目的 建立黄芪六一汤药效组分（黄芪总皂苷、黄芪总黄酮、黄芪多糖、甘草酸组成，HLDC）抗糖尿病肾病的药动学（PK）-药效动力学（PD）结合模型，阐明HLDC成分在糖尿病肾病大鼠体内的动态变化及与药效消长之间的相互关系。方法 采用高脂高糖饲料及尾iv链脲佐菌素（33 mg·kg^-1）方法建立糖尿病肾病大鼠模型。单次或多次igHLDC（1.320 g·kg^-1）后，采用HPLC-MS/MS测定不同时间点血浆中黄芪甲苷、芒柄花苷、芒柄花素、毛蕊异黄酮葡萄糖苷、毛蕊异黄酮和甘草酸6个入血原型成分的含量，并采用酶联免疫吸附法（ELISA）检测定不同时间点血样中肿瘤坏死因子-α （TNF-α）、血管内皮生长因子（VEGF）、白细胞介素-18 （IL-18）等炎症因子水平，获得时效曲线。并利用WinNonlin 8.2软件房室模型分析法拟合各成分的PK参数，固定PK，对时效关系进行拟合，得到PD；根据PD参数建立HLDC的PK-PD模型。结果 药动学结果显示，与单次给药相比，除了毛蕊异黄酮葡萄糖苷外，多次给药后模型大鼠的各成分的达峰时间（t_max）均提前；各成分的t_1/2均延后；除了甘草酸、毛蕊异黄酮外，各成分的AUC均升高。药效学结果显示，药物质量浓度逐渐降低的同时，药物的抑制作用先增大后减小，除了单次给药组中的黄芪甲苷，TNF-α、VEGF、IL-18的药物效应t_max均较血药浓度t_max长，存在滞后效应。PK-PD结合模型显示，单次及多次给药HLDC中6个成分的血药浓度与其药效数据均能较好地拟合。可通过黄芪甲苷等6个成分的血药浓度计算相应的药效值，也可以根据药效值计算相应的血药浓度。结论 PK-PD模型构建结果符合Sigmoid-Emax模型，TNF-α、VEGF、IL-18等炎症因子水平与HLDC中的黄芪甲苷等6个成分血药浓度有良好的相关性。HLDC活性成分发挥防治糖尿病肾病、显著改善肾功能的作用，可能与抑制TNF-α、VEGF、IL-18等炎症因子的分泌有关。

Objective The PK-PD binding model of effective component (total astragaloside, total flavonoids of astragalus, astragalus polysaccharide, and glycyrrhizic acid, HLDC) of Huangqi Liuyi Decoction against diabetic nephropathy was established, and the dynamic changes of HLDC component in diabetic nephropathy rats and the relationship between HLDC component and drug effect were clarified. Methods The rat model of diabetic nephropathy was established by high-fat and high-sugar feed and streptozotocin (33 mg· kg^-1) injected into the tail vein. After single or multiple oral administrations of HLDC (1.320 g· kg^-1), the plasma contents of astragaloside IV, ononin, formononetin, calycosin-7-O-beta-D-glucoside, calycosin and glycyrrhizic acid were determined by HPLCMS/MS at different time points, and tumor necrosis factor-α (TNF-α), vascular endothelial growth factor (VEGF) and interleukin were detected by enzyme-linked immunosorbent assay (ELISA). The pharmacokinetic (PK) parameters of each component were fitted by using WinNonlin 8.2 software, and the aging relationship was fitted to get PD. The PK-PD model of HLDC is established according to PD parameters. Results The results of pharmacokinetics showed that compared with a single administration, the t_max of each component of the model rats was advanced after repeated administration except calycosin-7-O-beta-D-glucoside. T_1/2 of each component was delayed. Except glycyrrhizic acid and calycosin, the AUC of all components increased. The pharmacodynamic results showed that the inhibitory effect of the drug increased first and then decreased while the drug concentration gradually decreased. Except for astragaloside IV, TNF-α, VEGF, and IL-18 in the single-dose group, the peak time of the drug effect was longer than that in the blood concentration, and there was a lag effect. PK-PD binding model showed that the blood concentration of six components in single and multiple doses of HLDC and their pharmacodynamic data could be well fitted. The corresponding drug effect value can be calculated by the blood concentration of six components such as astragaloside IV, and the corresponding blood concentration can also be calculated according to the drug effect value. Conclusion The results of the PK-PD model are by the Sigmoid-Emax model, and the levels of inflammatory factors such as TNF-α, VEGF, and IL-18 have a good correlation with the concentrations of six blood component drugs such as astragaloside IV in HLDC. The active components of HLDC play a role in preventing and treating diabetic nephropathy and significantly improving renal function, which may be related to inhibiting the secretion of inflammatory factors such as TNF-α, VEGF, and IL-18.

R285.5

贵州省卫生健康委科学技术基金项目（黔卫健函〔2024〕24号）;贵州省科技计划项目（黔科合基础-ZK［2023］一般416）;2023年度贵州中医药大学学术新苗项目（贵科合学术新苗［2023］-13号）