Allele-specific PCR for quantitative analysis of mutants in live viral vaccines

Quantitative RT-PCR Assays for Quantification of Undesirable Mutants in the Novel Type 2 Oral Poliovirus Vaccine

Emergence of mutations is an inherent property of RNA viruses with several implications for their replication, pathogenesis, and evolutionary adaptation. Oral poliovirus vaccine (OPV), developed by Albert Sabin, is composed of live attenuated polioviruses of three serotypes that can revert to neurovirulence during replication in cell culture and in vaccine recipients. Recently, a new modified variant of Sabin 2 virus was developed by introducing changes in its genome, making it more genetically stable to prevent the reversion. The new strain was used to manufacture novel OPV2 (nOPV2), which was approved by the World Health Organization for emergency use to stop outbreaks caused by circulating vaccine-derived poliovirus (cVDPV2). Manufacture of this improved vaccine requires close attention to the genetic heterogenicity to ensure that the levels of the undesirable mutations are limited. Preliminary studies using whole-genome Illumina sequencing (NGS) identified several genomic sites where mutations tend to occur with regularity. They include VP1-I₁₄₃T amino acid change at the secondary attenuation site; VP1-N₁₇₁D, a substitution that modestly increases neurovirulence in mice; and VP1-E₂₉₅K, which may reduce the immunogenicity of the nOPV2. Therefore, to ensure the molecular consistency of vaccine batches, the content of these mutants must be quantified and kept within specifications. To do this, we have developed quantitative, multiplex, one-step reverse-transcriptase polymerase chain reactions (qmosRT-PCRs) as simple methods for quantification of these mutations. Each method uses specific short TaqMan probes with different dyes for the analysis of both mutants and non-mutants in the same sample. The quantification is done using calibration curves developed using validated reference materials. To evaluate the sensitivity and the linearity of the qmosRT-PCR method, the mutant viruses were spiked in non-mutant viruses, and nOPV2 batches were used to validate the method. The spiked samples and the nOPV2 batches were analyzed by qmosRT-PCR and NGS assays. The results showed that qmosRT-PCR is sensitive enough to detect around 1% of mutants. The percentages of mutants determined by qmosRT-PCR correlate well with the results of the NGS. Further, the analysis of the nOPV2 batches showed that the results of qmosRT-PCR correlated well with the results of NGS. In conclusion, the qmosRT-PCR is a specific, sensitive, and linear method. It could be used for quality control of the nOPV2 batches.

Deep Sequencing for Evaluation of Genetic Stability of Influenza A/California/07/2009 (H1N1) Vaccine Viruses

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.