We first designed a series of 20-nt oligo-DNAs derived from the rice miR156 sequence (found at http://microrna.sanger.ac.uk) (Table 1) and FITC (fluorescein isothiocyanate)-labeled probe Osa-miD156* ((FITC)-5′-ACGT GCT CAC TCT CTT CTG TCA-3prime;), which contains two additional nucleotides at the 5prime; end [17]. All oligo-DNAs and oligo-RNAs were synthesized by Takara Biotechnology (Dalian) Co., Ltd. Osa-miD156* was hybridized with Osa-miD156 in liquid buffer (buffer 3, see below) at 42 °C for 60 min, followed by digestion of non-hybridized sequences with exonuclease I for 30 min. Non-denaturing PAGE electrophoresis showed that the hybridization signal was visible at FITC-labeled probe levels above 0.1 pmol (Figure 1a). The 0.01 pmol of hybridization signal can be recognized if performing Dot blot (data not shown).

To evaluate the specificity of the fluorescently labeled oligonucleotide probe in liquid Northern hybridization, we conducted a series of hybridizations with Osa-miD156* and Osa-miD156s (Table 1) that contain a one-base mismatch, three-base mismatch and five-base mismatch at 42, 57 and 60 ºC, respectively. The results indicated that the degree of tolerance of liquid Northern hybridization to the number of mismatches decreases as the temperature increases (Figure 1b), and only the hybridization signal from the perfect sequence of Osa-miD156 was visible at a hybridization temperature of 60 ºC, which was the melting temperature (Tm) of Osa-miD156* (Figure 1b, lower). This is of special significance, as it distinguishes the member that shares only a one-base difference with others in the miRNA family. However, the fluorescence intensity was stronger when the hybridization was performed 10 °C lower than the Tm (Figure 1c). Moreover, hybridization that proceeded for 0.5 h resulted in a visible fluorescence signal, and a hybridization time of greater than 1 h did not significantly enhance the fluorescence intensity (Figure 1d).

To screen for the appropriate hybridization buffer, we used small RNA samples from one-week-old rice seedlings to detect the five designed buffers: buffer 1 contained DEPC (diethylpyrocarbonate)- treated water + 10 mmol/L EDTA (ethylenediaminetetraacetic acid); buffer 2 contained 30 mmol/L Sodium phosphate buffer + 10 mmol/L EDTA; buffer 3 contained 0.3 mol/L NaCl + 30 mmol/L phosphate buffer + 10 mmol/L EDTA; buffer 4 contained 2 × SSC + 10 mmol/L EDTA; and buffer 5 contained Northern blotting solution [18]. Small RNAs were extracted from 0.2 g rice seedlings using an isolation buffer according to the method described previously [19], and we used 1ug small RNAs (89.69 ng/μL) in this experiment. The results indicated that the simple salt solution gave satisfactory results for liquid hybridization of miRNA because the hybridization signals were visible in all five buffers, and the signal intensity was not significantly different between buffers 3, 4 and 5 (Figure 2a).

We conducted further quantitative analyses of the electrophoresis results of liquid Northern hybridization of Osa-miR156 from rice seedlings using a Chemi Doc XRS System camera (Bio-Rad, US). A series of hybridizations of different concentrations of Osa-miD156* with synthesized Osa-miR156 was used to create a standard curve. The analysis showed that a good linear relationship exists between fluorescence intensity and miR156 concentration (C = 2.35E-5 * Int. −0.121, R2 = 0.998545), and the expression of Osa-miR156 in rice seedlings is 22 ng/mg fresh weight (Figure 2e).

For further verification, we performed an additional detection of miRNAs in 1 μg small RNAs (89.69 ng/μL) from one-week-old rice seedlings using 4 pmol of other miRNA probes (Table 1). The results showed that Osa-miR156 hybridization signals were also detected in different rice organs and other monocotyledonous and dicotyledonous species, for example, wheat, maize, tobacco and Arabidopsis (Figure 2b,c), by liquid Northern hybridization. Moreover, the fluorescence signal from the hybridization of Osa-miR156 with Osa-miD156* was brighter than miR394 with Osa-miD394* and miR528 with miD528* (Figure 2d), suggesting that the abundance of miR156 in rice seedlings is higher than that of miR394 or miR528. These experiments have documented that the liquid Northern hybridization technique that we developed is of universal applicability in detecting plant miRNAs. We also added two species of probes, Osa-miD156▴ and Osa-miD445* (24-nt long and labeled with Cy3), into a liquid hybridization system to simultaneously detect their expression levels. Non-denaturing 20% TBE-PAGE indicated that both Osa-miR156 and Osa-miR445 were detected in rice seedlings [20], and the expression of Osa-miR156 (lower in Figure 2d) was less than that of Osa-miR445 (upper in Figure 2d), suggesting that liquid Northern hybridization can provide simultaneous detection of multiple miRNAs in a single experiment.

Seeds of Oryza sativa L., Zea mays L. and Triticum aestivum L. were immersed in distilled water for two days and then germinated at room temperature. Two weeks later, seedlings that were approximately 5 cm high were used for RNA extraction. Some seedlings were grown in pots under natural conditions. Leaves, stems, roots and inflorescences or panicles were collected for RNA extraction when the plants reached maturity. Nicotiana tabacum plants, the leaves of which were used to isolate RNA, were grown in pots under natural conditions, while Arabidopsis thaliana (L.) Henyh.(Columbia ecotype) plants were grown in controlled environment chambers with a 16-h photoperiod at 24 °C. Whole plants were collected for RNA extraction when they grew to the 8-leaf stage.

Total RNA and small RNA were extracted from plant tissues using RNA extraction buffer (1.9 g/L macaloid dissolved in 50 mmol/L Tris-HCl (pH 7.6), 20 mmol/L sodium citrate, pH 7.0, 100 mmol/L NaCl, 10 mmol/L EDTA, 0.5% (w/v) SDS, and 1% (v/v) 2-mercaptoethanol), according to the method described by Chun et al. [19] with minor modifications. Extractions of small RNA were performed as follows: Fresh plant tissues (0.2 g) were ground into fine powder with a mortar and pestle in the presence of liquid nitrogen. The powder was immediately placed in an RNA extraction buffer at a ratio of extraction buffer to leaves of 4 mL/1 g and homogenized thoroughly. Then, 1/2 volume of water-saturated phenol and 1/2 volume of chloroform–isoamyl alcohol mixture (24:1, v:v) were added. The contents were mixed well and centrifuged at 12,000 rpm for 10 min at 4 °C. Pphenol/chloroform/isoamyl alcohol extraction was repeated once. The supernatant was mixed with 1/3 volume of 8 mol/L LiCl and incubated at −20 °C for at least 30 min, followed by centrifugation of15,000 g at 2 °C for 10 min. The resulting RNA supernatant was added to 50% PEG 8000 and 5 mol/L NaCl (50 μL of 50% PEG 8000 and 50 μL of 5 mol/L NaCl for every 400 μL of supernatant), mixed and incubated at −20 °C for at least 30 min, followed by centrifugation of 15,000 g at 2 °C for 10 min. The PEG-NaCl precipitation procedure was then repeated. The aqueous phase was transferred to a new 1.5-mL Eppendorf tube, and 1/10 volume of 1 mol/L MgCl₂ and 2.5 volume of absolute ethanol were added. The mixture was incubated at −20 °C for at least 2 hr, followed by centrifugation of 15,000 g at 2 °C for 30 min. The resulting small RNA pellets were washed with 75% (v/v) ethanol twice, air-dried and dissolved in DEPC-H₂O and stored at −70 °C.

Five rice miRNA sequences, miR156, miR167, miR394, miR528 and miR445 (Table 1) were found in miRBase (http://microrna.sanger.ac.uk). Their complementary probe sequences, which were labeled with fluorescein FITC or Cy3, and other series of oligonucleotides were synthesized by TaKaRa Biotechnology Co., Ltd., Dalian, China (Table 1).

A specific amount of oligonucleotide or plant small RNA (1 ug, isolated before) and synthesized probes were added to a 200-μL Eppendorf tube. Hybridization buffer (30 mmol/L Sodium phosphate buffer (pH 8.0), 0.3 mol/L of NaCl, 10 mmol/L of EDTA) was added up to 16 μL. After mixing thoroughly, the reaction mixture was heated to 94 °C for 5 min and incubated in a water bath at 42 °C for 60 min. When the hybridization reaction completed, non-hybridized single-strand sequences, including the hybridization probe, were digested with 10 U Exonuclease I at 37 °C for 30 min. Exonuclease I (NEW ENGLAND BioLabs, M0293) catalyzes the removal of nucleotides from single-stranded DNA in the 3prime;–5prime; direction.

A 0.8-mm thick, non-denaturing 10% TBE-polyacrylamide gel or 10% TAE-polyacrylamide gel was prepared according to Sambrook and Russell [21]. After gel solidification, the above hybridization products were loaded, and the gel was run at 90 V for approximately 40 min. After the run completed, the gel was moved onto a Dark Reader transilluminator (Clare Chemical Research, Inc.) for observation and photography with a Canon A650IS digital camera.

Hybridization of 10, 5, 2.5, 1.25, 0.625, 0.313, 0.156 or 0.078 pmol Osa-miD156* probes with an excess of Osa-miR156 and Osa-miD156* probes were performed with 6 μL small RNA from rice seedlings. After non-denaturing PAGE electrophoresis, the gel was placed on a citaBlue conversion screen in a Chemi Doc XRS System camera (Bio-Rad, US) equipped with a 560DF50nm, 62 mm Emssion filter and auto-exposure. The imaging data were analyzed with Quantity One software.