Institute of Biological Chemistry, Academia Sinica, Taipei 11529, Taiwan; Institute of Biochemical Sciences, National Taiwan University, Taipei 10617, Taiwan.
Max Planck Computing and Data Facility, 85748 Garching, Germany.
Cell. 2024 Feb 29;187(5):1296-1311.e26. doi: 10.1016/j.cell.2024.01.034.
Most membrane proteins are modified by covalent addition of complex sugars through N- and O-glycosylation. Unlike proteins, glycans do not typically adopt specific secondary structures and remain very mobile, shielding potentially large fractions of protein surface. High glycan conformational freedom hinders complete structural elucidation of glycoproteins. Computer simulations may be used to model glycosylated proteins but require hundreds of thousands of computing hours on supercomputers, thus limiting routine use. Here, we describe GlycoSHIELD, a reductionist method that can be implemented on personal computers to graft realistic ensembles of glycan conformers onto static protein structures in minutes. Using molecular dynamics simulation, small-angle X-ray scattering, cryoelectron microscopy, and mass spectrometry, we show that this open-access toolkit provides enhanced models of glycoprotein structures. Focusing on N-cadherin, human coronavirus spike proteins, and gamma-aminobutyric acid receptors, we show that GlycoSHIELD can shed light on the impact of glycans on the conformation and activity of complex glycoproteins.
大多数膜蛋白通过 N-和 O-糖基化通过共价添加复杂的糖进行修饰。与蛋白质不同,聚糖通常不采用特定的二级结构,并且仍然非常灵活,屏蔽了蛋白质表面的潜在大部分区域。高聚糖构象自由度阻碍了糖蛋白的完整结构阐明。计算机模拟可用于模拟糖基化蛋白,但需要在超级计算机上进行数十万小时的计算,因此限制了常规使用。在这里,我们描述了 GlycoSHIELD,这是一种简化方法,可以在个人计算机上实施,以便在几分钟内将真实的聚糖构象集合嫁接到静态蛋白质结构上。使用分子动力学模拟、小角度 X 射线散射、冷冻电子显微镜和质谱,我们表明这种开放访问工具包提供了增强的糖蛋白结构模型。我们专注于 N-钙粘蛋白、人类冠状病毒刺突蛋白和γ-氨基丁酸受体,表明 GlycoSHIELD 可以揭示聚糖对复杂糖蛋白构象和活性的影响。