Limited evolution of the yellow fever virus 17d in a mouse infection model

Figures & data

Figure 1. Mutation spectrum of YFV-17D and PLLAV following infection in AG129 mice. (A) Schematic presentation of study design. AG129 mice were inoculated i.p. with either a range of inocula of YFV-17D; 10⁻²–10⁻¹ PFU, or 10⁴ PFU of YFV-17D (n = 3/group) or with 20 µg of PLLAV (n = 7). Mice brains were harvested and homogenized for plaque purification prior to sequencing. (B) Mutation frequencies in brain-derived YFV-17D and PLLAV following neurotropic infection in AG129 mice.

Table 1. Nucleotide substitutions per genomic region of YFV-17D versus PLLAV

Gene	Mutation rate		Mutation rate
	# of Nucleotide changes*	YFV-17D	# of Nucleotide changes	PLLAV
5’ UTR	0	0	1	8.47E × 10^−3
Capsid protein	0	0	1	2.75E × 10^−3
prM/M	1	4.06 × 10^−3	1	4.06E × 10^−3
E	7	4.73 × 10^−3	1	6.8E × 10^−3
NS1	4	3.34 × 10^−3	5	4.0E × 10^−3
NS2A	3	6.0 × 10^−3	2	4.0E × 10^−3
NS2B	1	2.56 × 10^−3	0	0
NS3	8	4.81 × 10^−3	3	1.16 × 10^−3
NS4A	3	3.48 × 10^−3	2	2.32 × 10^−3
NS4B	3	8.93 × 10^−3	0	0
NS5	14	5.15 × 10^−3	4	1.47 × 10^−3
3’ UTR	3	5.89 × 10^−3	1	1.96 × 10^−3

*Sum total of all nucleotide changes observed in the virus population regardless of YFV-17D dose administered. Mutation rates were calculated by dividing the number of mutations observed in each genomic region by the number of nucleotides that constitute the genomic region. Pairwise comparison of the Shannon and Simpson entropy per genome region for each of the conditions (Stamaril® low and high, PLLAV) revealed no significant differences in entropy between regions (p-values >0.05, Mann-Whitney test).

Figure 2. Mutation patterns observed in YFV-17D infected mice. (A) Graphical presentation of genomes of 73 brain-derived, plaque-purified virus clones of YFV-17D following i.p. inoculation and neuroinvasion in AG129 mice (n = 6). (B) Graphical presentation of 86 similar virus clones following i.p. injection of 20 µg of PLLAV-YFV17D and neuroinvasion in AG129 mice (n = 7). Horizontal blue lines represent individual YFV-17D genomes with blue circles as 5’ and 3’ UTRs, respectively. Red, green and orange diamonds represent missense, silent and mixed mutations, respectively (mixed = parental + mutant nucleotide). For a full list of mutations see Supplementary Tables S4 and S5.

Figure 3. Diversity of virus variants in brains of mice inoculated with YFV-17D and PLLAV. (A) Hamming distances calculated for 10⁴ PFU [high inoculum] (n = 3, 29 clones), 10⁻¹ PFU + 10⁻² PFU [low inoculum] (n = 3, 44 clones) of YFV-17D, and PLLAV (n = 7, 86 clones). (B) Hamming distances computed for each virus inoculum. (C) The Shannon entropy and (D) Simpson’s index of diversity for each inoculum. Data for PLLAV are from 2 independent experiments (n = 3 each), bars presented median values. Student t-test for statistical comparison between groups.

Figure 4. Evolutionary analysis and diversity of YFV-17D before and after inoculation in AG129 mice. (A) Diversity in plaque-purified virus clones (n = 20) of Stamaril® (lot H5105) prior to injection in mice. A consensus sequence (Stamaril® P0) was generated from 20 plaque-purified virus clones. Evolutionary analysis in MEGA X [36] using Maximum Likelihood method and Kimura 2-parameter model [51]. Tree by Neighbor-Join and BioNJ algorithms to a matrix of pairwise distances (Maximum Composite Likelihood estimates), selecting the topology with superior log likelihood value (−15353.23). Branch lengths drawn to scale (scale bar: 0.0020 substitutions per site). The analysis involved n = 21 nucleotide sequences, including GenBank X03700 as reference genome, with a total of 10862 positions in the final dataset. (B) Graphical visualization of genetic diversity and segregation observed in YFV-17D populations following neurotropic infection in AG129 mice.

Figure 5. Phylogenetic tree of Stamaril® and PLLAV-derived YFV-17D clones from AG129 mice. Coloured dots represent the different inocula. Green: PLLAV 20 µg; blue dots: 10⁴ PFU Stamaril; yellow: 10⁻¹ PFU Stamaril; red: 10⁻² PFU Stamaril: purple: consensus sequence (P0) of Stamaril®. Evolutionary history inferred using the Neighbor-Joining method [52]. The optimal tree (sum of branch length = 0.00785397) drawn to scale. Evolutionary distances (in number of base substitutions/site) computed using the Kimura 2-parameter method [17], involving n = 162 nucleotide sequences (n = 10862 positions in the final dataset after pairwise deletion of ambiguous positions).

Kaptein SJ, De Burghgraeve T, Froeyen M, et al. A derivate of the antibiotic doxorubicin is a selective inhibitor of dengue and yellow fever virus replication in vitro. Antimicrob Agents Chemother. 2010;54(12):5269–5280.

 PubMed Web of Science ®Google Scholar

Drake JW, Charlesworth B, Charlesworth D, et al. Rates of spontaneous mutation. Genetics. 1998;148(4):1667–1686.

 PubMed Web of Science ®Google Scholar

Salmona M, Gazaignes S, Mercier-Delarue S, et al. Molecular characterization of the 17D-204 yellow fever vaccine. Vaccine. 2015;33(41):5432–5436.

 PubMed Web of Science ®Google Scholar

Stock NK, Boschetti N, Herzog C, et al. The phylogeny of yellow fever virus 17D vaccines. Vaccine. 2012;30(6):989–994.

 PubMed Web of Science ®Google Scholar