Detecting miRNAs by liquid hybridization and color development

Detecting miRNAs by liquid hybridization and color development

Methods 58 (2012) 151–155 Contents lists available at SciVerse ScienceDirect Methods journal homepage: www.elsevier.com/locate/ymeth Detecting miRN...

843KB Sizes 0 Downloads 28 Views

Methods 58 (2012) 151–155

Contents lists available at SciVerse ScienceDirect

Methods journal homepage: www.elsevier.com/locate/ymeth

Detecting miRNAs by liquid hybridization and color development Xiangqi Li a,b,⇑, Minjie Ni a, Yonglian Zhang a,b,⇑ a

Shanghai Key Laboratory for Molecular Andrology, State Key Laboratory of Molecular Biology, Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Shanghai 200031, China b Shanghai Institute of Planned Parenthood Research, Shanghai 200032, China

a r t i c l e

i n f o

Article history: Available online 3 August 2012 Communicated byGray Brewer Keywords: Method MicroRNA Liquid hybridization Avidin–biotin detection system Color development

a b s t r a c t Currently, two methods, PCR and Northern blot, are widely used to detect individual microRNAs (miRNA). Although PCR is highly sensitive, false positives and difficulties of primer design discourage its use. While a Northern blot is an effective tool, traditional Northern blot protocols are complicated, time-consuming, and usually inconvenient for users. Liquid Northern blot methods are rapid but require instruments for detection of fluorescent signals. Here, we describe an alternative protocol, liquid hybridization and color development (LHCD), based on the rapidity of liquid hybridization and the signal amplification of avidin– biotin complex (ABC) for detection. LHCD can distinguish a one-nucleotide difference within a miRNA family and allow for the sensitive detection of 2.5 f mol of miRNAs. Furthermore, LHCD is not only simple and rapid, but detection is visual and so it does not require expensive equipment. LHCD is easy to learn and convenient for miRNA analyses. Ó 2012 Elsevier Inc. All rights reserved.

1. Introduction MicroRNAs (miRNAs) are 21-nucleotide RNAs in plants, animals and microbes. They have emerged as key post-transcriptional regulators of gene expression and have revolutionized our comprehension of gene expression [1,2]. Predictions suggest that about one-third of all protein-coding genes are regulated by miRNAs [3]. In mammals, miRNAs are predicted to control the activity of 50% of all protein-coding genes, and are involved in the regulation of almost every cellular process investigated so far [2]. Importantly, their altered expression is associated with many human pathologies [2]. To date, release 18 of the miRBase sequence database contains 21,643 mature miRNAs from 168 species (http:// www.mirbase.org/), and novel molecules are constantly discovered. However, the biological characterization and the functional confirmation of most miRNAs remain to be investigated. Since the first miRNA gene was discovered in 1993 [4–6], great advances have been made in miRNA biology. However, the small Abbreviations: LHCD, liquid hybridization and color development; ABC, avidin– biotin complex; IHC, immunohistochemistry; AP, alkaline phosphatase; HRP, horseradish peroxidase; BCIP/NBT, 5-bromo-4-chloro-3-indolyl-phosphate/nitro blue tetrazolium; DAB, 3,30 -diaminobenzidine; ECL Prime, Amersham™ ECL™ prime Western blotting reagent; SSWF, Super signalÒ west femto; BSA, bovine serum albumin; TBS, tris-buffered saline. ⇑ Corresponding authors at: Shanghai Key Laboratory for Molecular Andrology, State Key Laboratory of Molecular Biology, Institute of Biochemistry and Cell Biology, 320 Yue-Yang Road, Shanghai Institutes for Biological Sciences, Shanghai 200031, China. Fax: +86 21 54921011. E-mail addresses: [email protected] (X. Li), [email protected] (Y. Zhang). 1046-2023/$ - see front matter Ó 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.ymeth.2012.07.025

size of miRNAs increases the technical difficulty of their identification, spatiotemporal detection, and functional confirmation. Generally, to investigate an individual miRNA, PCR and Northern blotting are used to measure miRNA expression levels in vitro. PCR is undoubtedly highly sensitive [7], however, the rates of false positives and difficulties designing primers limit its use. Northern blotting is the gold standard for directly examining the expression of miRNA [8]. Traditional Northern blot protocols include fractionating small RNAs by gel electrophoresis; transferring the separated RNA fragments onto a nylon membrane; over-night hybridization; and hours to days or even months of autoradiography [9,10]. The method is complex, lengthy, and inconvenient and in general, does not lend itself for simple and rapid analyses. The new technology of liquid Northern hybridization overcomes these shortcomings and allows quick and simple detection of miRNA [11]. However, the use of fluorescent probes can present problems of decreased sensitivity due to fluorescence quenching. Furthermore, instruments for detection of fluorescent signals must be available. Avidin–biotin complex (ABC) method is a standard tool in immunohistochemistry (IHC). The avidin–biotin interaction is one of the strongest known non-covalent interactions (Kd = 1015 M) [12], making it ideal for both purification and detection strategies [13]. Based on our experience with Northern blotting, IHC, and Western blotting methods, we developed an alternative method of liquid hybridization and color development (LHCD), which combines the rapid features of liquid hybridization and the amplified signals provided by avidin–biotin complex detection [14,15].

152

X. Li et al. / Methods 58 (2012) 151–155

We describe a LHCD protocol that provides an easy-to-learn and convenient-to-use tool for detecting miRNAs.

Fractionator System (Ambion, USA) can be used according to the manufacturer’s instructions.

2. Method

2.4.3. Prepare reagents and materials Prepare hybridization buffer, 50 biotinylated DNA probes, Exonuclease I, BSA, nylon membrane, ABC–AP or ABC–HRP, 1 TBS, and BCIP/NBT, DAB, ECL Prime, or SSWF. Hybridization buffer (Buffer A: 30 mmol/L Sodium phosphate buffer (pH 8.0), 0.3 mol/L of NaCl, 10 mmol/L of EDTA (11); or Buffer B: 1 Exonuclease I). 50 biotinylated DNA probes are cheaper, and can be easily synthesized by commercial company (e.g., Takara, Japan). Exonuclease I, BSA, ABC–AP, ABC–HRP, 1 TBS, Triton-X-100, Tween-20, BCIP/NBT, and DAB are common inexpensive reagents readily available from commercial sources. Here, Exonuclease I was ordered from NEB (USA); BSA, Triton-X-100, Tween-20, ABC–AP, ABC–HRP, BCIP/ NBT, and DAB from Boster (China), 1 TBS buffer from Dycent Biotech (China); ECL Prime from Amersham (USA). SSWF (Super SignalÒ West Femto) is expensive and is available from Thermo Scientific (USA). Nylon membrane is from Roche (Switzerland).

2.1. Outline Purified small RNAs are hybridized in buffer with 50 biotin-labeled DNA probes. The hybridized mixture is dotted onto a nylon membrane after non-hybridized probes are digested with Exonuclease I. The membrane is then incubated with ABC. Finally, the membrane is developed with BCIP/NBP or DAB to produce colorimetric end products; ECL prime can be used to produce light for detection (Fig. 1). 2.2. Liquid hybridization Hybridization buffer is pipeted into an Eppendorf tube. Synthesized or isolated small RNAs and 50 biotinylated probes are added into the hybridization buffer, mixed thoroughly, and heated at 94 °C for 4 min. The hybridization reaction is performed at 42– 65 °C for 60 min. Finally, digest non-hybridized probes with Exonuclease I at 37 °C for 30 min. 2.3. Color development Dot the digested hybridization mixture onto a nylon membrane, dry, and perform ultraviolet crosslinking. Block the membrane with 10% BSA, then incubate the membrane with ABC–AP or ABC–HRP at 37 °C for 30 min. Wash the membrane with TBS buffer six times. Finally, develop with BCIP/NBP, or DAB, or ECL Prime.

2.4.4. Liquid hybridization Up to 16 lL hybridization buffer is loaded into a 200 lL Eppendorf tube. A specific amount of small RNAs (such as 20 pmol) and 50 biotinylated probes (such as 20 pmol) is added into the hybridization buffer; mix thoroughly and heat the mixture to 94 °C for 4 min. Next, carry out the hybridization reaction by incubating the mixture in a water bath at 42–65 °C for 60 min. Finally, remove non-hybridized, single-strand probes thoroughly by incubating the reaction mixture with Exonuclease I (20U/ll, 2ll) at 37 °C for 30 min.

2.4.2. Purify small RNAs (optional procedure) Small RNAs are separated and purified from total RNAs by urea–PAGE (Urea–polyacrylamide gel electrophoresis) following protocols that are available from a number of laboratory websites (http://www.umassmed.edu/Content.aspx?id=154408&linkidentifier=id&itemid=154408; http://web.wi.mit.edu/bartel/pub/protocols.html). Alternatively, other methods, such as the FlashPAGE™

2.4.5. Color development The above digested mixture is 50% diluted with 1.5 M NaCl. Spot a specific volume (0.5 lL) of the diluted mixture onto a nylon membrane with a micropipettor, then air dry for 10 or 5 min at 50 °C. Crosslink with ultraviolet 90 s (Energy: 3,000, CL-1000 Ultraviolet crosslinker) at room temperature. Next, block the membrane with 10% BSA in 1 TBS (pH 7.2) for 30 min at room temperature. Then, incubate the membrane with a specific volume (1 ml) of ABC–AP or ABC–HRP at 37 °C for 30 min or 1 h at room temperature. (Prepare the ABC reagent according to the manufacturer’s instructions: we dilute it 1:1,000 with 5% BSA and 0.1% Triton-X100 in 1 TBS (pH7.2)). Wash the membrane in 1 TBS buffer containing 0.1% Triton-X-100 and 0.05% Tween-20 for 3 min, six times. Finally, cover the membrane with 1 ml of BCIP/NBT, DAB, or ECL

Fig. 1. Schematic diagram of procedures. Purified small RNAs are hybridized in buffer with 50 biotin-labeled DNA probes. The mixture is spotted on a nylon membrane after digestion of non-hybridized probes with Exonuclease I. The membrane is then incubated with ABC and is developed with BCIP/NBT, DAB, or ECL to produce colorimetric end products that can be seen by eye.

Fig. 2. Selection of hybridization buffers. (A) Buffer A is 30 mmol/L sodium phosphate buffer (pH 8.0), 0.3 mol/L NaCl, and 10 mmol/L EDTA. (B) Buffer B is 1 Exonuclease I buffer. The indicated amounts of a synthesized 22-nt small RNA (UCGGUCAGUCUGGGGCAGGCAA) and its antisense 50 biotin-labeled DNA probe (gtTTGCCTGCCCCAGACTGACCGA) were hybridized in the indicated buffers at 55 °C, digested with Exonuclease I, and 0.5 ll of each reaction was spotted on the membrane. Alkaline phosphatase-labeled ABC was employed and BCIP/NBT was selected for color development.

2.4. Experimental protocol 2.4.1. Prepare total RNA sample Isolate total RNAs by adding TRIzol (Invitrogen) to cells following the manufacturer’s protocol.

X. Li et al. / Methods 58 (2012) 151–155

Prime and incubate for 5 s–15 min. Then wash the membrane with running water or ddH2O in a dish to stop the color reaction.

3. Results and discussion

153

(defined here as Buffer B). We used 100 amol, 1 fmol, 10 fmol, and 100 fmol small RNAs commercially synthesized for this experiment. The results indicated that the signal intensity was not significantly different between Buffer A and B (Fig. 2). Thus, 1 Exonuclease I buffer (Buffer B) was chosen for use in subsequent experiments.

3.1. Selection of hybridization buffer 3.2. Screening of hybridization membranes An earlier report indicated that signal intensity was not significantly different between three buffers [11], which here, we define as Buffer A. We compared this buffer and 1 Exonuclease I buffer

We compared the performance of various commercial hybridization membranes in the miRNA detection system (Fig. 3). The

Fig. 3. Screening of hybridization membranes. (A) Positively charged nylon membrane from Roche applied science (cat no., 11417240001); (B) biodyne B nylon membrane from Pall Life Science (cat no., 60207); (C) Hybond-N + membrane optimized for nucleic acid transfer from Amersham Pharmacia/GE healthcare (cat no., RPN303B); (D) BrightStar™–Plus positively charged nylon membrane from AMBION (cat no., 0711004); (E) GeneScreen™–plus hybridization transfer membrane from PerkinElmer (cat no., NEF1017001PK); (F) Hybond–P PVDF membrane from Amersham Pharmacia/GE healthcare (cat no., RPN303F); (G) Whatman Protran™ nitrocellulose (NC) membranes from GE healthcare (cat no., 10401396). The indicated amounts of small RNA described in Fig. 2 were hybridized and spotted on the indicated membranes. Color was developed 3 min.

154

X. Li et al. / Methods 58 (2012) 151–155

Fig. 4. Sensitivity of hybridization with different color development reagents. (A) Color developed with BCIP/NBT. (B) Color developed with DAB. (C) Color developed with ECL Prime. Color was developed 5 min. The red arrow indicates the most sensitive detection. The hybridization treatments were as described in Fig. 2. NC is a negative control small RNA.

nylon membrane from Roche provided the greatest signal compared to background. NC membrane had little background, but its sensitivity was low; PVDF and NC had comparable sensitivity (to 25 fmol) and backgrounds but were less sensitive than the nylon membrane from Roche. However, PVDF requires a methanol pre-wetting step. We thus chose nylon membrane from Roche for hybridizations. 3.3. Sensitivity of hybridization Synthesized oligo RNAs were used to examine the sensitivity of the LHCD procedure. Hybridization signal was visible with AP-labeled ABC above 2.5 fmol (Fig. 4A), which is 40 times more sensitive than FITC-labeled DNA probes [11]; HRP-labeled ABC was also above 2.5 fmol with color development by DAB (Fig. 4B); and above 25 fmol with ECL prime (Fig. 4C). However, the background with DAB color development can be high. ECL color development is 4 times more sensitive than FITC-labeled DNA probes [11]. Although ECL color development is less sensitive than BCIP/ NBT color development, ECL always yields low backgrounds whether the time of color development is long or short. SSWF also has low backgrounds, allowing for sensitive detection of 25 fmol (data not shown).

Fig. 5. Specificity of hybridization with different color development reagents. Synthesized Rno–miR-29a/b/c miRNAs and the antisense 50 biotin-labeled DNA probe (gtTAACCGATTTCAGATGGTGCTA) for Rno–miR-29a were each hybridized in 1 Exonuclease I buffer at 61 °C. (A) The sequences of Rno–miR-29a/b/c are shown. Red letters indicate nucleotide differences compared to Rno–miR-29a. (B) Color developed with BCIP/NBT. (C) Color developed with DAB. (D) Color developed with ECL Prime. Color was developed 2 min. The spotted samples were 0.5 ll (500 fmol/ ll) each. NC denotes negative control.

3.4. Specificity of hybridization To evaluate the specificity of the liquid hybridization system, a series of hybridizations were performed with the antisense of Rno– miR-29a as probe to assess whether the probe could distinguish between the members of the Rno–miR-29 family. Compared to Rno–miR-29a, Rno–miR-29b and Rno–miR-29c contain five-base and one-base mismatches, respectively (Fig. 5A). The hybridization results indicated that Rno–miR-29a probe did not hybridize with the other member miRNAs (Fig. 5B–D). Thus, the procedure can distinguish single nucleotide differences in sequence, consistent with a report of FITC-labeled probes [11]. 3.5. Detection of tissue miRNAs Detection of Rno–miR-29a was performed with small RNAs from a variety of tissues. Eight nanograms of total small RNA from

rat epididymis, heart, intestine, liver, testis, seminal vesicle, and kidney were analyzed by liquid Northern hybridization at 61 °C. Expression of miRNA-29a was easily detected in these tissues (Fig. 6). Similar results were obtained with color development for 1 min with ECL prime, 30 s with BCIP/NBT, and 5 s with DAB. The background with BCIP/NBT was low, and less than DAB; ECL had little background as well. Additionally, expression of miRNA-21 was examined in 2 ng of total small RNAs from these tissues. However, signals were low, and ECL prime yielded no signals (data not shown). Consequently, to increase sample mass per spot, 12 ng of sample RNA was spotted onto the nylon membrane at the same spot 6 times (2 ng each time); consecutive spotting avoids pipetting too much volume onto the membrane at one time. While, ECL prime yielded no signals after 20 min development, SSWF provided ample signals for detection (data not shown). These

X. Li et al. / Methods 58 (2012) 151–155

155

Fig. 6. Detection of miR-29a in total small RNAs from rat tissues. Isolated total small RNAs (160 ng) from various rat tissues and 15 ng of the miR-29a probe were hybridized in 1 Exonuclease I buffer at 61 °C, treated with Exonuclease I, diluted 50% with 1.5 M NaCl, and 0.5 ll of each mixture was spotted onto a nylon membrane. Ep, epididymis; He, heart; In, intestine; Li, liver; Te, testis; Sv, seminal vesicle; Ki, kidney. (A) Color developed with BCIP/NBT. (B) Color developed with DAB. (C) Color developed with ECL Prime. The color development time was 30 s for BCIP/NBT, 5 s for DAB, and 1 min for ECL. The spotted sample was 0.5 ll (2.14 pmol/ll, i.e., 16 ng/ll) each. The probe amount was 0.1 pmol. NC denotes negative control.

experiments indicated that the liquid hybridization and color development system is applicable for detecting miRNAs from tissue samples. 4. Tips and troubleshooting Exonuclease I catalyzes the 30 –50 removal of nucleotides from single-stranded DNA. To confirm removal of non-hybridized, single-strand probes by Exonuclease I, it is important to use a negative control. When preparing miRNA samples, it is important to purify them by gel or by other methods to minimize cross-hybridization with their precursors. Usually, we cannot detect miRNA precursors for most of miRNA species, so purification procedures are generally not required. For hybridization buffer, we recommend 1 Exonuclease I buffer since it is very simple and effective. For hybridization time, 1 h is usually sufficient; incubation greater than 1h does not significantly enhance signal intensity [11]. For hybridization temperature, we recommend 42 °C to obtain high hybridization-signal intensity [11], but only if negative controls have little or no signals. However, for distinguishing members of miRNA families at singlenucleotide resolution, the hybridization temperature will require optimization depending on the sequence of the probe used, as well as the cell/tissue type. For instance, we can differentiate rat miRNA-29a and miRNA-29b/c with a 61 °C hybridization temperature (Fig. 5). Consistent with traditional Northern blotting, higher temperature yields less signal strength, while lower temperature yields higher signal strength. For color development, BCIP/NBT and DAB are widely used chromogenic substrates for IHC, while ECL and SSWF are commonly used chemiluminogenic substrates for Western blot. Whether using BCIP/NBT or DAB, blocking the membrane with BSA or other materials is required to minimize non-specific signals. ECL Prime and SSWF were used in this procedure for chromogenic substrates, not chemiluminescent substrates and provided low backgrounds in general. However, optimal signal-to-noise was obtained with BCIP/NBT. Finally, it is essential to keep the sample-spotted membrane wet at all times to prevent high backgrounds. Moreover, 0.1% Triton-X-100 must be included when ABC is used on the membrane, since it stabilizes the enzyme activity of AP and HRP.

5. Concluding remarks Traditional solid-phase Northern blotting is labor intensive and requires special facilities. Newly developed liquid-phase hybridization procedures overcome these problems. However, fluorescent labeling techniques raise concerns about impaired sensitivity and the need for fluorescence detectors. We have combined liquid hybridization and color development, LHCD, to take advantage of the rapidity of liquid hybridization and the amplification of signals provided by the avidin–biotin system. This makes detecting miRNAs more convenient because the signals are visible by eye and the color-developed membranes can serve as a permanent record. Furthermore, depending on the requirements, one can select special membranes or color development reagents. Moreover, there is little need for special buffer preparation. In conclusion, LHCD is rapid, simple, and visual, providing an easy-to-learn and convenient-to-use approach for assessing miRNA expression levels. Acknowledgments We thank all the members of our group for providing assistance and advice. This research was supported by Grants from the National Basic Science Research and Development Project of China (30930053) and the Chinese Academy of Sciences (CAS) Knowledge Innovation Program (KSCX2-EW-R-07). References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12]

D.P. Bartel, Cell 116 (2004) 281–297. J. Krol, I. Loedige, W. Filipowicz, Nat. Rev. Genet. 11 (2010) 597–610. B.P. Lewis, C.B. Burge, D.P. Bartel, Cell 120 (2005) 15–20. R.C. Lee, R.L. Feinbaum, V. Ambros, Cell 75 (1993) 843–854. B. Wightman, I. Ha, G. Ruvkun, Cell 75 (1993) 855–862. E. Berezikov, Nat. Rev. Genet. 12 (2011) 846–860. R. Shi, V.L. Chiang, Biotechniques 39 (2005) 519–525. K.A. Cissell, S.K. Deo, Anal. Bioanal. Chem. 394 (2009) 1109–1116. E. Varallyay, J. Burgyan, Z. Havelda, Nat. Protoc. 3 (2008) 190–196. E. Varallyay, J. Burgyan, Z. Havelda, Methods 43 (2007) 140–145. X. Wang, Y. Tong, S. Wang, Int. J. Mol. Sci. 11 (2010) 3138–3148. J.M. Teulon, Y. Delcuze, M. Odorico, S.W. Chen, P. Parot, J.L. Pellequer, J. Mol. Recognit. 24 (2011) 490–502. [13] E.P. Diamandis, T.K. Christopoulos, Clin. Chem. 37 (1991) 625–636. [14] S.M. Hsu, L. Raine, H. Fanger, J. Histochem. Cytochem. 29 (1981) 577–580. [15] V. Chi, K.G. Chandy, J. Vis. Exp. (2007) 308.