摘要

Ⅰ 黄酮类化合物对有机阳离子转运蛋白1的调控及分子机制研究

有机阳离子转运蛋白1（Organic cation transporter 1, OCT1）属于溶质载体（Solute carriers，SLC）SLC22A家族，是主要分布于人体肝脏的药物摄取转运蛋白，参与多种临床常用阳离子药物的肝脏摄取。OCT1的活性改变可影响其底物药物的肝脏摄取与系统暴露量，从而引起药物-药物相互作用（Drug-drug interaction, DDI）。利特昔替尼通过抑制OCT1减少健康受试者体内舒马曲坦的肝脏摄取，进而增加其体内暴露量，导致胃肠道不良反应的发生。对于具有肝毒性的底物药物来说，OCT1介导的肝脏摄取可导致底物药物在肝脏积累，从而引起或加重药物诱导的肝损伤（Drug-induced liver injury, DILI）。研究报道，奥沙利铂诱导的肝损伤与OCT1介导的肝脏摄取有关。阿托品在原代肝细胞中通过抑制OCT1显著降低奥沙利铂的细胞毒性。除了治疗药物外，一些天然化合物如小檗碱也被证明对OCT1有显著的抑制作用，能够缓解OCT1介导的DILI。因此，从天然产物中寻找高效低毒的OCT1抑制剂可能是缓解OCT1介导的DILI的有效途径。

黄酮是一类广泛分布于水果、蔬菜和中草药中的多酚类化合物，具有抗氧化、抗癌、抗炎、抗病毒等多种药理活性。目前，已有大量以黄酮为主要成分的保健品上市。研究表明，人类每日黄酮的摄取量高达1g。因此，黄酮产生的食物/草药-药物相互作用（food/herb-drug interactions, FDIs）的风险大大增加。已有研究表明，一些黄酮如芹菜素、木犀草素对OCT1具有较强的抑制作用，可减轻OCT1介导的DILI。因此，OCT1介导的黄酮类化合物相关的FDIs和DILI需要进一步研究。

 本研究应用人源化OCT1高表达的OCT1-HEK293细胞模型，研究了常见于食物和中草药中的96种黄酮类化合物对于OCT1活性的影响，筛选出了对OCT1具有强抑制作用的黄酮，并在细胞和整体动物水平上对黄酮抑制剂进行生物学效应评价，考察其对于奥沙利铂诱导的细胞毒性和倒千里光碱诱导的肝损伤的改善作用以及对舒马曲坦药代动力学特征的影响，然后构建药效团模型阐明黄酮类化合物与OCT1抑制作用之间的构效关系，为预测潜在的FDIs以及优化黄酮结构来缓解OCT1介导的DILI提供实验依据。

具体研究结果如下：

黄酮类化合物对药物转运体OCT1活性的影响

OCT1-HEK293细胞摄取实验的结果表明：在50μM或最大无毒浓度下，13种黄酮类化合物对OCT1具有显著的抑制作用（抑制率＞50%），包括α-萘黄酮、芹菜素、6-羟基黄酮、木犀草素、异水飞蓟宾、桔红素、川陈皮素、甜橙黄酮、刺甘草查尔酮、地奥司明、6-甲基黄酮、水飞蓟宾和穗花杉双黄酮。39种黄酮对OCT1的抑制作用较弱（抑制率为20-50%），其余的44种黄酮类对OCT1几乎没有抑制作用（抑制率<20%）。

在OCT1-HEK293细胞中，对抑制率大于50%的黄酮抑制剂进行了IC₅₀值测定，结果表明α-萘黄酮（IC₅₀ = 3.7μM）对OCT1的抑制作用最强，其次是芹菜素（IC₅₀ = 4.9μM）、6-羟基黄酮（IC₅₀ = 6.8μM）、木犀草素（IC₅₀ = 8.2μM）、异水飞蓟宾（IC₅₀ = 9.5μM）、桔红素（IC₅₀ = 10.1μM）、川陈皮素（IC₅₀ = 10.4μM）、甜橙黄酮（IC₅₀ = 13.8μM）、刺甘草查尔酮（IC₅₀ = 17.2μM）、地奥司明（IC₅₀ = 21.3μM）、6-甲基黄酮（IC₅₀ = 23.3μM）、水飞蓟宾（IC₅₀ = 32.3μM）和穗花杉双黄酮（IC₅₀ = 35.1μM）。

黄酮类化合物对药物转运体OCT1抑制作用的生物学效应评价

在奥沙利铂诱导的OCT1-HEK293细胞损伤模型中，阳性抑制剂多赛平和5种较强黄酮抑制剂（IC₅₀ < 10μM）α-萘黄酮、芹菜素、6-羟基黄酮、木犀草素和异水飞蓟宾可不同程度地降低奥沙利铂的细胞毒性，使细胞存活率升高。其中，α-萘黄酮、异水飞蓟宾和6-羟基黄酮的作用效果明显，强于阳性抑制剂多赛平。提示黄酮抑制剂可能通过抑制OCT1活性降低奥沙利铂的细胞摄取，从而减弱其细胞毒性。

在倒千里光碱诱导的KM小鼠肝损伤模型中，阳性抑制剂奎尼丁和上述5种黄酮抑制剂均能不同程度地降低KM小鼠血清ALT和AST水平，且使肝内倒千里光碱浓度与单独给药组相比减少43.6%-66.0%。其中，6-羟基黄酮和木犀草素的作用效果较为明显，与阳性抑制剂奎尼丁相当。上述结果提示，黄酮抑制剂可通过抑制OCT1减少倒千里光碱进入肝脏，从而发挥肝保护作用。

在KM小鼠体内，提前30min口服上述5种黄酮抑制剂后，与对照组相比，舒马曲坦的AUC_(0-t)和C_max分别增加了2.4%-44.5%和4.8%–57.5%，但弱于阳性抑制剂奎尼丁。提示黄酮抑制剂可通过调节OCT1而影响底物药物在KM小鼠的体内暴露量，导致FDI的发生。

黄酮类化合物抑制药物转运体OCT1的构效关系研究

黄酮类化合物与OCT1的药效团计算结果表明，黄酮类化合物的核心骨架上4，8，5’位置的氢键受体和疏水基团是黄酮类化合物对药物转运体OCT1产生抑制活性的关键药效团。结合体外活性筛选结果表明8位的甲氧基在OCT1抑制中也起到关键作用。

综上所述，本研究应用人源化高表达OCT1-HEK293细胞模型，在96种黄酮中筛选出13种对OCT1具有显著抑制作用的黄酮，它们可通过抑制OCT1减弱奥沙利铂的细胞毒性，缓解倒千里光碱所导致的肝损伤，增加舒马曲坦的系统暴露量。黄酮类化合物与OCT1的药效团计算结果表明，黄酮类化合物的核心骨架上4，8，5’位置的氢键受体和疏水基团是黄酮类化合物对药物转运体OCT1产生抑制活性的关键药效团。药效团模型的构建可初步阐明黄酮类化合物与OCT1之间的构效关系，为预测未经检测的黄酮类化合物和OCT1之间潜在的相互作用，缓解或避免临床中OCT1介导的DILI和FDIs提供参考依据。

Ⅱ 生物碱类化合物对多药及毒素外排转运蛋白1的调控及分子机制研究

多药及毒素外排转运蛋白1（Multidrug and toxin extrusion protein1, MATE1）是SLC家族中的一类重要的H⁺/有机阳离子外排转运蛋白，主要分布于肾脏近端小管刷状缘顶端膜上，在多种药物的肾脏排泄中发挥着重要作用。MATE1转运活性的改变可影响底物药物的系统暴露量，进而产生药物-药物相互作用（drug-drug interactions, DDI）。乙胺嘧啶通过抑制MATE1阻碍二甲双胍的肾脏排泄，增加健康志愿者的体内暴露量，进而导致乳酸酸中毒等不良反应的发生。对于具有肾毒性的底物药物来说，MATE1的活性抑制可影响MATE1底物药物的排泄过程，导致其在肾上皮细胞中的聚积，从而引起或加重药物引起的肾损伤（Drug-induced kidney injury, DIKI）。止吐药昂丹司琼通过抑制MATE1降低顺铂的肾脏排泄，增加小鼠的肾脏损伤。除了治疗药物，食物和草药中的天然产物也被报道对MATE1具有抑制作用，可能产生食物/草药-药物相互作用（food/herb-drug interactions, FDIs）。圣约翰草在健康志愿者体内通过抑制MATE1与二甲双胍产生FDI。因此鉴定天然产物对MATE1的抑制作用以预测并减少MATE1介导的FDIs和DILI具有重要的临床意义。

生物碱类化合物是一类广泛存在于植物、动物和微生物中含氮的有机化合物，具有解热抗炎、镇痛、抗癌、麻醉、抗病毒等多种药理活性。目前，已有多种以生物碱为主要成分的药物上市。除了用作治疗药物，生物碱经常作为保健品、饮品以及天然食品防腐剂出现在我们的日常生活中。鉴于生物碱在日常生活中频繁应用，其引发FDIs的风险也大大增加。已有研究表明，生物碱对于多种外排转运蛋白如P-糖蛋白（P-glycoprotein, P-gp）、乳腺癌耐药蛋白（Breast Cancer Resistance Protein, BCRP）、多药耐药相关蛋白2（Multidrug Resistance-associated Protein 2, MRP2）等具有调控作用，影响底物药物的体内处置，产生FDIs。而近期也有一些研究报道，生物碱类化合物如氯化两面针碱和金鸡纳碱对MATE1具有较强抑制作用，可加重MATE1介导的DIKI。因此，MATE1介导的生物碱化合物相关的FDIs和DIKI需要进一步研究。

本研究采用人源化MATE1高表达的MATE1-MDCK细胞模型，系统评估了食物和中草药中常见的120种生物碱类化合物对MATE1的抑制作用，筛选出对MATE1具有较强抑制活性的生物碱，并进一步考察其对顺铂诱导的细胞毒性和小鼠肾损伤及二甲双胍药代动力学特征的影响。最后建立药效团模型，阐明生物碱化合物对MATE1产生抑制作用的构效关系，为预测或避免MATE1介导的FDIs或DIKI风险提供实验依据。

具体研究结果如下：

120种生物碱类化合物对药物转运体MATE1活性的影响

MATE1-MDCK细胞摄取实验的结果表明：9种生物碱对MATE1具有显著的抑制作用（抑制率＞50%），包括去氢毛钩藤碱、盐酸去氢骆驼蓬碱、毛钩藤碱、缝籽嗪甲醚、浙贝丙素、盐酸黄连碱、千金藤素、盐酸巴马汀、紫蓳灵。21种生物碱类化合物对MATE1的抑制作用较弱（抑制率为20-50%），其余90种生物碱类化合物在100μM或最大无毒浓度下对MATE1无明显抑制作用（抑制率＜20%）。

在MATE1-MDCK细胞中，对抑制率＞50%的9个生物碱进行IC₅₀测定，结果表明去氢毛钩藤碱（IC₅₀ = 7.4μM）对MATE1的抑制作用最强，其次分别是盐酸去氢骆驼蓬碱（IC₅₀ = 7.9μM）、毛钩藤碱（IC₅₀ = 14.0μM）、缝籽嗪甲醚（IC₅₀ = 16.2μM）、盐酸黄连碱（IC₅₀ = 25.4μM）、浙贝丙素（IC₅₀ = 32.1μM）、千金藤素（IC₅₀ = 53.9μM）、盐酸巴马汀（IC₅₀ = 80.7μM）、紫蓳灵（IC₅₀ = 86.8μM）

生物碱类化合物对药物转运体MATE1抑制作用的生物学效应评价

在MATE1-MDCK细胞中，阳性抑制剂乙胺嘧啶和9种生物碱抑制剂均能增加顺铂的细胞毒性，使细胞存活率不同程度的降低。其中，盐酸去氢骆驼蓬碱的抑制作用最强，甚至在7.5-50 μM浓度范围内强于阳性抑制剂乙胺嘧啶。提示生物碱抑制剂可能通过抑制MATE1活性增加顺铂在细胞内的积累，从而增强其细胞毒性。

在顺铂诱导的ICR小鼠肾损伤模型中，5种较强生物碱抑制剂（IC₅₀ < 30μM）除盐酸黄连碱外，去氢毛钩藤碱、毛钩藤碱、缝籽嗪甲醚、盐酸去氢骆驼蓬碱均能显著升高血浆BUN和CRE水平，加重顺铂诱导的肾损伤，其中盐酸去氢骆驼蓬碱作用较强，但仍弱于阳性抑制剂乙胺嘧啶。提示生物碱抑制剂可通过抑制MATE1介导的顺铂肾脏清除从而加重肾损伤程度。

雄性ICR小鼠提前口服上述5种强生物碱抑制剂后，与对照组相比，二甲双胍的C_max和AUC_(0-t) 分别增加16.8%-158.3%和46.1%-170.8%，同时CL/F也降低了32.8%-63.0%。结果提示，生物碱抑制剂可通过抑制MATE1的外排作用而增加底物药物在小鼠体内的暴露量，导致FDI的发生。

生物碱类化合物抑制转运体MATE1的构效关系研究

药效团模型表明，生物碱类化合物中氢键受体是生物碱对MATE1产生抑制作用的关键药效团。此外，甲氧基、五元环等疏水基团的存在也对抑制MATE1产生重要影响。

综上所述，本研究应用人源化MATE1高表达的MATE1-MDCK细胞模型，在120种生物碱类化合物中筛选出9种生物碱对MATE1具有明显的抑制作用，它们可通过抑制MATE1增强顺铂的细胞毒性，加重顺铂诱导的肾损伤，增加二甲双胍的系统暴露量。生物碱化合物与MATE1的药效团计算结果表明氢键受体和疏水基团是生物碱类化合物对药物转运体MATE1产生抑制活性的关键药效团，阐明了生物碱抑制剂与MATE1之间的构效关系，为预测或避免临床中MATE1介导的DIKI和FDIs提供实验依据。

Ⅰ Regulation and Molecular Mechanisms of Flavonoids on Drug Transporter OCT1

Organic cation transporter 1 (OCT1), belonging to the SLC22 gene family, is an important uptake transporter mainly expressed in the liver, where it mediates the hepatic uptake of organic cationic drugs. Changes in OCT1 activity may affect the pharmacokinetic profiles of substrate drugs, leading to drug-drug interactions (DDIs). Ritlecitinib, by inhibiting OCT1, reduced the hepatic uptake and increased systemic exposure of sumatriptan, bringing about gastrointestinal side effects in healthy subjects. For hepatotoxic substrate drugs, OCT1-mediated hepatic uptake may lead to the accumulation of substrate drugs in the liver, thereby causing or exacerbating drug-induced liver injury (DILI). Previous studies indicated that oxaliplatin-induced liver injury was associated with OCT1-mediated hepatic uptake. In primary hepatocytes, atropine, by inhibiting OCT1, significantly reduced the cytotoxicity of oxaliplatin. Besides therapeutic drugs, natural products like berberine were also reported to have an inhibitory effect on OCT1, which can alleviate OCT1-mediated DILI. So, finding efficient and low-toxicity OCT1 inhibitors in natural products may be an effective approach for alleviating OCT1-mediated DILI.

Flavonoids, a group of polyphenols widely distributed in fruits, vegetables, and herbs, possess a wide range of pharmacological activities, such as anti-inflammatory, anticancer, antioxidant, antiviral and neuroprotective activities. Currently, many nutritional supplements containing flavonoids as the main ingredient are available on the market. Studies have shown that the daily intake of flavonoids in humans reaches up to 1g. Therefore, the risk of food/herb-drug interactions (FDIs) caused by flavonoids is increasing. Recent studies have indicated that flavonoids, such as apigenin and luteolin, have strong inhibitory effects on OCT1, which can alleviate OCT1-mediated DILI. Thus, flavonoid-related FDIs or DILI through inhibiting OCT1 need to be further studied.

In the present study, OCT1-HEK293 stably transfected with human OCT1 was employed to evaluate the inhibitory effects of 96 flavonoids that are rich in foods and herbs on OCT1. For flavonoids with strong inhibitory activity on OCT1, their biological effects were further explored in cell and animal models, including oxaliplatin-induced cytotoxicity, retrorsine-induced liver injury model, and sumatriptan pharmacokinetics. Finally, a pharmacophore model was established to elucidate the structure-activity relationship of flavonoids' inhibitory effects on OCT1, which provided experimental evidence for predicting potential FDIs and optimizing flavonoid structure to alleviate OCT1-mediated DILI in clinical practice.

The results were summarized as follows:

Effect of flavonoids on OCT1 transport activity

In the primary screening in OCT1-HEK293 cells, 13 of 96 flavonoids exhibited significant inhibitory effect on OCT1 (inhibition ratios > 50%), including α- naphthoflavone, apigenin, 6-hydroxyflavone, luteolin, isosilybin, tangeretin, nobiletin, sinensetin, echinatin, diosimin, 6-methylflavone, silymarin and amentoflavone. Another 39 flavonoids showed weaker inhibitory effects (20-50%), and the rest 44 flavonoids exhibited little or no inhibitory effects on OCT1 (<20%) in OCT1-HEK293 cells model at 50μM or a nontoxic concentration in our study.

The concentration-dependent inhibition on OCT1-mediated metformin uptake was further investigated by IC₅₀ assay. Among the tested flavonoids, the strongest inhibitory effect on OCT1 was α-naphthoflavone (IC₅₀ = 3.7μM), followed by apigenin (IC₅₀ = 4.9μM), 6-hydroxyflavone (IC₅₀ = 6.8μM), luteolin (IC₅₀ = 8.2μM), isosilybin (IC₅₀ = 9.5μM), tangeretin (IC₅₀ = 10.1μM), nobiletin (IC₅₀ = 10.4μM), sinensetin (IC₅₀ = 13.8μM), echinatin (IC₅₀ = 17.2μM), diosimin (IC₅₀ = 21.3μM), 6-methylflavone (IC₅₀ = 23.3μM), silymarin (IC₅₀ = 32.3μM) and amentoflavone (IC₅₀ = 35.1μM).

Evaluation of biological effects of flavonoids on OCT1

In oxaliplatin-induced OCT1-HEK293 cell damage model, positive inhibitor doxepin and five stronger flavonoid inhibitors (IC₅₀ < 10μM), including α-naphthoflavone, apigenin, 6-hydroxyflavone, luteolin and isosilybin, markedly reduced the cytotoxicity of oxaliplatin, resulting in an upward shift of cell viability curves. Among them, α-naphthoflavone, isosilybin, and 6-hydroxyflavone exhibited significant effects, even stronger than that of doxepin. These results indicate that the flavonoid inhibitors could reduce oxaliplatin’s cytotoxicity by inhibiting OCT1 transport activity.

In the retrorsine-induced liver injury model in KM mice, the above-mentioned 5 flavonoid inhibitors and quinidine (positive inhibitor of OCT1) could reduce the level of ALT and AST in the serum of KM mice to varying degrees. Furthermore, the flavonoid inhibitors decreased the liver concentration of retrorsine to different extents, from 43.6% to 66.0%. Notably, among these flavonoids, 6-hydroxyflavone and luteolin showed a relatively stronger inhibitory effect on OCT1, similar to quinidine. It is suggested that flavonoid inhibitors could protect the liver by inhibiting OCT1-mediated hepatic uptake of retrorsine.

 In flavonoid inhibitors pre-treated mice, the C_max and AUC_(0-t) of sumatriptan increased by 4.8% to 57.5% and by 2.4% to 44.5%, respectively. However, these effects were still weaker than those of quinidine. These results suggest that the inhibition of flavonoids on OCT1 could affect the exposure of OCT1 substrate drugs, leading to FDI.

The structure-activity relationship of flavonoids with OCT1

The structure-activity relationship between flavonoids and OCT1 was elucidated by pharmacophore calculation. The results indicate that hydrogen bond acceptors at the 4,8,5′ position were the essential pharmacophores of OCT1 inhibitors. Combined with the result of inhibitory effects, the methoxy group at 8-position may also play a vital role in the potency of inhibition on OCT1.

In conclusion, 13 of tested 96 flavonoids exhibited significant inhibition of OCT1 transport activity in OCT1-HEK293 cells. These flavonoid inhibitors remarkably decreased oxaliplatin cytotoxicity, alleviated retrorsine-induced liver injury, and increased systemic exposure of sumatriptan in mice by inhibiting OCT1. The pharmacophore model indicated that hydrogen bond acceptors in the 4, 8, and 5’ positions may play a critical role in the inhibitory effect of flavonoids on OCT1. Our findings elucidate the structure-activity relationship of flavonoid inhibition on OCT1, help to predict potential interactions between untested flavonoids and OCT1 and provide experimental evidence for alleviating or avoiding OCT1-mediated DILI and FDIs in clinical practice.

Ⅱ Regulation and Molecular Mechanisms of Alkaloids on Drug Transporter MATE1

Multidrug and toxin extrusion protein1 (MATE1), an important H+/organic cation efflux transporter belong to SLC family, is primarily located on the apical membranes of proximal tubular cells in the kidneys, where it is responsible for renal excretion of various drugs. The altered transport activity of MATE1 can affect the systemic exposure of substrate drugs, leading to drug-drug interactions (DDIs). Pyrimethamine, a specific inhibitor of MATE1, has been shown to impair the renal excretion of metformin and increase its systemic exposure, resulting in lactic acidosis in healthy volunteers. For nephrotoxic substrate drugs, inhibition of MATE1 activity can affect the excretion of MATE1 substrates, leading to their accumulation in renal epithelial cells, which may cause or exacerbate drug-induced kidney injury (DIKI). Ondansetron, a potent inhibitor for MATE1, enhanced cisplatin induced nephrotoxicity in mice by inhibiting cisplatin excretion. Besides therapeutic drugs, natural products from food and herbs have also been reported to inhibit MATE1, which may raise food/herb-drug interactions (FDIs). St. John's wort, by inhibiting MATE1, brought about FDI with metformin in humans. Therefore, screening and identifying potent MATE1 inhibitors from nature products can predict or minimize the risk of DIKI or FDIs.

Alkaloids are nitrogen-containing organic compounds widely found in plants, animals, and microorganisms, known for their diverse pharmacological activities such as antipyretic, anti-inflammatory, analgesic, anticancer, anesthetic, and antiviral effects. Currently, many alkaloid drugs are available on the market. Besides this, alkaloids are frequently found in our daily lives as nutritional supplements, drinks, and natural food preservatives. Given the ubiquity of alkaloids in our daily life, the risk of FDIs is accordingly increased. Alkaloids have been reported to modulate some efflux transporters, such as P-gp, BCRP, and MRP2, which affect drug disposition, leading to FDIs. Recent research has also revealed that certain alkaloids, like nitidine hydrochloride and quinine, strongly inhibit MATE1 activity, which may exacerbate MATE1-meidated DIKI. Thus, alkaloid-related FDIs or DIKI through inhibiting MATE1 need to be further explored.

In the present study, MATE1-MDCK stably transfected with human MATE1 was employed to investigate the inhibitory effects of 120 alkaloids that are rich in foods and herbs on MATE1. For alkaloids with strong inhibitory activity on MATE1, their biological effects were further evaluated in cell and animal models, including cisplatin-induced cytotoxicity and nephrotoxicity, and metformin pharmacokinetics. Finally, a pharmacophore model is established to elucidate the structure-activity relationship of alkaloid inhibitory effects on MATE1, which may provide experimental evidence for predicting or minimizing the risk of MATE1-mediated FDIs or DIKI in clinical practice.

The results were shown as follows:

Effect of 120 alkaloids on MATE1 transport activity

In the primary screening in MATE1-MDCK cells, 9 of 120 alkaloids exhibited significant inhibitory effect on MATE1 (inhibition ratio > 50%), including hirsuteine, harmine hydrochloride, hirsutine, geissoschizine methyl ether, zhebeirine, coptisine chloride, cepharanthine, palmatine hydrochloride, and corynoline.  Another 21 alkaloids showed weaker inhibitory effects (20-50%), and the rest 90 alkaloids exhibited little or no inhibitory effects on MATE1 (< 20%) at 100μM or a nontoxic concentration in our study.

In MATE1-MDCK cells, the IC₅₀ of alkaloids with potent inhibitory effect (> 50%) on MATE1 was determined. Among the tested alkaloids, the strongest inhibitory effect on MATE1 was hirsuteine (IC₅₀ = 7.4μM), followed by harmine hydrochloride (IC₅₀ = 7.9μM), hirsutine (IC₅₀ = 14.0μM), geissoschizine methyl ether (IC₅₀ = 16.2μM), coptisine chloride (IC₅₀ = 25.4μM), zhebeirine (IC₅₀ = 32.1μM), cepharanthine (IC₅₀ = 53.9μM), palmatine hydrochloride (IC₅₀ = 80.7μM) and corynoline (IC₅₀ = 86.8μM).

Evaluation of biological effects of alkaloids on MATE1

In cisplatin-induced MATE1-MDCK cell model, positive inhibitor pyrimethamine and the 9 alkaloid inhibitors increased cisplatin-induced cytotoxicity, reducing cell viability to varying extents in MATE1-MDCK cells. Among them, harmaline hydrochloride exhibited the strongest inhibitory effect, even stronger than that of pyrimethamine within the concentration range of 7.5–50 μM. This indicates that alkaloid inhibitors may enhance cisplatin accumulation in cells by inhibiting MATE1 activity, thereby increasing cisplatin’s cytotoxicity.

In the cisplatin-induced kidney injury model, five stronger alkaloid inhibitors (IC₅₀ < 30μM), except for coptisine chloride, hirsuteine, hirsutine, geissoschizine methyl ether, and harmine hydrochloride significantly elevated plasma BUN and CRE levels, aggravating cisplatin-induced kidney injury. Among them, harmine hydrochloride exhibited the strongest inhibitory effect, but still weaker than positive inhibitor pyrimethamine. These results suggest that alkaloid inhibitors may exacerbate kidney injury by inhibiting MATE1-mediated renal clearance of cisplatin.

In alkaloid inhibitors pre-treated mice, the C_max and AUC_(0-t) of metformin increased by 16.8–158.3% and 46.1–170.8%, respectively, while CL/F decreased by 32.8–63.0%. These findings indicate that alkaloid inhibitors can increase systemic exposure to substrate drugs by inhibiting MATE1 efflux activity, leading to FDIs.

The structure-activity relationship of alkaloids with MATE1

The pharmacophore model identified hydrogen bond acceptors as key functional groups contributing to the inhibitory effects of alkaloids on MATE1. Additionally, the presence of hydrophobic groups, such as methoxy groups and five-membered rings, was found to play a critical role in MATE1 inhibition.

To sum up, 9 of tested 120 alkaloids exhibited significant inhibitory effects on MATE1 in MATE1-MDCK cells. These alkaloid inhibitors, by inhibiting MATE1, enhanced the cytotoxicity of cisplatin, exacerbated cisplatin-induced nephrotoxicity, and increased the systemic exposure of metformin. The pharmacophore model indicated that hydrogen bond acceptors and hydrophobic groups were the essential pharmacophores of MATE1 inhibitors. These findings elucidate the structure-activity relationship of alkaloid inhibition on MATE1 and provide experimental evidence for predicting or avoiding MATE1-mediated DIKI and FDIs in clinical settings.