D. Pinkel[1,2], Donna Albertson[1,3], Y. Zhai[2], R. Segraves[2], D. Sudar[1], K. Ligtenberg[1], and J. Gray[1,2].
[1]Lawrence Berkeley National Laboratory, [2]University of California, San Francisco, [3]MRC Laboratory of Molecular Biology
Many genomic and genetic studies face the challenge of searching for differences in gene dosage or expression among cell populations. For example malignancies may involve the loss or gain of DNA sequences or abnormal expression levels of particular genes, reproductive abnormalities frequently result from loss or gain of chromosome segments, differential gene expression underlies development, and genetics frequently requires mapping duplicated or deleted regions of mutant genomes. Several years ago comparative genomic hybridization (CGH), which permits surveying entire genomes for regions of variant gene dosage was developed. In current practice two genomes, for example a test and a normal reference genome, are labeled with different fluorochromes and simultaneously hybridized to normal metaphase chromosomes. Regions of the test genome with elevated or reduced copy number are indicated by corresponding variations in the ratio of the hybridization signals along the target chromosomes. This technique has proven very powerful, but it has substantial limitations that stem from the use of metaphase chromosomes as the hybridization target. The most fundamental of these is that the genomic resolution is at best several Mb due to the complex packing of DNA in metaphase chromosomes. This determines the precision with which copy number variations can be mapped and the sensitivity with which changes affecting a small region can be detected.
To address these problems we are developing the ability to make arrays of cloned DNA that are suitable for comparative fluorescence hybridizations with probes of total mammalian genomic complexity. Currently each element of the array contains approximately 100 pg of target DNA in a 50-100 mm diameter spot on a glass microscope slide. Probes are labeled with fluorescein or Texas red. Test hybridizations in a model system consisting of lambda DNA targets hybridized with red and green labeled lambda DNA probes in various ratios indicate that the ratio of the fluorescence intensities is accurately proportional to the ratio of the two probe concentrations over a range of more than 103 in relative concentration. Signals can be detected with probe concentrations down to 2 pg/ml, which is equivalent to the concentration of a 50 kb length of DNA in a CGH hybridization involving mammalian genomic DNA. The amplification of chromosome 20q in breast cancer cell line BT474 was detected using a 4 element array containing total human genomic DNA, a P1 clone from chromosome 20q13, a P1 clone from chromosome 18, and lambda DNA. The fluorescence ratio on the 20q13 target was 2-4 times greater than that on the chromosome 18 and total genomic DNA targets. No hybridization was detectable on the lambda target. These measurements indicate the potential for using arrayed targets for quantitative comparisons of relative concentrations of multiple nucleic acid sequences in high complexity mixtures.
Supported by a grant from the director, Office of Energy Research and Development, Office of Health and Environmental Research, Department of Energy, under contract DE-AC-0376SF00098; Vysis Inc. by subcontract from National Institute of Science and Technology, Department of Commerce; and NIH grant CA45919.