Rahimi Abdessadek, Khibech Oussama, Benabbou Abdessamad, Merzouki Mohammed, Bouhrim Mohamed, Al-Zharani Mohammed, Nasr Fahd A, Ahmed Qurtam Ashraf, Abadi Said, Challioui Allal, Mimouni Mostafa, Elbekay Maarouf
Laboratory of Applied Chemistry and Environment (LCAE), Faculty of Science, University Mohammed Premier, Bd. Med VI B.P. 717, Oujda 60000, Morocco.
Laboratoires TBC, Laboratory of Pharmacology, Pharmacokinetics, and Clinical Pharmacy, Faculty of Pharmaceutical and Biological Sciences, 59000 Lille, France.
Pharmaceuticals (Basel). 2025 Jul 25;18(8):1110. doi: 10.3390/ph18081110.
Oncogenic KRAS drives ~30% of solid tumours, yet the only approved G12C-specific drugs benefit ≈ 13% of KRAS-mutant patients, leaving a major clinical gap. We sought mutation-agnostic natural ligands from Ziziphus lotus, whose stereochemically rich phenolics may overcome this limitation by occupying the SI/II (Switch I/Switch II) groove and locking KRAS in its inactive state. Phytochemical mining yielded five recurrent phenolics, such as (+)-catechin, hyperin, astragalin, eriodictyol, and the prenylated benzoate amorfrutin A, benchmarked against the covalent inhibitor sotorasib. An in silico cascade combined SI/II docking, multi-parameter ADME/T (Absorption, Distribution, Metabolism, Excretion, and Toxicity) filtering, and 100 ns explicit solvent molecular dynamics simulations. Pharmacokinetic modelling predicted oral absorption, Lipinski compliance, mutagenicity, and acute-toxicity class. Hyperin and astragalin showed the strongest non-covalent affinities (-8.6 kcal mol) by forging quadridentate hydrogen-bond networks that bridge the P-loop (Asp30/Glu31) to the α3-loop cleft (Asp119/Ala146). Catechin (-8.5 kcal mol) balanced polar anchoring with entropic economy. ADME ranked amorfrutin A the highest for predicted oral absorption (93%) but highlighted lipophilic solubility limits; glycosylated flavonols breached Lipinski rules yet remained non-mutagenic with class-5 acute-toxicity liability. Molecular dynamics trajectories confirmed that hyperin clamps the SI/II groove, suppressing loop RMSF below 0.20 nm and maintaining backbone RMSD stability, whereas astragalin retains pocket residence with transient re-orientation. Hyperin emerges as a low-toxicity, mutation-agnostic scaffold that rigidifies inactive KRAS. Deglycosylation, nano-encapsulation, or soft fluorination could reconcile permeability with durable target engagement, advancing phenolics toward broad-spectrum KRAS therapeutics.
致癌性KRAS驱动约30%的实体瘤，但唯一获批的G12C特异性药物仅使约13%的KRAS突变患者受益，这留下了一个重大的临床缺口。我们从莲雾中寻找与突变无关的天然配体，其富含立体化学的酚类物质可能通过占据SI/II（开关I/开关II）凹槽并将KRAS锁定在其无活性状态来克服这一限制。植物化学挖掘产生了五种常见的酚类物质，如（+）-儿茶素、金丝桃苷、紫云英苷、圣草酚，以及异戊烯基苯甲酸阿莫弗鲁汀A，并以共价抑制剂索托拉西布为基准。通过计算机模拟级联结合SI/II对接、多参数ADME/T（吸收、分布、代谢、排泄和毒性）筛选以及100纳秒的显式溶剂分子动力学模拟。药代动力学建模预测了口服吸收、符合Lipinski规则情况、致突变性和急性毒性类别。金丝桃苷和紫云英苷通过形成四齿氢键网络，将P环（Asp30/Glu31）与α3环裂缝（Asp119/Ala146）相连，显示出最强的非共价亲和力（-8.6千卡/摩尔）。儿茶素（-8.5千卡/摩尔）以熵经济性平衡了极性锚定。ADME预测阿莫弗鲁汀A的口服吸收最高（93%），但突出了其亲脂性溶解度限制；糖基化黄酮醇违反了Lipinski规则，但仍无致突变性，具有5类急性毒性倾向。分子动力学轨迹证实，金丝桃苷夹住SI/II凹槽，将环的均方根波动（RMSF）抑制在0.20纳米以下，并保持主链均方根偏差（RMSD）稳定状态，而紫云英苷则通过短暂的重新定向保持在口袋中。金丝桃苷是一种低毒性、与突变无关的支架，可使无活性的KRAS僵化。去糖基化、纳米封装或软氟化可以使渗透性与持久的靶点结合相协调，推动酚类物质成为广谱KRAS治疗药物。
