Introduction

Since human genetic variation likely accounts for genetic predispositions to common diseases as well as differential responses to drugs and environmental stimuli, there has been longstanding interest in this field of study. Research in human genetic variation fosters many avenues of genetics including prenatal diagnosis, detection of heterozygous carriers of genetic disease, risk evaluation for common disorders, paternity testing, forensics, and tissue typing. A genetic variant is defined as the occurrence of two or more alleles that appear at a frequency in a given population where the rarest allele can not be maintained by recurrent mutation processes alone. A genetic locus is considered polymorphic if one or more of the rare alleles have a frequency of at least 0.01 (1). The vast amount of genetic variability among humans was long thought to exist in the form of single nucleotide polymorphisms (SNPs). Estimates predicted at least 10 million SNPs in the human population (2,3), averaging one SNP every 300 nucleotides in the approximately 3 billion base pair human genome (4,5). Recently however, genome-wide, array comparative genomic hybridization (CGH) technologies have led to the discovery of a newly appreciated form of genetic variation: copy number variants (CNVs). These CNVs are typified as deletions and/or duplications of DNA segments 1 kb or larger and are ubiquitously found throughout the genome. Moreover, some of these variants are thought to have a significant impact on disease susceptibility and in general, our ability to adapt to the surrounding environment (6,7).

Array CGH: Scanning Genomes at High Resolution for Genetic Imbalances

Array-based CGH (array CGH) is a method that was developed to determine relative copy number of specific DNA sequences in one individual (test DNA) compared to a second, “normal” individual (reference DNA) in a single nonsubjective assay (8). Genomic DNA is isolated from blood (or other tissues) from both individuals, differentially labeled by random priming with specific fluorescent molecular tags (e.g., Cy™5-dCTP tags for the test DNA and Cy3-dCTP tags for the reference DNA) and co-hybridized onto an array containing carefully chosen DNA clones, PCR fragments, or oligonucleotides that recapitulate part of, or the entire, genome at a desired resolution (Figure 1).

Figure 1. Representation of an array-based comparative genomic hybridization (array CGH) experiment. (Click to enlarge)

Hybridization of the labeled DNA to its complementary nucleotide sequence targets on the array generally occurs for at least 16 h at 37°C in a humid chamber, permitting each labeled DNA probe to bind in a competitive and stoichiometric fashion. The hybridization efficiency of a given probe to these targets is proportional to its original concentration in the hybridization mixture. Unlabeled repetitive DNA, typically Cot-1 DNA, is usually added in excess to the probe mixture to reduce/eliminate hybridization signals that result from interspersed and highly repetitive DNA sequences (Figure 1).

After hybridization, the amount of fluorescence from each dye can be quantified at each target DNA spot on the array using an appropriate microarray reader, such as a laser scanner or imaging system equipped with a charge-coupled device (CCD) camera. Both fluorescent dyes are excited with the appropriate wavelengths (e.g., 635 nm for Cy5 and 532 nm for Cy3), and then the amount of emitted fluorescence from each dye is quantified for each DNA spot on the array. Subsequently, the software used to run the scanner provides a 635/532 fluorescence ratio for each DNA spot. Fluorescence ratios of approximately 1.0 are interpreted as similar copy numbers for the DNA sequence in both the test and reference genomes. At a given DNA spot on the array, increased fluorescence from the test DNA-labeled tag compared to the reference DNA-labeled tag (i.e., in the above scenario, the Cy5/Cy3 ratio would be >1.0) would be consistent with a duplication or amplification of that specific DNA sequence in the test individual's genomic DNA. Conversely, reduced fluorescence from the test DNA-labeled tag compared to the reference DNA-labeled tag (i.e., in the above mentioned scenario, the Cy5/Cy3 ratio would be

Many factors can affect the high signal-to-noise fluorescence ratio that one tries to achieve for each array CGH assay. Some of the more critical steps include (i) efficient DNA labeling: genomic DNA should be relatively clean with minimal protein and organic solvent residues prior to labeling. The efficiency of labeling can subsequently be determined by spectrophotometrically measuring the fluorescence of each dye in the probe and calculating the base-to-dye absorbance ratio parameter (representing the number of unlabeled nucleotides between any two labeled nucleotides). This number should ideally be between 40 and 80 for both Cy3 and Cy5 incorporations. The base-to-dye absorbance ratio can be calculated using the following formulas: