referred to above. The linear chromosome appeared to carry 35% of the genome's genes, including those encoding ribosomal and DNA-replication proteins, as well as 21 complete metabolic pathways. The presence of these genes confirmed the chromosome-like nature of the linear element; however, other genes on the same chromosome, especially those involved in conjugation, are reminiscent of a plasmid. Most interestingly, the sequence revealed that there was a repABC operon (a gene known to be associated with replication of circular chromosomes) near the centre of the chromosome, plus an inversion in the GC skew (see Section 3.1.3). This seems to indicate that the linear chromosome has a bidirectional, plasmid-like mode of replication. Why A. tumefa-ciens maintains such a complicated arrangement of its genome and the advantages of having multiple chromosomes remain a mystery. Multiple chromosomal elements are especially prominent among proteobacterial phyla and spirochaetes, and this in itself would suggest phylogenetic determination of the tendency to form extrachromosomal elements.

3.2.2 Lateral gene transfer

For many years it has been known that microorganisms can absorb DNA directly from the environment. The relative ease by which antibiotic resistance can be donated from one bacterium to another constitutes further proof that genetic information is not only transferred during cell division (vertical transmission), but also from one intact cell to another (lateral or horizontal transmission). Lateral gene transfer can occur via three mechanisms (Zhou and Thompson 2004), as follows.

Transformation. This is the uptake of DNA directly from the environment. If prokaryotic cells can do this they are called competent. Very few bacteria are competent during their whole life cycle, but some are during certain physiological stages. Transduction. Bacteriophages can transfer DNA between species if two host species share similar bacteriophage receptors. Transduction may concern random pieces of the host DNA, packaged during phage assembly in the lytic cycle, or it may be limited to the sequences flanking the insertion site.

Conjugation. Lateral gene transfer can occur via specialized plasmids during physical contact between F+ and F_ cells, as discussed above.

Lateral gene transfer is not limited to bacteria of the same species, it can also occur among species of widely different origin and even between Bacteria and Archaea. The latter phenomenon was discovered by Nelson et al. (1999), when they sequenced the genome of the thermophilic bacterium Thermotoga maritima (Fig. 3.12). Thermotoga derives its name from the sheath-like envelope that surrounds the cell (the 'toga') and the fact that it has a temperature optimum for growth of 80 °C. It is usually placed in a completely separate lineage of the Bacteria, called Thermotogales, because of some unique characteristics, including rRNA sequences that are unusual among the Bacteria and a set of fatty acids that is only found in this group. Phylogenetic analysis, aimed at enlightening the position of T. maritima within the Bacteria, resulted in a great lack of congruence when different genes were used as a basis for the comparison. Further analysis of the ORFs in the T. maritima genome showed that no less than 24% of the predicted genes were most similar to proteins in archaeal species, rather than to Bacteria. The Archaea-like genes were found to lie in clusters (islands) in the genome of T. maritima; in several of these islands even the archaeal gene order was conserved. These observations and those by Aravind et al. (1998) on another thermophilic bacterium, Aquifex aeolicus, provided great support to the theory that lateral gene transfer was not to be considered an oddity, but a very significant process for many microorganisms.

Koonin et al. (2001) performed a quantitative analysis of the frequency of lateral gene transfer by analysing the genomes of eight archaeal and 22 bacterial species. They estimated that the percentage of new genes acquired from another domain (Bacteria, Archaea, or Eukarya) was on average 0.9% for the bacterial genomes and 3.4% for the archaeal genomes. When looking at inter-species transfers within the Bacteria, the percentages of acquisition of new genes varied considerably, from 0.4 to 19.8%, depending on the group. A particularly high frequency of foreign DNA is found in the genomes of the Spirochaetales. In general, it turned out that bacteria living at high temperatures had more archaeal genes in their genomes than mesophilic bacteria, and bacteria with a parasitic life style more often had eukaryotic genes in their genome than non-parasitic bacteria. It therefore seems that lateral gene transfer is especially common among organisms that live in close proximity to each other.

Several authors have pointed out that not all evidence for lateral gene transfer is equally reliable. If lateral gene transfer is inferred only from

Figure 3.12 Electron micrograph of Thermotoga maritima, a thermophilic bacterium belonging to the group Thermotogales, which was isolated originally from a geothermal heated marine sediment at Vulcano, Italy. Courtesy of K.O. Stetter, University of Regensburg.

the genome sequence, showing that certain ORFs have a BLAST match outside the group considered, this is not sufficient evidence, because alternative explanations may be given, such as the loss of genes in some lineages or widely diverging rates of evolution across the groups compared (Eisen 2000; Nesbo et al. 2001). To prove that lateral gene transfer has occurred, one needs to conduct a gene-by-gene phylogenetic analysis. As an example we discuss the work of Deppenheimer et al. (2002).

Deppenheimer et al. (2002) sequenced the genome of Methanosarcina mazei (Euryarchaeota), an archaeon of great ecological importance since it derives its energy from fermenting simple organic substrates to methane. Methane production in underwater sediments and inundated land (notably rice paddies) is an important link in the global carbon cycle. The genome of M. mazei (4.1 Mbp) was more than twice as large as other methano-genic Archaea that had been sequenced completely. Of the 3371 identified ORFs, no less than 1043 (31%) had their closest homologue not in an archaeal but in a bacterial species. Unlike the situation in Thermotoga, the bacterial genes in Methanosarcina did not cluster together in islands, but were found scattered in the genome. Gene phylogenies in which the laterally transferred genes were compared with orthologues in other Bacteria and Archaea showed that the foreign genes clustered with bacterial homologues, rather than with Archaea (Fig. 3.13).

Detailed analysis of the metabolic role of the bacterial genes in Methanosarca showed that the imported genes had considerably enlarged the metabolic spectrum of the archaeon. The suggestion from the data was that the metabolism of M. mazei has evolved from a simple methanogenic pathway (using hydrogen and carbon dioxide to produce methane), to a much more versatile substrate spectrum, allowing the use of acetate, methanol, and methylamine. Interestingly, most of the laterally transferred genes seemed to have been obtained from obligate and facultative anaerobic bacteria; that is, from organisms that live in the same microenvironment as the methanogenic archaeon (sediments, inundated land). This is in line with observations on thermophilic communities showing that the greatest transfer is between organisms living close together.

As indicated by the examples above, lateral gene transfer is an important mechanism for recruiting new microbial functions. By lateral gene transfer, microorganisms may be able to exploit new ecological niches that were inaccessible prior to the event. However, not all lateral gene transfers are necessarily to be viewed within a purely adaptive framework. The presence of foreign genes in a genome might well be a consequence, rather than a cause, of adaptation (Nesbo et al. 2001). Mira et al. (2001) viewed the bacterial genome as resulting from an evolutionary balance between ongoing recruitment and removal processes. Lateral gene transfer is the most important recruitment process. If there is no pressure from natural selection to maintain a newly recruited gene, it will quickly be removed or inactivated. Assuming that in all bacterial genomes there is a bias towards a higher rate of deletions over insertions, this explains the small and streamlined genomes of many bacteria. According to G. Gottschalk (personal communication) the large genomes of methanogens may just be a consequence of the fact that removal of laterally transferred genes has not yet occurred. This argument is similar to the neutral theory of genome evolution by Lynch and Conery (2003), discussed in Section 3.1.

Are all genes equally subjected to lateral gene transfer? Jain et al. (1999) postulated the complexity hypothesis, which states that genes that have few interactions with other genes will integrate more easily into a new genomic background and are therefore more likely to be successful in lateral gene transfer, compared with genes that are part of a complicated network and dependent on many other genes. This hypothesis explains the greater tendency of operational genes to be transferred, compared to informational genes; however, there are many exceptions to the hypothesis (Zhou and Thompson 2004).

Although the actual rate of lateral gene transfer is considered to be low compared to the life cycle of microorganisms, lateral gene transfer may leave a permanent trace in the genome; the presence of

Methanococcus jannaschii



_ influenzae

0 0

Post a comment