Enhancing Influenza Vaccine Development

ZBiotechMicrobiologyEnhancing Influenza Vaccine Development

Microbiology

Enhancing Influenza Vaccine Development

Enhancing influenza vaccine development with glycan microarray technology: a case study using MAIVeSS.

Highlights

Array:Custom Sialylated Glycan Array

Field:Infectious Disease

Study:Vaccine Development

Influenza remains a significant global health challenge, necessitating the efficient development of effective vaccines. The process of selecting optimal seed viruses for vaccine production is challenging, involving the identification of viruses that not only match the antigenicity of circulating strains but also demonstrate high yield characteristics. The recent research paper, MAIVeSS: Streamlined Selection of Antigenically Matched, High-yield Viruses for Seasonal Influenza Vaccine Production, presents a breakthrough approach leveraging the Machine-learning Assisted Influenza VaccinE Strain Selection (MAIVeSS) framework. This study exemplifies the critical role of glycan microarrays in identifying high-yield influenza vaccine strains.

What is MAIVeSS Framework?

MAIVeSS utilizes advanced machine learning algorithms to analyze viral genomic sequences, predict antigenic and yield phenotypes, and select optimal candidate vaccine viruses (CVVs). In a landmark study, MAIVeSS was applied to A(H1N1)pdm09, yielding strains that demonstrated both high growth rates in vitro and strong antigenic matches to circulating viruses. This dual achievement is crucial for vaccine efficacy and timely production, addressing previous challenges in vaccine mismatch and production delays.

This MAIVeSS model is specifically developed to select optimal vaccine candidates that closely match the antigenic properties of circulating flu strains and demonstrate high growth potential. The process begins with the creation of a library of viruses, each modified at the receptor binding sites of the HA protein. These viruses are then subjected to a detailed analysis to evaluate their ability to trigger immune responses (serological reactivity), their replication efficiency in chicken eggs and cell cultures, and their interactions with cell surface sugars (glycan profiling using microarrays). A sparse learning model is utilized to identify key genetic features that correlate with these desired phenotypes. These features form the basis for developing predictive models that can accurately estimate how antigenically similar each virus is to current vaccine strains and predict their potential yield in production environments. This methodology is not only applicable to the initial targeted subtypes of influenza viruses but can also be adapted to include other subtypes and expanded to incorporate additional genetic elements like NA protein sequences. This integrated approach enhances the ability to swiftly identify and develop high-yield, antigenically matched vaccine candidates.

The image is reproduced from Gao, C. et al. ‘MAIVeSS: streamlined selection of antigenically matched, high-yield viruses for seasonal influenza vaccine production.’ Nat. Commun. 15, (2024).

Role of Glycan Microarray in MAIVeSS

A central component in this process is the glycan microarray technology provided by our company. Our custom-designed arrays featured 75 distinct glycoforms, enabling precise mapping of virus-receptor interactions. This technology facilitated the identification of unique glycan substructures that bind specifically to high-yield virus mutants, thus serving as a critical tool in MAIVeSS for predicting vaccine strain performance.

Each glycan in our array is selected to represent various glycan categories, providing comprehensive coverage of potential viral interactions. These arrays are printed on NHS-derivatized slides, allowing for high-fidelity binding assays. In the study, the application of our glycan arrays revealed that certain glycan substructures were preferentially bound by high-yield strains, supporting the selection of these strains for vaccine production.

This set of illustrations includes a heat map and a list of glycan substructures used in machine learning, providing insights into how different HA receptor binding site (RBS) mutants interact with specific glycans. The heat map visually represents the binding intensities of 189 mutant viruses to 75 glycoforms displayed on a glycan microarray. Each row in the heat map corresponds to a different HA RBS mutant, and each column represents an individual glycan, with the color intensity indicating the strength of the binding. This detailed visualization helps identify which virus mutants have stronger or weaker interactions with particular glycans, crucial for selecting high-yield strains in vaccine development.

Accompanying the heat map is a list categorizing 27 glycan substructures into three groups: terminal, internal, or basal. These classifications aid in understanding the specific roles these structures play in the binding preferences observed in the heat map analysis. The study found that all mutants exhibited a strong binding affinity to glycans ending in SA2,6 Gal. Further refinement using the categorized glycan substructures revealed distinct preferences, such as elevated affinities to Neu5Acα2-6Galβ1-4GlcNAc (6′SLN) and other specific structures like 3′SLN and sLeX. This comprehensive analysis highlights the correlation between glycan binding properties and high-yield traits in viruses, providing valuable insights for optimizing influenza vaccine production.

The images are reproduced from Gao, C. et al. ‘MAIVeSS: streamlined selection of antigenically matched, high-yield viruses for seasonal influenza vaccine production.’ Nat. Commun. 15, (2024).

Implications for Influenza Vaccine Development

The integration of glycan microarray technology into the MAIVeSS framework drastically reduces the timeline for selecting suitable vaccine candidates from months to days. This acceleration is possible due to the high-throughput and precise nature of glycan microarrays, which allow for rapid screening of numerous virus variants against a wide array of glycan structures. The ability to quickly identify viruses with optimal growth and antigenic characteristics can significantly impact global health by enabling faster responses to influenza outbreaks.

The recent study utilizing our glycan microarray product in conjunction with the MAIVeSS framework exemplifies the transformative potential of combining cutting-edge biological technologies with machine learning. Our glycan microarrays not only enhance the precision of vaccine strain selection but also contribute to a deeper understanding of virus-glycan interactions, which is crucial for future vaccine development and epidemiological research. This case study underscores the capabilities and benefits of our glycan microarray technology, proving its value as an indispensable tool in the ongoing battle against influenza.

By offering a robust and versatile platform for detailed glycan analysis, our product stands out as an essential component in the next generation of vaccine development strategies, ensuring that vaccines are not only produced swiftly but also maintain the highest standards of efficacy and safety. The successful application in the MAIVeSS study promises a new era of rapid, responsive vaccine production, aligning with the urgent needs of global health landscapes.

Reference

Gao, C. et al. MAIVeSS: streamlined selection of antigenically matched, high-yield viruses for seasonal influenza vaccine production. Nat. Commun. 15, (2024).