Discovery, evaluation, and understanding the biological relevance of single nucleotide polymorphisms (SNPs) and their associated phenotypes is relevant to many applications, including human disease diagnostics, pathogen detection, and identification of genetic traits impacting agricultural practices, both in terms of food quality and production efficiency. Validation of putative SNP associations in large-scale cohorts is currently impeded by the technical challenges and high cost inherent in analyzing large numbers of samples using available SNP genotyping platforms. We describe in this report the implementation of the 5′-exonuclease, biallelic PCR assay for SNP genotyping (TaqMan) in a nanofluidic version of a high-density microplate. System performance was assessed using a panel of 32 TaqMan SNP genotyping assays targeted to human polymorphisms. This functional test of the nanoliter fluidic SNP genotyping platform delivered genotyping call rates and accuracies comparable to the same larger volume reactions in microplate systems.
Single nucleotide polymorphisms (SNPs) are the most abundant and accessible class of biallelic polymorphisms for recording genetic variation in a population and offer considerable advantages over highly polymorphic, repetitive loci (i.e., short tandem repeats and microsatellites) because of their lower mutation rates and nearly uniform genomic distribution. The popularity of SNP genotyping has increased with the emergence and widespread availability of technologies for whole-genome SNP identification (1,2,3,4) and has resulted in a growing number of SNP databases for a wide range of organisms, including humans (www.ncbi.nlm.nih.gov/projects/SNP). The shifting focus from discovery to utilization of SNP genotypes for cohort phenotype classification and segmentation has led to applications in human disease diagnostics (5,6); pharmacogenetics (7,8); blood typing (9,10); DNA fingerprinting for criminal or parental verification (11); fisheries management (12,13); plant and human pathogen detection and identification (14,15); and marker-assisted crop and animal breeding (16,17,18).
A number of strategies have been proposed for scaling SNP genotyping to sample throughputs in the range of hundreds to thousands of specimens and analyzed with focused sets of ten to hundreds of predictive SNPs (19,20). Here, we describe an alternative high-throughput genotyping system based on a highly parallel implementation of well-established end-point TaqMan PCR SNP genotyping assays in a miniaturized nanofluidic system. TaqMan (Applied Biosystems, Foster City, CA, USA) is a homogenous, biallelic PCR assay utilizing 5′-hydrolysis of two different dark quenched fluorescent probes (21) to distinguish between SNPs on one or both alleles. Each TaqMan SNP assay used in this study is labeled with VIC, a fluorescent dye from Applied Biosystems, and 5-carboxyfluorescein (FAM). The combined specificity of PCR primer and probe leads to a high genotyping accuracy and call rate.
Although there have been numerous reports of PCR micro- and nanofluidic systems (22), here we employ a nanotiter plate, a configuration that readily scales to higher genotyping throughputs without the need for advanced automation and that avoids the significantly higher reagent consumption of PCR in microtiter plates (23,24). Previously described for use in SYBR Green–based real-time PCR measurement of gene expression levels (25), the nanotiter plate is similar in concept to a microtiter plate whereby a rectilinear array of 3072 through-holes grouped in 48 subarrays of 64 through-holes each is micromachined in a stainless steel platen the size of a microscope slide (26,27,28) (Figure 1). Each through-hole supports implementation of an independent 33 nL PCR assay, thus enabling a large number of PCR measurements to be made simultaneously on one or more samples. One advantage of genotyping a specimen with multiple, independent TaqMan PCR assays is that each assay in an assay panel can be designed and optimized under a uniform set of conditions for maximum sensitivity and specificity. Furthermore, if the SNP panel changes in the future, then assays can be independently replaced without the need to re-optimize the entire assay set.
Materials and methods Nanotiter plate system
The nanotiter plate, which is commercially available as the OpenArray system (BioTrove Inc., Woburn, MA, USA), features a stainless steel (317 stainless steel) platen the size of a microscope slide (25 mm × 75 mm × 0.3 mm) etched to form a rectilinear array of 3072 holes with diameter 320 µm, which are grouped in 48 subarrays of 64 through-holes each. The interior surface of each etched hole is made hydrophilic and the exterior surface of the plate is made hydrophobic by covalently linked polymer coatings. Wetting of the hydrophilic surface wicks fluid into each hole and capillary forces act against gravitational and inertial forces to retain the liquid in each hole in isolation.TaqMan SNP assays
TaqMan SNP assays were dispensed into microplates for transfer into a nanotiter plate at a final 1× concentration by an automated, four-axis robotic fluid transfer system. This system is based on an array of 48 slotted pins in which each pin transfers a 33-nL aliquot of assay reagent from one well in the source plate to multiple holes in the nanotiter plate. Assay solvent is evaporated under controlled environmental conditions and the loaded plates are stored at -20°C in a vacuum-sealed Mylar bag. Thirty-two TaqMan assays were deposited in duplicate in 64 through-holes of a subarray. A list of the TaqMan SNP assays and CEPH DNA samples used in the study are found in Supplementary Tables 1 and 2, respectively.