Universal influenza virus vaccines and therapeutics: where do we stand with influenza B virus?

Abstract

The development of a broadly protective or universal influenza virus vaccine is currently a public health priority worldwide. The vast majority of these efforts is exclusively focused on influenza A viruses. While influenza A viruses cause the majority of all influenza cases worldwide, influenza B viruses should not be ignored. Approximately 25% of all influenza cases are caused by influenza B viruses which circulate as two distinct B/Victoria/2/87-like and B/Yamagata/16/88-like lineages. In contrast to popular belief, influenza B cases frequently cause significant morbidity and mortality, especially in children. Similar to influenza A viruses, influenza B viruses drift antigenically and the influenza B components of current vaccines have to be reformulated almost on an annual basis. A broadly protective vaccine against influenza B viruses is therefore urgently needed. Here we review both broadly protective anti-influenza B antibodies as well as the sparse attempts to create a universal influenza B virus vaccine.

Figure 1. Influenza-positive clinical specimens by influenza virus type

Panel A shows seasonal surveillance data from public health and clinical laboratories reported to the United States Centers for Disease Control and Prevention (CDC) from the 1997–1998 season to the current 2017–2018 season with preliminary data up to and including week 6, ending February 10, 2018 (adapted from https://www.cdc.gov/flu/weekly/pastreports.htm). Panel B represents seasonal surveillance data from the World Health Organization (WHO) European Region from sentinel and non-sentinel specimen sources reported to the European Centre for Disease Prevention and Control (ECDC) from the 2014–2015 season to the current 2017–2018 season with preliminary data up to and including week 6, ending February 11, 2018 (adapted from http://flunewseurope.org/Archives). Panel C includes annual surveillance data submitted to the Global Influenza Surveillance and Response System (GISRS) and FluNet from 1999 (data from three countries) to 2018 (data from 109 countries) with preliminary data up to and including data submitted by February 19, 2018 (adapted from http://apps.who.int/flumart/Default?ReportNo=12).

Figure 2. Phylogenetic trees representing HA diversity of all influenza A subtypes and human H3 HAs compared to influenza B HAs

Panel A represents the diversity of the HA and phylogenetic footprint of influenza A virus HAs. Panels B and C show the same for H3 HAs and influenza B HAs sequences for human isolates over time, respectively. The ancestral B/Lee/1940 strain, and antigenically distinct B/Victoria/2/870-like and B/Yamagata/16/88-like lineages are indicated in panel C. Scale bars have been normalized across all three trees and represent 1% change in amino acid sequence. Sequences were obtained on FluDB or GISAID, aligned in Clustal Omega following sequence cleanup, aligned in FigTree, and compiled in Adobe Illustrator.

Figure 3. Reported antibody binding epitopes or footprints on influenza B HA

Antibody binding epitopes for CR8033 (yellow), CR8071 (orange), CR9114 (green), 5A7 (purple), C12G6 (red), and 46B8 (blue) are highlighted on a surface representation of the influenza B/Brisbane/60/2008 virus HA (PDB: 4FQM - [26]) using PyMOL. The globular head domain of the HA is highlighted in dark grey relative to the stalk domain which is shown in light grey. CR8033 and C12G6 show a similar binding footprint in proximity to the receptor binding site, while 46B8 and CR8071 show a similar footprint on the vestigial esterase domain.