Development of Lentiviral Vectors Pseudotyped With Influenza B Hemagglutinins: Application in Vaccine Immunogenicity, mAb Potency, and Sero-Surveillance Studies

Abstract

Influenza B viruses (IBV) cause respiratory disease epidemics in humans and are therefore components of seasonal influenza vaccines. Serological methods are employed to evaluate vaccine immunogenicity prior to licensure. However, classical methods to assess influenza vaccine immunogenicity such as the hemagglutination inhibition assay (HI) and the serial radial hemolysis assay (SRH), have been proven to have many limitations. As such, there is a need to develop innovative methods that can improve on these traditional assays and provide advantages such as ease of production and access, safety, reproducibility, and specificity. It has been previously demonstrated that the use of replication-defective viruses, such as lentiviral vectors pseudotyped with influenza A hemagglutinins in microneutralization assays (pMN) is a safe and sensitive alternative to study antibody responses elicited by natural influenza infection or vaccination. Consequently, we have produced Influenza B hemagglutinin-pseudotypes (IBV PV) using plasmid-directed transfection. To activate influenza B hemagglutinin, we have explored the use of proteases in increasing PV titers via their co-transfection during pseudotype virus production. When tested for their ability to transduce target cells, the influenza B pseudotypes produced exhibit tropism for different cell lines. The pseudotypes were evaluated as alternatives to live virus in microneutralization assays using reference sera standards, mouse and human sera collected during vaccine immunogenicity studies, surveillance sera from seals, and monoclonal antibodies (mAbs) against IBV. The influenza B pseudotype pMN was found to effectively detect neutralizing and cross-reactive responses in all assays and shows promise as an effective and versatile tool in influenza research.

Trial registration: ClinicalTrials.gov NCT00942071.

Keywords: influenza B; lentiviral (LV) vector; pseudotype neutralization; pseudotype viral particles; serosurveillance.

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Copyright © 2021 Ferrara, Del Rosario, da Costa, Kinsley, Scott, Fereidouni, Thompson, Kellam, Gilbert, Carnell and Temperton.

Figures

Figure 1

Phylogenetic tree of influenza B HA used for pseudotype production. Phylogenetic tree shows influenza B diverging into two distinct lineages, B/Yamagata (blue) and B/Victoria (red). Accession numbers are reported with the strain name on the tree tips. Nodes are shown at the ends of branches which represent sequences or hypothetical sequences at various points in evolutionary history. Branch lengths indicate the extent of genetic change. The tree generated was constructed with PhyML (48) and graphically elaborated with Archaeopteryx.js (https://sites.google.com/site/cmzmasek/home/software/archaeopteryx-js).

Figure 2

Production of influenza B pseudotypes employing different proteases. Various quantities of proteases were co-transfected with gag-pol packaging construct, firefly luciferase and influenza B HA-encoding plasmids in pI.18 (A) or pEVAC (B). (A) For transfection employing pI.18 HA plasmids, HAT, TMPRSS2 and TMPRSS4 were used in 250 and 125 ng quantities. Pseudotypes produced in the absence of protease (Δprotease) and those activated post-transfection and pre-titration with TPCK-Trypsin are also included. Plates were read using the Bright-Glo™ assay system (Promega) and GloMax Multi detection system luminometer (Promega). (B) For transfection employing pEVAC plasmids, only HAT and TMPRSS4 proteases in the range of 2.5 ng-10 ng were used. Titration plates were then read using the GloMax® Navigator (ProMega) using the Promega GloMax® Luminescence Quick-Read protocol. For both (A, B), pseudotype virus titers are expressed in relative luminescence units/mL (RLU/mL). Significant differences between titers obtained for each PV grouped into protease used and Δprotease are shown via brackets (*p=<0.05). Error bars show mean and standard deviation of four replicates.

Figure 3

Influenza B pseudotypes are expressed in different cell lines. (A) Titers of influenza B pseudotypes expressing firefly luciferase in HEK293T/17, MDCK, and A549 cells are shown in RLU/mL as the mean and standard deviation of 4 replicates. A ΔHA control representing non-HA pseudotyped virus was included for comparisons. No significant difference between PV titers and ΔHA is indicated by brackets (n.s. p>0.05). (B) Influenza B lentiviral pseudotype B/Florida/4/2006 containing an emerald GFP reporter is shown to successfully transduce HEK293T/17, MDCK and A549 cells. Scale bar = 6 µm.

Figure 4

Neutralization of influenza B pseudotypes by reference antisera and mouse sera from influenza vaccination studies. (A) Neutralization of pseudotypes from pI.18, B/Brisbane/60/2008, B/Florida/4/2006, and B/Hong Kong/8/1973 against reference antisera anti-B/Brisbane/60/2008 (NIBSC 11/36) was measured by a luciferase reporter assay. Reference sera was serially diluted two-fold from a starting dilution of 1:100. 1.0x106 RLU of IBV PV was then added to each well. For all plots, each point represents the mean and standard error of four replicates per dilution. (B)In vitro neutralization of influenza B pseudotypes from pEVAC, B/Phuket/3073/2013 and B/Washington/2/2019 by mouse sera vaccinated with Influenza B HA from B/Singapore/222/1972, B/Colorado/6/2017, B/Phuket/3073/2013, and PBS (negative control). Neutralization was determined via a luciferase-reporter assay and given as IC50 dilution values (In this case, IC50 is half maximal inhibitory serum dilution). Significant difference between the means of the IC50 values of the groups vaccinated with IBV HA against the PBS negative control group are shown via brackets (*p<0.05). Interleaved box plots show minimum, maximum, and mean values. Error bars show standard deviation of replicates. For mice vaccinated with B/Singapore/222/1972 HA, n = 5 mice/group; n = 6 mice/group for mice vaccinated with B/Colorado/6/2017 HA, B/Phuket/3073/2013 HA, and PBS.

Figure 5

Evaluation of IBV neutralization capacity of NCT0942071 clinical trial sera using influenza B pseudotype microneutralization assay (pMN). The neutralization capacity of human sera from clinical trial (NCT0942071) was assessed pre- and post-vaccination with Trivalent Influenza Vaccine (TIV) and then either a placebo saline solution or an MVA-NP fusion protein (n=17). All sera were serially diluted and incubated with each PV: (i) B/Brisbane/60/2008, (ii) B/Hong Kong/8/1973, and (iii) B/Florida/4/2006 PV before addition of HEK293T/17 cells. Residual luciferase activity was determined and % neutralization was calculated relative to the PV only control. From this, IC50 values were calculated as the reciprocal of the dilution required to neutralize 50% of initial input PV. Significant differences between the means of pre- and post-vaccination sera IC50 when tested against PV are indicated in brackets (*p(A). After data stratification on the basis of post-TIV vaccine regimens; TIV + placebo (n=8), TIV+MVA+NP+M1 (n=9), the fold-increase from pre-vaccination IC50 titers were compared to post-vaccination IC50(B). Data is compared using Tukey plots showing the median of the 25th and 75th percentile with outliers plotted as those which exceeded 1.5x quartile range. Virus pseudotype from B/Victoria lineage is indicated in red and B/Yamagata lineage in blue.

Figure 6

Neutralization of influenza B strains by lineage-specific monoclonal antibodies. Neutralization of influenza B pseudotypes B/Phuket/3073/2013 and B/Brisbane/60/2008 by monoclonal antibodies 3E8, 2F11, and 5A1, was measured by a luciferase reporter assay. Monoclonal antibodies were serially diluted two-fold from a starting concentration ranging from 2000 ng/mL to 0.0625 ng/mL against all pseudotypes tested. Influenza B pseudotypes (1.0x106 RLU) were then added to each well. For all plots, each point represents the mean and standard error of two replicates per dilution.