Crimean-Congo hemorrhagic fever virus genomics and global diversity

Abstract

Crimean-Congo hemorrhagic fever (CCHF) is a severe illness with high case fatality that occurs in Africa, Europe, the Middle East, and Asia. The complete genomes of 13 geographically and temporally diverse virus strains were determined, and CCHF viruses were found to be highly variable with 20 and 8%, 31 and 27%, and 22 and 10% nucleotide and deduced amino acid differences detected among virus S (nucleocapsid), M (glycoprotein), and L (polymerase) genome segments, respectively. Distinct geographic lineages exist, but with multiple exceptions indicative of long-distance virus movement. Discrepancies among the virus S, M, and L phylogenetic tree topologies document multiple RNA segment reassortment events. An analysis of individual segment datasets suggests genetic recombination also occurs. For an arthropod-borne virus, the genomic plasticity of CCHF virus is surprisingly high.

FIG.1.

Phylogenetic relationships of CCHF virus full-length S segments. Phylogenetic analysis was carried out using PAUP version 4.0b10 (Sinauer Associates, Inc., Sunderland, MA). Using maximum-likelihood criteria, the optimal evolutionary model employed utilized a transition/transversion (Ti/Tv) ratio of 5.80, a proportion of invariable sites of 0.29, and a γ distribution value of 0.5. Bootstrap values were calculated from analysis of 500 pseudoreplicates of the data set, and values greater than 50% are indicated at the appropriate nodes. Each virus sequence is designated by the name of the strain and its country of origin. Brackets indicate the virus genetic groupings. In addition, the different virus groups are color coded for clarity. (A) Phylogenetic tree generated from 15 S segment nucleotide sequences from viruses with complete genomes. The asterisk indicates the two strains that were not sequenced during this study within the 15 sets. (B) Phylogenetic tree constructed from all 32 complete S segment nucleotides available.

FIG. 1—

Continued.

FIG.2.

Phylogenetic relationships of CCHF virus full-length M segments. Using maximum-likelihood criteria the optimal evolutionary model employed utilized a transition/transversion (Ti/Tv) ratio of 6.05, a proportion of invariable sites of 0.13, and a γ distribution value of 0.5. Bootstrap values were calculated from analysis of 500 pseudoreplicates of the data set, and values greater than 50% are indicated at the appropriate nodes. (A) Phylogenetic tree based on the M segments from the 15 CCHF virus strains with available complete genome sequences. (B) Phylogenetic tree constructed from all 32 complete M segments available.

FIG. 2—

Continued.

FIG.3.

Phylogenetic relationships of CCHF virus full-length L segments. Using maximum-likelihood criteria the optimal evolutionary model employed utilized a transition/transversion (Ti/Tv) ratio of 8.94, a proportion of invariable sites of 0. 37, and a γ distribution value of 0.5. Bootstrap values were calculated from analysis of 500 pseudoreplicates of the data set, and values greater than 50% are indicated at the appropriate nodes. (A) Phylogenetic tree constructed based on L nucleotide sequences from the same 15 CCHF virus strains analyzed in Fig. 1 and 2. (B) Phylogenetic tree constructed from all 18 complete L segments available.

FIG. 3—

Continued.