The Neolithic revolution of bacterial genomes - UWC

Update

200

TRENDS in Microbiology

17 Holmstrom, A. et al. (2001) LcrV is a channel size-determining component of the Yop effector translocon of Yersinia. Mol. Microbiol. 39, 620–632 18 Sarker, M.R. et al. (1998) The Yersinia Yop virulon: LcrV is required for extrusion of the translocators YopB and YopD. J. Bacteriol. 180, 1207–1214 19 Lee, V.T. et al. (2000) LcrV, a substrate for Yersinia enterocolitica type III secretion, is required for toxin targeting into the cytosol of HeLa cells. J. Biol. Chem. 275, 36869–36875 20 Derewenda, U. et al. (2004) The structure of Yersinia pestis V-antigen, an essential virulence factor and mediator of immunity against plague. Structure 12, 301–306 21 Mota, L.J. et al. (2005) Bacterial injectisomes: needle length does matter. Science 307, 1278 22 Torruellas, J. et al. (2005) The Yersinia pestis type III secretion needle plays a role in the regulation of Yop secretion. Mol. Microbiol. 57, 1719–1733

Vol.14 No.5 May 2006

23 Journet, L. et al. (2003) The needle length of bacterial injectisomes is determined by a molecular ruler. Science 302, 1757–1760 24 Brubaker, R.R. (2003) Interleukin-10 and inhibition of innate immunity to Yersiniae: roles of Yops and LcrV (V antigen). Infect. Immun. 71, 3673–3681 25 Nilles, M.L. et al. (1997) Yersinia pestis LcrV forms a stable complex with LcrG and may have a secretion-related regulatory role in the lowCa2C response. J. Bacteriol. 179, 1307–1316 26 Sing, A. et al. (2005) A hypervariable N-terminal region of Yersinia LcrV determines Toll-like receptor 2-mediated IL-10 induction and mouse virulence. Proc. Natl. Acad. Sci. U. S. A. 102, 16049–16054

0966-842X/$ - see front matter Q 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.tim.2006.02.010

Genome Analysis

The Neolithic revolution of bacterial genomes Alex Mira, Ravindra Pushker and Francisco Rodrı´guez-Valera Evolutionary Genomics Group, Division of Microbiology, Universidad Miguel Hernandez, San Juan 03550, Alicante, Spain

Current human activities undoubtedly impact natural ecosystems. However, the influence of Homo sapiens on living organisms must have also occurred in the past. Certain genomic characteristics of prokaryotes can be used to study the impact of ancient human activities on microorganisms. By analyzing DNA sequence similarity features of transposable elements, dramatic genomic changes have been identified in bacteria that are associated with large and stable human communities, agriculture and animal domestication: three features unequivocally linked to the Neolithic revolution. It is hypothesized that bacteria specialized in human-associated niches underwent an intense transformation after the social and demographic changes that took place with the first Neolithic settlements. These genomic changes are absent in related species that are not specialized in humans.

The Neolithic revolution Before the Neolithic period, human survival was linked to the hunter-gatherer culture and populations were small and scattered [1]. Approximately 10 000 years ago, however, the advent of agriculture and animal husbandry brought the largest social revolution in the history of humankind [2]. Food resources were more abundant and constant, and the human species increased its population size at an extraordinary annual growth rate of 0.1% [3]. From the point of view of bacterial pathogens, humans suddenly became attractive hosts; they concentrated large populations on limited areas, which maximized the chance for transmission between longer-lived carriers. Thus, it is Corresponding author: Mira, A. ([email protected]). Available online 29 March 2006 www.sciencedirect.com

likely that human population growth and expansion during the Neolithic created a selective pressure that favoured pathogens that specialized in human hosts, originating what was probably the first wave of emerging human diseases [4]. One of the features that has emerged from fully sequenced genomes is that species that have specialized their niche undergo a process of reductive evolution through gene elimination [5–8]. Mobile elements such as insertion sequences (ISs) greatly increase in number following host restriction [9] and influence the process of genome reduction. After a niche change, many genes become obsolete and the selective pressure to preserve their function is dramatically reduced. In addition, host restriction can reduce the efficiency of selection in bacterial specialists because of reduced population sizes and genetic drift [10,11]. ISs might inactivate unnecessary or redundant genes in the new environment [12,13] by rapidly increasing the number of IS copies. The number of IS elements in bacterial pathogens could, therefore, indicate a recent host adaptation [14] or restriction [9]. Here, we show that recent expansions of ISs in bacteria are associated with humans or human activities that were developed in the Neolithic. The cultural revolution of the Neolithic can be dated precisely (for evolutionary time scales) at w10 000 years before present, which provides an outstanding model for prokaryotic evolution in the short-to-medium time scale. Recent expansions of mobile elements We investigated paralogous genes, including IS elements, across 255 fully sequenced prokaryotic genomes. Paralogues are defined as protein-coding sequences within a genome that have at least 30% sequence identity over

Update



201

(a)

100 90 80 70 60 50 40 30 0

0.5

1.0

1.5 2.0 2.5 3.0 Gene position (Mbp)

(b)

3.5

100 90 80 70 60 50 40 30 0

4.0

1

2 3 4 Gene position (Mbp)

Shigella flexneri 2a str. 301 100 90 80 70 60 50 40 30 0

0.5 1.0 1.5

2.0 2.5

3.0 3.5 4.0

4.5

Gene position (Mbp)

5

Escherichia coli K12 Sequence identity (%)

Sequence identity (%)

Bordetella bronchiseptica RB50 Sequence identity (%)


Bordetella pertussis Tohama I

100 90 80 70 60 50 40 30 0

0.5 1.0 1.5

2.0 2.5

3.0 3.5 4.0 4.5

Gene position (Mbp) TRENDS in Microbiology

Figure 1. IS expansions in human-related species. DNA sequence similarity between pairs of paralogous genes (blue crosses) and between pairs of IS elements (red crosses) is indicated across different bacterial species. Chromosomal location of the first gene for each pair is shown on the x-axis. (a) Bordetella pertussis Tohama I and Bordetella bronchiseptica RB50; (b) Shigella flexneri 2a and Escherichia coli K12. Bacteria specialized in humans are shown on the left, whereas related species that are not human specialists are shown on the right. Species associated with humans (B. pertussis and S. flexneri) show a recent explosion of IS elements across their genomes.

O60% of their lengths [15–17]. To estimate DNA sequence similarity, global alignments were performed on the gene pairs using Stretcher [18]. When a frequency histogram of paralogue sequence similarity was plotted, we observed that most paralogues across taxa display 40–60% DNA sequence similarity, followed by a second group of duplicated genes at 90–100% similarity. This second group presumably contains recently formed paralogues and a few genes with a high protein-dosage requirement, the DNA sequence of which is kept identical by gene conversion mechanisms [19]. However, a closer look reveals that most of the members in this group are ISs or IS-transposase genes. Interestingly, 69 out of 89 bacteria in which O75% of the recently formed paralogous genes correspond to ISs are species that are uniquely associated with humans (Figure 1). Most of the other cases correspond to host-associated species such as invertebrate symbionts or pathogens. These include the mutualist Photorhabdus luminescens, which is specifically associated with nematodes, and the reproductive parasite of insects Wolbachia pipientis (Figure 2a). This indicates that any niche specialization normally involves IS expansions [9]. A few cases of free-living species with recent IS expansions are also observed, such as the sediment-isolated species Shewanella oneidensis or Geobacter metallireducens (Figure 2b). Examples of pathogens specialized in humans (Figure 1) include Bordetella pertussis, a species that has recently www.sciencedirect.com

undergone an extraordinary paralogue expansion only as a result of ISs (Figure 1a); however, Bordetella bronchiseptica, a species that can infect different hosts and live freely in the environment, has a normal distribution of DNA similarity across paralogues. The IS expansion not only inactivates genes by insertion but also increases the rate of genome rearrangements [11]. The human pathogen Shigella flexneri, which is closely related to the saprophytic Escherichia coli K12 strain, has recently undergone another IS expansion (Figure 1b). We hypothesize that mobile element expansion took place in Neolithic times when ancestors of these two species independently restricted their host range to humans. Three lines of evidence support the hypothesis that the expansion occurred at that moment. First, IS expansions are also observed in bacteria associated with domestication processes typical of the Neolithic transition. Second, the substitution rate between IS elements is virtually zero, which is consistent with a single, recent event. And third, divergence estimates between IS-loaded human specialists and IS-free generalists are consistent with the Neolithic hypothesis. Pathogens of domesticated plants and livestock Pathogens of agricultural plants and domesticated animals also seem to have undergone recent episodes of IS expansion. Thus, the genomes burdened with ISs are those related to the three main changes that are

Update

202



(a)

Photorhabdus luminescens laumondii TTO1

Wolbachia pipientis wMel 100 Sequence identity (%)


100 90 80 70 60 50 40 30

90 80 70 60 50 40 30

0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0

1

2

Geobacter metallireducens GS-15

5

6

Shewanella oneidensis MR-1 Sequence identity (%)

100 Sequence identity (%)

4

Gene position (Mbp)

Gene position (Mbp) (b)

3

90 80 70 60 50 40 30

100 90 80 70 60 50 40 30

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

Gene position (Mbp)

0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0


Figure 2. IS expansions in species that are presumably unrelated to humans. DNA sequence similarity is indicated between pairs of paralogous genes (blue crosses) and between pairs of IS elements (red crosses). (a) Wolbachia pipientis (which causes cytoplasmic incompatibility in the fruit fly Drosophila melanogaster) and Photorhabdus luminescens (a mutualistic symbiont with a restricted host range to the nematode Heterorhabditis bacteriophora) are animal-associated. (b) Geobacter metallireducens and Shewanella oneidensis are metal reducers from the terrestrial subsurface.

unequivocally linked to the Neolithic transition: agriculture, animal domestication and the formation of larger human communities. For example, several sequenced species of Pseudomonas are versatile microorganisms that live in soil or water. However, Pseudomonas syringae DC3000 (a pathovar that is specialized for tomato plants [20]) has undergone a massive episode of IS expansion, with most mobile elements sharing O98% sequence similarity (Figure 3a). Because crops are abundant and concentrated on certain regions, they are likely to represent an attractive niche for plant pathogens. Such pathogens include different species of Xanthomonas such as Xanthomonas axonopodis pv. citri, which is specialized in citrus trees and contains an extraordinary number of IS elements [21]. The rice pathogen Xanthomonas oryzae is even more extreme and contains O800 ISs [22], and although the sequences of related species that are not plant specialists are unavailable for comparison, the reason behind this genomic change is likely to be related to the specialization of Xanthomonas in rice plants after the expansion of this staple food across Asia. The genomes of bacteria specialized in farm animals also seem to be affected. The species Burkholderia mallei is an obligate parasite of horses, donkeys and mules, with no other known natural reservoir [23]. Thus, it seems www.sciencedirect.com

likely that it specialized in these animals after the wild ancestors of horses were domesticated, which increased their population size and lifespan. The expansion of IS elements is also dramatic in this species (Figure 3b), whereas its sister species – the soil inhabitant Burkholderia pseudomallei – contains a limited number of IS elements [24]. Because the domestication of horses seems to have occurred relatively recently in human history (w4500 years ago [25]), the speed of these genomic changes mediated by ISs must have been extraordinary. Furthermore, bacteria associated with food production also seem to have undergone transposase amplification at a genomic scale. The domesticated lactic-acid bacterium Lactococcus lactis [26] is the most commonly used cheese starter in the world. This species has been used for generations to inoculate cheese, which has effectively maintained a domestic culture of the bacterium that is associated with human populations. When compared with sequenced related species, L. lactis shows a distinctive recent expansion of IS elements (Figure 3c). We predict that other bacterial species that are associated with traditional human food (such as those involved in yoghurt or wine production and the preservation of foods in brine or salt) might have undergone similar mobile-element expansions (e.g. Ref. [27]).

Update



203

(a)

Pseudomonas syringae tomato

Pseudomonas aeruginosa PA01 100 Sequence identity (%)


100 90 80 70 60 50 40 30

90 80 70 60 50 40 30

0

1

2

3

4

5

6

0

1

2

Gene position (Mbp)

3

4

5

6

Gene position (Mbp)

(b)

Burkholderia mallei ATCC 23344

Burkholderia pseudomallei K96243 100 Sequence identity (%)


100 90 80 70 60 50 40 30

90 80 70 60 50 40 30

0

0.5

1.0

1.5 2.5 2.0 Gene position (Mbp)

3.0

0

3.5

0.5

1.0

1.5 2.0 2.5 3.0 Gene position (Mbp)

3.5

4.0

(c)

Lactococcus lactis subsp. lactis

Lactobacillus acidophilus NCFM 100 Sequence identity (%)


100 90 80 70 60 50 40 30

90 80 70 60 50 40 30

0

0.5

1.0

1.5

2.0

Gene position (Mbp)

0

0.5

1.0

1.5

2.0


Figure 3. IS expansions in bacteria that are specialized in agricultural plants, in farm animals and in food processing. DNA sequence similarity is indicated between pairs of paralogous genes (blue crosses) and between pairs of IS elements (red crosses) across different bacterial species. (a) Pseudomonas syringae pv. tomato and Pseudomonas aeruginosa PA01; (b) Burkholderia mallei ATCC 23344 and Burkholderia pseudomallei K96243; (c) Lactococcus lactis subsp. lactis and Lactobacillus acidophilus NCFM. Specialized strains are shown on the left and non-specialized related strains are shown on the right.

Predicting niche specialization As more information accumulates about species that have experienced genomic changes linked to ISs, it can be used www.sciencedirect.com

in the reverse direction: recent lifestyle changes in bacteria could be predicted and ISs might even provide an approximate time scale for these genomic events. For

Update

204


Bradyrhizobium japonicum USDA 110

(a) 100

Mesorhizobium loti MAFF303099 Sequence identity (%)



90 80 70 60 50 40 30

90 80 70 60 50 40 30

0

1

2

(b)

3 4 5 6 Gene position (Mbp)

7

8

9

0

1

2

Halobacterium salinarum NRC-1 100

90

90

80 70 60 50 40 30

3 4 5 Gene position (Mbp)

6

7

Haloarcula marismortui ATCC 43049

100 Sequence identity (%)


100

80 70 60 50 40 30

0

0.5

1.0

1.5

2.0

Gene position (Mbp)

0

0.5

1.0

2.0 1.5 Gene position (Mbp)

2.5

3.0


Figure 4. IS expansions in species under potential human influence. DNA sequence similarity is indicated between pairs of paralogous genes (blue crosses) and between pairs of IS elements (red crosses) across selected bacterial species. Chromosomal location of the first gene for each pair is shown on the x-axis. (a) Bradyrhizobium japonicum USDA 110 and Mesorhizobium loti MAFF303099; (b) Halobacterium salinarum NRC-1 and Haloarcula marismortui ATCC 43049. Species on the left show recent IS element expansions that are not found in closely related species (right), which is hypothesized to be the result of niche specialization.

example, the genome of the legume symbiont Bradyrhizobium japonicum displays a large expansion of nearly identical transposases whereas another nodule-forming symbiont, Mesorhizobium loti, contains older transposases (Figure 4a). If our hypothesis is correct, the reason for this could be related to the formation of B. japonicum nodules in the widely cultivated soya bean. By contrast, M. loti has not been domesticated, is associated with several wild plants and has a wider host range [28]. The genome of the halophilic archaeon Halobacterium salinarum shows a recent IS expansion when compared with related species (Figure 4b). Although it could be claimed that this species is not related to humans, a closer look at its niche reveals that H. salinarum was isolated from salted fish. The extraction of salt is one of the oldest human activities for preserving food and was used by the ancient Chinese as long ago as 9000 BC [29]. As environments characterized by high salt concentrations expanded, a new human-associated niche extended across the globe and prokaryotes could have adapted to it. Similarly, genomic features such as IS expansions in other genomes could serve as hints of ecological changes*. The opposite scenario is also interesting. There might be cases of bacteria that are clearly specialized in humans but have no traces of IS expansion. Mycobacterium leprae (the causative agent of leprosy) could be one of the few * Plots for all studied species are available from the authors upon request. www.sciencedirect.com

examples. It contains no ISs even though they could have expanded across the two-thirds of the genome that are no longer necessary in its specialized niche [6]. Instead, the genome of M. leprae has undergone an incomplete reductive evolution process and has retained 1000 pseudogenes with a near-normal length. Perhaps the intracellular lifestyle has precluded it from incorporating ISs and the extraordinary number of pseudogenes is because of the absence of transposases, which has slowed down the pace of genome reduction. IS elements, when present, quickly inactivate genes [13] and regulatory elements [30], and serve as repetitive sequences that induce large deletions [31]. Thus, the expansion of ISs might function as a mechanism for facilitating ecological adaptation. Dating the IS expansion Dating divergence in bacterial lineages is a difficult task as a result of (among other things) the absence of a fossil record, the difficulty of calibrating a molecular clock and the meagre knowledge about generation times in nature [32,33]. The ribosomal 16S rRNA gene is estimated to diverge at a rate of w1–2% per 50 million years [32–35], using geological events that could be dated between 50 million and 500 million years. Although this gene seems to be too conserved to date relatively recent events, it is interesting to note that the observed divergences are similar among the different pairs studied here:

Update



205

Table 1. Sequence divergence in IS-interrupted ORFs and in fully sequenced genomes of putative Neolithic speciesa (a) Divergence between IS-interrupted ORFs and functional homologues in related strains Species with IS-interrupted ORFs Yersinia pestis CO92 Yersinia pestis CO92 Shigella flexneri 2a Burkholderia mallei Bordetella pertussis Shigella flexneri 301 Shigella flexneri 301 Pseudomonas syringae tomato

Fc

Average similarityd

He

Average similarityf

Average genomic dSg

Divergence estimateh

Refs

26

23

99.992

3442

99.872

0.00008G0.00002

6000 years

[37,38]

Yersinia pseudotuberculosis Escherichia coli K12

26

24

99.756

3413

99.154

0.00217G0.0002

11 000 years

[37,38]

77

56

97.399

3195

98.403

0.0436G0.00076

N/A

N/A

Burkholderia pseudomallei Bordetella parapertussis Shigella flexneri 2457T Shigella sonnei

43

41

99.584

2535

99.214

0.01647G0.00036

N/A

N/A

122

60

98.914

2880

98.713

0.0451G0.00124

N/A

N/A

77

58

99.693

3449

99.859

0.00047G0.00006

N/A

N/A

77

64

98.185

3034

98.361

0.03693G0.0008

N/A

N/A

43

21

86.549

3988

93.81

0.6279G0.0104

N/A

N/A

Species with functional homologue Yersinia pestis KIM

Pseudomonas syringae phaseolicola

Nb

(b) Synonymous divergence between fully sequenced genomes of putative Neolithic species and related strains

a

Abbreviation: N/A, not applicable. Number of genes interrupted by ISs. c Number of IS-interrupted ORFs with a functional homologue in a related strain. d Average DNA similarity between the IS-interrupted ORFs and functional homologues. e Number of homologous genes identified by reciprocal best-hit method. f Average DNA similarity of all homologues between the two genomes. g dS, synonymous substitution rate per synonymous site, estimated by maximum likelihood [42]. h Estimates in Y. pestis are included for reference. b

divergences in the 16S rRNA genes were 0.15% between Bu. mallei and Bu. pseudomallei, and 0.2% for the E. coli K12–S. flexneri pair and the Bo. pertussis–Bo. parapertussis pair. A larger divergence of 0.9% was observed for the P. syringae pathovars P. syringae syringae and P. syringae tomato. A 0.24% divergence was observed for the two sequenced strains of S. flexneri. A more direct estimate of IS expansions comes from studying the genes that have been inactivated by IS insertion. Under the assumption that these truncated genes lost functionality at the time of the insertion, the date of the IS expansion can be estimated by investigating the divergence of truncated open reading frames (ORFs). Being pseudogenes, all mutations can be considered to be selectively neutral [36]. We identified ORFs that are putative pseudogenes as a result of the insertion of ISs [13] and selected those for which a functional homologue could be found in related strains or species (Table 1a). The estimated divergence was then compared with that of the IS-interrupted genes in Yersinia pestis, one of the few species in which the age of diversification has been firmly calculated [37,38]. To extrapolate these estimations to other bacteria is necessarily a crude approximation because it assumes similar molecular substitution rates. Nevertheless, the levels of nucleotide divergence for the IS-generated pseudogenes are reasonably close to the Neolithic time frame, except for the pseudomonads. A similar result is obtained when divergence rates are investigated at synonymous sites in homologous genes in the species studied (Table 1b). Using complete sequenced genomes, we have calculated synonymous substitution rates (dS) for whole genome pairs. Again using the estimated divergence time for Y. pestis as a reference, the genomes of Shigella, Bordetella and Burkholderia www.sciencedirect.com

species seem to have expanded within a range of thousands of years in the past. Burkholderia showed more recent divergence estimates than the rest, which is consistent with the later domestication of horses. Outside the prokaryotes, it is interesting to note that the malaria parasite and human specialist Plasmodium falciparum also shows low levels of variation and is estimated to have originated as a clone w6000 years ago [39]. Although approximate, these estimates place the explosion of IS elements within the historical range rather than within longer, geological timescales. The speed reported here at which ISs can expand is supported by laboratory experiments in which IS elements were detected to replicate and insert across the genome as early as after 500 generations [40]. Specializing in a given environment such as the human host or human-related plants or animals probably rendered many genes unnecessary and widened the possible sites for non-lethal insertions. Therefore, bacteria could have undergone the Neolithic revolution as much as humans did, and the genomes of those bacteria still carry the tell-tale prints in the form of extended IS element families. Conclusions and future perspectives The work presented here shows that expansions of transposable elements in bacterial genomes can be reliable indicators of specialization to human-related niches. We anticipate that parallel genomic changes could be taking place in other human-related organisms such as domesticated fungal species involved in traditional food production or eukaryotic unicellular parasites. The remarkable speed at which bacteria have adapted to human cultural evolution underscores the evolutionary plasticity of the prokaryotic cell, which can

206

Update


probably be matched only by viral evolution [41]. The central role of IS elements described here and the use of their sequence divergence to identify recent niche specialization events could prove a useful approach to detect other (not necessarily human-related) recent environmental specializations. Furthermore, we propose that analysis of the type of genes inactivated in the IS insertion process might give hints as to the nature of the environmental change. Although the pseudogenes generated by IS elements are quickly eroded by deletional bias [8], their function could give useful information in recent specialization processes. In addition, older events would leave a trace of divergent ISs. The relatively new science of comparative genomics might provide a powerful tool to understand better the ecology and evolution of prokaryotes. This will be better elucidated when more genome sequences are available from non-human pathogens. Acknowledgements A.M. is a recipient of a ‘Ramo´n y Cajal’ research contract from the Spanish Ministry of Science and Education. Support from I.S. Carlos III (04/1319) and the EU project GEMINI (QLK3-CT-2002-02056) is also acknowledged. We thank S.B. Ingham for help with the figures and G. Nyiro and F. Kunst for providing information about host specificity in Wolbachia and Photorhabdus species.

References 1 Childe, V.G. (1925) The Dawn of European Civilization, 5th edn, Routledge & Kegan Paul 2 Bocquet-Appel, J.P. (2002) Paleoanthropological traces of Neolithic demographic transition. Curr. Anthropol. 43, 637–650 3 Eshed, V. et al. (2004) Has the transition to agriculture reshaped the demographic structure of prehistoric populations? New evidence from the Levant. Am. J. Phys. Anthropol. 124, 315–329 4 McKeown, T. (1988) The Origins of Human Disease, 1st edn, Blackwell 5 Andersson, J.O. and Andersson, S.G. (1999) Insights into the evolutionary process of genome degradation. Curr. Opin. Genet. Dev. 9, 664–671 6 Cole, S.T. et al. (2001) Massive gene decay in the leprosy bacillus. Nature 409, 1007–1011 7 Moran, N.A. and Mira, A. (2001) The process of genome shrinkage in the obligate symbiont Buchnera aphidicola. Genome Biol. 2, research0054 8 Mira, A. et al. (2001) Deletional bias and the evolution of bacterial genomes. Trends Genet. 17, 589–596 9 Moran, N.A. and Plague, G.R. (2004) Genomic changes following host restriction in bacteria. Curr. Opin. Genet. Dev. 14, 627–633 10 Moran, N.A. (1996) Accelerated evolution and Muller’s rachet in endosymbiotic bacteria. Proc. Natl. Acad. Sci. U. S. A. 93, 2873–2878 11 Parkhill, J. et al. (2003) Comparative analysis of the genome sequences of Bordetella pertussis, Bordetella parapertussis and Bordetella bronchiseptica. Nat. Genet. 35, 32–40 12 Wei, J. et al. (2003) Complete genome sequence and comparative genomics of Shigella flexneri serotype 2a strain 2457T. Infect. Immun. 71, 2775–2786 13 Lerat, E. and Ochman, H. (2004) J-F: exploring the outer limits of bacterial pseudogenes. Genome Res. 14, 2273–2278 14 Parkhill, J. et al. (2001) Genome sequence of Yersinia pestis, the causative agent of plague. Nature 413, 523–527 15 Altschul, S.F. et al. (1997) Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Res. 25, 3389–3402 16 Pushker, R. et al. (2004) Comparative genomics of gene-family size in closely related bacteria. Genome Biol. 5, R27 17 Brown, S.M. (2000) Bioinformatics: A Biologist’s Guide to Biocomputing and the Internet (1st edn), pp. 57–84, Eaton Publishing

www.sciencedirect.com


18 Rice, P. et al. (2000) EMBOSS: the European Molecular Biology Open Software Suite. Trends Genet. 16, 276–277 19 Abdulkarim, F. and Hughes, D. (1996) Homologous recombination between the tuf genes of Salmonella typhimurium. J. Mol. Biol. 260, 506–522 20 Buell, C.R. et al. (2003) The complete genome sequence of the Arabidopsis and tomato pathogen Pseudomonas syringae pv. tomato DC3000. Proc. Natl. Acad. Sci. U. S. A. 100, 10181–10186 21 da Silva, A.C. et al. (2002) Comparison of the genomes of two Xanthomonas pathogens with differing host specificities. Nature 417, 459–463 22 Lee, B.M. et al. (2005) The genome sequence of Xanthomonas oryzae pathovar oryzae KACC10331, the bacterial blight pathogen of rice. Nucleic Acids Res. 33, 577–586 23 Alibasoglu, M. et al. (1986) Glanders outbreak in lions in the Istanbul zoological garden. Berl. Munch. Tierarztl. Wochenschr. 99, 57–63 24 Nierman, W.C. et al. (2004) Structural flexibility in the Burkholderia mallei genome. Proc. Natl. Acad. Sci. U. S. A. 101, 14246–14251 25 Jansen, T. et al. (2002) Mitochondrial DNA and the origins of the domestic horse. Proc. Natl. Acad. Sci. U. S. A. 99, 10905–10910 26 Bolotin, A. et al. (2001) The complete genome sequence of the lactic acid bacterium Lactococcus lactis ssp. lactis IL1403. Genome Res. 11, 731–753 27 Prust, C. et al. (2005) Complete genome sequence of the acetic acid bacterium Gluconobacter oxydans. Nat. Biotechnol. 23, 195–200 28 Jordan, D.C. (1984) Family Rhizobiaceae. In Bergey’s Manual of Systematic Bacteriology (Vol. 1, 1st edn) (Krieg, N.R. and Holt, J.G., eds), pp. 234–244, Williams & Wilkins 29 Kurlansky, M. (2002) Salt: A World History, 1st edn, Vintage Canada 30 Mira, A. and Pushker, R. (2005) The silencing of pseudogenes. Mol. Biol. Evol. 22, 2135–2138 31 Schneider, D. et al. (2000) Long-term experimental evolution in Escherichia coli. IX. Characterization of insertion sequence-mediated mutations and rearrangements. Genetics 156, 477–488 32 Ochman, H. and Wilson, A.C. (1987) Evolution in bacteria: evidence for a universal substitution rate in cellular genomes. J. Mol. Evol. 26, 74–86 33 Ochman, H. et al. (1999) Calibrating bacterial evolution. Proc. Natl. Acad. Sci. U. S. A. 96, 12638–12643 34 Clark, M.A. et al. (1999) Sequence evolution in bacterial endosymbionts having extreme base compositions. Mol. Biol. Evol. 16, 1586–1598 35 Wernegreen, J.J. and Moran, N.A. (1999) Evidence for genetic drift in endosymbionts (Buchnera): analyses of protein-coding genes. Mol. Biol. Evol. 16, 83–97 36 Gojobori, T. et al. (1982) Patterns of nucleotide substitution in pseudogenes and functional genes. J. Mol. Evol. 18, 360–369 37 Achtman, M. et al. (1999) Yersinia pestis, the cause of plague, is a recently emerged clone of Yersinia pseudotuberculosis. Proc. Natl. Acad. Sci. U. S. A. 96, 14043–14048 38 Achtman, M. et al. (2004) Microevolution and history of the plague bacillus, Yersinia pestis. Proc. Natl. Acad. Sci. U. S. A. 101, 17837–17842 39 Rich, S.M. et al. (1998) Malaria’s Eve: evidence of a recent population bottleneck throughout the world populations of Plasmodium falciparum. Proc. Natl. Acad. Sci. U. S. A. 95, 4425–4430 40 Schneider, D. and Lenski, R.E. (2004) Dynamics of insertion sequence elements during experimental evolution of bacteria. Res. Microbiol. 155, 319–327 41 Zanotto, P.M. et al. (1996) Population dynamics of flaviviruses revealed by molecular phylogenies. Proc. Natl. Acad. Sci. U. S. A. 93, 548–553 42 Yang, Z. (1997) PAML: a program package for phylogenetic analysis by maximum likelihood. Comput. Appl. Biosci. 13, 555–556

0966-842X/$ - see front matter Q 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.tim.2006.03.001