Chlamydial evolution

Update

[Under construction]

[This section based on an article by Stephens 2002 presented at the International Chlamydia Conference in Turkey. This article placed within the context of chlamydial evolution some of the original thinking of Ochman and others (see references). This article should be read in conjunction with the detailed review of chlamydial evolution by Moulder on this site].

The evolution of free living bacteria is different from that of chlamydiae in the following respects [Stephens, 2002]:

Free living bacteria	Chlamydiales
Marked genetic exchange leading to massive sampling of the bacterial gene pool	Relative genetic isolation
Recombination events leading to horizontal gene transfer	Point mutation
Substantial gene loss and gain	Gene loss or duplication
Significant phenotypic change	Long periods of time without genotypic change

Among free living bacteria, horizontal gene transfer between E. coli and Salmonella lead to the exchange of over 3 million nucleotides and was a major force in diversification. Chlamydiae by contrast have substantial numbers of eukaryotic plant-related genes which must have been derived from colonisation of unicellular protists [see: plant genes in Chlamydia]. Since the divergence of eukaryotic organisms occurred approximately 2 billion years ago, this indicates that protochlamydia must have interacted since then with eukaryotic protists. However, there is little evidence of horizontal gene transfer in chlamydiae apart from intraserovar recombination events in the major outer membrane protein [see: MOMP recombination]. The rarity of horizontal gene transfer in chlamydiae is suggested by the >92% identity in the C. pneumoniae / C. trachomatis genome sequences and by biases in codon usage and in AT / GC ratios. This relative genetic isolation of chlamydiae is probably the result of the organism being sequestered in an intracellular inclusion and by the relative metabolic inertness of the chlamydial elementary body. It means that, since chlamydiae became intracellular pathogens approximately 800 million years ago, chlamydial evolution has been incremental rather than, as in free living bacteria, characterised by recombination events leading to rapid species divergence.

Whereas Moulder in his classic early paper considered chlamydial evolution to have taken place in a largely hostile and thus selective intracellular environment [Moulder, 1974], Stephens 2002 considered this same environment to be protective and thus characterised by only weak selection. Weak selective pressures in the absence of exogenous gene acquisition mean that a higher proportion of mutations would be only mildly deleterious and thus would become fixed in the genome. Bacterial endosymbionts typically have a 1.7 to 2.7 times faster rate of accumulation of fixed mutations compared with free living bacteria [Moran 1996; see: Muller's ratchet]. Accumulation of these mildly deleterious mutations might lead to reduced protein function which might be compensated by an increase in the amount of chaperone proteins, such as GroEL, [see dictionary GroEL ] to salvage damaged protein. This might account for the comparative infrequency of mutations in the GroEL (hsp60) gene.

Plasticity zones, regions of genomic deterioration in the chlamydial genome, are evidenced by frame shifts in the toxin B gene [Belland et al., 2001] and in the HKD gene family and by the trpA mutations and deletions affecting tryptophan pathways [Fehlner-Gardiner et al., 2002]. Chlamydial TrpA sequences unlike those of other prokaryotes display numerous mutations. All ocular serovars of C. trachomatis contain a deletion mutation resulting in a truncated TrpA protein, which lacks alpha reaction activity. In contrast the TrpA protein from the genital serovars retains conserved amino acids required for catalysis but has mutated several active site residues involved in substrate binding. These mutations result in a TrpA protein that is unable to utilize indole glycerol 3-phosphate as substrate. However, the chlamydial TrpB protein can carry out the beta reaction, involving the formation of tryptophan from indole and serine. but displaying an absolute requirement for full-length TrpA. Thus genital, but not ocular, serovars are capable of utilizing exogenous indole for the biosynthesis of tryptophan indicating that mutations in the TrpA gene may have played a role in speciation [Fehlner-Gardiner et al., 2002].

Studies by Stephens and colleagues reported at the conference indicate that in the chlamydiae there is a relatively high ratio of non synonymous to synonymous nucleotide changes in chlamydial point mutations compared with either E. coli or Buchnera (see below). A non synonymous nucleotide substitution is one which introduces a new amino acid. The ratios of non synonymous to synonymous nucleotide changes observed were:

E. coli:

Buchnera

Chlamydia

20: 1

4: 1

1.5: 1

Thus in Chlamydia there are many mutations occurring in coding regions.

The evolutionary clock

For free-living bacteria, 16srRNA mutations serve as a kind of phylogenetic clock, permitting an estimation, from sequence data, of when major evolutionary changes might have occurred. The divergence rate in 16S rRNA sequences for free living bacteria is thought to be approximately 1% per 40 - 50 million years [Ochman et al., 1999]. Whether this clock ticks at a different speed in chlamydiae and intracellular endosymbionts compared with free living organisms is a moot point. It seems likely that the actual mutation rate will be similar to free living bacteria but that there will be a higher apparent rate of sequence divergence because more mutations are retained. For a discussion of the evolution of symbiosis see Ochman & Moran, 2001.

Stephens suggested that 2 billion years ago the chlamydial ancestor was probably a free living bacterium. The eukaryotes diverged some 800 million to 1 billion years ago, when the chlamydial ancestor may have interacted with protists and gained its eukaryotic genes. He suggests that divergence of C. trachomatis from C. pneumoniae probably occurred about 100 million years ago and speculated that LGV may have diverged when new primate hosts evolved following the dinosaur extinction. Divergence of genital and ocular serovars of C. trachomatis might have occurred with the appearance of new humanoid primate hosts. Normally a 5 - 6% difference in 16SrRNA sequence is considered sufficient grounds for genus differentiation in a free living bacterium (and indeed is the basis of the differentiation of the Chlamydiaceae into the genera Chlamydia and Chlamydophila) However the normal assumptions made for free living bacteria are unlikely to be valid for intracellular bacteria like chlamydiae, where there is little exogenous gene acquisition but a very high rate of point mutation. [see: new taxonomy].

Previous comparison of a relatively small set of homologous genes from Escherichia coli and Salmonella typhimurium revealed that genes nearer to the origin of replication had substitution rates lower than genes closer to the replication terminus. The recently completed sequences of numerous bacterial genomes have allowed us to test whether this effect of distance from the replication origin on substitution rates, as observed for the E. coli-S. typhimurium comparison, is a general feature of bacterial genomes. Extending the analysis to all 3,000 E. coli-S. typhimurium homologs confirmed the significant association between chromosomal position and synonymous site divergence. However, the effect, though still significant, is not as dramatic as originally thought. A similar association between relative chromosomal location and synonymous substitution rate was detected in the majority of other bacterial species comparisons within alpha- and gamma- Proteobacteria, and Firmicutes but was absent in Chlamydiales. The opposite trend, i.e., a decrease in synonymous divergence with distance from the replication origin, was detected in Mycobacteria. Analysis of the patterns of nucleotide substitutions revealed that the distance effect is not affected by gene orientation and is mainly caused by an increase in rates of transversions, suggesting that this effect may not be caused by recombinational repair or biased gene conversion, as originally suggested [Mira & Ochman, 2002].

[MEW, February 2003]

References

Belland, R. J., Scidmore, M. A., Crane, D. D., Hogan, D. M., Whitmire, W., McClarty, G. & Caldwell, H. D. (2001). Chlamydia trachomatis cytotoxicity associated with complete and partial cytotoxicity genes. Proceedings of the National Academy of Sciences of the U. S. A. 98, 13984 - 13989. Full article

Fehlner-Gardiner, C., Roshick, C., Carlson, J. H., Hughes, S., Belland, R. J., Caldwell, H. D. & McClarty, G. (2002). Molecular basis defining human Chlamydia trachomatis tissue tropism. A possible role for tryptophan synthase. Journal of Biological Chemistry 277, 26893 - 26903. Full article

Mira, A. & Ochman, H. (2002). Gene location and bacterial sequence divergence. Molecular Biology & Evolution 19, 1350 - 1358.

Moulder, J. W., (1974). Intracellular parasitism: life in an extreme environment. J Infect. Dis. 130, 300 - 306.

Ochman, H., Elwyn, S. & Moran, N. A. (1999). Calibrating bacterial evolution. Proceedings of the National Academy of Sciences U S A. 96, 12638 - 12643. Full article

Stephens, R. S. (2002). Chlamydiae in evolution: a billion years and counting. In: Chlamydial Infections. Proceedings of the 10th International Symposium on Human Chlamydial Infections. pp 3 - 12. Published by International Chlamydia Symposium, San Francisco ISBN 0-9664383-1-0

NEXT: Chlamydia - plant relationships & eukaryote organelle evolution