Keys Allison M, Kastner David W, Kiessling Laura L, Kulik Heather J
Computational and Systems Biology Program, Massachusetts Institute of Technology, Cambridge, Massachusetts, USA; Department of Chemical Engineering, MIT, Cambridge, Massachusetts, USA; Department of Chemistry, MIT, Cambridge, Massachusetts, USA.
Department of Chemical Engineering, MIT, Cambridge, Massachusetts, USA; Department of Chemistry, MIT, Cambridge, Massachusetts, USA; Department of Biological Engineering, MIT, Cambridge, Massachusetts, USA.
J Biol Chem. 2025 Jun 14;301(8):110379. doi: 10.1016/j.jbc.2025.110379.
Protein-carbohydrate binding plays an essential role in biological processes including cellular recognition and immune signaling. However, glycans are hydrophilic with limited hydrophobic surfaces, a challenge for selective recognition by proteins. CH-π stacking interactions are pervasive in protein-carbohydrate binding sites and have emerged as critical drivers of protein-carbohydrate recognition. These interactions are highly favorable and have a broad orientational landscape. However, it is unknown how the orientations of CH-π stacking interactions are influenced by the protein environment; their functional interplay with hydrogen bonds in protein-carbohydrate binding is also unclear. Here, we employ well-tempered metadynamics simulations to obtain binding free energy landscapes for a set of protein-β-D-galactoside complexes with CH-π stacking interactions. Our data show that the favored orientation of a CH-π stacking interaction is controlled by the location of hydrogen bonds in the protein binding site. Complexes with extended carbohydrate ligands that form additional hydrogen bonds have more specific orientational dependencies, while protein variant complexes with fewer hydrogen bonds have broader free energy landscapes with glycan ligands adopting multiple CH-π stacking interaction orientations. We also show that forming multiple CH-π stacking interactions facilitates the dynamics necessary for the translocation of oligosaccharide ligands within a processive enzyme. Our findings underscore the cooperative nature of hydrogen bonds and CH-π stacking interactions, demonstrating that tuning the number and positions of these interactions through evolution or protein engineering can alter ligand recognition or support ligand movement.
蛋白质 - 碳水化合物结合在包括细胞识别和免疫信号传导在内的生物过程中起着至关重要的作用。然而，聚糖具有亲水性，疏水表面有限，这对蛋白质的选择性识别构成了挑战。CH-π堆积相互作用普遍存在于蛋白质 - 碳水化合物结合位点，并已成为蛋白质 - 碳水化合物识别的关键驱动因素。这些相互作用非常有利，并且具有广泛的取向格局。然而，尚不清楚CH-π堆积相互作用的取向如何受到蛋白质环境的影响；它们在蛋白质 - 碳水化合物结合中与氢键的功能相互作用也不清楚。在这里，我们采用温度调控元动力学模拟来获得一组具有CH-π堆积相互作用的蛋白质 - β-D-半乳糖苷复合物的结合自由能景观。我们的数据表明，CH-π堆积相互作用的有利取向由蛋白质结合位点中氢键的位置控制。与形成额外氢键的延伸碳水化合物配体的复合物具有更特定的取向依赖性，而氢键较少的蛋白质变体复合物具有更广泛的自由能景观，聚糖配体采用多种CH-π堆积相互作用取向。我们还表明，形成多个CH-π堆积相互作用促进了寡糖配体在进行性酶内转运所需的动力学。我们的研究结果强调了氢键和CH-π堆积相互作用的协同性质，表明通过进化或蛋白质工程调整这些相互作用的数量和位置可以改变配体识别或支持配体移动。
