98c Microfluidic Integration of RNA Sample Preparation and Amplification for Global Gene Expression Profiling of Small Samples

Jason G. Kralj1, Audrey Player2, Ernest Kawasaki2, Dan Edelman2, and Laurie E. Locascio1. (1) Biochemical Sciences, National Institute of Standards and Technology, 100 Bureau Dr., MS 8313, Gaithersburg, MD 20899, (2) National Cancer Institute/National Institutes of Health, 8717 Grovemont Circle, Bethesda, MD 20877

We present the integration of the Eberwine process in a microfluidic system, incorporating small sample mRNA capture, reverse transcription, RNA digestion with second strand synthesis, and linear amplification using T7 RNA polymerase. Each step of the process has been shown to scale down to the single-cell level efficiently, and this is the first demonstration of all steps incorporated into a single device. Multilayer softlithography was used to fabricate 9-channel devices incorporating sieve valve structures, allowing functionalized microbeads to be used for immobilizing the nucleic acids. Here, Jurkat T-cell total RNA was used as a model sample with amounts from 100 ng down to approximately 100 pg. Amplified product from conventional benchtop and microfluidic methods was compared based on transcript length using electrophoresis, efficiency of transcription based on PCR of selected genes, and global gene expression profiles using microarrays. Early results indicate about 850 more genes were detected using the microfluidic process at the 1 ng level (100 cell equivalents), suggesting an order of magnitude improvement in yield of aRNA than conventional methods. Samples as small as 100 pg of total RNA (10 cell equivalents) have been successfully amplified. The main advantages of microfluidic processing include immobilization of the cDNA onto a stationary bead column, thereby eliminating sample transfer steps; RNAse contamination is minimized because the channel is sealed; and higher numbers of genes are detected on microarrays from samples 1 ng and smaller than amplification using benchtop methods.