The Study Of Applying Array-based Comparative Genomic Hybridization In Clinical Cytogenetic Diagnosis

Object: To investigate the possibility and superiority of array-based comparative genomic hybridization (array-CGH) in clinical cytogenetic diagnosis.Method: Both G-banding karyotype analysis (reference method) and whole genomic array-CGH were performed on 10 clinical cytogenitic disorder cases to ascertain their karyotypes, fluorescence quantitative polymerase chain reaction (FQ-PCR) and fluorescence in situ hybridization (FISH) were used to verify the results of G-banding karyotype analysis and array-CGH.Results: By comparing the results between G-banding karyotype analysis and array-CGH, we found that:①array-CGH indicated that 18p11.21→pter was deleted and 18p11.21→qter was duplicated in case 1, and the breakpoint was located at 18p11.21 (between 12104527 bp - 12145199 bp), thus, the derivative chromosome in case 1 was ascertained to be idic(18)(p11.21→qter) rather than i(18q) showed by G-banding karyotype analysis;②G-banding karyotype analysis showed a derivative chromosome similar to short arm of X chromosome was presented in case 2, however, array-CGH confirmed that the derivative chromosome was normal Y chromosome and indicated an ~ 1.43 mb duplication of 8p23.1 between the olfactory receptor/defensin repeats (ORDRs) (between 10245882 bp - 11676699 bp) in case 2, which was associated with the abnormal phenotypes of case 2. Therefor, the case 2 was diagnosed to be a rare 8p23.1 duplication syndrome patient, whose duplicated region reduced the 8p23.1 duplication syndrome critical region and whose heart defect was associated with GATA4 gene duplication;③G-banding karyotype analysis showed that the karyotype of case 3 was complex chromosomal rearrangement (CCR) involving 4, 5, 15 chromosomes, but it can not be determined whether there were submicroscopic genomic imbalances presented in the breakpoint regions (4q23, 5p15, and 15q23); as natural limitation, array-CGH could not detect the CCR, but confirmed that no submicroscopic genomic imbalances presented in the breakpoint regions, indicating the CCR was balanced;④G-banding karyotype analysis showed that case 4 was a pure 4q25→qter partial trisomy patient, both her father and grandmother were balanced translocation t(4;10)(q25;q26) carriers; however, array-CGH not only indicated that 4q26→qter was duplicated in case 4, the breakpoint was located at 4q26 (between 115596658 bp - 118785802 bp) rather than 4q25, but also detected an ~ 0.54mb microdeletion del(10)(q26.3) (between 134750859 bp - 135286223 bp) from the case 4. QF-PCR and FISH confirmed that del(10)(q26.3) was also presented in her father and grandmother. Because the phenotypes of both father and grandmother were normal, it seem that no phenotypic effect was produced by del(10)(q26.3), therefor, the abnormal phenotypes of case 4 could only be attributed to 4q26→qter trisomy;⑤G-banding karyotype analysis showed that both karyotypes of cases 5, 6 were normal, however, array-CGH detected an ~ 80 kb microdeletion del(4)(q13.2) (between 70183990 bp - 70264889 bp) from both cases, this deletion only contains UGT2B28 gene, the deletion of which was associated with two patients'primary amenorrhea and hyperandrogenism;⑥case 7 might be a new type of acrofacial dysostosis syndrome patient, G-banding karyotype analysis was normal, however, two microduplications were detected by array-CGH: dup(1)(p36.33)(between 784258 bp - 1556626 bp, ~ 722 kb) and dup(1)(q21.3-22)(between 153182506 bp - 153318761 bp, ~ 136 kb). Two genes, PYGO2 and VWA1, located in both duplicated regions repectively, may be the pathogenic genes of patient's abnormal phenotypes;⑦both cases 8, 9 suffered from mental retardation, developmental delay, speech impairment, standing and walking disability, no similar cases have been reported, which means this condition might be a new syndrome. No abnormalities were determined by G-banding karyotype analysis, but array-CGH detected a microdeletion del(2)(p13.2) (between 73007487 bp - 73385899 bp, ~ 378 kb) from both cases. This deletion, containing 9 protein coding genes, was associated with the abnormal phenotypes of both cases, the EMX1 gene might be the pathogenic gene;⑧G-banding karyotype analysis indicated a mosaic small supernumerary marker chromosome (sSMC) in case 10, but the origin and characteristics could not be determined, only after array-CGH analysis, this sSMC was ascertained to be 18q21.1→pter.In addition, array-CGH also detected a large number of submicroscopic copy number variations (CNVs) undetectable by G-banding karyotype analysis from 10 cases, the minimal was only 12.87 kb. The sizes, breakpoints and genomic locations of these CNVs were accurately determined by array-CGH, which contributed to the connection between these CNVs and corresponding genes and facilitated the characteristics identification and phenotypic effects analysis of CNVs. Among these CNVs, 5 CNVs, including dup(8)(p23.1) (case 2), del(4)(q13.2) (cases 5, 6), dup(1)(p36.33) (case 7), dup(1)(q21.3-q22) (case 7) and del(2)(p13.2) (cases 8, 9), were potential pathogenic CNVs and may be association with abnormal phenotypes of corresponding cases; the rest of CNVs might be benign and did not produce phenotypic effects. FQ-PCR and FISH confirmed that the results of array-CGH were accurate.Conclusions: Due to limited resolution, the detectivity of G-banding karyotype analysis is unsatisfactory, so that it is difficult for G-banding karyotype analysis to determine the derivative chromosomes and marker chromosomes, and the "normal" results diagnosed by G-banding karyotype analysis may be unreliable and posses submicroscopic genomic imbalances. Compared with conventional cytogenetic technique, array-CGH has many advantages. Array-CGH can be comprehensive (genome-wide), high resolution, sensitive, rapid, accurate, amenable to automation, and less sample demand, which facilitates accurately determining karyotype, identifying derivative chromosome, screening out pathogenic CNVs/genes and investigating karyotype-phenotype correlation. Although there are still the cost and technical problems, array-CGH can serve as a useful complement for G-banding to be used in the clinical cytogenetic diagnosis.