S added to the total RNA as a template. The Cq value showed no significant difference 15900046 with or without the presence of genomic DNA (student’s t-test, P.0.05) (Figure 3A B), and produced one peak during the melting curve analysis, respectively (Figure 3C). In contrast, the linear adapter (miScript) gave a different result; double-stranded DNA had a significant effect on the Cq values (Figure 3A B) forming two peaks in the melting curve analysis (Figure 3C). For ulterior validation and application, 102?07 cells of the bone marrow samples and SiHa cells wereheat-lysed as described in the Methods, and added directly as the substrate for polyuridination. We found a significant correlation between the Cq values and the added cell number (R2.0.993) (Figure 3D). Melting curve analysis produced only one peak (Figure 3E). Figure 3D presented good concordance of the Cq values between the purified total RNA and Table 1. Comparison of amplification Epigenetics efficiency of proposed assay with miRCURY method on SiHa cell total RNA.Advantage of Poly(U) TailWe used a poly(U) tail instead of the usual poly(A) tail, and it provided more convenience and specificity. miRNAs were tailed by poly(U) and lacked the poly(A) tail, and therefore would not anneal to the ordinary oligo(T) RT primer. Conversely, the poly(A) tail of the mRNA was still present, and could bind to the oligo(T) RT primer. To prove the advantage of the U-tailing, we performed a multiplex RT reaction of mRNA and miRNAs within the same run. As shown in Figure 5, mRNA existence did not alter Cq values of microRNAs (student’s t-test, P.0.05), indicating that mRNA and miRNAs did not affect each other when combined in the same RT reaction. The inhibitor dissociation curve showed a similar situation. Many commercially available strategies of mRNA RT reactions exploit the characteristics of the mRNA-poly(A) tail. The employment of poly(U) tail of miRNAs maked it convenient to use these strategies in the quantification of mRNA of target genes and miRNAs in the same system.microRNA miR-455 miR-32 miR-181a miR-181b miR-126 let-7a let-7b let-7c let-7d let-7fproposed assay ( ) miRCURY ( ) 96 99 97 98 92 98 90 105 101 94 93 
91 89 91 87 95 86 88 92 85 90 (3.2)difference ( ) 2.8 7.9 8.1 6.8 9.2 3.1 4.2 16.7 9.2 8.9 7.7 (4.0)average (SD) 97 (4.3) doi:10.1371/journal.pone.0046890.tFacile and Specific Assay for Quantifying MicroRNAFigure 3. Effect of double-stranded genomic DNA on the miRNA assay. (A) Amplification plot of hsa-miR-32 miRNA from SiHa cell total RNA with and without double-stranded genomic DNA using stem-loop or linear adapter RT primer. (B) Comparison with and without genomic DNA in the two RT reaction systems for real-time quantification of 6 miRNAs. (C) Dissociation curve analysis of the same experiment of panel A. (D) ComparisonFacile and Specific Assay for Quantifying MicroRNAof heat-treated SiHa cells, heat-treated bone marrow and purified total RNA of bone marrow for real-time quantification of hsa-miR-32 miRNA. (E) Dissociation curve analysis of the same experiment of panel D. (B D) The level of miRNA expression is measured in the quantification cycle (Cq). Each value represents the mean (6 SD) of three measurements. doi:10.1371/journal.pone.0046890.gmiRNA Expression Profile of Four miRNAs in Mouse TissuesOptimization of the proposed miRNA quantification technique was required for practical applications. As well as experimental validation of the assay, it needed to be validated with biological sample.S added to the total RNA as a template. The Cq value showed no significant difference 15900046 with or without the presence of genomic DNA (student’s t-test, P.0.05) (Figure 3A B), and produced one peak during the melting curve analysis, respectively (Figure 3C). In contrast, the linear adapter (miScript) gave a different result; double-stranded DNA had a significant effect on the Cq values (Figure 3A B) forming two peaks in the melting curve analysis (Figure 3C). For ulterior validation and application, 102?07 cells of the bone marrow samples and SiHa cells wereheat-lysed as described in the Methods, and added directly as the substrate for polyuridination. We found a significant correlation between the Cq values and the added cell number (R2.0.993) (Figure 3D). Melting curve analysis produced only one peak (Figure 3E). Figure 3D presented good concordance of the Cq values between the purified total RNA and Table 1. Comparison of amplification efficiency of proposed assay with miRCURY method on SiHa cell total RNA.Advantage of Poly(U) TailWe used a poly(U) tail instead of the usual poly(A) tail, and it provided more convenience and specificity. miRNAs were tailed by poly(U) and lacked the poly(A) tail, and therefore would not anneal to the ordinary oligo(T) RT primer. Conversely, the 
poly(A) tail of the mRNA was still present, and could bind to the oligo(T) RT primer. To prove the advantage of the U-tailing, we performed a multiplex RT reaction of mRNA and miRNAs within the same run. As shown in Figure 5, mRNA existence did not alter Cq values of microRNAs (student’s t-test, P.0.05), indicating that mRNA and miRNAs did not affect each other when combined in the same RT reaction. The dissociation curve showed a similar situation. Many commercially available strategies of mRNA RT reactions exploit the characteristics of the mRNA-poly(A) tail. The employment of poly(U) tail of miRNAs maked it convenient to use these strategies in the quantification of mRNA of target genes and miRNAs in the same system.microRNA miR-455 miR-32 miR-181a miR-181b miR-126 let-7a let-7b let-7c let-7d let-7fproposed assay ( ) miRCURY ( ) 96 99 97 98 92 98 90 105 101 94 93 91 89 91 87 95 86 88 92 85 90 (3.2)difference ( ) 2.8 7.9 8.1 6.8 9.2 3.1 4.2 16.7 9.2 8.9 7.7 (4.0)average (SD) 97 (4.3) doi:10.1371/journal.pone.0046890.tFacile and Specific Assay for Quantifying MicroRNAFigure 3. Effect of double-stranded genomic DNA on the miRNA assay. (A) Amplification plot of hsa-miR-32 miRNA from SiHa cell total RNA with and without double-stranded genomic DNA using stem-loop or linear adapter RT primer. (B) Comparison with and without genomic DNA in the two RT reaction systems for real-time quantification of 6 miRNAs. (C) Dissociation curve analysis of the same experiment of panel A. (D) ComparisonFacile and Specific Assay for Quantifying MicroRNAof heat-treated SiHa cells, heat-treated bone marrow and purified total RNA of bone marrow for real-time quantification of hsa-miR-32 miRNA. (E) Dissociation curve analysis of the same experiment of panel D. (B D) The level of miRNA expression is measured in the quantification cycle (Cq). Each value represents the mean (6 SD) of three measurements. doi:10.1371/journal.pone.0046890.gmiRNA Expression Profile of Four miRNAs in Mouse TissuesOptimization of the proposed miRNA quantification technique was required for practical applications. As well as experimental validation of the assay, it needed to be validated with biological sample.
