Department of Chemistry, School of Science , The University of Tokyo , 7-3-1 Hongo , Bunkyo , Tokyo 113-0033 , Japan.
Department of General Internal Medicine 4 , Kawasaki Medical School , 2-6-1 Nakasange , Kita-ku , Okayama 700-8505 , Japan.
J Am Chem Soc. 2019 Dec 11;141(49):19193-19197. doi: 10.1021/jacs.9b09106. Epub 2019 Dec 2.
l-Carboranylalanine (Cba) is a unique artificial amino acid containing a cluster of 10 boron atoms. Since the three-dimensional aromaticity and charge distributions of the carborane side chain are quite different from any side chains of proteinogenic amino acids, there is no report whether Cba can be a substrate for the translation machinery. Here, we report studies on the ribosomal incorporation of Cba into peptide via initiation and elongation using the flexizyme-assisted translation system. Our results indicate that only the initiation step could tolerate Cba incorporation, but the elongation steps could not, very likely due to its steric bulkiness of the side chain. Based on this knowledge, we have designed a library of macrocyclic peptides initiated by α--(2-choloroacetyl)-l-carboranylalanine (ClAc-Cba) and selected molecules capable of binding to human epidermal growth factor receptor (hEGFR). Two peptides that were forwarded to deeper studies exhibited affinities with values at 16 and 20 nM against hEGFR. Computational modeling of one of the peptides suggested that the carborane side chain might be directly involved in the interaction with the hydrophobic β-sheet core in the EGF binding site of hEGFR, which is consistent with the mutational data where replacing Cba residue with Phe completely eliminated the binding activity. Cell lines that stably express hEGFR could be stained by incubation with the C-terminal fluorescein-labeled peptides, whereas hEGFR-negative cells could not be stained. This study provides a general strategy for the de novo discovery of carborane-containing macrocyclic peptides targeting various tumor biomarker proteins, potentially applicable to boron neutron capture therapy.
L-卡硼烷丙氨酸（Cba）是一种独特的人工氨基酸，含有 10 个硼原子簇。由于卡硼烷侧链的三维芳香性和电荷分布与任何蛋白质氨基酸的侧链都有很大的不同，因此目前还没有报道 Cba 是否可以作为翻译机制的底物。在这里，我们使用柔性酶辅助翻译系统，报道了 Cba 通过起始和延伸掺入肽中的核糖体研究。我们的结果表明，只有起始步骤可以容忍 Cba 掺入，但延伸步骤不能，很可能是由于其侧链的庞大体积。基于这一知识，我们设计了一个由α-（2-氯乙酰基）-L-卡硼烷丙氨酸（ClAc-Cba）起始的大环肽文库，并筛选出能够与人类表皮生长因子受体（hEGFR）结合的分子。两个被推进到更深入研究的肽具有 16 和 20 nM 的亲和力值针对 hEGFR。对其中一种肽的计算建模表明，卡硼烷侧链可能直接参与与 hEGFR 的 EGF 结合位点的疏水性β-折叠核心的相互作用，这与突变数据一致，其中用 Phe 取代 Cba 残基完全消除了结合活性。稳定表达 hEGFR 的细胞系在用 C 末端荧光标记的肽孵育后可以被染色，而 hEGFR 阴性细胞则不能被染色。该研究为靶向各种肿瘤生物标志物蛋白的含硼大环肽的从头发现提供了一种通用策略，可能适用于硼中子俘获治疗。
