少量细胞模板下多重置换扩增技术的保真度评估

Evaluation of the fidelity of multiple displacement amplification from small number of cells

目的应用单核苷酸多态性（single nucleotide polymorphism，SNP）芯片，检测以少量细胞（1～10个）为模板的多重置换扩增（multiple displacement amplification，MDA）产物的保真度。方法结合10K2．0SNP芯片平台与MDA，以纤维母细胞系GM02732（47，XY，＋18）作为模板，共分6组进行实验，其中A组与B组分别为阳性对照组（细胞gDNA）与阴性对照组（无模板MDA）；C-F组为实验组，分别以1、2、5和10个细胞为模板进行MDA扩增，产物与芯片杂交，检测并比较各组产物的基因组覆盖率、等位基因杂合性缺失（loss of heterozygosity，LOH）率以及等位基因脱扣（allele dropout，ADO）率。结果阴性对照组的MDA产物与gDNA序列有3．2％的重叠。随着起始模板从单细胞提升至10细胞，MDA产物的基因组覆盖率从86．4％逐步上升至96．4％，平均LOH率和ADO率则逐渐下降，各组之间比较差异均有统计学意义（P〈0．05）。结论多重置换扩增技术是一种高效而可靠的全基因组扩增方法，随着模板细胞的增加，MDA产物的保真度可获得明显的改善。10K 2．0SNP芯片可快速准确地在全基因组水平对DNA扩增产物进行保真度分析，但应注意区分LOH中的ADO和等位基因优势扩增，以避免产生误差。

Objective To evaluate the fidelity of multiple displacement amplification （MDA） from small number of cells （1-10 cells） by 10K 2.0 SNP mapping array. Methods A fibroblast cell line （Tri-18; GM02732, 47, XY, ＋18） was used as the template, and 6 groups were set up in the study. Groups A and B were positive and negative control, respectively; groups C-F were experimental groups involving the MDA products from 1, 2, 5 and 10 cells respectively. In combination of single nucleotide polymorphism （SNP） array, the product of each group was assessed based on the genome coverage, loss of heterozygosity （LOH） rate and allele dropout （ADO） rate. Results The nonspecific product of negative control presented an average call rate of 3.2%. The genome coverage of the MDA product increased from 86.4% to 96.4% with the increasing number of template from 1 to 10 cells, while the LOH rate and ADO rate decreased significantly （P〈0. 05）. Conclusion MDA is a highly efficient and reliable method for whole genome amplification. The fidelity of MDA will be improved significantly with the increasing number of template cells. 10K 2.0 SNP mapping array is a quick, accurate and comprehensive method to evaluate the fidelity of amplified DNA products, but the ADO SNPs should be distinguished from those of preferential amplification from the LOH loci to avoid errors.