Metagenomic sequencing, molecular characterization, and Bayesian phylogenetics of imported type 2 vaccine-derived poliovirus, Spain, 2021

• Published in: Frontiers in cellular and infection microbiology Vol. 13; p. 1168355
• Main Authors: Fernandez-Garcia, Maria Dolores, Faye, Martin, Diez-Fuertes, Francisco, Moreno-Docón, Antonio, Chirlaque-López, Maria Dolores, Faye, Ousmane, Cabrerizo, Maria
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Media S.A 02.05.2023
• Subjects: Bayes Theorem; Cellular and Infection Microbiology; Humans; metagenomics; molecular epidemiology; OPV (oral polio vaccine); Phylogeny; Poliomyelitis - epidemiology; Poliomyelitis - prevention & control; Poliovirus - genetics; Poliovirus Vaccine, Oral; recombination; Spain - epidemiology; vaccine-derived polio virus; whole-genome; Spain; Bayesian phylogenetic analysis; molecular epidemiology; recombination; OPV (oral polio vaccine); whole-genome; metagenomics; vaccine-derived polio virus
• ISSN: 2235-2988; 2235-2988
• DOI: 10.3389/fcimb.2023.1168355

In 2021, a type 2 vaccine-derived poliovirus (VDPV2) was isolated from the stool of a patient with acute flaccid paralysis (AFP) admitted to Spain from Senegal. A virological investigation was conducted to characterize and trace the origin of VDPV2. We used an unbiased metagenomic approach for the whole-genome sequencing of VDPV2 from the stool (pre-treated with chloroform) and from the poliovirus-positive supernatant. Phylogenetic analyses and molecular epidemiological analyses relying on the Bayesian Markov Chain Monte Carlo methodology were used to determine the geographical origin and estimate the date of the initiating dose of the oral poliovirus vaccine for the imported VDPV2. We obtained a high percentage of viral reads per total reads mapped to the poliovirus genome (69.5% for pre-treated stool and 75.8% for isolate) with a great depth of sequencing coverage (5,931 and 11,581, respectively) and complete genome coverage (100%). The two key attenuating mutations in the Sabin 2 strain had reverted (A481G in the 5'UTR and Ile143Thr in VP1). In addition, the genome had a recombinant structure between type-2 poliovirus and an unidentified non-polio enterovirus-C (NPEV-C) strain with a crossover point in the protease-2A genomic region. VP1 phylogenetic analysis revealed that this strain is closely related to VDPV2 strains circulating in Senegal in 2021. According to Bayesian phylogenetics, the most recent common ancestor of the imported VDPV2 could date back 2.6 years (95% HPD: 1.7-3.7) in Senegal. We suggest that all VDPV2s circulating in 2020-21 in Senegal, Guinea, Gambia, and Mauritania have an ancestral origin in Senegal estimated around 2015. All 50 stool samples from healthy case contacts collected in Spain (n = 25) and Senegal (n = 25) and four wastewater samples collected in Spain were poliovirus negative. By using a whole-genome sequencing protocol with unbiased metagenomics from the clinical sample and viral isolate with high sequence coverage, efficiency, and throughput, we confirmed the classification of VDPV as a circulating type. The close genomic linkage with strains from Senegal was consistent with their classification as imported. Given the scarce number of complete genome sequences for NPEV-C in public databases, this protocol could help expand poliovirus and NPEV-C sequencing capacity worldwide.