Our proposal is in response to NCI's solicitation PAR-98-066, Innovative technologies for the molecular analysis of cancer. We are developing a DNA microarray synthesizer capable of synthesizing DNA oligonucleotides in situ on two dimensional substrates. This technology will provide cost-effective, medium density, flexible design hybridization arrays not currently available in any other array format to support gene- expression monitoring, genotyping and sequence checking applications. Our system combines drop-on-demand inkjet reagent delivery with a chemically modified substrate. Surface tension constrains active chemistry to predefined locations. We plan to extend the method to include covalent immobilization of presynthesized nucleic acids. Preliminary results from experiments with a prototype synthesizer show successful in situ synthesis of 20mer homopolymers, 20mer mixed sequence oligonucleotides and a 40mer-oligonucleotide sequence. This was confirmed by capillary electrophoresis of synthesized products and by hybridizing with complementary oligonucleotides and PCR products. We have synthesized 400 and 1600 feature arrays on 16 cm2 surfaces. This DNA microarray technology is unique in its ability to support array design flexibility, manufacturing economy and incorporation of customized nucleotide chemistries. We propose to engineer the prototype synthesizer into a robust, automated manufacturing unit. We also plan to more fully characterize and optimize substrate surface-tension chemistry and oligonucleotide synthetic chemistry into robust manufacturing processes.
DNA microarrays have become the method of choice for performing highly parallel nucleic acid analysis whether it is for gene expression monitoring or genotyping applications. In addition, microarrays are proving to be useful adjuncts for rapid sequence checking and polymorphism discovery. Methods currently used to manufacture arrays are expensive, primarily due to set up costs for each array design. This may include collecting oligonucleotide or cDNA libraries or designing a set of photolithographic masks. The most cost-effective way to apply these methods is to make many copies of single array designs. The method we propose reduces array design and synthesis to a four-color printing process using inexpensive, commercially available reagents. The low expense and high flexibility should make arrays widely available to accelerate the pace of acquiring gene expression and genotype data. In addition, the array dimensions and feature sizes do not demand expensive, custom optical scanning instrumentation. These arrays are well suited to simple, generic data acquisition and analysis approaches that should enable wider accessibility and broader based adoption of hybridization array based genetic analysis.
