Large-Scale Generation of Synthetic DNA Libraries: Sequence-Specific Priming of Reverse Transcription.

Abstract

Custom-designed DNA and RNA oligonucleotide collections are used as building blocks in synthetic biology, complementary probes for targeted sequencing, and nucleic acid aptamers. They encode information for technologies like RNA interference, protein engineering and DNA-encoded chemical libraries. These applications require an economical source of diverse libraries.

High-throughput microarray technology produces hundreds of thousands of diverse sequences on a single planar substrate at low cost (<$200 for 20,000 oligonucleotides). However, the quantity and quality of individual oligonucleotides obtained from this technology is low (~ 5 fmol). In order to produce feasible amounts of high quality oligonucleotides (10 – 100 pmol), it is necessary and more economical to go through a molecular amplification procedure. Existing methods of amplification lead to the formation of chimeric products. We developed a 2-step amplification process which gives an overall ~450 fold increase in the amount of single-stranded oligonucleotides. The first-step, emulsion polymerase chain reaction (PCR), provides the initial amplification while preserving the library complexity (~25 μM product). However, the overall product yield from a single reaction after removal of primer binding sequences and single-strand DNA formation is limited (~15 pmol) as a result of restrictions imposed on the input reactant quantity. Therefore, this step alone is labor intensive to produce large amounts of products due to lack of scalability. This limitation is addressed via transcription-reverse transcription of the emulsion PCR amplicons. This two stage process allows scalability of microarray synthesized oligonucleotides for preparation of large quantities of ssDNA libraries (~6500 pmol).

As an application of the amplification technology we developed an integrated method for cDNA library construction with sequence-specific primers incorporating a unique tag and universal primer sequence. The method suffers from the formation of three types of false positives that need to be sufficiently removed to reduce contribution of false-positives signals. A 3 step process is implemented to reduce the false positives contributors and still detect differential expression of yeast genes in galactose and glucose conditions. The sequence-specific primers libraries can be used for applications not limited to detection of low abundance and rare RNA and identification of aberrant splicing variants and gene-fusions.