Fig. 6: Virus escape generation and in vivo protection with co-administered mAbs.
a, DMS library-based selection of mAb escape. Lentivirus-based pseudovirus expressing H5 DMS HA library was used to enrich mAb escape variants. DMS escape score is shown in purple gradient. The structurally defined epitope of each antibody is shown as a black outline. See the interactive version of deep mutational scanning data at https://dms-vep.org/Flu_H5_American-Wigeon_South-Carolina_2021-H5N1_DMS/htmls/VRC_antibody_escape_faceted.html. The residues K120 and G225 are highlighted in green and orange outlines, respectively. b, HA haplotype frequency of TX/24 virus in the absence or presence of 326-289.74 pressure by single-genome sequencing. Samples from input TX/24 virus stock as well as virus culture without mAb (no mAb) were used as controls. In each successive round, an increased amount of selection mAb was used. c, Neutralization sensitivity of TX/24 viruses selected with 326-289.74. Neutralization curves of early (R1) and late (R5) TX/24 viruses against mAbs. MEDI8852 was used as a control. d, HA haplotype frequency of TX/24 virus in the absence or presence of 310-12D03 pressure by single-genome sequencing. e, Neutralization sensitivity of TX/24 viruses selected with 310-12D03. Neutralization curves of early (R1) and late (R6) TX/24 viruses against mAbs. f, Viral replication fitness of TX/24 and its neutralization resistant variant carrying either K120T or G225E mutation. HA haplotype frequency after 1:1 mix of TX/24 virus and either 326-289.74-selected virus (left) or 310-12D03-selected virus (right) (R1) and after serial passages in the absence of antibody pressure. R3 and R5 samples were taken from 2 independent replicates after 3 and 5 passages, respectively. g, Protection against H5 TX/24 challenge of combinations of mAbs. Total mAb dose including all mAbs in each condition was 0.6 mg kg−1. Statistical significance of survival was determined by log-rank (Mantel–Cox) test with Bonferroni–Sidak adjustment. All mAb combinations were statistically significant relative to VRC01 (P < 0.001); P = 0.0002 for 310-7D11 + 310-12D03, P = 0.0006 for 310-12D03 + MEDI8852, P = 0.0008 for 310-7D11 + MEDI8852 and P = 0.0002 for 310-7D11 + 310-12D03 + MEDI8852. h, Maximum weight change (%) post challenge with same mAb combinations as in g as indicated. There was no significant difference between groups by Kruskal–Wallis test.
