三级结构Protein tertiary structure

主要驱动力是疏水效应：非极性侧链（亮氨酸、缬氨酸、苯丙氨酸等）聚集在内部形成疏水核心，极性和带电残基分布在表面与水接触。此外还有侧链间的氢键、盐桥（离子键）、范德华力，以及半胱氨酸之间的共价二硫键。折叠后的自由能仅比未折叠态低约 20–60 kJ/mol，是许多弱相互作用的微妙平衡。

The main driving force is the hydrophobic effect: nonpolar side chains such as leucine, valine and phenylalanine gather inside as a hydrophobic core, while polar and charged residues face the water. Side-chain hydrogen bonds, salt bridges, van der Waals forces and covalent disulfide bonds between cysteines also contribute. The folded state is only about 20–60 kJ/mol more stable than the unfolded one, a delicate balance of many weak interactions.

1958 年肯德鲁用 X 射线晶体学解析出抹香鲸肌红蛋白的三维结构，1959 年佩鲁茨解析出血红蛋白结构，二人共获 1962 年诺贝尔化学奖。1969 年莱文塔尔指出，若随机搜索所有构象，折叠所需时间将远超宇宙年龄，而实际蛋白质在毫秒到秒内折叠，这一“莱文塔尔悖论”推动了能量漏斗理论的提出。

In 1958 Kendrew solved the three-dimensional structure of sperm-whale myoglobin by X-ray crystallography, and Perutz solved haemoglobin in 1959; they shared the 1962 Nobel Prize in Chemistry. In 1969 Levinthal pointed out that a random search of all conformations would take longer than the age of the universe, yet proteins fold in milliseconds to seconds; this Levinthal paradox led to the energy-landscape (folding funnel) theory.

较大的蛋白质常由若干独立折叠的结构域组成，常见折叠类型有 TIM 桶、罗斯曼折叠、免疫球蛋白折叠等。体内分子伴侣（如热休克蛋白）帮助新生肽链正确折叠。蛋白质数据库（PDB）已收录超过 20 万个实验结构，AlphaFold 等人工智能方法已预测出几乎所有已知蛋白质的结构。

Larger proteins are often built from several independently folding domains, with common folds such as the TIM barrel, the Rossmann fold and the immunoglobulin fold. In cells, molecular chaperones such as heat-shock proteins help new chains fold correctly. The Protein Data Bank holds more than 200,000 experimental structures, and AI methods such as AlphaFold have predicted structures for almost every known protein.