目的 介绍大鼠及小鼠 Pig-a基因突变试验方法，并汇总国家药物安全评价监测中心 2015—2022年开展的基于免疫磁珠检测法的大鼠及小鼠 Pig-a 基因突变试验背景数据。方法 阴性物质包括超纯水和 0.5% 羧甲基纤维素钠（CMCNa），雄性 C57BL/6J 小鼠间隔 24 h ig 0.5% CMC-Na，连续 7 d；雄性 SD 大鼠间隔 24 h ig 0.5% CMC-Na，连续 14 d；大鼠间隔24 h ig 超纯水，连续 3 d。阳性对照为已知致细菌突变化合物，包括 N-乙基-N-亚硝基脲（ENU，10、40 mg·kg^-1）、盐酸丙卡巴肼（PCZ，60、150 mg·kg^-1）、乌拉坦（EC，300、800 mg·kg^-1）、N- 亚硝基二甲胺（NDMA ，1.5 mg·kg^-1）、N- 亚硝基二乙胺（NDEA，15 mg·kg^-1）。小鼠间隔 24 h ig ENU 40 mg·kg^-1，连续 3 d；间隔 24 h ig NDMA 1.5 mg·kg^-1、NDEA 15 mg·kg^-1，连续 7 d。大鼠间隔 24 h ig PCZ 150 mg·kg^-1、EC 800 mg·kg^-1、ENU 40 mg·kg^-1，连续 3 d ；间隔 24 h ig PCZ 60 mg·kg^-1、EC300 mg·kg^-1、ENU 10 mg·kg^-1，连续28 d。分别于给予受试物前，首次给予后14、28 d采集外周血，用流式细胞术检测大鼠红细胞表面 CD59蛋白的结合情况，结合免疫磁性计数微球技术计算网织红细胞（RETs）占总红细胞的百分率（%RET）（作为外周血毒性考察指标）、总红细胞中CD59表达为阴性细胞（RBC^CD59-，即突变的总红细胞）发生率和RETs中CD59表达为阴性细胞（RET^CD59-，即突变的 RETs）发生率。结果 各试验%RET数值均无大幅增加。SD大鼠和 C57BL/6J 小鼠的阴性对照组RBC^CD59-和 RET^CD59-突变率均低于 5×10-6，小鼠的背景值相对不稳定。连续 3 d ig给予小鼠 40 mg·kg^-1的 ENU，RBC^CD59-和RET^CD59-发生率自给药后2周开始均大幅增加（P＜0.05），给药后4周进一步增加（P＜0.01、0.001）；给予小鼠NDMA后2、4周，RBC^CD59-发生率略有增加，但仍在阴性背景范围内，但RET^CD59-发生率在给药后第2周大幅增加（P＜0.001），给药后第4周则大幅回落；给予小鼠NDEA后2周，RBC^CD59-和RET^CD59-发生率均有所增加（P＜0.05、0.001），给药后第 4 周则有所降低。连续3 d ig给予大鼠40 mg·kg^-1 ENU，或连续28 d ig给予大鼠10 mg·kg^-1 ENU，RBC^CD59-、RET^CD59-发生率自给药后第2周开始均大幅增加（P＜0.001），给药后第 4 周进一步增加（P＜0.001）；连续 3、28 d ig 给予大鼠不同剂量的 PCZ 或 EC 后，RBC^CD59-和RET^CD59-发生率的变化趋势与 ENU类似，但 EC诱发的突变细胞率低于 ENU和 PCZ。结论 体内 Pig-a基因突变试验可在首次给药后4周内有效检出致细菌突变化合物ENU、PCZ、EC、NDMA、NDEA的致突变性。提供了大鼠和小鼠Pig-a基因突变试验的背景值范围，为标准化试验方法的建立和研究结果的判定提供借鉴。

Objective The method of Pig-a gene mutation assay in rats and mice was introduced, and the background data based on immunomagnetic beads carried out by the National Center for Safety Evaluation of Drugs from 2015 to 2022 were summarized.Methods The negative substances included ultrapure water and 0.5% carboxymethyl cellulose sodium (CMC-Na). Male C57BL/6J mice received 0.5% CMC-Na at an interval of 24 h for seven consecutive days. Male SD rats were given 0.5% CMC-Na at an interval of 24 h for 14 consecutive days. Rats were given ultrapure water at an interval of 24 hours for three consecutive days.Positive controls were known bacterial mutagenic compounds, including N-ethyl-N-nitrosourea (ENU, 10 and 40 mg·kg^-1), procarbazide hydrochloride (PCZ, 60 and 150 mg·kg^-1), uratan (EC, 300 and 800 mg·kg^-1), N-nitrosodimethylamine (NDMA, 1.5 mg·kg^-1), and N-nitrosodiethylamine (NDEA, 15 mg·kg^-1). Mice were given ENU 40 mg·kg^-1 at an interval of 24 h for three consecutive days, and given NDMA 1.5 mg·kg^-1 and NDEA 15 mg·kg^-1 at an interval of 24 h for seven consecutive days. Rats were given PCZ 150 mg·kg^-1, EC 800 mg·kg^-1, and ENU 40 mg·kg^-1 at 24 hour intervals for three consecutive days, and given PCZ 60 mg·kg^-1, EC 300 mg·kg^-1, and ENU 10 mg·kg^-1 at 24 hour intervals for 28 consecutive days. Peripheral blood was collected before the administration of the test substance, 14 and 28 days after the first administration, and the binding of CD59 protein on the surface of rat red blood cells was detected by flow cytometry. The percentage of reticulocyte (RETs) in total red blood cells (%RET, as an indicator for peripheral blood toxicity), the incidence of CD59 expression in total red blood cells (RBC^CD59-, i.e. mutated total red blood cells) and the incidence of CD59 expression in RETs (RET^CD59- , i. e. mutated RETs) were calculated by using immunomagnetic counting microsphere technology.Results There was no significant increase in the % RET values for each experiment. The spontaneous RBC^CD59- and RET^CD59- mutation rates were lower than 5×10-6 in SD rats and C57BL/6J mice. The background values of mice are relatively unstable. After three consecutive days of ig administration of 40 mg·kg^-1 of ENU to mice, the incidence of RBC^CD59- and RET^CD59- increased significantly from two weeks after administration (P < 0.05), and further increased at four weeks after administration (P < 0.01 and 0.001). After two and four weeks of administration of NDMA to mice, the incidence of RBC^CD59- slightly increased, but remained within the negative background range. However, the incidence of RET^CD59- increased significantly (P < 0.001) in the second week after administration and decreased significantly in the fourth week after administration. The incidence of RBC^CD59- and RET^CD59- increased two weeks after administration of NDEA to mice (P < 0.05 and 0.001), but decreased four weeks after administration. The incidence of RBC^CD59- and RET^CD59- increased significantly (P < 0.001) after three consecutive days of ig administration of 40 mg·kg^-1 ENU to rats, or 28 consecutive days of ig administration of 10 mg·kg^-1 ENU to rats, starting from the 2nd week after administration and further increasing at the 4th week after administration. After continuous administration of different doses of PCZ or EC to rats for 3 and 28 days, the trend of changes in the incidence of RBC^CD59- and RET^CD59- was similar to that of ENU, but the mutation cell rate induced by EC was lower than that of ENU and PCZ.Conclusion In vivo Pig-a gene mutation assay can effectively detect the mutagenicity of bacterial mutant compounds ENU, PCZ, EC, NDMA, NDEA within four weeks after the first administration. This study provides the background value range of Pig-a gene mutation assay in rats and mice, which provides reference for the establishment of standardized test methods and the determination of research results.

R965.3

国家十三五“重大新药创制”专项课题（2018ZX09201017）；国家自然科学基金资助项目（81503347）