Insights into Comparative Modeling of VHH Domains

• Published in: International journal of molecular sciences Vol. 22; no. 18; p. 9771
• Main Authors: Vattekatte, Akhila Melarkode, Cadet, Frédéric, Gelly, Jean-Christophe, de Brevern, Alexandre G.
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 01.09.2021; MDPI
• Subjects: Algorithms; Antibodies; Antigens; Biochemistry, Molecular Biology; Chains; Datasets; Domains; homology modeling; Immunoglobulins; Life Sciences; Light; Modelling; Molecular structure; nanobodies; protein structure; Proteins; secondary structures; structural alphabet; Secondary structures; Nanobodies; Protein structure; Structural alphabet; Homology modeling
• ISSN: 1422-0067; 1661-6596; 1422-0067
• DOI: 10.3390/ijms22189771

In the particular case of the Camelidae family, immunoglobulin proteins have evolved into a unique and more simplified architecture with only heavy chains. The variable domains of these chains, named VHHs, have a number of Complementary Determining Regions (CDRs) reduced by half, and can function as single domains making them good candidates for molecular tools. 3D structure prediction of these domains is a beneficial and advantageous step to advance their developability as molecular tools. Nonetheless, the conformations of CDRs loops in these domains remain difficult to predict due to their higher conformational diversity. In addition to CDRs loop diversity, our earlier study has established that Framework Regions (FRs) are also not entirely conformationally conserved which establishes a need for more rigorous analyses of these regions that could assist in template selection. In the current study, VHHs models using different template selection strategies for comparative modeling using Modeller have been extensively assessed. This study analyses the conformational changes in both CDRs and FRs using an original strategy of conformational discretization based on a structural alphabet. Conformational sampling in selected cases is precisely reported. Some interesting outcomes of the structural analyses of models also draw attention towards the distinct difficulty in 3D structure prediction of VHH domains.