 WT sxy-1 sxy-2 sxy-3 sxy-4 sxy-5 log-phase late-log 5 hypercompetent sxy mutants CTACTGACT TCA 5 4 3 Gln A 2 1 Val Ile
- 46. So do Neisseria species. Bacteria in the Pasteurellaceae preferentially take up ‘self’ DNA.
- 47. 734 copies in + orientation 731 copies in - orientation Smith et al . 1995 Science 269:538-540 The H. influenzae genome contains 1471 copies of the uptake sequence AAGTGCGGT.
- 48. Uptake sequences are motifs, not mobile elements. AAGTGCGGT SAV SAL CAV SAI H. influenzae protein Homologs in species without uptake sequences Pattern found with the Gibbs motif sampler program
Editor's Notes
- #4: Why is sexual reproduction so common in eukaryotes?
- #5: We don’t understand why their genomes are so much more fit…. Species are a consequence of sexual reproduction…
- #6: Bacteria don't have sexual reproduction. What about Archaea? Sex = production of recombinant genotypes By this much broader definition they certainly do have 'sex'.
- #7: Talk at SFU: My analyses and experiments show that the genes that cause the parasexual processes are very strongly selected for other (‘housekeeping’) functions, and provide no evidence of selection for the ability to cause genetic exchange. In general, the evolutionary biologists I’ve presented this to have such a strong commitment to the idea that randomizing alleles is beneficial that they can’t accept this conclusion. Instead they think I should keep searching for evidence of selection for the genetic exchange consequences. Given the strength of selection for the housekeeping functions, I think this would be a waste of time.
- #8: What some philosophers of evolution call ‘functional design’ analysis.
- #12: Genes discovered and named and originally studied because of effect on recombination. But recombination is not why they exist. No evidence that actions have been modified by selection for ability to produce recombination.
- #13: Parasexual processes were discovered by microbiologists looking for genetic recombination So they naturally interpreted these processes as the equivalent of eukaryotic sexual reproduction. The assumption that their function is genetic exchange has influenced research into the function of sexual reproduction. Time for a reevaluation. READ TEXT ON NEXT SLIDE
- #14: Many bacteria can take up DNA. If from close relative may recombine. If changes genotype, transformation. Before details about DNA uptake, why recombination?
- #15: Many ATPs to make an ATP Benefits are potentially much higher than genetic benefits, because much more frequent and much more reliable. 100 µg/ml DNA in our body too. (back of envelope calculation). Is this DNA also a good source of genes (for sex or repair)? Probably not - most of it comes from strangers & losers.
- #16: Model organism for studies of competence. Chosen because we have good evidence that DNA uptake is frequent in natural populations.
- #20: Explain the figs on the left first: DNA (promoter), RNA polymerase, CRP bends DNA CRP turns genes on when it receives cAMP, a sugar-is-running-out signal.
- #31: We put our grant proposals on line as soon as they're submitted.
- #32: Explanations of meiotic sex must solve problems that bacteria don’t have. Why do eukaryotes need so much more recombination than bacteria?
- #41: Remember that only about half of these peptides are encoded by perfect uptake sequences.
- #45: Evidence from genome sequences. E. coli and other enteric bacteria are closest relatives, have many of same genes, so comp may be even older.
- #48: This is a BIG evolutionary issue. If we assume that bacteria take up DNA as a form of sex, then the USS work like mate-choice signals. But why else would they be there? First some info about them.
- #49: Only 24% of sequences had gaps anywhere, and most of these were not in USS cores. Each USS in the genome evolves independently.