Combining computational modeling and experimental library screening to affinity-mature VEEV-neutralizing antibody F5

Abstract

Engineered monoclonal antibodies have proven to be highly effective therapeutics in recent viral outbreaks. However, despite technical advancements, an ability to rapidly adapt or increase antibody affinity and by extension, therapeutic efficacy, has yet to be fully realized. We endeavored to stand-up such a pipeline using molecular modeling combined with experimental library screening to increase the affinity of F5, a monoclonal antibody with potent neutralizing activity against Venezuelan Equine Encephalitis Virus (VEEV), to recombinant VEEV (IAB) E1E2 antigen. We modeled the F5/E1E2 binding interface and generated predictions for mutations to improve binding using a Rosetta-based approach and dTERMen, an informatics approach. The modeling was complicated by the fact that a high-resolution structure of F5 is not available and the H3 loop of F5 exceeds the length for which current modeling approaches can determine a unique structure. A subset of the predicted mutations from both methods were incorporated into a phage display library of scFvs. This library and a library generated by error-prone PCR were screened for binding affinity to the recombinant antigen. Results from the screens identified favorable mutations which were incorporated into 12 human-IgG1 variants. The best variant, containing eight mutations, improved KD from 0.63 nM (parental) to 0.01 nM. While this did not improve neutralization or therapeutic potency of F5 against IAB, it did increase cross-reactivity to other closely related VEEV epizootic and enzootic strains, demonstrating the potential of this method to rapidly adapt existing therapeutics to emerging viral strains.

Keywords: affinity‐maturation; anti‐viral; computational modeling; library screening; therapeutic antibodies.

FIGURE 1

The structural model used for mutational analysis by Sequence Tolerance, FlexddG, and dTERMen.

FIGURE 2

(a) Heavy chain and (b) light chain mutation predictions and results of library screening. The predicted mutations from Sequence Tolerance, dTERMen, and FlexddG mutations are shown along with mutations that were highly selected in the library screening. The mutations that were included in the library are shown in the fourth column. Mutations that were not predicted but were a consequence of including the predicted mutations are shown in cyan in the fourth column. The mutations predicted by Sequence Tolerance and dTERMen that were included in the library are shown in gold. For the FlexddG predictions, mutations predicted to be strongly favorable (>1.0 REU), moderately favorable (0.5–1.0 REU), and marginally favorable (0.2–0.5 REU) are shown in red, purple, and green, respectively.

FIGURE 3

ELISA results for binding of clones to the E1E2 heterodimer for the (a) directed library and (b) the random library.

FIGURE 4

Summary of beneficial mutations for (a) directed library and (b) random library. The criteria used to assign a mutation as being clearly beneficial was that a mutation must have appeared in greater than three clones and roughly 50% or greater of all instances of that mutation must have occurred within the clones that gave higher binding than parental. The nine mutations shown in (a) meet those criteria whereas only two of the mutations in (b) meet those criteria.

FIGURE 5

Sequences of 12 fabricated IgG variants of F5. (a) Sequence of parental and mutant light and heavy chains, with mutated residues highlighted in red. (b) Combinations of each mutated light and heavy chain sequence incorporated into 12 fabricated IgG variants.

FIGURE 6

ELISA for IgGs binding to recombinant TC‐83 E1E2 heterodimer.

FIGURE 7

Bilayer interferometry for the IgGs binding to the TC‐83 E1E2 heterodimer. (a) Association/dissociation curves for each IgG, and (b) calculated KD, Ka, and Kd values based on 1:1 Global kinetic model.

FIGURE 8

Gyros binding analysis of parental F5 and SNL1‐1 to E1/E2 heterodimer from related VEEV Subtypes. Binding screens conducted on Gyrolabs Xplore against purified E1E2 from VEEV subtypes IAB (TC‐83), IC (P676), ID (3880), IE (MenaII), and IIIA (MUCV). Fit curves generated using nonlinear least squares regression.

FIGURE 9

Parental and modified antibodies display similar therapeutic efficacy during lethal VEEV TC‐83 infection. Mice (n = 10) were infected intranasally with 5 × 10⁷ PFU VEEV TC‐83 and dosed with hF5, SNL1‐1, SNL1‐13, Abs1, Abs13, or isotype control antibodies at (a) +24 h and (b) +48 h post infection. Morbidity and mortality were assessed daily. **p < 0.01, ****p < 0.0001.