Real-Time Polymerase Chain Reaction MicroRNA Detection Based on

May 26, 2009 - of a whole family of miRNA (such as let-7) at one time. Finally ..... (25) Viswanathan, S. R.; Daley, G. Q.; Gregory, R. I. Science 200...
0 downloads 0 Views 306KB Size
Download PDF
Anal. Chem. 2009, 81, 5446–5451

Real-Time Polymerase Chain Reaction MicroRNA Detection Based on Enzymatic Stem-Loop Probes Ligation Juan Li,† Bo Yao,‡,§ Huang Huang,‡ Zhao Wang,‡ Changhong Sun,‡ Yu Fan,‡ Qing Chang,‡ Shaolu Li,‡ Xiang Wang,*,† and Jianzhong Xi*,‡ School of Electronic and Information Engineering, Beihang University, Beijing, China, 100191, and Department of Biomedical Engineering, College of Engineering, Yan Nan Yuan 60, Peking University, Beijing, China, 100871 MiRNAs (microRNAs) are a group of endogenous, small noncoding RNA with the length of 18-25 nucleotides, which have recently been demonstrated to play important roles in a wide range of biological processes. In this work, we developed a simple, sensitive, specific, and inexpensive assay through the combination of enzymatic probe ligation and real-time PCR amplification for the measurement of mature miRNAs. A couple of novel DNA probes with a stem-loop structure were implemented to reduce nonspecific ligation by at least 100-fold. The assay has several remarkable features including wide dynamic range, low total RNA input (0.02-0.2 ng), distinct antiinterference from precursor miRNAs (signal-to-noise ratio > 500), and single-base mismatch discrimination among miRNA sequences. In addition, a one-tube assay could be accomplished by designing a couple of universal probes, which makes it feasible to examine the expression of a whole family of miRNA (such as let-7) at one time. Finally, we validated the method for quantifying the expression of four mature miRNAs including miR-122, miR-1, miR-34a, and let-7a across 10 mouse tissues, where U6 snRNA could be simultaneously examined as an endogenous control. Thus, this method revealed a great potential for miRNA quantitation in ordinary laboratory studies and clinical diagnoses. MiRNAs, as a group of small noncoding RNA with 18∼25 nucleotides (nt) in length, widely distribute in diverse organisms and post-transcriptionally regulate protein production through mRNA cleavage or translation inhibition. Over the past decade,1-3 several thousand of conserved and nonconserved miRNAs have been identified in different species from algae to animals. Although many evidence clearly indicate that miRNAs play important roles * To whom correspondence should be addressed. E-mail, [email protected]; phone/fax, +86-10-82338874 (X.W.). E-mail, [email protected]; phone/fax, +8610-62760698 (J.X.). † Beihang University. ‡ Peking University. § Present address: Department of Chemistry, Zhejiang University, Hangzhou, 310058, China. (1) Lagos-Quintana, M.; Rauhut, R.; Lendeckel, W.; Tuschl, T. Science 2001, 294, 853–858. (2) Lau, N. C.; Lim, L. P.; Weinstein, E. G.; Bartel, D. P. Science 2001, 294, 858–862. (3) Lee, R. C.; Ambros, V. Science 2001, 294, 862–864.

5446

Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

in a wide range of biological processes including proliferation, development, tumorigenesis. and viral infection,4-6 our understanding of the physiological functions of miRNAs still remains in its infancy. The explosion of interest into unveiling the role of miRNAs in complicated signal regulation networks has created a great demand to develop sensitive and accurate miRNA detection methods. A reliable and inexpensive miRNA detection kit may become more attractive because of its great potential in biological research and clinical diagnosis. However, the detection of miRNAs is a challenging task given their small size, and the sequence similarity between some miRNA members (i.e., let-7 family). Although Northern blot analysis was usually regarded as a gold standard method in early miRNA profiling studies,7-9 it is time and labor consuming and moreover demands a large amount of total RNA sample especially when monitoring the expression of low-abundant miRNAs. Besides that, there were a large number of miRNA detection methods reported in past few years.10-15 Compared with those approaches, realtime PCR is no doubt a powerful tool for accurate and quantitative gene expression. To overcome the aforementioned difficulties, miRNA was extended by adding poly(A) tail16 or utilizing a long DNA probe.17,18 However, these methods had to employ modified (4) Ma, L.; Teruya-Feldstein, J.; Weinberg, R. A. Nature 2007, 449, 682–689. (5) van Rooij, E.; Sutherland, L. B.; Qi, X.; Richardson, J. A.; Hill, J.; Olson, E. N. Science 2007, 316, 575–579. (6) Yi, R.; Poy, M. N.; Stoffel, M.; Fuchs, E. Nature 2008, 452, 225–230. (7) Lee, R. C.; Feinbaum, R. L.; Ambros, V. Cell 1993, 75, 843–854. (8) Lagos-Quintana, M.; Rauhut, R.; Yalcin, A.; Meyer, J.; Lendeckel, W.; Tuschl, T. Curr. Biol. 2002, 12, 735–739. (9) Va´lo´czi, A.; Hornyik, C.; Varga, N.; Burgya´n, J.; Kauppinen, S.; Havelda, Z. Nucleic Acids Res. 2004, 32, e175. (10) Neely, L. A.; Patel, S.; Garver, J.; Gallo, M.; Hackett, M.; McLaughlin, S.; Nadel, M.; Harris, J.; Gullans, S.; Rooke, J. Nat. Methods 2006, 3, 41–46. (11) Jonstrup, S. P.; Koch, J.; Kjems, J. RNA 2006, 12, 1747–1752. (12) Maroney, P. A.; Chamnongpol, S.; Souret, F.; Nilsen, T. W. RNA 2007, 13, 930–936. (13) Thomson, J. M.; Parker, J.; Perou, C. M.; Hammond, S. M. Nat. Methods 2004, 1, 1–7. (14) Liang, R. Q.; Li, W.; Li, Y.; Tan, C.; Li, J. X.; Jin, Y. X.; Ruan, K. C. Nucleic Acids Res. 2005, 33, e17. (15) Lee, I.; Ajay, S. S.; Chen, H.; Maruyama, A.; Wang, N.; McInnis, M. G.; Athey, B. D. Nucleic Acids Res. 2008, 36, e27. (16) Shi, R.; Chiang, V. L. Biotechniques 2005, 39, 519–525. (17) Chen, C.; Ridzon, D. A.; Broomer, A. J.; Zhou, Z.; Lee, D. H.; Nguyen, J. T.; Barbisin, M.; Xu, N. L.; Mahuvakar, V. R.; Andersen, M. R. Nucleic Acids Res. 2005, 33, e179. (18) Raymond, C. K.; Roberts, B. S.; Garrett-Engele, P.; Lim, L. P.; Johnson, J. M. RNA 2005, 11, 1737–1744. 10.1021/ac900598d CCC: $40.75  2009 American Chemical Society Published on Web 05/26/2009

or LNA modified probe, was utilized to quantify PCR amplification products, thus considerably reducing the detection cost. We finally validated this method for quantifying the expression of four mature miRNAs across 10 mouse tissues, as well as direct measurement of U6 snRNA as an endogenous control.

Figure 1. Schematic diagram of the enzymatic ligation-based realtime PCR assay for measurement of mature miRNAs. A couple of oligonucleotide probes with a stem-loop structure (probe1 and probe2) were created to hybridize with target miRNA. Then the probes were ligated in the presence of T4 DNA ligase at 16 °C overnight. The joint DNA strand was used as a template in the following PCR amplification. Quantification was performed on an Opticon2 DNA Engine by real time data acquisition.

probes such as TaqMan or LNA probes to ensure the detection specificity, thus dramatically increasing the experimental cost. In this article, we presented a real-time PCR miRNAs detection method based on enzymatic ligation of DNA stem-loop probes. T4 DNA ligase efficiently catalyzed the joining of 5′-phosphomonoester group and 3′-hydroxyl group of two stem-loop probes in the presence of miRNA as a template. The implementation of stemloop probes not only prevented their self-ligation and nonspecific hybridization with precursor miRNAs but also allowed precise distinction of a single mutant in the miRNA sequence. Our result clearly showed that pre-miRNAs (precursor miRNAs) could produce no more than 0.2% signal of an equal amount of their mature miRNAs. Furthermore, SYBR Green I, rather than TaqMan

EXPERIMENTAL PROCEDURES Tissue Preparation and Total RNA Extraction. Dissected organs including brain, heart, liver, lung, kidney, spleen, pancreas, small intestine, skeletal muscle, and testis from 8-week-old mice were used for preparing mouse total RNA samples. All animal procedures were in accordance with Institutional Animal Care and Use Committee (IACUC) and OECD guidelines. Total RNA was extracted from ∼30 mg of tissues using TRNzol (Tiangen, Beijing, China) according to the manufacturer’s protocol. The concentration of total RNA was quantified by the absorbance at 260 nm with a NanoDrop Spectrophotometer (ND-1000, ThermoFisher, MA). Mature and Precursor miRNAs, Probes, and Primers. Precursor and mature miRNA genes including the let-7 family, mir-1, mir-34a, and mir-122 were selected from the Sanger Center miRBase at http://microrna.sanger.ac.uk/sequences. Synthetic mature miRNA oligonucleotides were obtained from Shanghai GenePharma (Shanghai, China). All precursor miRNAs were produced by in vitro transcription. DNA probes and primers for miRNA recognition and PCR amplification were purchased from Invitrogen (Shanghai, China). All oligonucleotides were purified by polyacrylamide gel electrophoresis, and the sequences are available in the section of the Supporting Information (Tables S1 and S2). Nucleic acid concentrations were determined by absorption readings at 260 nm on a NanoDrop Spectrophotometer (ND100, ThermoFisher, MA). In Vitro Transcription Reaction. A total of 50 pmol of forward and reverse DNA oligomers (Supporting Information, Table S3) incubates at 75 °C for 5 min, then slowly cools to room temperature (∼30 min). The fills-in reaction for forming dsDNA templates is performed in a 20 µL volume containing 10 mM Tris-HCl, 50 mM NaCl, 10 mM MgCl2, 1 mM DTT, 2.5 µM dNTPs, and 5 U Klenow fragment (3′-5′ exo) (New England BioLabs, Eastwin Scientific, China) at 37 °C for 1 h. Then the reaction mixture is heated at 75 °C for 20 min to inactivate the enzyme and then slowly cooled to room temperature for dsDNA annealing.

Figure 2. Nonspecific ligation of probes with different structures was estimated by real time PCR amplification. Probes with a stem-loop structure (Sp 1/2) and a linear structure (Lp 1/2) were exposed to the ligation reaction and PCR amplification in the presence of 107 copies of miRNA 122, respectively. Pure Sp 1/2 and pure Lp 1/2 were the negative controls where no target miRNA was added. Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

5447

A volume of 20 µL of aforementioned dsDNA templates mixture is added into 30 µL of in vitro transcription buffer containing 0.5 mM NTPs, 40 mM Tris-HCl, 6 mM MgCl2, 10 mM DTT, 2 mM spermidine, 40∼200 U ribonuclease inhibitor (TAKARA, Dalian, China), and 50 U T7 RNA polymerase (New England BioLabs, Eastwin Scientific, China). The reaction runs at 37 °C for 4 h. After that, 1 U of RNase-Free DNase I (Fermentas, Shenzhen, China) is added to digest DNA templates. Finally, the precursor miRNAs in the transcription mixture are purified with phenol/chloroform extraction by disposing of the extra salt and proteins. The transcripted precursor miRNAs are examined by 2% agarose gel, and the concentration is determined from the absorption at 260 nm with a NanoDrop spectrophotometer. Ligation Reactions. A total of 25 fmol of DNA probes (probe1, probe 2 with a 5′-phosphorylated group) were first mixed with 1 µL of detection samples and incubated at 65 °C for 3 min, then slowly cooled to room temperature (∼45 min), and finally placed on ice for the formation of nicked heteroduplex structures. Enzymatic ligation was carried out as previously described20 with slight modification. Freshly prepared buffer of 10 mM MnCl2, 10 mM Tris HCl pH 7.5, 10 µM ATP, 0.5 µL RNase inhibitor (TAKARA, Dalian, China), and 70 U/µL T4 DNA ligase (1.9 weiss unit/µL, TAKARA, Dalian, China) was added to the above mixture to a final volume of 10 µL, and the solution was incubated at 16 °C overnight. The ligation reaction was ended by heating at 70 °C for 20 min. Real Time PCR Assay. Real-time PCR was performed on an Opticon2 DNA Engine (Bio-Rad Laboratories, CA). The ligation product was pipetted as PCR templates, and 10 µL of TransStartTM SYBR Green qPCR Supermix (TransGen Biotech, Beijing, China) and 0.8 µL of 10 µM forward and reverse primers were mixed and fixed to a final volume of 20 µL. The reactions were incubated in a 96-well plate at 95 °C for 2 min, followed by 40 cycles of 94 °C for 15 s, 60 °C for 30 s, and 72 °C for 30 s. All reactions were run in triplicate. The threshold cycle (CT) was defined as the fractional cycle number at which the fluorescence passed the fixed threshold. Data Processing Methodology. For data analysis, a 10-fold dilution series of synthetic miRNAs were used as a template for realtime PCR to generate a plot of log copy numbers of miRNAs versus the corresponding CT. The slope of the linear plot was defined as -(1/log E), where E was the amplification efficiency. U6 sn RNA was employed as the house keeping gene and detected to demonstrate equal loading. The quantity of miRNA was calculated from the calibration curve, and relative quantification of miRNA expression was based on the Pfaffl formula.19 ratio )

(EmicroRNA)∆CTmicroRNA(calibrator-sample) (EU6)∆CTU6(calibrator-sample)

(1)

where, EmicroRNA and EU6 were the amplification efficiency of target miRNA and U6, respectively. Real time PCR efficiency was calculated according to E ) 10[-1/slope]

(2)

%E ) (E - 1) × 100%

(3)

and

5448

Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

Figure 3. Electrophoresis diagram of ligation products for mir-122 and several other miRNAs containing the same sequence of 2-8 nt across the ligation site. Sequence of miRNAs were laid out (top), where the same sequences were highlighted in red. The arrow denotes the ligation site designed for the mir-122 detection probes. Ligation products were examined by 10% polyacrylamide gel electrophoresis at 200 V for 30 min (bottom). A total of 1.2 × 1013 copies of miRNAs were added to each ligation reaction.

The miRNA sample with the lowest CT value, thus the highest expression level, was selected as the calibrator, of which expression level represents 1 for normalization in each comparison. RESULTS AND DISCUSSION The schematic diagram of our method was shown in Figure 1, including hybridization, ligation, and PCR amplification steps. Obviously, the key step of this method was the ligation of two probes when their hybridizing to target miRNAs. Moreover, the detection sensitivity and specificity would be highly dependent on the ligation efficiency and selectivity, which are to a large degree determined by the sequences of both stem-loop probes and their joint position relative to the sequence of target miRNA. As to the ligation efficiency, although Nilsson et al. optimized the conditions of RNA-templated ligation of DNA probes by T4 DNA ligase,20,21 we and other researchers still found that T4 DNA ligase could catalyze the connection of single DNA strands even in the absence of miRNA targets.22 Albeit with substantially lower ligation yield, those nonspecific background signals would be considerably intensified in the subsequent PCR amplification, especially when low-abundant miRNA target was characterized, thus leading to a (19) Pfaffl, M. W. Nucleic Acids Res. 2001, 29, 2002–2007. (20) Nilsson, M.; Barbany, G.; Antson, D. O.; Gertow, K.; Landegren, U. Nat. Biotechnol. 2000, 18, 791–793. (21) Nilsson, M.; Antson, D. O.; Barbany, G.; Landegren, U. Nucleic Acids Res. 2001, 29, 578–581. (22) Kuhn, H.; Frank-Kamenetskii, M. D. FEBS J. 2005, 272, 5991–6000. (23) Shell, S.; Park, S. M.; Radjabi, A. R.; Schickel, R.; Kistner, E. O.; Jewell, D. A.; Feig, C.; Lengyel, E.; Peter, M. E. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 11400–11405. (24) Johnson, S. M.; Grosshans, H.; Shingara, J.; Byrom, M.; Jarvis, R.; Cheng, A.; Labourier, E.; Reinert, K.; Brown, D.; Slack, F. Cell 2005, 120, 635– 647. (25) Viswanathan, S. R.; Daley, G. Q.; Gregory, R. I. Science 2008, 320, 97– 100.

Figure 4. Mouse let-7 family assays: (a) eight closely related sequences of mouse let-7 family members including let-7a∼7 g and let-7i. Different bases among those miRNAs were labeled in red. (b) Electrophoresis diagram of ligation products for each let-7 assay. A total of 1.2 × 1013 copies of synthetic miRNA was added to each ligation reaction (10 µL). (c) The relative detection (%) by each assay was calculated on a CT difference between target miRNA (normalized to 100) and other let-7 members. A volume of 0.2 nL of ligation products in part b was used as a template in PCR amplification. (d) Electrophoresis diagram of ligation products of each let-7 member by a whole let-7 family ligation assay with a couple of universal probes.

noticeable background signal. Therefore, an improved probe with a stem-loop structure (with a 4-6 nt duplex stem) was designed here. To illustrate the advantages of the stem-loop probes over the linear probes, we synthesized two pairs of probes on purpose to quantify miRNA-122, including a pair of stem-loop probes (spmiRNA-122-1/2) and a pair of linear probes (lp-miRNA-122-1/ 2) (see Figure 2). Each couple of probes were exposed to T4 ligation reaction, respectively, and followed to PCR amplification with or without synthetic mir-122. As shown in Figure 2, ∆CT between mir-122 and its negative control (∆CT ) CT,mir-122 CT,NC) was evidently increased when a couple of stem-loop probes were employed, thus reducing the nonspecific ligation at least 100-fold. We hypothesized that those stem-loop probes would unlock and hybridize to target miRNA during the annealing process, while those residual probes unbound to the targets might recover to the stem-loop structure or hybridize with themselves so that the self-ligation reaction between probe1 and probe2 could be dramatically restrained due to steric hindrance (Supplementary Figure 1 in the Supporting Information). We also investigated the ligation efficiency of DNA probes by T4 ligase in mediation of a synthetic DNA or RNA oligomer and found the comparable efficiency in either case (Supplementary Figure 2 in the Supporting Information). T4 DNA ligase was previously proven to be able to distinguish a single mismatch from a nucleotide sequence homologous to target miRNA,22 but the mismatch position has a dramatic effect on its discrimination capability. So far, it was still not clear whether a small RNA molecule containing a fragment exactly the same as one of target miRNA could cause any observable effect on the specificity of T4 DNA ligase. In fact, it is always true that there are other miRNAs sharing the same sequence of 2-7 nt across the ligation site when designing probes for certain miRNA. We addressed this concern by examining the discrimination of miRNA-122 from several natural miRNAs including aga-mir-210, has-mir-549, has-mir-214, hasmir-21, and an artificial miRNA named syn-aa (Figure 3). The synthetic mir-122 and above-mentioned miRNAs were put into the

Table 1. Discrimination between miRNA-122, miRNA-1, and Their pre-miRNA Precursors identification mir-122

mir-1

synthetic miRNA (no. of copies)

synthetic precursor (no. of copies)

CT miRNA

1 × 108 0 0 0 1 × 108 1 × 108 1 × 108 1 × 108 0 0 0 1 × 108 1 × 108 1 × 108

0 1 × 108 1 × 109 1 × 1010 1 × 108 1 × 109 1 × 1010 0 1 × 108 1 × 109 1 × 1010 1 × 108 1 × 109 1 × 1010

16.54 25.70 23.95 23.12 17.31 17.63 17.47 13.26 22.19 19.61 18.13 12.80 13.16 13.84

ligation reaction composed of stem-loop probes (sp-miRNA-122-1/ 2) and T4 DNA ligase, respectively. Then the ligation products were analyzed by gel electrophoresis (Figure 3, bottom). Two stem-loop DNA probes were ligated together only in the presence of mir-122 as predicted, whereas no ligation products were observed when using other miRNAs including syn-aa, indicating that the ligation is highly specific for the perfectly matched duplex. To further test the specificity of this ligation assay, DNA probes were designed to discriminate among eight closely related sequences in the mouse let-7 family including let-7a∼7g and let-7i (Figure 4a), where each pair of DNA probes (Supporting Information, Table S2) was tested to detect special miRNA as well as other family members, respectively. This ligation assay was examined by gel electrophoresis and then evaluated through the subsequent real-time PCR amplification (Figure 4b,c). Distinct sequence specificity was achieved only with a few exceptions, in which nonspecific signal were observed above 1% (i.e., 7d, 7e in the 7a assay, 7b in the 7c assay, and 7a in the 7f assay). Further optimization of the probes might improve the specificity of those assays. Furthermore, multimembers of one family Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

5449

Figure 5. Dynamic range and sensitivity of the mmu-mir-122 assay: (a) amplification plot of synthetic mir-122 over 6 orders of magnitude. The synthetic mir-122 input ranged from 105 to 1010 copies. NC was a control reaction in which no mir-122 existed in the ligation reaction, and NTC was another control reaction where no ligation product was added into the PCR reaction. (b) Standard curve of mir-122. (c) Correlation of total RNA input to the CT values for four miRNA assays. Mouse liver (mir-122), lung (let-7), heart (mir-1), and testis (mir-34a) total RNA input ranged from 0.02 ng to 2 µg per ligation reaction.

Figure 6. The relative signal intensity of U6. The black and white rectangles represented the relative U6 intensities detected by real time RT-PCR (RT-PCR) and our assay (T4 ligation PCR), respectively. The expression amount of U6 in testis was normalized to 1.

could be quantitatively analyzed as a whole in a one-tube assay by designing a pair of universal probes. For example, eight pieces of the let-7 family were totally ligated in one pot by a pair of probes (Figure 4d). This attribute might be useful in the case where it is unnecessary to discriminate the relative expression level of each family member given the hypothesis that family members might work together to tune the expression of a group of functionally related proteins.23-26 (26) Yu, F.; Yao, H.; Zhu, P.; Zhang, X.; Pan, Q.; Gong, C.; Huang, Y.; Hu, X.; Su, F.; Lieberman, J.; Song, E. Cell 2007, 131, 1109–1123.

5450

Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

Pre-miRNA containing an entire miRNA sequence may disturb mature miRNA detection. Considering that the predicted melting temperature of pre-miRNAs is in general above 75 °C, we carried out the hybridization reaction at 65 °C to ensure two stem-loop probes to hybridize to miRNA, rather than to pre-miRNA. As shown in Table 1, pre-miRNAs such as precursors of mir-122 or mir-1 could only produce as much as 0.2% signal of an equal amount of their mature miRNAs. Even if 100 times more precursors were added than the mature miRNAs, it could only contribute to less than 1.5% signal intensity. These results clearly implied that this assay can efficiently prevent precursor from hybridizing to the nucleotide probes and is highly specific for mature miRNAs. The dynamic range and sensitivity of the miRNA quantification were also evaluated. Mir-122 was quantified based on the A260 value and diluted over 6 orders of magnitude. Amplification plot showed excellent linearity between the log of target input and the CT value, thus demonstrating that the assay has a dynamic range of at least 6 logs from 105 to 1010 miRNA input for each ligation reaction (Figure 5a,b). Additionally, separate assays were performed to validate the detection limit of total RNA samples for four miRNAs (Figure 5c). The RNA input ranged from 0.02 ng to 2 µg. The CT values correlated to the RNA input (R2 > 0.994) over 5 orders of magnitude. Therefore, we believed that this assay exhibited great dynamic range and sensitivity even in the use of total RNA samples. Normalization of miRNA level between samples relative to endogenous control is extremely critical for miRNA quantification

Figure 7. Relative expression profiles of four miRNAs across 10 mouse tissues. All data were normalized according to the Pfaffl formula (eq 1), and the highest signal was set to 1.

in real-time PCR experiment. Since U6 snRNA is usually used as an endogenous control in conventional miRNA or other RNA detection, the relative expression levels of U6 snRNA across 10 mouse tissues were quantified by our ligation-based real-time PCR and further validated by traditional real time RT-PCR. As shown in Figure 6, both of the methods exhibited similar expression patterns of U6 across all tissues. This result strongly supported the assertion that our assay can be used for quantitative characterization of miRNAs among or across tissues. Next, we applied this assay to detect relative expression levels of mir-122, mir-1, mir-34a, and let-7a across 10 mouse tissues (Figure 7). Data were normalized according to the expression of the small nuclear RNA U6 as described in the Experimental Procedures. As expected, mir-1 and mir-122 were specifically expressed in muscles and liver, respectively, while miRNA-34a and let7a was widely distributed in various mouse tissues. CONCLUSIONS There is still an increasing need for the development of a novel method for miRNA quantification, which is not only sensitive and specific but also simple and low cost. In this study, we demonstrated such a new assay for monitoring the expression of mature miRNAs in tissues. Unlike previously reported assays, our approach relied on the enzymatic ligation reaction to lengthen a miRNA strand by employing a pair of DNA probes with a stem-loop structure, which significantly reduced nonspecific signals and enhanced the efficiency of ligation. This assay demonstrated several remarkable features including single nucleotide discrimination and anti-interference from their precursors. In addition, not only individual miRNA but also several members of a family with related sequences (i.e., let-7 family)

could be detected simultaneously in a one-tube assay, which makes it feasible to examine the expression of a whole family of miRNA at one time. MiRNA in the total RNA sample could be quantitatively analyzed without further enrichment. What is also noticed, SYBR Green I, rather than TaqMan or LNA modified probes, was sufficient for the acceptable miRNA quantification, which could extremely reduce the cost of this assay and might be helpful to widely apply. Together, we believe this method based on enzymatic ligation might be a promising tool for detection of miRNA and other small nucleic acids, especially for studies on clinical diagnosis, cancer and disease research, and drug development. ACKNOWLEDGMENT J.L. and B.Y. contributed equally to this study. This research was supported by projects of NSFC (Grant No. 30600142 and Grant No. 90607004), Ministry of Science and Technology of P.R. China (Grant No. 2007AA04Z313), and the Postdoc Foundation of China (Grant No. 20060400004). The authors thanked Dr. Peter Peizhuo Zhang at Shanghai GenePharma Co. Ltd. for RNA synthesis and technical consultations. SUPPORTING INFORMATION AVAILABLE Additional information as noted in text. This material is available free of charge via the Internet at http://pubs.acs.org.

Received for review March 23, 2009. Accepted April 27, 2009. AC900598D

Analytical Chemistry, Vol. 81, No. 13, July 1, 2009

5451