Emergence of influenza A ( H 1 N 1 ) pdm 09 genogroup 6 B and drug resistant virus , India , January to May 2015

To investigate the aetiology of the 2015 A(H1N1)pdm09 influenza outbreak in India, 1,083 nasopharyngeal swabs from suspect patients were screened for influenza A(H1N1)pdm09 in the state of Madhya Pradesh. Of 412 positive specimens, six were further characterised by phylogenetic analysis of haemagglutinin (HA) sequences revealing that they belonged to genogroup 6B. A new mutation (E164G) was observed in HA2 of two sequences. Neuraminidase genes in two of 12 isolates from fatal cases on prior oseltamivir treatment harboured the H275Y mutation.

To investigate the aetiology of the 2015 A(H1N1)pdm09 influenza outbreak in India, 1,083 nasopharyngeal swabs from suspect patients were screened for influenza A(H1N1)pdm09 in the state of Madhya Pradesh.Of 412 positive specimens, six were further characterised by phylogenetic analysis of haemagglutinin (HA) sequences revealing that they belonged to genogroup 6B.A new mutation (E164G) was observed in HA2 of two sequences.Neuraminidase genes in two of 12 isolates from fatal cases on prior oseltamivir treatment harboured the H275Y mutation.
An epidemic of influenza A(H1N1)pdm09, affecting over 39,000 persons and causing more than 2,500 deaths occurred in India in 2015 [1].We show that genotype 6B strains forming two sub-lineages circulated during the outbreak.Comparison of the sequences of six outbreak strains recovered in this work, to other published genotype 6B sequences, also reveals a unique combination of previously-reported mutations in the haemagglutinin (HA) gene.Two of the six sequences additionally display a E164G mutation in HA2, which has not been reported to date, moreover a N129D mutation in HA1 is observed for two sequences derived from patients with severe disease.Among strains analysed from 12 fatal cases on prior oseltamivir treatment, two harbour the H275Y mutation in the neuraminidase (NA) gene, which confers resistance to this antiviral.

Sampling and testing for influenza A(H1N1) pdm09
A total of 1,083 acute phase nasopharyngeal swab specimens from patients suspected of influenza (as prior defined [2]), were referred by 13 district health authorities of Madhya Pradesh, India between 29 January and 7 May 2015.Upon specimen collection, the travel history, treatment status, and symptoms of the patients were recorded in addition to age, sex and place of residence.The samples were handled in a designated biosafety level (BSL) 3 laboratory and viral RNA was extracted using QIAamp viral RNA mini kit (Qiagen).The RNA samples were screened by World Health Organization (WHO)-Centers for Disease Control and Prevention (CDC) approved quantitative real-time reverse transcription-polymerase chain reaction (qRT-PCR) for influenza A(H1N1)pdm09 [3].

Molecular analyses of the strains
Six clinical samples testing positive for influenza A(H1N1)pdm09 by qRT-PCR were selected based on patients' disease severity category A (n=2; A/India/ DRDE_GWL897/2015 and A/India/DRDE GWL721/2015), B (n=2; A/India/DRDE_GWL703/2015, A/India/ DRDE_GWL989/2015), and C (n=2; A/India/DRDE GWL719/2015 and A/India/DRDE_GWL812/2015) as previously described [2], and used for direct nucleotide (nt) sequencing of the haemagglutinin (HA) gene.A phylogenetic analysis was performed by comparing with nt sequence of 45 globally diverse influenza A(H1N1)pdm09 viruses retrieved from GenBank (as further shown in the phylogenetic tree) and the Global Initiative on Sharing Avian Influenza Data (GISAID) (Table 1).The phylogenetic tree in this analysis was constructed with maximum likelihood and bootstrap analysis of 1,000 replicates using Mega 5.03 software [4].Further the amino acid substitutions were marked at the major branches for better clarity.
Influenza A(H1N1)pdm09 HA amino-acid sequences were inferred from the genetic sequences obtained in this study, and the protein structures were modelled using Modeller software and compared to prototype A/ California/07/2009 through Discovery studio client 4.1.
The qRT-PCR positive samples from 12 fatal cases, all on prior oseltamivir therapy, were also tested for a mutation (H275Y) conferring resistance to this antiviral by PCR-restriction fragment length polymorphism (RFLP)  analysis of the NA gene [5].Of these 12 cases, one (with corresponding sample: A/India/DRDE GWL719/2015) also belonged to the group of six patients, from whom the HA gene was sequenced.The PCR-RFLP positive samples were further confirmed through nt sequencing of the target sites of the NA gene.

Results of screening for influenza A(H1N1) pdm09
A total of 1,083 patients, including 525 males, were screened for influenza A(H1N1)pdm09 by qRT-PCR.The age range of these patients varied from 0 to 90 years-old, with age groups between 21 and 30 year-old (n=284) as well as between 31 and 40 year-old (n=179) representing 26% and 17% of the total respectively (Table 2).Of the 1,083 clinical samples tested, 412 (38%) were found positive for influenza A(H1N1)pdm09 virus.Similar to patients screened, most of those testing positive were from young age groups, with 21 to 30 years-old (n=104; 25%) representing the majority, followed by 31 to 40 year-olds (n=69; 17%).The positivity rate among the different age groups varied from 25 to 59% (Table 2).The female to male sex ratio of PCR positive patients was found to be 1.20:1.

Molecular characteristics of outbreak strains
In addition to the six substitutions defining genotype 6B, all HA-sequenced viruses in this study presented five mutations compared to prototype A/ California/07/2009, namely, P83S, I321V in HA1, as well as E47K, S124N, and E172K in HA2 (Figure 2).Further to these total 11 mutations, N129D was found in HA1 sequences of two specimens (A/India/DRDE GWL719/2015 and A/India/DRDE GWL812/2015) from patients with severe disease (both category C including one fatal case).Also, E164G was found in HA2 of A/India/DRDE GWL721/2015 and A/India/DRDE GWL812/2015.

Figure 2
Three-dimensional quaternary structure of trimeric haemaglutinin protein, identifying mutations compared to A/California/07/2009 in the proteins in this study, India, 2015 In this structure two monomers are coloured green and red.In the third monomer, residues are coloured blue and amino acid residues differing from A/California/07/2009 are denoted as yellow spheres.The mutations are listed to the right of the molecule, along with the four structural regions that contain the mutations.Mutations characterising genogroup 6B are underlined.We acknowledge the authors, originating and submitting laboratories of the sequences from GISAID's EpiFlu Database on which this research is based.All submitters of data may be contacted directly via the GISAID website www.gisaid.org.
A thorough in silico analysis revealed that all of the 11 mutations common to the 2015 Indian sequences studied here, have been reported in different strains of influenza A(H1N1)pdm09 virus isolated from various parts of the world in the past [8][9][10].However, to date, no single strain was reported to possess all these 11 mutations together, except the Indian 2015 strains sequenced in this study.Moreover, the E164G mutation found in HA2 of A/India/DRDE GWL812 and A/India/ DRDE GWL721 has not previously been reported.
Modelling reveals that mutations are found in the head, stalk and tail region of HA protein but the majority were found in the head region which covers the major antigenic binding region.The HA2 E172K mutation showed distinct structural changes in the tail region compared to the influenza A(H1N1)pdm09 virus prototype [11].
Two influenza A(H1N1)pdm09 strains from 12 fatal cases were found to possess H275Y oseltamivir resistance mutation.

Discussion
Influenza A viruses have been responsible for four influenza pandemics in last century viz., Spanish influenza (H1N1) in 1918, Asian influenza (H2N2) in 1957, Hong Kong influenza (H3N2) in 1968 and pandemic influenza (H1N1) in 2009, which was caused by influenza A(H1N1)pdm09.During the 2009 pandemic period (2009-2010), India was affected with around 50,000 cases and a case fatality of 6% [12].After the end of the 2009 pandemic, the virus continued to circulate at low level in the population, and during the period from 2011 to 2014 the circulation of the virus declined [13].
From January to May 2015 however, over 39,000 persons in India were affected by a new epidemic of influenza A(H1N1)pdm09, with more than 2,500 deaths [1].The outbreak spread across 22 of the 29 states in the country, making it the largest since 2009.This sudden re-emergence and wide spread simultaneous reporting of influenza A(H1N1)pdm09 along with higher number of hospitalisations and deaths was a major public health concern.
By further characterising the strains infecting patients positive for influenza A(H1N1)pdm09 through HA phylogeny, this study finds that sequences of genogroup 6B were circulating during the 2015 epidemic.The genogroup 6B was found to evolve from a Russian isolate (A/Moscow-Oblast/CRIE-08/2013) and is since then circulating in many parts of the world.However, this is the first report from India regarding circulation of genogroup 6B, coinciding with a large scale outbreak [1].
Researchers from Massachusetts Institute of Technology (MIT) have recently reported mutations D225N, and T200A in a 2014 Indian strain (A/ India/6427/2014, which also clusters with genogroup 6B sequences in the phylogenetic tree Figure 1) making the virus more infectious [14].Although we did not find these two mutations in our study, all the sequences that we characterised harboured five mutations (P83S, I321V in HA1, as well as E47K, S124N, and E172K in HA2), which although previously described, have not been reported in combination.Moreover, two isolates from patients with severe disease harboured a N129D mutation in HA1 and two isolates had a mutation in HA2, E164G, that has not been observed to date.These unique features of the viruses found here may have played a role in shaping the large scale epidemic with cases of severe disease.On the other hand, the 2015 epidemic in India may be attributed to lack of immunity among an immune-naïve population.It is also noteworthy that seasonal influenza vaccination is not very common in India.
Some limitations of the study include that the samples were only tested for influenza A(H1N1)pdm09 virus, whereby only 38% of samples tested were positive.Therefore, co-circulation of other influenza subtypes or types could not be ruled out.Moreover the sequence analysis was conducted with only few positive samples that did not cover other gene segments than the HA and NA genes.
The influenza A(H1N1)pdm09 virus represents a quadruple reassortment of two swine, one human, and one avian strain of influenza virus [15].The largest proportion of genes comes from swine influenza viruses (30.6% from North American swine strains, 17.5% from Eurasian swine strains), followed by North American avian strains (34.4%) and human influenza strains (17.5%).It will be interesting to investigate the involvement of any gene reassortment in the 2015 outbreak in India through complete genome sequencing.
Two of 12 strains from fatal cases were found to harbour a mutation conferring resistance to oseltamivir.Learning more about the 2015 strains circulating in India could help public health officials determine treatment options and inform on vaccines for the next influenza season, which is likely to include currently circulating strains [16].
Our findings show the importance of systematic molecular surveillance to provide insight into strains circulating during influenza epidemics.

Figure 1
Figure 1Phylogenetic analysis of influenza A(H1N1)pdm09 viral sequences derived from an outbreak in India, 2015

Table 1
Details of the A(H1N1)pdm09 sequences retrieved from the Global Initiative on Sharing Avian Influenza Data (GISAID)'s EpiFlu Database for complete haemagglutinin-gene-based phylogenetic analysis in this study S: segment.

Table 2
Age distribution of persons with confirmed influenza A(H1N1)pdm09 positive samples, Madhya Pradesh, India, 29 January-7 May 2015 (n=412) Where in each age group, n is the number of positive samples and N the total number of samples screened and the percentage is the positivity rate. a