Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2009 Jun;5(6):e1000472.
doi: 10.1371/journal.ppat.1000472. Epub 2009 Jun 12.

Molecular Evolutionary Consequences of Niche Restriction in Francisella Tularensis, a Facultative Intracellular Pathogen

Free PMC article

Molecular Evolutionary Consequences of Niche Restriction in Francisella Tularensis, a Facultative Intracellular Pathogen

Pär Larsson et al. PLoS Pathog. .
Free PMC article


Francisella tularensis is a potent mammalian pathogen well adapted to intracellular habitats, whereas F. novicida and F. philomiragia are less virulent in mammals and appear to have less specialized lifecycles. We explored adaptations within the genus that may be linked to increased host association, as follows. First, we determined the genome sequence of F. tularensis subsp. mediasiatica, the only subspecies that had not been previously sequenced. This genome, and those of 12 other F. tularensis isolates, were then compared to the genomes of F. novicida (three isolates) and F. philomiragia (one isolate). Signs of homologous recombination were found in approximately 19.2% of F. novicida and F. philomiragia genes, but none among F. tularensis genomes. In addition, random insertions of insertion sequence elements appear to have provided raw materials for secondary adaptive mutations in F. tularensis, e.g. for duplication of the Francisella Pathogenicity Island and multiplication of a putative glycosyl transferase gene. Further, the five major genetic branches of F. tularensis seem to have converged along independent routes towards a common gene set via independent losses of gene functions. Our observations suggest that despite an average nucleotide identity of >97%, F. tularensis and F. novicida have evolved as two distinct population lineages, the former characterized by clonal structure with weak purifying selection, the latter by more frequent recombination and strong purifying selection. F. tularensis and F. novicida could be considered the same bacterial species, given their high similarity, but based on the evolutionary analyses described in this work we propose retaining separate species names.

Conflict of interest statement

The authors have declared that no competing interests exist.


Figure 1
Figure 1. Whole genome phylogeny among 17 Francisella strains based on 1,104,129 aligned nucleotide positions.
Panel (A) depicts relationships among major clades within the Francisella genus and panel (B) relationships within the species F. tularensis. The evolutionary tree was inferred using the Neighbor-Joining method. Bootstrap support values (500 replicates) are shown next to branches. Scale bars indicate the number of base substitutions per site.
Figure 2
Figure 2. Illustration of one of many recombination events detected in basal parts of the Francisella phylogeny.
(A) shows that an accumulation of phylogenetically informative SNPs is correlated with homoplastic SNPs. (B) is a magnification illustrating that the SNP patterns are incongruent with the overall whole genome SNP tree. (C) shows that the patterns indicate multiple recombination events in the region, as seen from several conflicting tree topologies.
Figure 3
Figure 3. Likelihood estimates of dN/dS ratios across the Francisella phylogeny obtained using the software package Hyphy.
The analysis indicates statistically significant differences in dN/dS values for different tree branches. Different colors represent the maximum number of rate classes estimated from the data using the Akaike information criterion (AIC) with a genetic algorithm. Values on the branches represent local optima, with confidence intervals in brackets. To optimize the evolutionary models comparisons were separately optimized for the phylogeny of (A) F. tularensis strains, and (B) the Francisella genus, including F. philomiragia and F. novicida.
Figure 4
Figure 4. Illustrative plot of pairwise dN/dS values versus sequence divergence in intergenic regions among 13 F. tularensis strains.
The high overall dN/dS values are indicative of inefficient purifying selection, and time-dependence at intergenic distances <0.15% is apparent.
Figure 5
Figure 5. The distribution of gene function losses projected for the Francisella phylogeny.
Total numbers of gene function losses are indicated on branches, and values in brackets indicate numbers of homoplastic gene function losses. Homoplastic loss means that these genes have been inactivated independently in different branches.
Figure 6
Figure 6. Gene order alterations in F. tularensis have mainly occurred with breakpoints at IS-elements.
Results of BLASTN analyses of sequences flanking each ISFtu1 and ISFtu2 element in eight completed Francisella genomes, showing “hits” mapped along a linear depiction of the F. novicida U112 chromosome, with 53 local collinear sequence blocks within the 1,910 kb sequence. A black line in the upper panel corresponds to a flanking sequence of an ISFtu2 and a black line in the lower panel to a flank of an ISFtu1 element. The positions of rRNA genes and the FPI in the F. novicida U112 sequence are indicated.
Figure 7
Figure 7. Phylogeny based on gene orders in six completed F. tularensis genomes (SCHU S4, WY-96, FSC147 and the identical orders of OSU18, LVS and FTA) and one F. novicida genome (U112).
Inversion distances are indicated on the branches. Reconstructions were generated both without a constraining topology (A), and with the “true” topology, as determined by SNP analysis (B).
Figure 8
Figure 8. A proposed model for duplication of the FPI.
(A) In an ancestral genome similar to F. novicida U112, the FPI was present as a single copy. (B) Insertion of copies of the ISFtu1 element adjacent to the present FPI, and adjacent to a second rRNA operon, enabled (C) duplication of the FPI by non-reciprocal recombination. (D) The two copies of the region have been inherited by all accepted F. tularensis subspecies.

Similar articles

See all similar articles

Cited by 65 articles

See all "Cited by" articles


    1. Dennis DT, Inglesby TV, Henderson DA, Bartlett JG, Ascher MS, et al. Tularemia as a biological weapon: medical and public health management. Jama. 2001;285:2763–2773. - PubMed
    1. Rotz LD, Khan AS, Lillibridge SR, Ostroff SM, Hughes JM. Public health assessment of potential biological terrorism agents. Emerg Infect Dis. 2002;8:225–230. - PMC - PubMed
    1. Ochman H, Davalos LM. The nature and dynamics of bacterial genomes. Science. 2006;311:1730–1733. - PubMed
    1. Sjöstedt AB. Family XVII. Francisellaceae, Genus I. Francisella. In: Brenner DJ, Krieg NR, Staley JT, Garrity GM, editors. New York: Springer; 2005. pp. 200–210.
    1. Jensen WI, Owen CR, Jellison WL. Yersinia philomiragia sp. n., a new member of the Pasteurella group of bacteria, naturally pathogenic for the muskrat (Ondatra zibethica). J Bacteriol. 1969;100:1237–1241. - PMC - PubMed

Publication types