Reverse transcription-free digital-quantitative-PCR for microRNA analysis

Abstract

MicroRNAs (miRNAs) are non-coding RNA sequences that regulate many biological processes and have become central targets of biomedical research. However, their naturally low abundances in biological samples necessitates the development of sensitive analytical techniques to conduct routine miRNA measurements in research laboratories. Digital PCR has the potential to meet this need because of its single-molecule detection capabilities, but PCR analyses of miRNAs are slowed by the ligation and reverse transcription steps first required to prepare samples. This report describes the development of a method to rapidly quantify miRNA in digital microwell arrays using base-stacking digital-quantitative-PCR (BS-dqPCR). BS-dqPCR expedites miRNA measurements by eliminating the need for ligation and reverse transcription steps, which reduces the time and cost compared to conventional miRNA PCR analyses. Under standard PCR thermocycling conditions, digital signals from miRNA samples were lower than expected, while signals from blanks were high. Therefore, a novel asymmetric thermocycling program was developed that maximized on-target signal from miRNA while minimizing non-specific amplification. The analytical response of BS-dqPCR was then evaluated over a range of miRNA concentrations. The digital PCR dimension increased in signal with increasing miRNA copy numbers. When the digital signal saturated, the quantitative PCR dimension readily discerned miRNA copy number differences. Overall, BS-dqPCR provides rapid, high-sensitivity measurements of miRNA over a wide dynamic range, which demonstrates its utility for routine miRNA analyses.

Figure 1.

Cartoon showing the sequential steps in Cycle 1 of BS-PCR. All components are first sealed into individual microwells, and PCR is then initiated. After melting, miRNA (blue) anneals to the DNA guide (orange), which stabilizes annealing of forward primer (black/green) to the guide. DNA polymerase binds to this base-stacking complex, displaces the miRNA, and extends the guide sequence into a full-length amplicon (red/black/green).

Figure 2.

(A) Top-down SEM image of a region of a microwell array at 4,000x magnification. (B) Tilted SEM image of an array device loaded with beads at 4,000x magnification. The dimensions of the microwells permit only a single bead to be loaded into each microwell. (C) Top-down fluorescence image of a region of a microwell array after BS-PCR at 8x magnification. The autofluorescence of beads is observed in every loaded microwell. Microwells originally containing a target miRNA also exhibit fluorescence in the signal acquisition region after BS-PCR amplification.

Figure 3.

DNA guide number was evaluated by measuring the digital signal from positive controls (4 copies of let-7a per microwell) and negative controls (0 copies of let-7a per microwell). Increasing guide numbers yielded more base-stacking complex in the positive controls but also caused more non-specific amplification in the negative controls. A 58 °C PCR annealing temperature was used in this study.

Figure 4.

The Cycle 1 annealing time and temperature were evaluated to determine their combined effect on the digital signal of a BS-PCR assay. The highest P/N ratio was attained using a cold-long-start where the Cycle 1 annealing time was 300 s and the temperature was 53 °C. Annealing for Cycles 2-35 occurred at 69 °C.

Figure 5.

qPCR curves for positive controls (green) and negative controls (gray) using uniform thermocycling (dashed lines) with annealing at 58 °C for 20 s and asymmetric thermocycling (solid lines) with a 300-s long-start combined with a 53 °C cold-start for Cycle 1 and annealing at 69 °C for 20 s for Cycles 2-35. Each trace depicts the average fluorescence from all active wells at each thermocycle.

Figure 6.

(A) Digital signal (blue circles) and C_q values (red triangles) from BS-PCR were measured over a range of let-7a copy numbers in microfluidic microwell arrays. Concentration-dependent responses are observed in both the dPCR and qPCR dimensions. (B) C_q values from bulk BS-qPCR analyses over a range of let-7a copy numbers analyzed using a commercial qPCR instrument. A concentration-dependent response was not observed over this broad concentration range.

Scheme 1.

Side-view cartoon illustrating the steps of a BS-PCR assay (sizes not to scale). (A) Beads conjugated with DNA guides are loaded into microwells. (B) PCR master mix containing the target miRNA and primers is introduced into the device. (C) Loaded microwells are sealed with oil to create individual reaction partitions. (D) The PCR hot-start step simultaneously activates the PCR polymerase (as in conventional PCR) and releases DNA guides into solution within the individually sealed microwells.