Recipient 2021: Moïra Dion
A pipeline to improve phage host predictions using CRISPR spacers
Moïra B. Dion1, E. Zufferey1, P.-L. Plante2, S.A. Shah3, L. Deng4, J.L.C. Mejia4, M.-A. Petit5, D.S. Nielsen4, H. Bisgaard3, J. Corbeil2, S. Moineau1*,
1Département de biochimie, de microbiologie et de bio-informatique, Université Laval; 2Big Data Research Center, Université Laval, Québec, Canada; 3Copenhagen Prospective Studies on Asthma in Childhood, University of Copenhagen; 4Department of Food Science, University of Copenhagen, Denmark; 5INRA, Jouy-en-Josas, France
Phages are viruses that infect bacteria. Most of the newly discovered phages from viral metagenomics have no match in reference databases: this is the viral dark matter obstacle. To begin unraveling the impact of phages on their ecosystem, it is essential to improve phage-host predictions, even with unknown phages, using bioinformatics approaches. CRISPR spacers in the bacterial chromosome are useful for host predictions because they are molecular archives of past phage infections. Indeed, many bacteria use their CRISPR spacers, together with Cas proteins, to recognize invading phages and block subsequent infections. Since spacer sequences are identical to a short target sequence found in the phage genome, they can be used to link a phage to its host. The performance of CRISPR spacers-based host predictions largely depend on the extensiveness of the database against which phages are compared. We implemented a phage host prediction pipeline by building a database of more than 11 million CRISPR spacers. Using phages infecting known bacterial hosts, we identified biologically meaningful criteria that optimize the recall and precision of the predictions, reaching more than 80% accuracy at the bacterial family level. We built a web platform to explore the CRISPR spacers database and developed a command line tool to perform host predictions. This tool allowed us to uncover phages that infect bacteria with previously no known phages in the human gut virome.
Recipient 2020: Dr Ximena Zottig
Self-assembled peptide nanorod vaccine confers protection against influenza A virus
Ximena Zottig a,b,c,d, Soultan Al-Halifa a,b,c,d, Mélanie Côté-Cyr a,b,c,d, Cynthia Calzas e, Ronan Le Goffic e, Christophe Chevalier e, DenisArchambault c,d and SteveBourgault a,b,d
aChemistry Department, Université du Québec à Montréal, Montreal, Canada, bQuebec Network for Research on Protein Function, Engineering and Applications (PROTEO), Quebec, Canada, cDepartment of Biological Sciences, Université du Québec à Montréal, Montreal, Canada, dThe Swine and Poultry Infectious Diseases Research Centre (CRIPA), Sainte-Hyacinthe, Canada, eUR892 VIM, Equipe Virus Influenza, Université Paris-Saclay, INRAE, Jouy-en-Josas, France
Proteinaceous nanostructures have emerged as a promising strategy to develop safe and efficient subunit vaccines. The ability of synthetic β-sheet self-assembling peptides to stabilize antigenic determinants and to potentiate the epitope-specific immune responses have highlighted their potential as an immunostimulating platform for antigen delivery. Nonetheless, the intrinsic polymorphism of the resulting cross-β fibrils, their length in the microscale and their close structural similarity with pathological amyloids could limit their usage in vaccinology. In this study, we harnessed electrostatic capping motifs to control the self-assembly of a chimeric peptide comprising a 10-mer β-sheet sequence and a highly conserved epitope derived from the influenza A virus (M2e). Self-assembly led to the formation of 100–200 nm long uniform nanorods (NRs) displaying the M2e epitope on their surface. These cross-β assemblies differed from prototypical amyloid fibrils owing to low polydispersity, short length, non-binding to thioflavin T and Congo Red dyes, and incapacity to seed homologous amyloid assembly. M2e-NRs were efficiently uptaken by antigen presenting cells and the cross-β quaternary architecture activated the Toll-like receptor 2 and stimulated dendritic cells. Mice subcutaneous immunization revealed a robust M2e-specific IgG response, which was dependent on self-assembly into NRs. Upon intranasal immunization in combination with the polymeric adjuvant montanide gel, M2e-NRs conferred complete protection with absence of clinical signs against a lethal experimental infection with the H1N1 influenza A virus. These findings indicate that by acting as an immunostimulator and delivery system, synthetic peptide-based NRs constitute a versatile self-adjuvanted nanoplatform for the delivery of subunit vaccines.
