Biology subjects

Misasi, J.

Publications and source records attributed to Misasi, J..

7 recordsLinked to original sources

bioRxiv2025

CDR-aware masked language models for pairedantibodies enable state-of-the-art bindingprediction

BackgroundTherapeutic antibodies are a leading class of biologics, yet their unique architecture poses challenges for computational modeling. Each antibody comprises paired heavy and light variable domains with conserved framework regions that maintain structure and hypervariable complementarity-determining regions (CDRs) that directly contact antigens. This functional asymmetry, where CDRs determine binding specificity while frameworks provide scaffolding, suggests that region-aware training strategies could yield superior representations. Existing protein language models treat all regions uniformly, potentially missing critical features present in CDRs. MethodsWe developed a region-aware pretraining strategy for paired variable domain sequences using two protein language models: a 3 billion parameter model (ESM2) and a compact 600 million parameter model (ESM C). We compared three masking approaches: uniform whole-chain masking, CDR-focused masking, and a hybrid strategy. Final models were trained on over 1.6 million paired antibody sequences and evaluated on binding affinity datasets with over 90,000 antibody variants across six antigens, including single-mutant panels and combinatorial libraries. ResultsHere we show that CDR-focused training produces embeddings with superior predictive performance for antibody-antigen binding. Our approach achieves up to 27% improvements in binding affinity prediction compared to benchmarked antibody models. Remarkably, training exclusively on paired sequences proves sufficient; pretraining on billions of unpaired sequences provides no measurable benefit. Our compact model matches or exceeds larger antibody-specific baselines. ConclusionsThese findings establish that prioritizing paired sequences with CDR-aware supervision over scale and complex training schemes achieves both computational efficiency and predictive accuracy, providing a practical framework for next generation antibody language models.

bioinformatics↗

bioRxiv2022

A multispecific antibody confers pan-reactive SARS-CoV-2 neutralization and prevents immune escape

Summary ParagraphDespite effective countermeasures, SARS-CoV-2 persists worldwide due to its ability to diversify and evade human immunity1. This evasion stems from amino-acid substitutions, particularly in the receptor-binding domain of the spike, that confer resistance to vaccines and antibodies 2-16. To constrain viral escape through resistance mutations, we combined antibody variable regions that recognize different receptor binding domain (RBD) sites17,18 into multispecific antibodies. Here, we describe multispecific antibodies, including a trispecific that prevented virus escape >3000-fold more potently than the most effective clinical antibody or mixtures of the parental antibodies. Despite being generated before the evolution of Omicron, this trispecific antibody potently neutralized all previous variants of concern and major Omicron variants, including the most recent BA.4/BA.5 strains at nanomolar concentrations. Negative stain electron microscopy revealed that synergistic neutralization was achieved by engaging different epitopes in specific orientations that facilitated inter-spike binding. An optimized trispecific antibody also protected Syrian hamsters against Omicron variants BA.1, BA.2 and BA.5, each of which uses different amino acid substitutions to mediate escape from therapeutic antibodies. Such multispecific antibodies decrease the likelihood of SARS-CoV-2 escape, simplify treatment, and maximize coverage, providing a strategy for universal antibody therapies that could help eliminate pandemic spread for this and other pathogens.

immunology↗

bioRxiv2022

Convergent epitope specificities, V gene usage and public clones elicited by primary exposure to SARS-CoV-2 variants

An important consequence of infection with a SARS-CoV-2 variant is protective humoral immunity against other variants. The basis for such cross-protection at the molecular level is incompletely understood. Here we characterized the repertoire and epitope specificity of antibodies elicited by Beta, Gamma and ancestral variant infection and assessed their cross-reactivity to these and the more recent Delta and Omicron variants. We developed a high-throughput approach to obtain immunoglobulin sequences and produce monoclonal antibodies for functional assessment from single B cells. Infection with any variant elicited similar cross-binding antibody responses exhibiting a remarkably conserved hierarchy of epitope immunodominance. Furthermore, convergent V gene usage and similar public B cell clones were elicited regardless of infecting variant. These convergent responses despite antigenic variation may represent a general immunological principle that accounts for the continued efficacy of vaccines based on a single ancestral variant.

immunology↗

bioRxiv2021

Molecular probes of spike ectodomain and its subdomains for SARS-CoV-2 variants, Alpha through Omicron

Since the outbreak of the COVID-19 pandemic, widespread infections have allowed SARS-CoV-2 to evolve in human, leading to the emergence of multiple circulating variants. Some of these variants show increased resistance to vaccines, convalescent plasma, or monoclonal antibodies. In particular, mutations in the SARS-CoV-2 spike have drawn attention. To facilitate the isolation of neutralizing antibodies and the monitoring the vaccine effectiveness against these variants, we designed and produced biotin-labeled molecular probes of variant SARS-CoV-2 spikes and their subdomains, using a structure-based construct design that incorporated an N-terminal purification tag, a specific amino acid sequence for protease cleavage, the variant spike-based region of interest, and a C-terminal sequence targeted by biotin ligase. These probes could be produced by a single step using in-process biotinylation and purification. We characterized the physical properties and antigenicity of these probes, comprising the N-terminal domain (NTD), the receptor-binding domain (RBD), the RBD and subdomain 1 (RBD-SD1), and the prefusion-stabilized spike ectodomain (S2P) with sequences from SARS-CoV-2 variants of concern or of interest, including variants Alpha, Beta, Gamma, Epsilon, Iota, Kappa, Delta, Lambda, Mu, and Omicron. We functionally validated probes by using yeast expressing a panel of nine SARS-CoV-2 spike-binding antibodies and confirmed sorting capabilities of variant probes using yeast displaying libraries of plasma antibodies from COVID-19 convalescent donors. We deposited these constructs to Addgene to enable their dissemination. Overall, this study describes a matrix of SARS-CoV-2 variant molecular probes that allow for assessment of immune responses, identification of serum antibody specificity, and isolation and characterization of neutralizing antibodies.

immunology↗

bioRxiv2021

Structural basis for potent antibody neutralization of SARS-CoV-2 variants including B.1.1.529

With B.1.1.529 SARS-CoV-2 variants rapid spread and substantially increased resistance to neutralization by vaccinee and convalescent sera, monoclonal antibodies with potent neutralization are eagerly sought. To provide insight into effective neutralization, we determined cryo-EM structures and evaluated potent receptor-binding domain (RBD) antibodies for their ability to bind and neutralize this new variant. B.1.1.529 RBD mutations altered 16% of the RBD surface, clustering on a ridge of this domain proximal to the ACE2-binding surface and reducing binding of most antibodies. Significant inhibitory activity was retained, however, by select monoclonal antibodies including A19-58.1, B1-182.1, COV2-2196, S2E12, A19-46.1, S309 and LY-CoV1404, which accommodated these changes and neutralized B.1.1.529 with IC50s between 5.1-281 ng/ml, and we identified combinations of antibodies with potent synergistic neutralization. Structure-function analyses delineated the impact of resistance mutations and revealed structural mechanisms for maintenance of potent neutralization against emerging variants. Summary SentenceWe show potent B.1.1.529 neutralization by select antibodies and use EM structures to reveal how potency can be retained.

immunology↗

bioRxiv2021

High-Throughput, Single-Copy Sequencing Reveals SARS-CoV-2 Spike Variants Coincident with Mounting Humoral Immunity during Acute COVID-19

Tracking evolution of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) within infected individuals will help elucidate coronavirus disease 2019 (COVID-19) pathogenesis and inform use of antiviral interventions. In this study, we developed an approach for sequencing the region encoding the SARS-CoV-2 virion surface proteins from large numbers of individual virus RNA genomes per sample. We applied this approach to the WA-1 reference clinical isolate of SARS-CoV-2 passaged in vitro and to upper respiratory samples from 7 study participants with COVID-19. SARS-CoV-2 genomes from cell culture were diverse, including 18 haplotypes with non-synonymous mutations clustered in the spike NH2-terminal domain (NTD) and furin cleavage site regions. By contrast, cross-sectional analysis of samples from participants with COVID-19 showed fewer virus variants, without structural clustering of mutations. However, longitudinal analysis in one individual revealed 4 virus haplotypes bearing 3 independent mutations in a spike NTD epitope targeted by autologous antibodies. These mutations arose coincident with a 6.2-fold rise in serum binding to spike and a transient increase in virus burden. We conclude that SARS-CoV-2 exhibits a capacity for rapid genetic adaptation that becomes detectable in vivo with the onset of humoral immunity, with the potential to contribute to delayed virologic clearance in the acute setting. Author SummaryMutant sequences of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) arising during any individual case of coronavirus disease 2019 (COVID-19) could theoretically enable the virus to evade immune responses or antiviral therapies that target the predominant infecting virus sequence. However, commonly used sequencing technologies are not optimally designed to detect variant virus sequences within each sample. To address this issue, we developed novel technology for sequencing large numbers of individual SARS-CoV-2 genomic RNA molecules across the region encoding the virus surface proteins. This technology revealed extensive genetic diversity in cultured viruses from a clinical isolate of SARS-CoV-2, but lower diversity in samples from 7 individuals with COVID-19. Importantly, concurrent analysis of paired serum samples in selected individuals revealed relatively low levels of antibody binding to the SARS-CoV-2 spike protein at the time of initial sequencing. With increased serum binding to spike protein, we detected multiple SARS-CoV-2 variants bearing independent mutations in a single epitope, as well as a transient increase in virus burden. These findings suggest that SARS-CoV-2 replication creates sufficient virus genetic diversity to allow immune-mediated selection of variants within the time frame of acute COVID-19. Large-scale studies of SARS-CoV-2 variation and specific immune responses will help define the contributions of intra-individual SARS-CoV-2 evolution to COVID-19 clinical outcomes and antiviral drug susceptibility.

microbiology↗

bioRxiv2020

Structure-Based Design with Tag-Based Purification and In-Process Biotinylation Enable Streamlined Development of SARS-CoV-2 Spike Molecular Probes

Biotin-labeled molecular probes, comprising specific regions of the SARS-CoV-2 spike, would be helpful in the isolation and characterization of antibodies targeting this recently emerged pathogen. To develop such probes, we designed constructs incorporating an N-terminal purification tag, a site-specific protease-cleavage site, the probe region of interest, and a C-terminal sequence targeted by biotin ligase. Probe regions included full-length spike ectodomain as well as various subregions, and we also designed mutants to eliminate recognition of the ACE2 receptor. Yields of biotin-labeled probes from transient transfection ranged from [~]0.5 mg/L for the complete ectodomain to >5 mg/L for several subregions. Probes were characterized for antigenicity and ACE2 recognition, and the structure of the spike ectodomain probe was determined by cryo-electron microscopy. We also characterized antibody-binding specificities and cell-sorting capabilities of the biotinylated probes. Altogether, structure-based design coupled to efficient purification and biotinylation processes can thus enable streamlined development of SARS-CoV-2 spike-ectodomain probes.

immunology↗