IRCM, Institut de Recherche en Cancérologie de Montpellier, Montpellier, F-34298, France; INSERM U896, Montpellier, F-34298, France; Université Montpellier 1, Montpellier, F-34298, France; Institut Régional du Cancer Montpellier, Montpellier, F-34298, France.
Integrated Structural Biology Department, Institute of Genetics and Molecular and Cellular Biology, INSERM U964/CNRS UMR 7104/Université de Strasbourg, Illkirch F-67404, France.
J Mol Biol. 2014 Nov 11;426(22):3729-3743. doi: 10.1016/j.jmb.2014.08.013. Epub 2014 Aug 29.
Antibody molecules are able to recognize any antigen with high affinity and specificity. To get insight into the molecular diversity at the source of this functional diversity, we compiled and analyzed a non-redundant aligned collection of 227 structures of antibody-antigen complexes. Free energy of binding of all the residue side chains was quantified by computational alanine scanning, allowing the first large-scale quantitative description of antibody paratopes. This demonstrated that as few as 8 residues among 30 key positions are sufficient to explain 80% of the binding free energy in most complexes. At these positions, the residue distribution is not only different from that of other surface residues but also dependent on the role played by the side chain in the interaction, residues participating in the binding energy being mainly aromatic residues, and Gly or Ser otherwise. To question the generality of these binding characteristics, we isolated an antibody fragment by phage display using a biased synthetic repertoire with only two diversified complementarity-determining regions and solved its structure in complex with its antigen. Despite this restricted diversity, the structure demonstrated that all complementarity-determining regions were involved in the interaction with the antigen and that the rules derived from the natural antibody repertoire apply to this synthetic binder, thus demonstrating the robustness and universality of our results.
抗体分子能够以高亲和力和特异性识别任何抗原。为了深入了解这种功能多样性的分子多样性来源,我们编译并分析了 227 种抗体-抗原复合物的非冗余对齐集合。通过计算丙氨酸扫描定量了所有残基侧链的结合自由能,从而首次对抗体变区进行了大规模的定量描述。这表明,在 30 个关键位置中,仅 8 个残基就足以解释大多数复合物中 80%的结合自由能。在这些位置上,残基分布不仅与其他表面残基不同,而且还取决于侧链在相互作用中所起的作用,参与结合能的残基主要是芳香族残基,而其他残基则主要是 Gly 或 Ser。为了质疑这些结合特征的普遍性,我们使用带有仅两个多样化互补决定区的偏向性合成库通过噬菌体展示分离出一个抗体片段,并解析了其与抗原复合物的结构。尽管这种多样性受到限制,但该结构表明所有互补决定区都参与了与抗原的相互作用,并且从天然抗体库中得出的规则适用于这种合成配体,从而证明了我们的结果具有稳健性和普遍性。