直接多肽结合试验组合人细胞系活化试验预测皮肤致敏物的探讨

Preliminary study for integrating DPRA with h-CLAT to predict skin sensitizers

目的建立直接多肽结合试验(DPRA)和人细胞系活化试验(h-CLAT)的皮肤致敏组合检测方法,对化学品及植物提取物的致敏性进行初筛。方法选择12种化学品和7种植物提取物为受试物,把不同受试物分别与两种肽(半胱氨酸肽和赖氨酸肽)共孵育24 h,采用高效液相色谱法分析反应后多肽消耗。同时将不同浓度的受试物与体外培养的人急性单核细胞白血病细胞(THP-1)共孵育24 h,通过流式细胞仪检测暴露后细胞表面标志物CD86和CD54的变化。再进一步比较DPRA和h-CLAT预测结果的一致性。结果 DPRA方法和h-CLAT方法准确区分了12种化学品的皮肤致敏性,其中2种判定为阴性,10种判定为阳性,两种方法检测结果一致。DPRA对植物提取物致敏性的预测中,除绿茶提取物无法判定外,其余6种均为皮肤致敏疑似物质。h-CLAT预测中,除绿茶提取物、马齿笕提取物和人参果提取物为非致敏物,其余4种植物提取物为致敏物。DPRA与h-CLAT预测的一致性为0.57。结论 DPRA与h-CLAT的简单组合可以实现对单一化合物的准确预测,对于复杂混合物可实现初步预测,确认需要进一步组合其他方法。

Objective To establish a detection method integrating DPRA（ direct peptide reactivity assay） with hCLAT（ human cell line activation test） to screen the skin sensitization potency of chemicals and plant extracts. Methods12 chemicals and 7 plant extracts were chosen as the test substances. Firstly,the test substances were incubated together with two different peptides（ cysteine and lysine） respectively for reaction for 24 h. The peptide consumptions were analyzed by HPLC. Simultaneously,THP-1 cells were cultured in vitro and then exposed to different concentrations of test substances for 24 h to examine the cell viability,cell surface markers CD54 and CD86 were assessed by flow cytometry. The predicting results were compared further between DPRA and h-CLAT. Results 12 chemicals were distinguished correctly by DPRA classified as 2 non-sensitizers and 10 sensitizers. The results of DPRA were in accordance with h-CLAT. Predicting the sensitization potency of plant extracts by DPRA showed that 6 plant extracts were determined as suspected sensitizers except for green tea extract. But using the method of h-CLAT,4 plant extracts were examined as suspected sensitizers except for green tea extract,herba portulacae extract and ginseng fruit extract. The coherence of DPRA and h-CLAT was0. 57. Conclusion This detection method integrating DPRA with h-CLAT can predict single compound accurately. As for complex compound,it can achieve preliminary prediction and need other integrating methods to make a further identification.