High-throughput droplet-based analysis of influenza A virus reassortment by single-virus RNA sequencing, for pandemic risk assessment and Candidate Vaccine Viruses optimisation – microFLU-REASSORT

Virology and cell biology methods, including production of viral stocks and labelling and sorting of cells. Molecular biology methods for the production of viral transcripts, synthesis of beads carrying barcoded primers, and generation of libraries for Illumina sequencing. High-throughput droplet-based microfluidic methods.

The main result of the microFlu-REASSORT project to date is i) the establishment of an experimental and bioinformatic «pipeline« to generate and process data from single-cell sequencing; and ii) the demonstration of the feasibility of the detection of the eight segments of the influenza genome associated within each drop, in a specific manner and with almost no cross-assignments between segments derived from the two viral strains. A reassortment experiment between two human seasonal influenza viruses H1N1 and H3N2 will soon be launched and will complete the proof-of-concept.

The experimental and bioinformatics pipeline will be used to quantify experimentally and in a statistically robust manner the result of genetic reassortment involving swine and/or avian IAVs considered to be of concern regarding pandemic risk, in order to add a predictive element to the pandemic risk assessment tools used by WHO and CDC and to optimize Candidate Vaccine Viruses (CVVs).

no publications or patents so far

Influenza A viruses (IAVs) are human respiratory pathogens that represent a major worldwide public health threat due to their ease of transmission, ability to cause severe respiratory symptoms and persistence in animal reservoirs. They recur every year as “seasonal influenza”, but antigenically novel and potentially dangerous, pandemic viruses are also generated episodically through genetic reassortment between animal and human IAVs: when a cell is co-infected with two or more IAVs, the 8 genomic RNA segments from the parental viruses can reassort and lead to progeny viruses with unique and unpredictable phenotypes. Although the theoretical number of genotypes from reassortment between two strains is 256 (2e8), the full panel has never been observed in vitro or in vivo, and certain genes tend to co-segregate, suggesting that genetic reassortment is biased. However, with available virology techniques, reassortment cannot be studied comprehensively or quantitatively, and it is currently impossible to predict the outcome of reassortment events and the emergence of new strains.

The aim of µFlu-REASSORT is to implement an innovative droplet-based microfluidics approach in order to provide, in an experimental setting of co-infection with two IAVs, a statistically relevant ranking of the probability of reassortant viruses to emerge, and to add a predictive component to the existing pandemic risk assessment tools of the WHO and CDC.
The two first goals of our project are (i) to develop and (ii) validate a high-throughput droplet-based single-virus RNA sequencing system (the µFlu system) to analyze a large number of reassortant IAVs in BSL2 and BSL3 confinement. We aim to adapt a proven droplet-based microfluidics system for high-throughput single-cell RNA-seq (scRNA-seq) to achieve massively parallelized barcoded next-generation sequencing (NGS) of > 10e5 individual viral genotypes resulting from genetic reassortment upon co-infection with two IAVs.
The third goal is to use the µFlu system to assess the likelihood of emergence of potentially zoonotic/pandemic IAVs upon genetic reassortment between a human IAV and avian-origin H9N2 or H6N6 viruses that circulate in swine in Asia and, in the case of the H9N2, have caused sporadic cases in human. The potential of these viruses to reassort with a human virus, followed by the acquisition of better transmissibility and/or fitness in humans raises many concerns. Systematic evaluation of their potential for reassortment will contribute to the selection of appropriate candidate vaccine viruses (CVVs).
Finally, the µFlu system will also be used to improve the yield and antigen content of CVVs produced by reverse genetics against a HxNy virus of interest by investigating reassortment between the PR8 vaccine and the HxNy strain and identifying an optimal combination of gene segments: the H1N1pdm9 virus A/California/04/2009, an early isolate from the 2009 pandemic, will be used as a model HxNy virus.

The success of the µFlu-REASSORT project relies on the complementary expertise of two teams: (i) N. Naffakh (coordinator, Institut Pasteur, Paris) with strong expertise in IAV biology, molecular virology and virus engineering and solid connections in the areas of human and animal IAV surveillance and (ii) A. Griffiths (ESPCI Paris) with extensive experience of sc-phenotyping and sorting using droplet-based microfluidics, which are now being combined with scRNA-seq. The recent, non ANR-funded, addition of the expertise of H. Isambert (Institut Curie, Paris) to the consortium will integrate surveillance and µFlu-generated data into new machine-learning algorithms to predict the likelihood of genetic reassortment.
The µFlu system should prove valuable to study reassortment in other segmented viruses and will foster a strong and unique competitive research network centered around high-throughput microfluidics for scRNA-seq of prokaryotic and eukaryotic pathogens requiring BSL2/3 confinement.