High Yield (HY) Reassortants Bioinformatics

When developing influenza vaccine candidate viruses, whether by classical reassortment or reverse genetics techniques, it is desirable to identify and select a set of donor genes that, when rescued together with a known wild type (wt) HA-NA gene pair, predictably confer a high yield phenotype for HA protein production.
Doris Bucher.s group at the New York Medical College have shown that classical reassortment followed by selection of high yield wt HA-NA vaccine candidates often results in 5:3 and occasionally 4:4 reassortants, as well as the expected 6:2 reassortants (Fulvini et al.). Work presented recently by Dr. Ruben Donis at the CDC showed that the yield and growth capacity of influenza A/PR-8:wt 6:2 reassortants varied significantly depending on both the six internal PR8 gene set and the wt HA-NA genes. Together, these results suggest that it may be possible to use informatics approaches to identify a collection of donor gene sets that will predictably confer high yield on related HA-NA gene pairs.
This is proposal seeks to identify useful donor gene sets that will predictably confer high yield on related HA-NA gene pairs. Our aims are 1) sequence analyses of CDC and NYMC vaccine candidate viruses, 2) comparison of HA protein (or analogous) yield and 3) preparation and evaluation of test reassortants.
This project builds on a foundation of analysis previously performed by Dr. Ruben Donis at CDC and Dr. Andrew Fulvini and Dr. Doris Bucher at NYMC. In both of their studies, HA yield for a wt HA-NA gene pair vaccine seed candidate was impacted by the constellation of internal gene segments. In the initial NYMC analysis, the origin of internal genes were identified (either from wt virus that contributed HA-NA gene pair or from PR8 donor) for 57 different high yield (HY) reassortants, and the change in HI titer between the high yield reassortant and wt viruses was measured. In the CDC analysis, the HA yield of seven different wt HA-NA gene pairs was measured when rescued in three different PR8 donor gene constellations using reverse genetics. We plan to examine the existing genome segment sequence data from high yield reassortants, primarily generated through the influenza genome sequencing project, in order to identify statistically significant sequence features that when taken as a group correlate to a high yield phenotype. Identifying correlations between wt HA-NA gene pair sequences and the .internal gene segments., will allow vaccine seed stock producers to select the best HY donor internal gene constellations for a particular HA-NA gene pair prior to generating the vaccine seed candidate by either reverse genetics or classical reassortment approaches. This would be a major advance, as currently trial and error along with mutation and biological selection are used to select the high yield candidates, which periodically results in vaccine seeds that produce lower yields than required for optimal vaccine production. 

Funding