Bioinformatics analysis and genetic diversity of the poliovirus

Transcription

1 Journal of Medical Microbiology (2014), 63, DOI /jmm Bioinformatics analysis and genetic diversity of the poliovirus Yanhan Liu, Tengfei Ma, Jianzhu Liu, Xiaona Zhao, Ziqiang Cheng, Huijun Guo, Shujing Wang and Ruixue Xu Correspondence Jianzhu Liu or Received 4 August 2014 Accepted 19 September 2014 College of Veterinary Medicine, Research Center for Animal Disease Control Engineering Shandong Province, Shandong Agricultural University, Tai an , PR China Poliomyelitis, a disease which can manifest as muscle paralysis, is caused by the poliovirus, which is a human enterovirus and member of the family Picornaviridae that usually transmits by the faecal oral route. The viruses of the OPV (oral poliovirus attenuated-live vaccine) strains can mutate in the human intestine during replication and some of these mutations can lead to the recovery of serious neurovirulence. Informatics research of the poliovirus genome can be used to explain further the characteristics of this virus. In this study, sequences from 100 poliovirus isolates were acquired from GenBank. To determine the evolutionary relationship between the strains, we compared and analysed the sequences of the complete poliovirus genome and the VP1 region. The reconstructed phylogenetic trees for the complete sequences and the VP1 sequences were both divided into two branches, indicating that the genetic relationships of the whole poliovirus genome and the VP1 sequences are very similar. This branching indicates that the virulence and pathogenicity of poliomyelitis may be associated with the VP1 region. Sequence alignment of the VP1 region revealed numerous mutation sites in which mutation rates of.30 % were detected. In a group of strains recorded in the USA, mutation sites and mutation types were the same and this may be associated with their distribution in the evolutionary tree and their genetic relationship. In conclusion, the genetic evolutionary relationships of poliovirus isolate sequences are determined to a great extent by the VP1 protein, and poliovirus strains located on the same branch of the phylogenetic tree contain the same mutation spots and mutation types. Hence, the genetic characteristics of the VP1 region in the poliovirus genome should be analysed to identify the transmission route of poliovirus and provide the basis of viral immunity development. INTRODUCTION Poliomyelitis, caused by the poliovirus, is an acute viral infectious disease that attacks the circulatory system and invades the nervous system. This disease primarily affects children aged one to six years. Clinical manifestations vary and the main symptoms include fever, general malaise, severe physical pain and total paralysis (Tavares et al., 2013). Humans are the only primary natural hosts; however, primates and Old World monkeys can be experimentally infected with this virus. Poliovirus is a typical picornavirus, and has a small non-enveloped virion of approximately 27 nm in diameter. The icosahedral capsid particle is composed of 60 copies and each of the copies contains capsid proteins VP1, VP2, VP3 and VP4 (Roberts et al., 2012). Polioviruses comprise three serotypes (PV1 3) and immunity against one type does not inhibit infection by the Abbreviations: AFP, acute flaccid paralysis; OPV, oral poliovirus attenuated-live vaccine; WHO, World Health Organization; VDPD, vaccine-derived poliovirus. two other types (Odoom et al., 2012). The whole viral genome is approximately 7440 nt in length and 90 % of this genome comprises a large open reading frame encoding a single polyprotein (Tavares et al., 2013). The VP1 region, which contains 906 nt and encodes the major capsid protein, is thought to contribute to the virulence and pathogenicity of the whole virus (Pan et al., 2006). Viral transmission usually occurs via the faecal oral route; the virus multiplies mainly in the small intestine and throat, although exact histological sites have not yet been established (John et al., 2011). In 1 2 % of infected individuals, the virus enters the central nervous system and replicates in the motoneurons in the spinal cord, brainstem, or motor cortex producing muscle paralysis (Racaniello, 2006). The oral poliomyelitis vaccine (OPV) was the first tool used in the attempt to eradicate poliomyelitis worldwide (Rhoden et al., 2013). In the winter of 1963, annual campaigns were conducted to administer live-attenuated OPV (Luo et al., 2013). Since 1978, OPV has been included in the World Health Organization s (WHO) Expanded Program on G 2014 The Authors Printed in Great Britain

2 Bioinformatics analysis of the poliovirus Immunization and the incidence rate of poliomyelitis has rapidly decreased (Kew et al., 2002). Poliovirus is susceptible to sequence drift; after individuals have been vaccinated for a relatively long period, the poliovirus is excreted into the environment and OPVs can efficiently eradicate wild polioviruses. However, the risks associated with this vaccination have increased (Blomqvist et al., 2012). Although OPVs were developed to prevent infection, OPVs can induce poliomyelitis infection, which can then be transmitted from one person to another (Minor, 2009). Attenuated poliovirus is highly adaptable and can mutate to increase its virulence; as a result, the incidence and spread of poliomyelitis are increasing and levels of circulating vaccine-derived poliovirus (VDPV) isolates are enhanced (Shimazawa & Ikeda, 2012). Circulating VDPVs are recombinant isolates that include sequences either from OPV strains or other enteroviruses circulating in the environment (Pastoret, 2010). In microcosmic environments, the VP1 region contains distributed mutation points in an increasing number of polio vaccine-related strains with mutation rates of,1% (Haddad-Boubaker et al., 2008). In the regional distribution of this disease, viruses are present in low concentrations and exhibit limited circulation in local areas (Dimitriou et al., 2013). To determine the relationship of polio strains at a molecular level, we selected 100 representative strains and analysed the corresponding sequences. We also reconstructed phylogenetic trees using the whole poliovirus genome sequence and the VP1 region sequence. To determine whether genetic mutations affect the distribution of strains in the evolutionary trees, we performed sequence alignment to search for mutation hotspots in the VP1 sequences. METHODS A total of 100 poliovirus isolate sequences were downloaded from GenBank using the BLAST program ( BLAST). We first selected a single serotype (PV1) for analysis utilizing the PubMed database ( For our bioinformatic analyses, 100 sequences were selected at random from the numerous complete PV1 genomes available in the databases. The representative sequences were from three geographic areas: 61 from the USA, 23 from Asia and 16 from Europe (Table 1). All of the sequences were separated from people who had been vaccinated and had been isolated from acute flaccid paralysis (AFP) cases or stool specimens of healthy individuals in accordance with WHO standard protocols. All of the isolates, which were at least 95 % identical in VP1 sequence, belonged to the same PV1 serotype from different strains. Phylogenetic analyses based on the whole gene sequences and VP1 sequences were conducted using the distance-based neighbourjoining method using MEGA v. 4.0 software (Pan et al., 2012). Mutation hotspots were identified by aligning the sequences with the reference Sabin strain (GenBank accession no. AY184219) using DNASTAR software. GENEDOC software was used for sequence alignment and to identify the location and the type of mutation points present (Schindler & Fischer, 2012). The mechanism by which these mutations occurred (i.e. A to G or A to T) was also investigated. In addition, we determined whether isolates with similar mutation spots and mutation types were located in the same branch of the phylogenetic tree to estimate the relationship between gene mutation and genetic evolution of the poliovirus strains. RESULTS Analysis of the phylogenetic trees The phylogenetic tree reconstructed using the whole genome sequences of 100 poliovirus strains was divided into two branches (Fig. 1); the phylogenetic tree reconstructed based on the VP1 sequences was divided into three branches and there was one branch with only two isolates (Fig. 2). We found that the shape and the branches of the two phylogenetic trees were substantially identical, and the sequences on each branch are similarly identical and, as only highly homologous sequences were located in the same branch or close to each other, the phylogenetic tree of the whole genome sequence might resemble the shape of the phylogenetic tree based on the VP1 protein because VP1 is the major virulence protein of the virus. Therefore, the VP1 protein may be highly associated with the virulence of the poliovirus. By contrast, VP1 gene mutations or restructuring could lead to the transition of a strain type. Table 2 indicates the neurotoxicity points that were found in the VP1 region. Several strains collected from different areas were detected in the same branch but exhibited slight differences. Such differences indicate that the polioviruses infecting individuals worldwide are generally the same. Sequence alignment VP1 sequences were compared and analysed. The results revealed multiple mutations with a total of 12 mutation hotspots: nt 81, nt 221, nt 274, nt 290, nt 292, nt 319, nt 406, nt 526, nt 541, nt 733, nt 817 and nt 892. Numerous isolates were mutated in these sites at a mutation rate of.30 %. Table 2 shows the specific positions and types of nucleotide mutations identified in the VP1 region when the isolates were compared with the Sabin 1 (GenBank accession no. AY184219) sequence as reference. Hotspots were searched by conducting a comparative analysis; the results showed that a group of strains isolated from the USA exhibited similar mutations at several sites (nt 330, A to G; nt 826, C to T). These mutation positions and types were exactly the same; however, no mutations were found in these sites in other isolates. Only rare mutations were found in these positions in the other isolates. In the phylogenetic tree, mutations occurred in these sites to a higher extent in the group of isolates from the USA than in other groups; hence, a closer genetic relationship was observed in these isolates or those in the same branch. DISCUSSION Although OPV has been used worldwide to successfully treat polio, derived strains remain sporadically distributed worldwide because of this vaccine (Nathanson & Fine, 2002). Poliovirus is retained in the environment and can infect the body, thereby causing atypical polio (Smallman- Raynor & Cliff, 2013) that could impede advances in the global polio-free status (Iliyasu et al., 2013; Rahimi et al.,

3 Y. Liu and others Table 1. Origin of poliovirus isolate sequences acquired from GenBank No. GenBank accession no. Isolate Serotype Sequence length (bp) Country of origin 1 AY Sabin1 PV USA 2 VO1150 Sabin1 PV USA 3 GQ S302 PV CN 4 EU A21 PV UK 5 EU A49 PV UK 6 EU A63 PV UK 7 EU A102 PV UK 8 EU A84 PV UK 9 EU A91 PV UK 10 EU A88 PV UK 11 EU A118 PV UK 12 EU A105 PV UK 13 EU A199 PV UK 14 VO1149 Sabin1 PV USA 15 FJ CHN/8184/GZ/CHN/2004 PV CN 16 FJ CHN/8229-1/GZ/CHN/2004 PV CN 17 FJ CHN/8223c/GZ/CHN/2004 PV CN 18 FJ CHN/8225c/GZ/CHN/2004 PV CN 19 FJ CHN/8248c/GZ/CHN/2004 PV CN 20 FJ CHN/8229-2/GZ/CHN/2004 PV CN 21 AJ CoxI PV USA 22 FJ CHN/8229-3/GZ/CHN/2004 PV CN 23 AJ CHAT10A-11 PV USA 24 EU A449 PV UK 25 VO1148 Sabin1 PV USA 26 EU A649 PV UK 27 AF TCDC PV CN 28 AY d PV USA 29 AY d PV USA 30 AF TCDC PV CN 31 AF TCDCE PV CN 32 EF USA10770 PV USA 33 EF USA10774 PV USA 34 EF USA10777 PV USA 35 AY d PV USA 36 EF USA10785 PV USA 37 EF USA10779 PV USA 38 EF USA10771 PV USA 39 AY d PV USA 40 EF USA10772 PV USA 41 EF USA10784 PV USA 42 EF USA10776 PV USA 43 EF USA10775 PV USA 44 EF USA10778 PV USA 45 EF USA10773 PV USA 46 AY d PV USA 47 EF USA10781 PV USA 48 EF USA10783 PV USA 49 EF USA10780 PV USA 50 EF USA10782 PV USA 51 EF USA10786 PV USA 52 AF TCDC PV CN 53 AJ / PV GER 54 FJ PV CN 55 FJ c PV CN 56 FJ c PV CN 1726 Journal of Medical Microbiology 63

4 Bioinformatics analysis of the poliovirus Table 1. cont. No. GenBank accession no. Isolate Serotype Sequence length (bp) Country of origin 57 FJ c PV CN 58 FJ c PV CN 59 FJ c PV CN 60 FJ c PV CN 61 AF RUS PV RUS 62 AF PV USA 63 AJ / PV GER 64 AF DOR00013 PV USA 65 AF DOR00015 PV USA 66 AF DOR00023C PV USA 67 AF HAI01002 PV USA 68 AF DOR00028 PV USA 69 AF DOR00028C PV USA 70 AF HAI01007 PV USA 71 AF DOR00024 PV USA 72 AF DOR00042C1 PV USA 73 AF DOR01001C2 PV USA 74 AF DOR00016 PV USA 75 AF DOR01001 PV USA 76 AF DOR01001C1 PV USA 77 AF DOR01002 PV USA 78 AF DOR00042 PV USA 79 AF DOR01002C PV USA 80 AF DOR00042C2 PV USA 81 AF DOR00025 PV USA 82 AB Mindanao-01-1 PV JPN 83 AF DOR00044 PV USA 84 AB Luzon-01-1 PV JPN 85 AF HAI01001 PV USA 86 AF DOR00041C2 PV USA 87 AB Luzon-01-2 PV JPN 88 AF DOR00041C3 PV USA 89 AF DOR00041C1 PV USA 90 AF DOR01012 PV USA 91 AF HAI01009 PV USA 92 AF HAI01008C2 PV USA 93 AF HAI01008C1 PV USA 94 AF HAI01008 PV USA 95 AF HAI00003 PV USA 96 AB Luzon-01-2c PV JPN 97 AF HAI01015 PV USA 98 AF HAI PV USA 99 EF /b/97 PV Greece 100 DQ /EDU00B3 PV GER CN, China; GER, Germany; JPN, Japan; RUS, Russia. 2009). In this study, poliovirus strains were subjected to bioinformatics analyses involving the reconstruction of phylogenetic trees and genetic diversity analysis. In contrast to the genetic relationship between the whole sequences of the two groups, three genetic groups were found in the VP1 region. The results also indicated that polioviruses found in different regions were closely related to the isolated strains of polioviruses from Europe, Asia and North America and were also found in the same branch of the phylogenetic tree. Hence, no evidence was observed of variations and gene reorganization in polioviruses, indicating that the re-entry of wild strains into circulation could be prevented. In addition, the distribution of the whole sequence from the phylogenetic tree is the same as that of VP1; this result is consistent with that of a previous study (Jiang & Huang, 2010). VP1 can be

5 Y. Liu and others EF USA EF USA EF Greece AF USA VO1148 USA AY USA DQ GER AY USA AF CN AF CN AY USA AY USA AY USA FJ CN FJ CN EF USA EF USA EF USA EF USA FJ CN FJ CN FJ CN FJ CN FJ CN EF USA EF USA EF USA AF USA AF RUS EF USA AF USA AF USA EF USA EF USA AF USA AF USA AJ GER EF USA AF USA AF USA AJ GER AF USA AF USA AF USA AF USA AF USA EU UK AF USA AF USA EU UK EU UK AF USA EU UK AF USA AF USA EU UK AF USA EU UK AF USA EU UK AF USA AF USA EU USA EU UK EU UK AF USA AF USA EU UK EU UK AF USA AF USA AF CN FJ CN FJ CN AF USA AF USA AF USA FJ CN FJ CN FJ CN FJ CN AB Japan AF USA AF USA AF USA FJ CN VO1150 USA AY USA GQ CN VO1149 USA AJ USA AJ USA AF CN EF USA EF USA EF USA EF USA AB Japan AB Japan AB Japan Fig. 1. Phylogenetic tree reconstructed from whole genome sequences of 100 poliovirus isolates by the distance-based neighbour-joining method using MEGA v. 4.0 software ( Bootstrap values were calculated on 1000 replicates of the alignment. The whole genome sequence of the poliovirus evolved into two branches. CN, China; GER, Germany; JPN, Japan; RUS, Russia. considered as the primary research target for the global eradication of the poliovirus. Scattered mutation points are present in VP1 sequences. The P1 region encoding capsid proteins may be an important variable in the whole genome among different strains of polioviruses (Blomqvist et al., 2010; Luo et al., 2013). In theory, each nucleic acid of the poliovirus genomic RNA molecule could undergo mutation; however, the distribution of mutations in the genome was not random. By contrast, several mutation types were observed at a higher than average frequency. Mutation hotspots were observed in the VP1 coding regions of the genomes but were not completely randomized; each nucleic acid also exhibited a different mutation probability. Some mutation hotspots are also associated with neurotoxicity (Rekand et al., 2003). For example, the mutation hotspot related to neurotoxicity is found at nt 2749 in type I poliovirus. Mutant strains, 1728 Journal of Medical Microbiology 63

6 AF USA Bioinformatics analysis of the poliovirus EF USA EF USA EF USA EF USA EF USA EF USA EF USA EF USA EF USA EF USA EF USA EF USA EF USA EF USA AB Japan AB Japan AB Japan AB Japan AF CN AF CN EF USA EF USA EF USA AY USA AY USA AY USA AY USA AF CN EU UK EU UK EU UK V01150 USA AY USA AF CN AJ GER AY USA AF RUS EU USA EU UK AJ GER EU UK EU UK EU UK EU UK EU UK EU UK EU UK AF USA FJ CN AF USA FJ CN AF USA FJ CN FJ CN AF USA FJ CN AF USA FJ CN FJ CN AF USA AF USA AF USA V01149 USA VO1148 USA AF USA AF USA FJ CN AJ USA AJ USA AF USA AF USA AF USA AF USA FJ CN FJ CN FJ CN FJ CN AF USA AF USA AF USA AF USA AF USA EF Greece FJ CN FJ CN AF USA AF USA AF USA AF USA AF USA AF USA AF USA AF USA AF USA AF USA AF USA DQ GER AF USA Fig. 2. Phylogenetic tree reconstructed using the poliovirus VP1 region sequences. A neighbour-joining phylogenetic tree of 99 sequences taken from 906 bp of the poliovirus VP1 region was reconstructed by the distance-based neighbour-joining method using MEGA v. 4.0 software. Bootstrap values were calculated on 1000 replicates of the alignment (bar, 0.005). VP1 sequences of poliovirus evolved into three branches with only two isolates on one branch. Therefore, this distribution is consistent with the evolutionary distribution of the whole genome sequence of the poliovirus. CN, China; GER, Germany; JPN, Japan; RUS, Russia. produced from an A to G mutation at the nt 2749 site, are virulence back-mutation strains. As such, this site is exposed to extreme selection pressure and easily undergoes mutation, reproducing wild-type strains. Moreover, this process is likely to cause a series of changes in virulence that may paralyse many patients, resulting in a rapid increase in AFP cases. Several similar exchanges are caused naturally by mutated hotspots between amino acids (Li et al., 1996). Approximately 12 mutation hotspots were found in 100 isolates and.30 % of these isolates were mutated at these mutation points. Such mutations may affect the genetic relationships and distributions of the poliovirus strains in the phylogenetic tree and may be associated with changes in virus virulence. For example, a group of strains isolated in the USA were found on similar branches of the phylogenetic tree because they mutated at the same sites; other isolates did not mutate and exhibited the same mutation types. Therefore, changes in specific gene loci may affect the relationships and distribution of strains in the phylogenetic tree. Changes in nucleotide sequences may cause changes in protein properties and affect the biological characteristics of the poliovirus. These characteristics should be investigated further. Genetic recombination, as one of the mechanisms of the molecular evolution of RNA viruses, can help the virus to adapt easily in host systems and evade host immunity. The observation of genetic recombination suggests that the

7 Y. Liu and others Table 2. Nucleotide variations in the VP1 region of the poliovirus from all isolates Mutational site in VP1 (nt) monitoring of the region s epidemiology and laboratory should be strengthened to maintain a high level of immunization coverage of OPV (Stevens et al., 2013). The monitoring procedures for the poliovirus in external environments and the detection and prevention of the possible circulation of VDPVs should be improved. The relationship between restructured sites and neurotoxicity has not yet been determined in vaccine-related polio types or in mutant virus isolated from healthy patients. Mutations are likely to occur in immunocompromised children or individuals (Zhang & Hou et al., 2010). Studies have shown that mutation is not the primary cause of increasing virulence and propagation; instead, this increase in virulence and propagation occurs in a specific population during the virus life cycle. Therefore, mutation is highly important in the immune system to stimulate vaccine immunization coverage in a region, but a polio-free environment poses a great challenge to researchers. For this reason, further studies should be conducted to determine whether mutation can increase the virulence of viruses. Studies should also be conducted to prevent potential risks associated with increased virulence. In conclusion, this study provides evidence that the genetic evolutionary relationship of poliovirus strains has been determined to a great extent by the VP1 protein. In VP1 sequences, 12 mutation hotspots were found in which the mutation rate was greater than 30 %. Using bioinformatics analyses, we found that the selected strains containing the same mutation spots and mutation types were located on the same branches of the phylogenetic tree. REFERENCES Nucleotide substitution Mutation rate (%) 81 C A T G A A T A C A A T G A A G A T T A C G A A C A T C A T G A T G A A 34 Blomqvist, S., Savolainen-Kopra, C., Paananen, A., El Bassioni, L., El Maamoon Nasr, E. M., Firstova, L., Zamiatina, N., Kutateladze, T. & Roivainen, M. (2010). Recurrent isolation of poliovirus 3 strains with chimeric capsid protein Vp1 suggests a recombination hot-spot site in Vp1. Virus Res 151, Blomqvist, S., El Bassioni, L., El Maamoon Nasr, E. M., Paananen, A., Kaijalainen, S., Asghar, H., de Gourville, E. & Roivainen, M. (2012). Detection of imported wild polioviruses and of vaccine-derived polioviruses by environmental surveillance in Egypt. Appl Environ Microbiol 78, Dimitriou, T. G., Pliaka, V., Kyriakopoulou, Z., Ruether, I. G., Tsakogiannis, D., Fountoucidou, P., Gartzonika, C., Levidiotou- Stefanou, S. & Markoulatos, P. (2014). PCR assays for the identification of rare recombination types from VP1 to 3D genomic region of vaccine derived poliovirus strains. Mol Cell Probes 28, Haddad-Boubaker, S., Ben Yahia, A., Bahri, O., Morel, V., Balanant, J., Delpeyroux, F. & Triki, H. (2008). Genetic features of polioviruses isolated in Tunisia, J Clin Virol 41, Iliyasu, Z., Nwaze, E., Verma, H., Mustapha, A. O., Weldegebriel, G., Gasasira, A., Wannemuehler, K. A., Pallansch, M. A., Gajida, A. U. & other authors (2014). Survey of poliovirus antibodies in Kano, Northern Nigeria. Vaccine 32, Jiang, Y. Y. & Huang, L. (2010). Survey of immune level of healthy population to poliomyelitis virus in border areas of Guangxi. China Trop Med 10, John, T. J., Jain, H., Ravishankar, K., Amaresh, A., Verma, H., Deshpande, J., Pallansch, M. A., Singh, A. P., Sreevatsava, M. & other authors (2011). Monovalent type 1 oral poliovirus vaccine among infants in India: report of two randomized double-blind controlled clinical trials. Vaccine 29, Kew, O., Morris-Glasgow, V., Landaverde, M., Burns, C., Shaw, J., Garib, Z., André, J., Blackman, E., Freeman, C. J. & other authors (2002). Outbreak of poliomyelitis in Hispaniola associated with circulating type 1 vaccine-derived poliovirus. Science 296, Li, J., Zhang, L. B., Hou, X. H., Zheng, H., Fang, Y., Minoru, H. & Akio, H. (1996). [Genetic analysis of vaccine polioviruses isolated in China]. Chinese Journal of Virology 12, (in Chinese). Luo, H. M., Zhang, Y., Wang, X. Q., Yu, W. Z., Wen, N., Yan, D. M., Wang, H. Q., Wushouer, F., Wang, H. B. & other authors (2013). Identification and control of a poliomyelitis outbreak in Xinjiang, China. N Engl J Med 369, Minor, P. (2009). Vaccine-derived poliovirus (VDPV): impact on poliomyelitis eradication. Vaccine 27, Nathanson, N. & Fine, P. (2002). Virology. Poliomyelitis eradication a dangerous endgame. Science 296, Odoom,J.K.,Forrest,L.,Dunn,G.,Osei-Kwasi,M.,Obodai,E.,Arthur- Quarm,J.,Barnor,J.,Minor,P.D.&Martin,J.(2012).Interruption of poliovirus transmission in Ghana: molecular epidemiology of wild-type 1 poliovirus isolated from 1995 to JInfectDis206, Pan, W. Y., Cai, Z. K., Yang, Y. & other authors (2006). Study on molecular epidemiology in Fujian Province. China Pub Health 22, Pan, W., Gao, Y., Qin, L., Ni, W., Liu, Z., Yun, B., Wang, Y., Qi, X., Gao, H. & Wang, X. (2012). Genetic diversity and phylogenetic analysis of glycoprotein GP85 of ALV-J isolates from Mainland China between 1999 and 2010: coexistence of two extremely different subgroups in layers. Vet Microbiol 156, Pastoret, P. P. (2010). Human and animal vaccine contaminations. Biologicals 38, Racaniello, V. R. (2006). One hundred years of poliovirus pathogenesis. Virology 344, Rahimi, P., Tabatabaie, H., Gouya, M. M., Mahmudi, M., Musavi, T., Rad, K. S., Azad, T. M. & Nategh, R. (2009). Direct identification of non-polio enteroviruses in residual paralysis cases by analysis of VP1 sequences. J Clin Virol 45, Rekand, T., Male, R., Myking, A. O., Nygaard, S. J. T., Aarli, J. A., Haarr, L. & Langeland, N. (2003). Detection of viral sequences in archival 1730 Journal of Medical Microbiology 63

8 Bioinformatics analysis of the poliovirus spinal cords from fatal cases of poliomyelitis in J Virol Methods 114, Rhoden, E., Liu, H. M., Wang-Chern, S. W. & Oberste, M. S. (2013). Anti-poliovirus activity of protease inhibitor AG-7404, and assessment of in vitro activity in combination with antiviral capsid inhibitor compounds. Antiviral Res 98, Roberts, J. A., Kuiper, M. J., Thorley, B. R., Smooker, P. M. & Hung, A. (2012). Investigation of a predicted N-terminal amphipathic a-helix using atomistic molecular dynamics simulation of a complete prototype poliovirus virion. J Mol Graph Model 38, Schindler, C. & Fischer, W. B. (2012). Sequence alignment of viral channel proteins with cellular ion channels. J Comput Biol 19, Shimazawa, R. & Ikeda, M. (2012). The vaccine gap between Japan and the UK. Health Policy 107, Smallman-Raynor, M. & Cliff, A. D. (2013). The geographical spread of the 1947 poliomyelitis epidemic in England and Wales: spatial wave propagation of an enigmatic epidemiological event. J Hist Geogr 40, Stevens, H., Bertelsen, M. F., Sijmons, S., Van Ranst, M. & Maes, P. (2013). Characterization of a novel polyomavirus isolated from a fibroma on the trunk of an African elephant (Loxodonta africana). PLoS ONE 8, e Tavares, F. N., da Costa, E. V., Kew, O. M. & da Silva, E. E. (2013). Complete genome sequence of the last representative genotype of wild indigenous poliovirus type 1, which circulated in Brazil. Genome Announc 1, e Zhang, L. B., Hou, X. H., Zhu, H. & other authors (2010). The circulation and pathogenesis of type 2 recombinant polioviruses in China. China Plan Immunization 6,