Correction
2 Nov 2015: The PLOS ONE Staff (2015) Correction: Deep Sequencing for Evaluation of Genetic Stability of Influenza A/California/07/2009 (H1N1) Vaccine Viruses. PLOS ONE 10(11): e0142248. https://doi.org/10.1371/journal.pone.0142248 View correction
Figures
Abstract
Virus growth during influenza vaccine manufacture can lead to mutations that alter antigenic properties of the virus, and thus may affect protective potency of the vaccine. Different reassortants of pandemic "swine" H1N1 influenza A vaccine (121XP, X-179A and X-181) viruses as well as wild type A/California/07/2009(H1N1) and A/PR/8/34 strains were propagated in embryonated eggs and used for DNA/RNA Illumina HiSeq and MiSeq sequencing. The RNA sequences of these viruses published in NCBI were used as references for alignment of the sequencing reads generated in this study. Consensus sequences of these viruses differed from the NCBI-deposited sequences at several nucleotides. 121XP stock derived by reverse genetics was more heterogeneous than X-179A and X-181 stocks prepared by conventional reassortant technology. Passaged 121XP virus contained four non-synonymous mutations in the HA gene. One of these mutations (Lys226Glu) was located in the Ca antigenic site of HA (present in 18% of the population). Two non-synonymous mutations were present in HA of viruses derived from X-179A: Pro314Gln (18%) and Asn146Asp (78%). The latter mutation located in the Sa antigenic site was also detected at a low level (11%) in the wild-type A/California/07/2009(H1N1) virus, and was present as a complete substitution in X-181 viruses derived from X-179A virus. In the passaged X-181 viruses, two mutations emerged in HA: a silent mutation A1398G (31%) in one batch and G756T (Glu252Asp, 47%) in another batch. The latter mutation was located in the conservative region of the antigenic site Ca. The protocol for RNA sequencing was found to be robust, reproducible, and suitable for monitoring genetic consistency of influenza vaccine seed stocks.
Citation: Majid L, Zagorodnyaya T, Plant EP, Petrovskaya S, Bidzhieva B, Ye Z, et al. (2015) Deep Sequencing for Evaluation of Genetic Stability of Influenza A/California/07/2009 (H1N1) Vaccine Viruses. PLoS ONE 10(9): e0138650. https://doi.org/10.1371/journal.pone.0138650
Editor: Florian Krammer, Icahn School of Medicine at Mount Sinai, UNITED STATES
Received: March 24, 2015; Accepted: September 2, 2015; Published: September 25, 2015
This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication
Data Availability: All data underlying the findings in this study are freely available in this manuscript and its supplemental file. The sequencing data are publically available in the following links: http://dx.doi.org/10.7910/DVN/C85WNZ and https://hive.biochemistry.gwu.edu/dna.cgi?cmd=home&folder=/HIVE Space/California 2009 pandemic vaccine.
Funding: The authors have no support or funding to report.
Competing interests: The authors have declared that no competing interests exist.
Introduction
The influenza A/California/07/2009 (H1N1) virus emerged in Mexico, and was identified in the United States at the beginning of 2009. It caused outbreaks of respiratory illness in Mexico and in the United States, [1] and was shown to be genetically related to swine influenza viruses [2]. Subsequently, the virus spread to several other countries in North America, and eventually the outbreak reached pandemic proportions, which was recognized by the World Health Organization (WHO) in June of 2009 [3, 4].
The quick emergence and spread of A/California/07/2009 (H1N1) virus has been of great concern globally. Therefore, the development of A /California/07/2009 vaccine was an essential component of the global preventive strategy for the pandemic preparedness and response. WHO recommended that pandemic influenza A (H1N1) vaccines must contain A/California/7/2009 (H1N1)-like virus [5].
To date, several vaccines have been developed against A/California/07/2009 (H1N1) virus. Some vaccine viruses were prepared by classical reassortment (e.g. X-179A and X-181); others by reverse genetics (e.g. 121XP). They were manufactured as seasonal influenza vaccines manufactured i.e. after virus cultivation; the allantoic fluid of the egg is harvested and is then usually clarified by centrifugation to remove cellular debris. Then whole virus subjected to purification, concentration, and inactivation steps.—See more detail at: http://www.ifpma.org/resources/influenza-vaccines/influenza-vaccines/vaccine-manufacture-virus-processing.html#sthash.5GUYGsGs.dpuf.
The A/California/07/2009 (H1N1) strain is an influenza A virus, a single-stranded RNA virus from Orthomyxoviridae family [6]. The family contains four other species: influenza B virus, influenza C virus, thogotovirus and isavirus. The genome of influenza A virus consists of eight negative-sense RNA segments that encode at least 10 proteins. Influenza A virus genome is highly variable due to low fidelity of RNA polymerase and reassortment between co-infecting strains [7]. New virus mutants emerge continuously, allowing viruses to evade host immunity and cause cyclical annual outbreaks, and occasionally pandemics.
Spontaneous mutations easily emerge in RNA viruses, and the consequences of their accumulation must be identified to ensure the safety and potency of vaccines. The genetic diversity within a viral population is important, as RNA viruses in particular have high mutation rates, can evolve rapidly, and may exhibit adaptation to specific hosts and/or tissues [8]. Because of the genetic and antigenic plasticity, influenza vaccines have to be reformulated frequently to include antigens of the dominant circulating strains. Most live and inactivated influenza vaccines are produced by reassortment of the current strains with high-growth reference strains, to enable high yields in eggs or cell substrates used for vaccine production [9, 10]. Adaptation to growth in different cells can lead to changes in the viral receptor-binding region, as well as in protective epitopes, because they significantly overlap each other [11, 12]. Therefore, it is important to monitor genetic changes of viruses used in vaccine manufacture to ensure that their antigenic structures remain intact.
Most mutants are present in viral populations at relatively low levels, making them difficult to detect using conventional sequencing methods. A highly sensitive mutant analysis by PCR and restriction enzyme cleavage (MAPREC) [13, 14] and quantitative allele-specific PCR (asqPCR) [15] were developed for detection of a low level of mutants in viral vaccines. However, these methods are only suitable for analysis of one mutation at a time, and cannot be used to screen for unknown mutations.
Other approaches based on analysis of electrophoretic mobility in gels [16, 17] do not always allow precise location of mutations. Matrix-assisted laser desorption/ionization-time of flight (MALDI-TOF), mass spectrometry [18], and hybridization with microarrays of short oligonucleotides [19–21] are sensitive, but are relatively laborious and may require follow-up by direct sequencing.
Recently, several versions of high-throughput sequencing technology, known also as deep or massively parallel sequencing (MPS), enabled rapid generation of large amounts of sequence information [22]. These new technologies were shown to be suitable for analysis of heterogeneities in viral populations [23]. Previously, we have demonstrated that MPS can be used to monitor genetic stability of Oral Polio Vaccines (OPV) [24], and could replace the WHO-recommended MAPREC assay for lot release of OPV. In this study, Illumina sequencing technology was used for screening mutations in viruses of A/California/07/2009 (H1N1) vaccine strains passaged in eggs. We found that this approach could be used to identify and to accurately quantify small amounts of mutant viruses present in preparations of inactivated influenza vaccines. We also found that sequence heterogeneity of seed viruses depended on the methods used for their preparation. The ability to quantify potentially undesirable mutations in vaccine batches makes this method suitable for quality control to ensure manufacture of safe and effective vaccines.
Materials and Methods
Viruses and reference sequences
Viruses X-179A, X-181 and 121XP were passaged in eggs under similar conditions to those of the manufacturers [25, 26], and A/Puerto Rico/08/34 (H1N1) reference and A/California/07/2009 (H1N1) pandemic virus stocks (Table 1), obtained from the strain collection at the Center for Biologics Evaluation and Research, US Food and Drug Administration were grown in eggs. All these viruses (Table 1) were propagated in 10 days old embryonated hen’s eggs by inoculation of 0.2 ml of diluted virus stock containing about 104 pfu at 33°C. Allantoic fluid was harvested at 72 hours post infection and clarified by spinning at 4000 rpm for 10 min at 4°C. The clarified viruses were saved at -80°C.
Note: CBER: Center for Biologics Evaluation and Research at US Food and Drug Administration.
Amplification of the entire influenza virus genome
Total RNA was isolated from 140 μl of virus suspension using QIAamp Viral RNA Mini Kit (Qiagen, Valencia, CA) according to the manufacturer’s protocol. The RNA was eluted in a final volume of 60 μl of sterile RNase-free water. cDNA was synthesized using Superscript III Reverse Transcriptase (SSIII, Invitrogen, Carlsbad, CA) with Universal_F primer: 5’ GACTAATACGACTCACTATAGGGAGCAAAAGCAGG 3’. cDNA synthesis was performed in a reaction containing 10 μl of isolated viral RNA, 1.6 μM of Universal_F primer, 600 units of SSIII, 0.5 mM dNTPs, 50 mM DTT, and first strand buffer in a total volume of 50 μl. The reaction mixture was incubated at 55°C for two hours, followed by SSIII inactivation at 70°C for 20 min.
Two different PCR protocols were used to simultaneously amplify all genome segments of influenza virus.
Phusion protocol: Four microliters of viral cDNA was used as a template for PCR amplification in a total reaction volume of 50 μl, containing 0.2 μM of each universal primers: Universal_F (see above) and Universal_R: 5’GACATTTAGGTGACACTATAGAAGTAGAAACAAGG 3’, 200 μM of dNTP, 3 mM MgCl2, and 1 unit of Phusion DNA polymerase (New England BioLabs, Ipswich, MA). PCR cycling conditions were as follows: an initial denaturation at 98°C for 30 sec followed by 35 cycles of 98°C for 10 sec, 55°C for 30 sec and 72°C for 5 min, and the final elongation at 72°C for 10 min.
Extensor protocol: Four microliters of viral cDNA was used as a template for PCR amplification in a total reaction volume of 50 μl containing 0.4 μM of each universal primer (Universal_F and Universal_R), 500 μM of dNTP, 5 μl of 10x Extensor Buffer 2, and 2.5 units of Extensor PCR Enzyme Mix (5U/μl) (Dharmacon, Lafayette, CO). PCR cycling conditions were as follows: an initial denaturation at 94°C for 2 min followed by 35 cycles of 94°C for 10 sec, 60°C for 30 sec and 68°C for 5 min, and the final elongation at 68°C for 10 min.
The resulting eight DNA amplicons representing all segments of viral genome were separated by electrophoresis in 1% agarose gels with ethidium bromide (Lonza, Rockland, ME) and visualized using the Kodak Gel Logic 200 Imaging System and Kodak Molecular Imaging Software (Carestream Health, Inc. Rochester, NY).
Preparation of samples for Illumina sequencing
The PCR product was purified by QIAquick PCR Purification Kit (Qiagen, Valencia, CA) and fragmented by focused ultrasonicator (Covaris) to generate the optimal fragment sizes of 350–400 bp suitable for Illumina sequencing.
The DNA fragments were analyzed for DNA size by the 2100 Bioanalyzer system using a high-sensitivity kit (Agilent Technologies, Inc., Santa Clara, CA). Next, the fragmented DNAs were used for library preparation with NEBNext® DNA Sample Prep Reagent Set 1 (New England BioLabs, Ipswich, MA). The primers containing multiplexing index adapters (barcodes) to identify sequencing reads belonging to each sample were synthesized according to the specifications provided by Illumina Inc. for multiplex pair-end sequencing. Deep sequencing was performed at Macrogen (Seoul, Korea) using HiSeq2000 (Illumina) producing 101 bp-long paired-end reads, or at our laboratory using MiSeq (Illumina) producing 100 to 300 bp-long paired-end reads.
NEBNext mRNA Sample Prep Master Mix Set 1 (New England BioLabs, Ipswich, MA) was used for preparation of Illumina sequencing libraries from total RNA contained in the sample (whole-RNA library). Briefly, 0.5 μg total RNA (extracted as mentioned above) was used for fragmentation by focused ultrasonicator (Covaris) to generate the optimal fragment sizes of 200–300 bp suitable for RNA Illumina sequencing. The fragmented RNA was used for cDNA synthesis using ProtoScript II Reverse Transcriptase and random primers. The cDNA was further converted into double stranded cDNA with end-repair procedure (Klenow fragment, T4 polynucleotide kinase and T4 polymerase), and ligated to Illumina paired end (PE) adaptors. Size selection was performed using 2% agarose gel electrophoresis, generating cDNA libraries ranging from 275 to 325 bp in size. Finally, the libraries were expanded using 15 cycles of PCR with multiplex indexed primers and purified with magnetic beads using Agencourt Ampure Beads (Beckman Coulter).
Data analysis
Reference RNA sequences of X-179A, X-181, 121XP, A/California/07/2009 (H1N1), and A/PR/08/34 viruses were downloaded from GenBank. Accession numbers for segments 1 to 8 of X-179A were CY058516 to CY058523 respectively. The same sequence references were used for X-181 except for segments 2, 4, and 6 were replaced by GQ906800—GQ906802 respectively. For 121XP virus, the reference sequences for segments 1 to 8 were GQ454863, GQ454864, GQ454865, GQ280121, GQ454866, GQ280122, GQ454867, and GQ454868 respectively. As references for A/PR/8/34 viruses, we used the sequences downloaded from GenBank; for segments 1–7 the accession numbers were EF467818- EF467824 respectively and EF467817 for segment 8. For A/California/07/2009 (H1N1) the following NCBI sequences were used for segments 1 to 8: CY121687, CY121686, CY121685, CY121680, CY121683, CY121682, CY121681, and CY121684 respectively.
The RNA sequences were used as references for alignment of the viral sequence reads generated in this study. First, sequencing reads with low quality (phred) scores were removed from the data set. The remaining sequences aligned with reference influenza virus sequences using custom software that is implemented in a highly integrated virtual environment (HIVE) computer cluster (https://hive.biochemistry.gwu.edu/dna.cgi?cmd=main).
To eliminate the false mutations, which could be an artifact produced during PCR amplification or sequencing procedures, the bias mutations were identified by Shannon entropy as a measure of randomness of distribution of mutations along individual sequencing reads [27]. Generally, a bias of mutation is considered significant if its entropy value is lower than one sigma (≈0.67). Also, only sequencing reads that have more than 75% of their expected size were aligned to the reference sequences, to eliminate nonspecific reads.
Finally, aligned sequencing reads were used to compute single-nucleotide polymorphism (SNP) profiles, read depth and the entropy for the entire viral genomes.
Results
Robustness of simultaneous influenza A genome segment amplification
To amplify all segments in one reaction, the reverse transcriptase Superscript III was used for cDNA synthesis, and two different PCR amplification protocols were used for DNA amplification.
The agarose gel electrophoresis results showed that all eight segments were amplified efficiently from passaged A/California/07/2009 (H1N1) pandemic vaccine viruses (The used viruses are presented in Table 1) by using Phusion DNA polymerase (New England BioLabs, Ipswich, MA) (Fig 1A). In contrast, use of Thermo Scientific Extensor Mix resulted in nonspecific bands and poor amplification of some segments (Fig 1B). To determine which of the two PCR protocols ensured unbiased amplification of all segments of influenza virus genome, the amplified samples (Fig 1A and 1B) were subjected to Illumina HiSeq sequencing. The percentage of sequencing reads obtained for each of the segments is presented in Fig 1C, along with the whole-RNA library that was run by HiSeq that was prepared as described in section 2.3 of the materials and methods paragraph. This result demonstrated that adequate distribution of sequencing reads between segments were obtained from a DNA library prepared from samples amplified by Phusion DNA polymerase (DNA library), and by whole-RNA library.
Consistency of Illumina sequencing, and sequencing analysis of A/PR/8/34 and A/California/07/2009 strains
Center for Biologics Evaluation and Research (CBER) stock of A/PR/8/34 was kindly provided by Dr. Peter Palese at Mount. Sinai School of Medicine and was passaged in CBER to produce CBER stock, this virus was sequenced five times by MiSeq and HiSeq using whole-RNA library and DNA library approaches. The results of sequencing the RNA library showed high levels of reproducibility. There was a slight difference between MiSeq and HiSeq results only for mutations present at low levels (around 5%, Table 2). In contrast, the five repeats of DNA library sequencing were less consistent and more heterogeneous. (S1 Table).
Notes: MiSeq values that differ from HiSeq's values are presented in parentheses. These mutation percentages were calculated per comparison to their published sequences. (See accession numbers in Materials and Methods paragraph.) nt: Nucleotide, aa: Amino-acid.
Most of A/PR/8/34 segments contained several complete mutations in comparison to their published sequences, and most of these mutations were silent. This virus was less heterogeneous, as few low substitutions were observed along the genome.
Another A/PR/8/34 virus stock was provided by National Institute for Biological Standards and Control (NIBSC) and was propagated 13 times in eggs and then was grown (passage 14) separately in two different eggs, and the resulting stocks were designated as A/PR/8/34-1 and A/PR/8/34-3 (Table 1). Both of them were used for whole-RNA library preparation, followed by HiSeq sequencing. The results in Table 3 showed that these two viruses shared almost all mutations, and the percentages of mutations were slightly higher in the A/PR/8/34-1 than in the A/PR/8/34-3 virus. They were more heterogeneous than CBER A/PR/8/34 virus stock, and all three A/PR/8/34 viruses shared several complete mutations along their genomes (Tables 2 and 3). They also shared one low-level mutation C1094A (Thr350Lys) in the NP gene which was present in about 20% of the RNA sequences in all three A/PR/8/34 viruses.
Note: These mutation percentages were calculated per comparison of A/PR/8/34 strains 1 and 3 genomes with the corresponding published sequences. (See accession numbers in Materials and Methods paragraph.) nt: Nucleotide, aa: Amino-acid.
A/California/07/2009 (H1N1) strain was sequenced by HiSeq using the whole-RNA library approach, and the summary of the results is presented in Table 4. No complete substitutions were observed, and the highest percentage of mutants observed was 66% for mutation A57G (Glu18Gly) in the PA segment. No mutations were observed in the NA segment. Three low-level non-synonymous mutations were found in HA: A456G (Asn146Asp, 11%), A736G (Asp239Gly, 16%) and G739A (Arg240Gln, 16%). The Asn146Asp mutation is located in the Sa antigenic site.
Note: These mutation percentages were calculated per comparison of A/California/07/2009 (H1N1) genome with its published sequence. (See accession numbers in Materials and Methods paragraph.) aa: Amino acid, nt: Nucleotide.
3.3. Deep sequencing analysis of 121XP virus passaged in eggs
The 121XP virus was generated by the reverse genetics method followed by serial passages in chicken eggs; this virus contains the HA and NA genes from A/California/07/2009 (H1N1) and all other genes from A/PR/8/34 (H1N1) [26].
To evaluate its genetic stability, the 121XP virus was passaged ten times in eggs under manufacture conditions (Table 1). Mutations revealed by deep sequencing in HA and NA genes of the virus are summarized in Table 5. Several non-synonymous mutations emerged in NA and HA, including the mutations Asn16Ser and Lys226Glu that were present at low levels (36% and 18% of mutants, respectively). The latter mutation is located in the Ca antigenic site in HA. Mutation Lys136Asn, located close to the antigenic site Sa, emerged and accumulated to about 78% of the population.
Note: These mutation percentages were calculated per comparison with the corresponding expanded wild-type progenitor. (See accession numbers in Materials and Methods.) aa: Amino-acid, nt: Nucleotide.
Ten mutations were found in the NA gene, and eight of them led to amino acid changes. Interestingly, three of them were found in codon 88. Sequencing of other genomic segments revealed several heterogeneities in most of them, with some mutations leading to amino acid changes. There was a good agreement between results obtained by sequencing of the whole-RNA library by MiSeq and HiSeq, as well as the DNA library by HiSeq. For more details, see S2 to S5 Tables.
Deep sequencing analysis of X-179A viruses passaged in eggs
The X-179A virus is a reassortant prepared by New York Medical College (NYMC) using classical reassortant methodology from A/California/7/2009 (H1N1) virus and NYMC X-157 virus, with the HA, NA and PB1 genes originating from A/California/7/2009 (H1N1), and the other internal genes from A/PR/8/34 (H1N1) [26]. This virus was passaged several times and adapted to grow in eggs. Viruses derived by passaging of X-179A virus in chicken eggs under manufacture conditions (Table 1) were similarly analyzed by deep sequencing. The results of deep sequencing of HA and NA genes of these viruses are summarized in Table 6. In sample X-179A-M5, two non-synonymous mutations emerged in HA: Pro314Gln (17%) and Asn146Asp (78%). The latter mutation located in the Sa antigenic site was present as low level mutation in A/California/07/2009 (H1N1). The confirmation of the presence of this mutation by Sanger sequencing method is presented on Fig 2. Sample X-179A-M2 contained a low level (9%) of Lys328Thr mutation.
Notes: The percentages obtained from RNA library followed by HiSeq sequencing are presented without parentheses or brackets. The percentages obtained from RNA library followed by MiSeq sequencing are presented in parentheses. The percentages obtained from DNA library followed by HiSeq sequencing are presented in brackets. These mutation percentages were calculated per comparison with the corresponding expanded progenitor. (See accession numbers in Materials and Methods.) aa: Amino-acid, nt: Nucleotide.
One non-synonymous mutation, Ser95Ile, was present at a low level (7%) in NA of sample X179A-M3.
In general, samples derived from X-179A virus were heterogeneous and contained several complete substitutions compared to its published sequence.
HiSeq sequencing of DNA libraries prepared from X-179A viruses revealed more heterogeneity than HiSeq/MiSeq sequencing of RNA libraries prepared from the same samples. Mutations found in other genomic segments are presented in S5 to S7 Tables.
Deep sequencing analysis of X-181 viruses passaged in eggs
Similar to X-179A virus, X‐181 virus was developed by New York Medical College using conventional reassortment technology. It has a 5:3 gene constellation with three genes (PB1, HA and NA) coming from A/California/07/2009 (H1N1) and the remaining five genes from A/PR/8/34 [26]. The A/PR/8/34 genes were provided by reassorting NYMC X‐157 with A/California/07/2009 (H1N1). This virus is produced to grow well in eggs, and was passaged several times in eggs.
Viruses derived by passaging of the X-181 virus in chicken eggs under manufacture conditions (Table 1) were analyzed by deep sequencing. The results of deep sequencing of HA and NA genes of these viruses are summarized in Table 7. Two mutations were found in HA genes: A1398G (silent, 31%) in the X-181-M3 virus and G756T (Glu252Asp, 47%) in the X-181-M4 virus. The latter mutation is located in the conserved region of the antigenic site Ca. One non-synonymous mutation, Thr466Ala, was present at a low level (8%) in the NA gene of X-181-M2. Interestingly, the mutation Asn146Asp is located in the Sa antigenic site and exists as a complete substitution in X-181 virus which is derived from X-179A virus. As mentioned above, this mutation was observed in X-179A-M5 at a high level (78%) and at a low level (11%) in the wild-type A/California/07/2009 (H1N1) virus.
Notes: The percentages obtained from RNA library followed by HiSeq sequencing are presented without parentheses or brackets. The percentages obtained from RNA library followed by MiSeq sequencing are presented in parentheses. The percentages obtained from DNA library followed by HiSeq sequencing are presented in brackets. These mutation percentages were calculated per comparison with the corresponding expanded progenitor. (See accession numbers in Materials and Methods.) aa: Amino-acid, nt: Nucleotide.
Viruses derived by passaging of X-181 in chicken eggs were heterogeneous and contained several complete mutations in PB2, NP and NS segments (S8 to S10 Tables).
As for X-179A viruses, HiSeq sequencing of DNA libraries prepared from X-181 viruses revealed more heterogeneity than HiSeq/MiSeq sequencing of whole-RNA libraries prepared from the same samples. Mutations found in other genomic segments are presented in S8 to S10 Tables.
Discussion
Rapid emergence of mutations is an inherent property of RNA viruses, leading to genetic heterogeneity of viral populations [8]. This allows viruses to rapidly adapt to changing growth conditions by selecting from the preexisting variants with higher fitness [28]. For some vaccines, virus growth in cell cultures during vaccine manufacture leads to reversion of attenuated phenotype to virulence; therefore, their content must be monitored as a part of vaccine quality control [13]. Mutations emerging during virus growth may also alter the antigenic properties, and as a consequence affect protective potency of live and inactivated vaccines. Antigenic properties of influenza A are notoriously prone to rapid changes. Hemagglutinin (HA) and neuraminidase (NA) are the two most important influenza virus envelope proteins that elicit immune response, and therefore are under immunological pressure that leads to their continuous change. The antigenic drift reflects the accumulation of amino acid substitutions in the globular domain of HA, the principal target of neutralizing antibodies, as well as in NA [29, 30]. The primary function of HA is the attachment to host cell receptors and mediation of viral entry [31, 32]. The NA is involved in progeny virion release and spread of the virus, and also may have a role in HA-mediated membrane fusion [33, 34]. Adaptation of viruses to replication in new cell types (e.g. growth in cell substrates, such as embryonated eggs or mammalian cells used for vaccine manufacture) involves changes in HA. Since the receptor-binding region on the HA molecule overlaps with the major antigenic site, the adaptation may also be accompanied by changes in antigenic properties [33, 35, 36].
The larger number of eggs used during manufacture process would decrease the level (percentage) of each individual mutant, but could nevertheless increase the overall heterogeneity of virus population. Cumulatively, they could lead to a similar loss of antigenic properties as one mutation present at a higher level. Therefore, it may be important to monitor the changes that take place in major protective epitopes of the seed virus stocks to avoid losses in protective efficacy of the vaccine. Not all mutations that emerge in viruses are harmful to vaccine properties. Mutations that consistently accumulate have higher fitness and could increase virus yields and vaccine potency. Therefore, studying mutations emerging in vaccine strains could be used for both quality control, and for optimization of the genetic composition of vaccine strains.
Seed viruses used in influenza vaccine manufacture are produced either by reassortment technology that uses co-cultivation of a high-growth reference strain with the current virus isolate leading to exchange of genomic segments [37], or by reverse genetics based on transfection of susceptible cells with a combination of cDNA plasmids coding for the appropriate segments of the virus [38]. Both approaches have their strengths and weaknesses, and can potentially lead to the emergence of mutants. One of the purposes of this study was to determine which of them produces more homogeneous virus stocks with a lower number of mutants. We have used three pandemic influenza A/California/07/2009 (H1N1) virus vaccine strains and virus stocks produced from these viruses by passaging them on eggs. X-179A was prepared at the New York Medical College on eggs as a reassortant between A/California/07/2009 and A/PR/8/34. After reports of a low yield of production virus from X-179A a second reassortant, X-181, was made. X-181 contains the HA, NA and PB1 segments from A/California/07/2009. 121XP was made using reverse genetics followed by serial passages in eggs and only has the HA and NA segments from A/California/07/2009 virus [26].
To reveal sequence heterogeneity in viral stocks we have used a deep sequencing approach that was recently shown to be an effective tool for quantifying all mutants in viral populations [24]. Mutations that were identified by deep sequencing and were present at a significant level were also confirmed by the analysis of Sanger sequencing electrophoregrams that revealed sequence heterogeneities. Amplification of the entire genome of influenza viruses presents a challenge because it consists of eight genomic segments of different lengths and sequences. Previously, we described PCR conditions that allowed us to amplify all genomic segments of influenza A virus in one reaction [27]. In this work, the PCR amplification was optimized to amplify all influenza segments simultaneously, efficiently and equally, which resulted in an adequate distribution of sequencing reads between all segments (Fig 1A and 1C). We have also used whole-RNA libraries prepared without specific amplification of viral cDNA. Illumina deep-sequencing of viral cDNA libraries and whole-RNA libraries were used to determine quantitative profiles of mutations along the entire genome of viruses of influenza A/California/07/2009 (H1N1) vaccines.
The deep sequencing results revealed several heterogeneities in most genomic segments, and several mutations led to amino acid changes. No big differences were observed between the results obtained by sequencing of whole-RNA libraries using MiSeq and HiSeq, and HiSeq sequencing of DNA libraries (S2 to S10 Tables). Deep sequencing of whole-RNA libraries was found to be more reproducible than sequencing of DNA libraries (Table 2). But the HiSeq sequencing of DNA libraries were less consistent and produced more heterogeneous sequences (S1 Table). This may be due to errors introduced by PCR amplification and non-specific alignment of primers.
Deep sequencing of several stocks of high-growth reference strain A/PR/8/34 revealed also some mutations and sequence heterogeneities. The CBER master stock of this strain contained several complete mutations in comparison to the published sequence, and most of these mutations were silent. There were also low levels of additional mutations. This indicates that this A/PR/8/34 virus is genetically stable, and the use of this virus as backbone for development of influenza vaccine may improve the vaccine production.
Another A/PR/8/34 virus stock was passaged 13 times on eggs and was grown (passage 14) separately in two different embryonated chicken eggs, the harvested 2 viruses (designated as A/PR/8/34-1 and A/PR/8/34-3) were deep-sequenced using HiSeq protocol. These two viruses shared most sequence heterogeneities, but the percentages of mutants were slightly different (Table 3). Therefore, virus growth in eggs introduced some additional sequence heterogeneity, but caused no complete nucleotide substitutions.
The parental wild type A/California/07/2009 (H1N1) virus was sequenced by HiSeq using the whole RNA library (Table 4). Its consensus matched the published nucleotide sequence. There was some sequence heterogeneity observed at select positions, reaching up to 66% (Glu18Gly) of the viral population, this mutation is very rare and nothing known about its biological function.
The Deep sequencing of the X-179A passaged viruses revealed several mutations in HA and NA. Two non-synonymous mutations in HA (Pro314Gln present at 17% and Asn146Asp present at 78% of the population) were identified in X-179A-M5 virus. Interestingly, the latter is located in the antigenic site Sa; it was detected at a low level (11%) in A/California/07/2009 (H1N1) strain and was present as a complete substitution in the X-181 virus stock derived from the X-179A virus (Fig 2), suggesting that it may be involved in the adaptation of X-181 virus. Virus X-179A-M2 contained Lys328Thr mutation at a low level (9%). Viruses derived from X-179A were heterogeneous and contained some complete nucleotide substitutions in comparison to their published sequences in PB2, PB1, NP, and in NS segments (see S5 to S7 Tables).
The X-181 virus was developed from the X-179A seed lot by another round of reassortment, and also is subjected to several passages in eggs (Table 1). Deep sequencing results showed that the G756T (Glu252Asp, present at 47%) mutation emerged in HA of the X181-M4 virus, and it is located in the conserved region of the antigenic site Ca.
Unlike the two previous viruses, 121XP was developed by reverse genetics [26]. Deep sequencing of virus derived from it showed several non-synonymous mutations in NA and HA. Mutations leading to Asn16Ser and Lys226Glu amino acid substitutions emerged at relatively low levels (36% and 18% of mutants, respectively) in HA. The latter mutation was located in the Ca antigenic site in HA, next to codon 225 that was shown to participate in the modulation of HA receptor avidity/specificity [39]. In addition, mutation of residue 225 was found to enable H3 viruses to switch from avian to human specificity [40, 41]. Mutation Lys136Asn was present at a high level (78%). It is located close to the antigenic site Sa in close proximity within the sialic acid-binding pocket, suggesting its functional importance [42]. Interestingly, in the NA segment, three mutations occurred in codon 88, suggesting that this codon represents a hot spot. Also, the results of passages 10 showed that the overall number of mutations and sequence heterogeneities present in virus stocks derived by reverse genetics was higher than the number of mutations in virus stocks created by conventional reassortment procedure. This observation is counterintuitive, since reverse genetics is based on homogeneous cDNA plasmids coding individual viral segments. One possible explanation for increased sequence heterogeneity of cDNA-derived RNA viruses is that transcription of cDNA plasmids to rescue the virus followed by its passages to adapt to grow in chicken eggs led to higher sequence heterogeneity.
In conclusion, our deep sequencing approach based on RNA library preparation was shown to be effective and reproducible for detection of low quantities of mutants in the entire genome of influenza viruses. It revealed that the viruses derived from the three of pandemic A/Ca/07/2009 (H1N1) vaccine viruses have varying levels of sequence heterogeneities. It can be used for assessment of the quality of seed viruses, and for ensuring that they do not contain mutations that could affect immunogenic properties of the vaccine.
Supporting Information
S1 Table. Percentage of mutations (≥5%) emerged in the genome of A/PR/8/34 strain (CBER stock) obtained by sequencing five times its DNA library by HiSeq.
https://doi.org/10.1371/journal.pone.0138650.s001
(XLS)
S2 Table. Percentage of mutations (≥5%) emerged in 121XP-M4 virus obtained by using RNA library preparation followed by MiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s002
(XLS)
S3 Table. Percentage of mutations (≥5%) emerged in 121XP-M4 virus obtained by using RNA library preparation followed by HiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s003
(XLS)
S4 Table. Percentage of mutations (≥5%) emerged in 121XP-M4 virus obtained by using DNA library preparation followed by HiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s004
(XLS)
S5 Table. Percentage of mutations (≥5%) emerged in X-179A viruses obtained by using RNA library preparation followed by MiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s005
(XLS)
S6 Table. Percentage of mutations (≥5%) emerged in X-179A viruses generated by using RNA library preparation followed by HiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s006
(XLS)
S7 Table. Percentage of mutations (≥5%) emerged in X-179A viruses generated by using DNA library preparation followed by HiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s007
(XLS)
S8 Table. Percentage of mutations (≥5%) emerged in X-181 viruses generated by using RNA library preparation followed by MiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s008
(XLS)
S9 Table. Percentage of mutations (≥5%) emerged in X-181 viruses generated by using RNA library preparation followed by HiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s009
(XLS)
S10 Table. Percentage of mutations (≥5%) emerged in X-181 viruses generated by using DNA library preparation followed by HiSeq sequencing.
https://doi.org/10.1371/journal.pone.0138650.s010
(XLS)
Author Contributions
Conceived and designed the experiments: ML ZY KC. Performed the experiments: ML TZ EP SP BB. Analyzed the data: ML VS KC. Contributed reagents/materials/analysis tools: ML TZ SP KC. Wrote the paper: ML EP ZY KC.
References
- 1. Fraser C, Donnelly CA, Cauchemez S, Hanage WP, Van Kerkhove MD, Hollingsworth TD, et al. Pandemic potential of a strain of influenza A (H1N1): early findings. Science. 2009;324(5934):1557–61. Epub 2009/05/13. doi: 1176062 [pii] pmid:19433588.
- 2. Dawood FS, Jain S, Finelli L, Shaw MW, Lindstrom S, Garten RJ, et al. Emergence of a novel swine-origin influenza A (H1N1) virus in humans. N Engl J Med. 2009;360(25):2605–15. Epub 2009/05/09. doi: NEJMoa0903810 [pii] pmid:19423869.
- 3. Garten RJ, Davis CT, Russell CA, Shu B, Lindstrom S, Balish A, et al. Antigenic and genetic characteristics of swine-origin 2009 A(H1N1) influenza viruses circulating in humans. Science. 2009;325(5937):197–201. Epub 2009/05/26. doi: 1176225 [pii] pmid:19465683; PubMed Central PMCID: PMC3250984.
- 4. Monto AS, Black S, Plotkin SA, Orenstein WA. Response to the 2009 pandemic: effect on influenza control in wealthy and poor countries. Vaccine. 2011;29(38):6427–31. Epub 2011/07/19. doi: S0264-410X(11)01007-3 [pii] pmid:21763381.
- 5.
WHO. Characteristics of the emergent influenza A (H1N1) viruses and recommendations for vaccine development. 2009.
- 6.
Lamb RA, Krug MA. Orthomyxoviridae: the viruses and their replication. Knipe D. M. HPM, editor. Philadelphia: Lippincott Williams & Wilkins; 2007. 1487–531 p.
- 7. Steinhauer DA, Domingo E, Holland JJ. Lack of evidence for proofreading mechanisms associated with an RNA virus polymerase. Gene. 1992;122(2):281–8. Epub 1992/12/15. pmid:1336756.
- 8. Domingo E, Martinez-Salas E, Sobrino F, de la Torre JC, Portela A, Ortin J, et al. The quasispecies (extremely heterogeneous) nature of viral RNA genome populations: biological relevance—a review. Gene. 1985;40(1):1–8. Epub 1985/01/01. pmid:3912262.
- 9. Girard MP, Steele D, Chaignat CL, Kieny MP. A review of vaccine research and development: human enteric infections. Vaccine. 2006;24(15):2732–50. Epub 2006/02/18. doi: S0264-410X(05)01049-2 [pii] pmid:16483695.
- 10. McCarthy MW, Kockler DR. Trivalent intranasal influenza vaccine, live. Ann Pharmacother. 2004;38(12):2086–93. Epub 2004/10/21. doi: aph.1E191 [pii] pmid:15494383.
- 11. Raymond FL, Caton AJ, Cox NJ, Kendal AP, Brownlee GG. The antigenicity and evolution of influenza H1 haemagglutinin, from 1950–1957 and 1977–1983: two pathways from one gene. Virology. 1986;148(2):275–87. Epub 1986/01/30. pmid:3942036.
- 12. Wiley DC, Wilson IA, Skehel JJ. Structural identification of the antibody-binding sites of Hong Kong influenza haemagglutinin and their involvement in antigenic variation. Nature. 1981;289(5796):373–8. Epub 1981/01/29. pmid:6162101.
- 13. Chumakov KM. Molecular consistency monitoring of oral poliovirus vaccine and other live viral vaccines. Dev Biol Stand. 1999;100:67–74. Epub 2000/01/01. pmid:10616177.
- 14. Chumakov KM, Powers LB, Noonan KE, Roninson IB, Levenbook IS. Correlation between amount of virus with altered nucleotide sequence and the monkey test for acceptability of oral poliovirus vaccine. Proc Natl Acad Sci U S A. 1991;88(1):199–203. Epub 1991/01/01. pmid:1846038; PubMed Central PMCID: PMC50777.
- 15. Bidzhieva B, Laassri M, Chumakov K. Allele-specific PCR for quantitative analysis of mutants in live viral vaccines. J Virol Methods. 2014;201:86–92. Epub 2014/03/13. doi: S0166-0934(14)00055-X [pii] pmid:24607431.
- 16. Muyzer G, de Waal EC, Uitterlinden AG. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol. 1993;59(3):695–700. Epub 1993/03/01. pmid:7683183; PubMed Central PMCID: PMC202176.
- 17. Orita M, Iwahana H, Kanazawa H, Hayashi K, Sekiya T. Detection of polymorphisms of human DNA by gel electrophoresis as single-strand conformation polymorphisms. Proc Natl Acad Sci U S A. 1989;86(8):2766–70. Epub 1989/04/01. pmid:2565038; PubMed Central PMCID: PMC286999.
- 18. Amexis G, Oeth P, Abel K, Ivshina A, Pelloquin F, Cantor CR, et al. Quantitative mutant analysis of viral quasispecies by chip-based matrix-assisted laser desorption/ ionization time-of-flight mass spectrometry. Proc Natl Acad Sci U S A. 2001;98(21):12097–102. Epub 2001/10/11. doi: 211423298 [pii]. pmid:11593021; PubMed Central PMCID: PMC59774.
- 19. Cherkasova E, Laassri M, Chizhikov V, Korotkova E, Dragunsky E, Agol VI, et al. Microarray analysis of evolution of RNA viruses: evidence of circulation of virulent highly divergent vaccine-derived polioviruses. Proc Natl Acad Sci U S A. 2003;100(16):9398–403. Epub 2003/07/25. doi: 1633511100 [pii]. pmid:12878723; PubMed Central PMCID: PMC170930.
- 20.
Laassri M, Cherkasova E, Abu-Asab MS, Chumakov K. Microarray Techniques for Evaluation of Genetic Stability of Live Viral Vaccines. In: Garcia ML RV, editor. Viral Genomes—Molecular Structure, Diversity, Gene Expression Mechanisms and Host-Virus Interactions 2012. p. 181–94.
- 21. Laassri M, Chizhikov V, Mikheev M, Shchelkunov S, Chumakov K. Detection and discrimination of orthopoxviruses using microarrays of immobilized oligonucleotides. J Virol Methods. 2003;112(1–2):67–78. Epub 2003/09/03. doi: S0166093403001939 [pii]. pmid:12951214.
- 22. Rogers YH, Venter JC. Genomics: massively parallel sequencing. Nature. 2005;437(7057):326–7. Epub 2005/09/16. doi: 437326a [pii] pmid:16163333.
- 23. Victoria JG, Kapoor A, Li L, Blinkova O, Slikas B, Wang C, et al. Metagenomic analyses of viruses in stool samples from children with acute flaccid paralysis. J Virol. 2009;83(9):4642–51. Epub 2009/02/13. doi: JVI.02301-08 [pii] pmid:19211756; PubMed Central PMCID: PMC2668503.
- 24. Neverov A, Chumakov K. Massively parallel sequencing for monitoring genetic consistency and quality control of live viral vaccines. Proc Natl Acad Sci U S A. 2010;107(46):20063–8. Epub 2010/11/03. doi: 1012537107 [pii] pmid:21041640; PubMed Central PMCID: PMC2993378.
- 25. Jing X, Phy K, Li X, Ye Z. Increased hemagglutinin content in a reassortant 2009 pandemic H1N1 influenza virus with chimeric neuraminidase containing donor A/Puerto Rico/8/34 virus transmembrane and stalk domains. Vaccine. 2012;30(28):4144–52. Epub 2012/05/09. doi: S0264-410X(12)00619-6 [pii] pmid:22561313.
- 26. Robertson JS, Nicolson C, Harvey R, Johnson R, Major D, Guilfoyle K, et al. The development of vaccine viruses against pandemic A(H1N1) influenza. Vaccine. 2011;29(9):1836–43. Epub 2011/01/05. doi: S0264-410X(10)01817-7 [pii] pmid:21199698.
- 27. Bidzhieva B, Zagorodnyaya T, Karagiannis K, Simonyan V, Laassri M, Chumakov K. Deep sequencing approach for genetic stability evaluation of influenza A viruses. J Virol Methods. 2014;199:68–75. Epub 2014/01/11. doi: S0166-0934(13)00520-X [pii] pmid:24406624.
- 28. Domingo E, Martin V, Perales C, Grande-Perez A, Garcia-Arriaza J, Arias A. Viruses as quasispecies: biological implications. Curr Top Microbiol Immunol. 2006;299:51–82. Epub 2006/03/30. pmid:16568896.
- 29. Das SR, Hensley SE, Ince WL, Brooke CB, Subba A, Delboy MG, et al. Defining influenza A virus hemagglutinin antigenic drift by sequential monoclonal antibody selection. Cell Host Microbe. 2013;13(3):314–23. Epub 2013/03/19. doi: S1931-3128(13)00074-7 [pii] pmid:23498956; PubMed Central PMCID: PMC3747226.
- 30.
Webster RG, Laver W.G., and Kilbourne E.D. Antigenic variation of influenza viruses. In The Influenza Viruses and Influenza. New York: Academic Press. 1975:269–314.
- 31. Steinhauer DA, Skehel JJ. Genetics of influenza viruses. Annu Rev Genet. 2002;36:305–32. Epub 2002/11/14. 052402.152757 [pii]. pmid:12429695.
- 32. Zambon M, Hayden FG. Position statement: global neuraminidase inhibitor susceptibility network. Antiviral Res. 2001;49(3):147–56. Epub 2001/06/29. doi: S0166354201001243 [pii]. pmid:11428241.
- 33. Hughes MT, McGregor M, Suzuki T, Suzuki Y, Kawaoka Y. Adaptation of influenza A viruses to cells expressing low levels of sialic acid leads to loss of neuraminidase activity. J Virol. 2001;75(8):3766–70. Epub 2001/03/27. pmid:11264365; PubMed Central PMCID: PMC114867.
- 34. Kobasa D, Wells K, Kawaoka Y. Amino acids responsible for the absolute sialidase activity of the influenza A virus neuraminidase: relationship to growth in the duck intestine. J Virol. 2001;75(23):11773–80. Epub 2001/11/02. pmid:11689658; PubMed Central PMCID: PMC114763.
- 35. Gambaryan AS, Marinina VP, Tuzikov AB, Bovin NV, Rudneva IA, Sinitsyn BV, et al. Effects of host-dependent glycosylation of hemagglutinin on receptor-binding properties on H1N1 human influenza A virus grown in MDCK cells and in embryonated eggs. Virology. 1998;247(2):170–7. Epub 1998/08/26. doi: S0042682298992248 [pii]. pmid:9705910.
- 36. Schild GC, Oxford JS, de Jong JC, Webster RG. Evidence for host-cell selection of influenza virus antigenic variants. Nature. 1983;303(5919):706–9. Epub 1983/06/23. pmid:6190093.
- 37. Maassab HF, DeBorde DC. Development and characterization of cold-adapted viruses for use as live virus vaccines. Vaccine. 1985;3(5):355–69. Epub 1985/12/01. doi: 0264-410X(85)90124-0 [pii]. pmid:3909681.
- 38. Hoffmann E, Neumann G, Kawaoka Y, Hobom G, Webster RG. A DNA transfection system for generation of influenza A virus from eight plasmids. Proc Natl Acad Sci U S A. 2000;97(11):6108–13. Epub 2000/05/10. 100133697 [pii]. pmid:10801978; PubMed Central PMCID: PMC18566.
- 39. Stevens J, Blixt O, Tumpey TM, Taubenberger JK, Paulson JC, Wilson IA. Structure and receptor specificity of the hemagglutinin from an H5N1 influenza virus. Science. 2006;312(5772):404–10. Epub 2006/03/18. doi: 1124513 [pii] pmid:16543414.
- 40. Connor RJ, Kawaoka Y, Webster RG, Paulson JC. Receptor specificity in human, avian, and equine H2 and H3 influenza virus isolates. Virology. 1994;205(1):17–23. Epub 1994/11/15. doi: S0042-6822(84)71615-1 [pii] pmid:7975212.
- 41. Matrosovich M, Tuzikov A, Bovin N, Gambaryan A, Klimov A, Castrucci MR, et al. Early alterations of the receptor-binding properties of H1, H2, and H3 avian influenza virus hemagglutinins after their introduction into mammals. J Virol. 2000;74(18):8502–12. Epub 2000/08/23. pmid:10954551; PubMed Central PMCID: PMC116362.
- 42. Varghese JN, McKimm-Breschkin JL, Caldwell JB, Kortt AA, Colman PM. The structure of the complex between influenza virus neuraminidase and sialic acid, the viral receptor. Proteins. 1992;14(3):327–32. Epub 1992/11/01. pmid:1438172.