The evolution of bacterial diseases

Obscured behind current media focus on the Human Genome Project, there is considerable activity in sequencing bacterial genomes. There are presently more than 30 published eubacterial genome sequences (data from TIGR). Comparison of the sequences is leading to new insights into how bacteria have acquired new genes which allow them to take advantage of developing ecological niches.

Bacteria are responsible for causing a number of serious human diseases. Many of these diseases depend on high human population density for their success. However, humans are not naturally herd animals and therefore it seems likely that the pathogens have "taken advantage" of the development of civilisation as an opportunity to adopt a new host. What is the evidence for this and is evolution of new diseases still occurring?

Over the past decade comparisons of gene and more recently genome sequences of pathogenic bacteria and their relatives have given an insight into ways in which human diseases have developed. For example tuberculosis is now believed to have first developed about 15.000 years bp (before present) when domestication of cattle brought humans into close contact with these hosts. The ancestral TB organism probably originated from a soil bacterium. This conclusion is based on the difference in gene DNA sequences from different groups of Mycobacterium tuberculosis and  M. bovis (the bovine TB bacterium) (9).

DNA sequences can provide evidence for evolutionary routes in several ways. Chromosomal DNA accumulates mutations at a predictable rate as a result of errors in replication and environmental mutagens, which are not corrected by DNA repair systems. Comparing sections of DNA in related species or strains allows deductions to be made on the length of time since they diverged from a common ancestor. Choice of sequence to apply this "molecular clock" is important. Mutations which change amino acid sequence at important positions in key proteins are likely to be lethal and hence are selected against; comparison of such sequences will probably give underestimates of evolutionary time. In Escherichia coli a synonymous site (where no amino acid change is caused by a base-changing mutation) is estimated to mutate once in approximately 10¹⁰ years.

Sequence comparison has also revealed that bacteria carry sections of DNA which seem to have come from other species in the evolutionarily recent past. This suggests that there is a significant level of "gene sharing" between bacterial species; this is called horizontal gene transfer. A crucial example is that pathogens carry blocks of genes which code for proteins involved in pathogenicity within their chromosomes. Sequence analysis suggests that in some cases these "pathogenicity islands" have been transferred from other species. Thus it is possible for a harmless bacterium to acquire the genes necessary to make it a pathogen in one step, by conjugational transfer of a plasmid or transfer by bacteriophage.

Bubonic plague is an interesting example of a bacterial disease which has probably recently evolved. It is caused by the bacterium Yersinia pestis, and has led to the death of large sections of human populations since the 6th Century AD. Y. pestis, Y. pseudotuberculosis, (which normally causes gut infections with occasional septicemia) and Y. enterocolitica (also an enteric food/water borne pathogen) were compared (1) by sequencing fragments of six genes and analysing sequence differences. The results indicated that there was little genetic difference between Y. pestis (Y. pst) and Y, pseudotuberculosis (Y. ptb), although only Y. pst is a serious pathogen. The differences between Y. pst and Y. enterocolitica suggested that their most recent common ancestor existed 41-186Myrs ago.

The similarity of tested chromosomal sequence between Y. pst and Y. ptb makes accurate dating uncertain, however the authors conclude that divergence occurred recently (1500 - 20 000 years ago). Although this is is a wide range, the most recent date ties up with the first recorded pandemic of plague (Justinian's plague) between 541-767 AD; suggesting that Y. pst appeared just prior to this. So what is the difference between these genetically almost identical bacterial species that leads to their vastly different pathogenicities?

Although Y. ptb is a gastrointestinal pathogen, it can cause fatal septicemia in stressed animals, and can be transmitted to fleas. However, injection of the bacterium into the blood of a new host is unlikely because of lack of retention in the flea gut, and probable low dissemination in the host after injection during biting. Acquisition of two plasmids removed these barriers.

The pPla (= pPCP) plasmid of Y. pst encodes a plasminogen activator which is important (but not essential for) subcutaneous dissemination allowing better movement of the bacterium through the circulatory system of the infected individual. The pFra (= pMT1) plasmid carries the gene for a murine toxin which aids colonisation of the flea midgut (4). This could have been transferred from another bacterial species during residence of Y. ptb in flea gut. The chromosomal gene hms which is shared with Y. ptb and codes for a protein which blocks the flea proventriculus, would increase retention of Y. pst infected blood in the crop and so passage of infected blood to bitten individuals. Subsequent mutations in the bacterial chromosome may have removed the (now unecessary) ability of the bacterium to colonise the mammalian host's gut. These changes in capability would start the progress to development of the two current species; genomes would be exposed to different selection pressures and would retain different mutations. If these evolutionary changes occurred 1,500 years ago they would have led to the first recorded plague.

The second pandemic occurring in the middle ages in Europe was caused by a different strain of Y. pestis and most of the present day disease is caused by a third strain. A recent paper (8) reported PCR and sequence analysis of Y. pst DNA fragments from teeth from a 14th century mass grave in France, adding to evidence that bubonic plague was prevalent during this period. The three strains appear to have originated in different parts of the world as determined by differences in genome composition.

Studies such as this, which attempt to piece together the genetic events which led to the development of diseases are relevant to current concerns about the appearance of new and unsuspected epidemics against which we have little protection. An example is Escherichia coli strain O157. This differs from more familia,r less harmful, strains in causing severe diarrhea and possibly death. The complete genomes of two non-pathogenic strains of E. coli have been sequenced (5). Now the O157 sequence is being analysed and compared with the other E. coli sequences. So far the comparison has revealed the gene for shiga toxin from the diarrhea-causing bacterium Shigella (on a bacteriophage) (7), virulence genes carried on plasmids and a pathogenicity island. There is also evidence for a higher than normal mutation rate due to defective repair of mismatches in sequence between DNA strands. Such defects also make it more likely that foreign DNA will be incorporated into the chromosome. This potential for accelerated genomic change could increase the rate of acquisition of pathogenic characteristics.

The emergence of this strain as a health issue may be an example of exploitation of the bacterium of a new niche; the mass preparation of food. Comparison of partial and complete genome sequences can identify the genes which determine pathogenicity, and the origin and means of acquisition of the genes. This knowledge can then be used in developing effective defences against old and new diseases.

1. Achtman M et al (1999) Yersinia pestis, the cause of plague, is a recently emerged clone of Yersinia pseudotuberculosis. Proc. Natl. Acad Sci. USA 96, 14043.
2. Finlay BB Falkow S (1997) Common Themes in Microbial Pathogenicity Revisited. Micro. Mol. Biol. Rev. 61, 136
3. Groisman EA Ochman H (1996) Pathogenicity islands: bacterial evolution by leaps and bounds. Cell 87, 791
4. Hu P et al (1998) Structural organisation of virulence-associated plasmids of Yersinia pestis. J. Bacteriol. 180, 5192.
5. Perna NT et al (2001) Genome sequence of enterohaemorrhagic Escherichia coli O157:H7. Nature 409, 529
6. Perry RD Fetherston JD (1997) Yersinia pestis - etiologic agent of plague. Clin. Microbiol. Rev. 10, 35.
7. Plunkett G et al (1999) Sequence of shiga toxin 2 phage 933W from Escherichia coli O157:H7: shiga toxin as a late-gene product. J. Bacteriol 181, 1767
8. Raoult D et al (2000) Molecular identification by "suicide PCR" of Yersinia pestis as the agent of medieval Black Death. Proc. Natl. Acad. Sci. USA 97, 12800.
9. Sreevatsan S et al (1997) Restricted structural gene polymorphism in the Mycobacterium tuberculosis complex indicates evolutionarily recent global dissemination. Proc. Natl. Acad. Sci. USA 94, 9869

see also: BioBeat online magazine feature on ref. 7
Boyd EF et al (2001) Bacteriophage-bacteriophage interactions in the evolution of pathogenic bacteria. Trends in Microbiology 9, 137.

Web Links

TIGR genome resource

Proceedings of the National Academy of Sciences USA

Nature genome gateway

The Bad Bug Book

Virtual Library of Microbiology and Virology

Site Links

Home

Bioupdates-to-go

Classroom

Courses

Useful links

glossary