We studied 561NPEV isolates obtained from 2,530 AFP patients, 89 from 383 HFMD patients, and 1 from 23 encephalitis patients in Mumbai and surrounding areas during 2008–2012. Partial sequencing viral protein 1 (VP1) was used to identify the NPEV (sero) types as described previously (14). Fourteen EV-A71 isolates were thus identified: 10 from AFP patients, 2 from HFMD patients, and 1 each from patients with encephalitis and febrile illness. Five EV-A71 isolates identified at various times since 2002 from northern Indian states were also included in this study (J. Deshpande, unpub. data). The VP1 region of EV-A71–positive isolates was sequenced as described previously (15). Sequencing was done by using Big Dye Terminator v3.1 Cycle Sequencing kit in accordance with the manufacturer’s instructions (Applied Biosystems, Foster City, CA, USA). Sequences were resolved on ABI 3130xl Genetic Analyzer (Applied Biosystems) and edited by using Sequencher v4.10.1 (Gene Codes, Ann Arbor, MI, USA). Complete VP1 sequences were aligned by using ClustalW (http://www.ebi.ac.uk). Phylogenetic analysis was conducted in MEGA5 by using maximum-likelihood and neighbor-joining methods (http://www.megasoftware.net). Genetic distances were calculated by Kimura 2-parameter method. A genotype was defined by >85% nt sequence similarity in the complete VP1 region (12). Sequences were deposited in GenBank under accession nos. KF906416–KF906434.

Median age of patients from whom the EV-A71 strains were isolated was 28 months; 10 (56%) were girls (Table 1). Three EV-A71–positive patients reported in Mumbai had lived in Uttar Pradesh. EV-A71–positive HFMD cases were non-neurologic, and the encephalitis patient recovered completely without neurologic deficit. Medical examination results 60 days after onset were not available for the AFP patients.

Figure 1

Figure 1. Neighbor-joining tree of enterovirus (EV) A71 using complete viral protein 1 (VP1) sequences. Genetic distances calculated by Kimura 2-parameter method are shown above the branches, and bootstrap values (1,000 replicates) are...

Nine isolates clustered with the strain R-13223-IND-01 (GenBank accession no. AY179600) (Figure 1), which was the only representative of EV-A71 genogroup D. A mean genetic distance of 11% and presence of multiple clusters within this genogroup indicated continuous evolution of genogroup D viruses over several years. Eight isolates (5 from AFP patients, 2 from HFMD patients, and 1 from an encephalitis patient) clustered with subgenogroup C1. Sequence similarity of >99% of 6 of the 8 isolates indicated a focal outbreak caused by C1 subgenogroup. Clustering of the subgenogroup C1 isolates with those from Germany, the Netherlands, and Azerbaijan suggested an epidemiologic link with the EV-A71 circulating in the European Region. V08–5327 and V11–2209 had sequence similarity of 89.5% between them and >18% divergence from all other known EV-A71 genogroups. We propose a new genogroup G designation for EV-A71 isolates V08–5327 and V11–2209. Table 2 shows estimates of evolutionary divergence of genogroup G in global context.

Figure 2

Figure 2. Neighbor-joining tree of enterovirus (EV) A71 strains isolated in India. Partial viral protein 1 (VP1) sequences (707 nt) were used for the analysis. Genetic distances calculated by Kimura 2-parameter method are...

To study the genetic diversity among EV-A71 strains isolated in India, we extracted partial VP1 nucleotide sequences of various lengths from GenBank. We detected 50 EV-A71 strains among NPEV isolates from AFP patients in Uttar Pradesh, Kerala, and Karnataka during 2007–2009 (9). Four EV-A71 isolates were reported from AFP patients from Uttar Pradesh studied during 2009–2010 (10). We used 21 sequences of isolates from Rao (9) and 1 of Laxmivandana (10) for determining the genetic relationship of EV-A71 isolated in India. Neighbor-joining tree using partial VP1 (707 nt) sequences (Figure 2) showed that 11 EV-A71 strains assigned to genogroup F by Rao et al. actually clustered within isolates of genogroup D, and the remaining 10 EV-A71 strains clustered within the new genogroup G. None of the EV-A71 isolates of Rao et al. (9) clustered in the subgenogroup C1 (Technical Appendix).

Bessaud et al. very recently reported (12) that 20 partial sequences reported by Rao et al. did not cluster into any known subgenogroup, which suggests more genogroups in India. Our current study confirms that some of them cluster within genogroup G. The remaining sequences from the study of Rao et al. need further analysis on the basis of longer nucleotide sequences.

Our study reports the discovery of a new genogroup G of EV-A71 in India. Genogroups D and G may be endemic in India because they were isolated from AFP patients in wide geographic areas; however, these genogroups have not been implicated in any specific outbreaks. Moreover, the 2 genogroups appear to be indigenous to India because they have not been detected in any other country. EV-A71 strains of subgenogroup C1 were isolated only in 2011–2012, indicating very recent circulation. Subgenogroup C1 was associated with time- and space-clustered cases of AFP, HFMD, and encephalitis. Multiple genogroups and high sequence divergence within genogroups D and G showed that EV-A71 has been spreading across the country. Therefore, systematic efforts should be made to understand the impact of the virus on public health in India.