2Biochemistry & Microbiology, University of Victoria, Victoria, BC, Canada
p19 RNA binding protein from the Carnation Italian ringspot virus (CIRV) is an RNA-silencing suppressor that binds small interfering RNA (siRNA) with high affinity. We created a bifunctional p19 fusion protein with an N-terminal maltose binding protein (MBP), for protein purification, and a C-terminal chitin binding domain (CBD) to bind p19 to chitin magnetic beads. The fusion protein binds dsRNAs in the size range of 20–23 nucleotides, but does not bind ssRNA or dsDNA. Relative affinities of the p19 fusion protein for different-length RNA and DNA substrates were determined. Binding specificity of the p19 fusion protein for small dsRNA allows detection of miRNA:RNA probe duplexes. Using radioactive RNA probes, we were able to detect low levels of miRNAs in the sub-femtomole range and in the presence of a million-fold excess of total RNA. Detection is linear over three logs. Unlike most nucleic acid detection methods, p19 selects for RNA hybrids of correct length and structure. Rules for designing optimal RNA probes for p19 detection of miRNAs were determined by in vitro binding of 18 different dsRNA oligos to p19. These studies demonstrate the potential of p19 fusion protein to detect miRNAs and isolate endogenous siRNAs.
MicroRNAs (miRNAs) are cleaved from RNA hairpins to generate dsRNAs 21–22 nucleotides long that associate with the RNA-induced silencing complex (RISC). microRNAs generally act by binding to a complementary region in the 3′ untranslated region of mRNA to reduce protein expression. In plants, miRNA binding directs specific cleavage of the mRNA (1). Biological systems regulated by miRNAs include embryonic differentiation, hematopoiesis, cardiac hypertrophy, and a variety of cancers. Measuring miRNA levels in cancer is important in both diagnosis and prognosis of the disease (2).
To evaluate the role of miRNAs in gene expression, an accurate quantitative method of detection is needed. Existing methods using microarrays (3,4,5), single-molecule detection with lasers (6), or microsphere detection of miRNAs (7) often require expensive equipment. Other methods that use multiple enzymatic steps including RNA ligation, amplification, and gel electrophoresis (8,9,10,11) are susceptible to quantitative differences in detection. In this report, we describe a novel and simplified method for miRNA detection that utilizes the small interfering RNA (siRNA) binding properties of the p19 protein from the Carnation Italian ringspot virus (CIRV), a plant tombusvirus. The protein's binding activity was discovered by its ability to block RNAi in plants (12,13,14). The dimeric p19 binds dsRNA in a size-dependent sequence-independent manner (15,16).
Hybridization of a miRNA-specific probe to a single-stranded target miRNA creates dsRNA that tightly binds the p19 fusion protein. The amino terminus of p19 was fused to maltose binding protein (MBP) to aid purification (17) and the carboxy terminus was fused to chitin binding domain (CBD) for tight binding to chitin magnetic beads (18). The recombinant fusion protein retains the ability to selectively bind siRNA. The p19 fusion protein did not bind ssRNA, rRNA, or dsDNA, and showed the greatest affinities for 19- and 20-nucleotide dsRNAs with blunt ends, as well as 21-nucleotide siRNA with 5′ phosphates and 3′ 2-nucleotide extensions. We describe a novel miRNA detection method which is based on selective binding of p19 fusion to siRNAs. The method is linear over a three-log range and has a sensitivity of 0.1 fmoles. It is significantly faster and more sensitive than Northern blot analysis.
Materials and methods
In vitro RNA binding with p19 fusion
Size-specific binding of dsRNA to p19 fusion was measured using a gel mobility shift. Increasing amounts of MBP-p19-CBD fusion protein (0–3 µg) were bound to 90 ng of an equal mixture of 17-, 21-, and 25-nucleotide dsRNAs. The MBP-p19-CBD fusion protein was obtained from New England BioLabs (NEB; Ipswich, MA, USA). All dsRNAs contained 5′ phosphates and 3′ 2-nucleotide extensions. RNA was bound to the p19 fusion at room temperature (RT) for 2 h in 1× p19 binding buffer [20 mM Tris-HCl, pH 7.0, 1 mM EDTA, 1 mM TCEP [tris (2-carboxy-ethyl) phosphine], 100 mM NaCl, 0.02% Tween-20] with 10 units of murine RNase inhibitor (NEB) and 1× BSA (0.1 mg/mL) in a total volume of 20 µL. The reactions were mixed with 5 µL 5× loading buffer (15% Ficoll 400 and 5 mM EDTA) and analyzed by electrophoresis on 20% TBE acrylamide gel (Invitrogen, Carlsbad, CA, USA) for 2 h at 100 V in 1× TBE buffer followed by ethidium bromide staining.
siRNA isolation using p19 beads
The selectivity of siRNA binding to p19 chitin magnetic beads (NEB) in a large excess of total RNA was measured with 21-bp siRNA oligos (Table 1, row 1). Up to 55 µg rat liver RNA and 10 ng 21-mer siRNA was incubated with 10 µL p19 beads suspended in 1× p19 binding buffer with RNase inhibitor and BSA (see “In vitro RNA binding with p19 fusion” section), and shaken at RT for 1–2 h in an Orbis shaker (MarketLab, Caledonia, MI, USA). Unbound RNA was removed by washing 6 times in 600 µL 1× p19 wash buffer (identical to 1× p19 binding buffer except without Tween-20 and TCEP). For each wash, the beads were shaken for 5 min at 37°C. The bound siRNA was eluted from beads in 20 µL 1× p19 elution buffer (20 mM Tris-HCl pH 7.0, 100 mM NaCl, 1 mM EDTA, and 0.5% SDS) by mixing on an Eppendorf thermo-mixer (New Brunswick Scientific, Edison, NJ, USA) for 20 min at 37°C. The eluted RNA was analyzed by electrophoresis on a 20% acrylamide gel in 1× TBE buffer.