Validation of a multiplex reverse transcription and pre-amplification method using TaqMan® MicroRNA assays

METHODS ARTICLE published: 26 November 2014 doi: 10.3389/fgene.2014.00413 Validation of a multiplex reverse transcription and pre-amplification method usingTaqMan® MicroRNA assays Joane Le Carré1 , Séverine Lamon 2 and Bertrand Léger 1 * 1 Institute for Research in Rehabilitation, SuvaCare Rehabilitation Clinic, Sion, Switzerland 2 Centre for Physical Activity and Nutrition, School of Exercise and Nutrition Sciences, Deakin University, Burwood, VIC, Australia Edited by: Since the discovery of microRNAs (miRNAs), different approaches have been developed Florent Hubé, Université Paris to label, amplify and quantify miRNAs. The TaqMan® technology, provided by Applied Diderot, France Biosystems (ABIs), uses a stem-loop reverse transcription primer system to reverse Reviewed by: Matthias S. Leisegang, Johann transcribe the RNA and amplify the cDNA. This method is widely used to identify global Wolfgang Goethe-Universität differences between the expression of 100s of miRNAs across comparative samples. This Frankfurt am Main, Germany technique also allows the quantification of the expression of targeted miRNAs to validate Richard Danger, King’s College observations determined by whole-genome screening or to analyze few specific miRNAs London, UK on a large number of samples. Here, we describe the validation of a method published by *Correspondence: Bertrand Léger, Institute for Research ABIs on their web site allowing to reverse transcribe and pre-amplify multiple miRNAs and in Rehabilitation, SuvaCare snoRNAs simultaneously. The validation of this protocol was performed on human muscle Rehabilitation Clinic, Avenue and plasma samples. Fast and cost efficient, this method achieves an easy and convenient Grand-Champsec 90, 1951 Sion, way to screen a relatively large number of miRNAs in parallel. Switzerland e-mail: bertrand.leger@crr-suva.ch Keywords: miRNA, qRT-PCR, multiplexing, plasma, human skeletal muscle BACKGROUND (Chen et al., 2008). miRNAs are reported to be highly stable in both MicroRNAs (miRNAs) are recently discovered small non-coding plasma and serum (Mitchell et al., 2008) and circulating miRNAs RNAs (∼22 nucleotides) regulating protein expression in animals expression is altered in pathological conditions. Although their and plants (Bartel et al., 2004). miRNAs can alter cellular func- role in circulation is not yet clear, miRNAs are promising biomark- tion by binding the 3 -UTR of target mRNA and therefore inhibit ers for the diagnostic of various pathologies, injuries and health the expression of the corresponding protein by either repressing conditions (Chen et al., 2008; Baggish et al., 2011; Zampetaki et al., protein translation or promoting mRNA degradation (Krol et al., 2012b). 2010). miRNAs can be highly and specifically enriched in spe- The expression of specific miRNAs in tissues, including skeletal cific tissues and each miRNA can target multiple mRNA species muscle and plasma, can be assessed using the reverse transcription (Lim et al., 2005; Sood et al., 2006). It is now well established that quantitative real-time polymerase chain reaction (RT-qPCR). Mei miRNAs play a pivotal regulatory role in many cellular processes et al. (2012) already described a broad range of commercially avail- including cell growth, proliferation, differentiation, and apopto- able miRNAs RT-qPCR assays. The classical approach involves the sis (Bueno et al., 2008; Subramanian and Steer, 2010). miRNAs use of predesigned individual assays. The TaqMan® technology dysregulation is reflective of physiological and pathological adap- provided by Applied Biosystems (ABI) uses a target-specific stem- tation processes and aberrant miRNAs expression is a hallmark of loop RT primer that extends the 3 end of the targeted miRNA numerous disease conditions such as cancer, cardiovascular, neu- to produce a cDNA template, which can then be amplified and rological, and autoimmune disorders (Lawrie et al., 2008; Gidron quantified by real-time qPCR (Chen et al., 2005). This method et al., 2010; Caporali and Emanueli, 2011; Dai and Ahmed, 2011; is suitable for targeted quantification and validation of miRNAs Rasheed et al., 2013). Environmental factors including nutrition, profiling results. In contrast, miRNAs arrays collectively allow for sleep, exercise, hypoxia, and stress also contribute to the modu- the accurate quantification of 100s of miRNAs; a highly efficient lation of miRNAs expression (Chan et al., 2009; Wang and Cui, method to establish the extended miRNAs expression profile of 2012; Zacharewicz et al., 2013). multiple tissue samples. However, miRNAs arrays are not suitable Skeletal muscle is one of the largest organ of the body, making for the analysis of a small number of miRNAs on a large number up approximately 40% of the whole body mass. Skeletal muscle of samples. Although the use of individual assays seems to be the is a highly plastic tissue able to adapt its size, structure and func- most appropriate approach for this type of analyzes, the reverse tion in response to various internal and external stimuli, such as transcription (RT) and quantification of each individual miRNA acute exercise, hypoxia, and training. miRNAs have been recently is time and reagent consuming. Recently, ABI described on their identified as novel, essential regulators of skeletal muscle health web site (Protocol for Creating Custom RT and Preamplification (Zacharewicz et al., 2013) and may account for specific regulation Pools using TaqMan® MicroRNA Assays User Bulletin (Pub. no. of muscle growth and differentiation (Buckingham and Rigby, 4465407 Rev. C) – cms_094060.pdf, 2014) a method allowing to 2014). miRNAs localization is, however, not restricted to cells and multiplex the RT and pre-amplification (PA) steps. To our knowl- some miRNAs produced in cells are secreted in the bloodstream edge, this method was never validated in the literature. The aim www.frontiersin.org November 2014 | Volume 5 | Article 413 | 1 Le Carré et al. Multiplexing reverse transcription for microRNAs quantification of the present study was to validate and adapt this protocol on Plasma human muscle and plasma samples. Blood samples were obtained from four healthy subjects partici- pating in a study previously published by our group (Faiss et al., DESCRIPTION OF METHODS 2013). Peripheral blood samples were collected from the antecu- METHOD DESIGN bital vein in 2 × 2.6 ml EDTA tubes (Sarstedt S-Monovette) using In this report, we describe and validate a modified TaqMan® Small a butterfly device. The samples were immediately centrifuged at RNA Assay method allowing the simultaneous RT followed by 3500 g at 4◦ C for 10 min and the upper phase was collected. Plasma the PA of multiple miRNAs in both human muscle and plasma samples were then frozen in liquid nitrogen and stored at −80◦ C samples. The different steps of the method for each type of sample until further processing. are depicted in Figure 1. The low amount of RNA in plasma justifies the addition of a PA step prior to the real-time qPCR RNA EXTRACTION to enhance the sensitivity of the reaction. Irrespectively of the Muscle tissue used ABI suggests to perform a PA step for any starting Total RNA from skeletal muscle sample (approximately 25 mg of RNA amount smaller than 350 ng. Previous experiments from muscle) was isolated using a commercially available preparation, our lab demonstrated that a PA step is not necessary with skeletal TriReagent® (Molecular Research Center, Inc., Cincinnati, OH, muscle tissue as muscle miRNAs concentration is generally high USA), following the manufacturer’s instructions and as previously enough to yield reliable results. We consequently decided to apply published by our group (Léger et al., 2006). Briefly, tissue samples the PA step to plasma tissue only. The RT primer pool designed to were homogenized in 500 μl TriReagent® with a power homog- analyze muscle and plasma miRNAs consisted in 11 (miR-1, miR- enizer (Polytron® System PT2100, Kinematica AG, Lucerne) and 15a, miR-16, miR-21, miR-126, miR-133a, miR-210, miR-221, incubated for 5 min at room temperature (R.T). Following this, miR-222, RNU44, and RNU48) and 8 (miR-16, miR-20a, miR-21, 100 μl of chloroform were added and the sample was mixed dur- miR-126, miR-133a, miR-146a, miR-210, and miR-454) miRNA ing 20 s before being incubated for 10 min at R.T. After a 20 min specific primers sets, respectively. These miRNAs were selected on centrifugation at R.T (12’000 g), 250 μl isopropanol were added the basis of their relevance and expression levels in human muscle to the aqueous phase and the sample was mixed for 20 s. Sam- and plasma, respectively (Mitchell et al., 2008; Baggish et al., 2011; ples were precipitated overnight at −20◦ C and then centrifuged Davidsen et al., 2011; Zampetaki et al., 2012a; Aoi et al., 2013; Bye for 30 min at 4◦ C (12’000 g). RNA pellet was washed with 75% et al., 2013; Nielsen et al., 2014). In each type of sample, four ethanol before being resuspended in 20 μl dH2 O and stored at miRNAs were further selected to complete the validation process. −80◦ C until use. SAMPLES COLLECTION AND HANDLING Plasma Muscle Plasma aliquots were thawed on ice and centrifuged at 10’000 g for Biopsies from the vastus lateralis muscle were obtained from four 10 min at 4◦ C to remove any remaining cellular contents. 400 μl healthy subjects participating in a study previously published by of plasma were used for total RNA extraction using the mirVana our group (Faiss et al., 2013). PARIS kit (Life Technologies, Ambion, #AM1556) following the FIGURE 1 | Method design. Frontiers in Genetics | Non-Coding RNA November 2014 | Volume 5 | Article 413 | 2 Le Carré et al. Multiplexing reverse transcription for microRNAs quantification manufacturer’s protocol with minor modifications. Briefly, 400 μl 2 min at 55◦ C, 2 min at 72◦ C, followed by 13 cycles of 15 s at 95◦ C, of 2×denaturing solution and 800 μl of acid-phenol:chloroform 4 min at 60◦ C and 10 min at 99.9◦ C. At the end of the run, the PA were added to the plasma. Following this, the samples were vor- products were diluted 4× in 0.1× TE buffer pH 8.0 and stored at texed for 60 s and incubated for 10 min on ice. After 20 min of −20◦ C. centrifugation at 13’000 g at 4◦ C, the supernatant was collected and 1 ml of 100% ethanol was added. The lysate/ethanol mixture Individual cDNA pre-amplification mix was then transferred onto a filter cartridge and centrifuged for Protocol for the individual cDNA PA mix was as described by 15 s at 10’000 g at R.T. Filter was washed and total RNA was eluted manufacturer. Briefly, the PA reaction combined 2 μl of RT prod- with 80 μl of pre-heated (95◦ C) elution solution provided by the uct with 10 μl 2× TaqMan® PreAmp Master Mix (#4391128), manufacturer and then frozen at −80◦ C. 2 μl 10× Megaplex Preamp primers V2.1 (#4399233) and 6 μl DEPC-treated water. REVERSE TRANSCRIPTION Individual miRNA RT Multiple cDNAs pre-amplification mix miRNAs were reverse-transcribed using the TaqMan® microRNA To simultaneously pre-amplify multiple cDNAs, we created a RT kit (Applied Biosystems, USA, #4366596) and the associated PA primer pool targeting the same miRNAs that were reverse- miRNA-specific stem-loop primers (TaqMan® microRNA assay transcribed and containing 5 μl of each individual 20× TaqMan® kit, #4427975). Total RNA from muscle was diluted at a concentra- Small RNA Assay (part of #4427975) diluted in 500 μl 1× TE. The tion of 12.5 ng/μl and 4 μl of RNA were added to the reaction mix reaction mix was prepared by combining 3.75 μl of PA primer pool containing 0.15 μl 100 mM dNTP, 1 μl enzyme (50 U/μl), 1.5 μl with 2.5 μl of RT product, 12.5 μl of 1× TaqMan® Universal PCR 10× RT buffer, 0.19 μl RNase inhibitor (20 U/μl), 1.5 μl 5× RT MasterMix (2×), no UNG (#4440040) and 6.25 μl DEPC-treated specific-primer and 7.66 μl DEPC-treated water to obtain a final water. volume of 15 μl. The used primer concentration is twofold lower than the concentration recommended by ABI. In order to save REAL-TIME QUANTITATIVE PCR reagent, we evaluated the possibility to reduce the primer concen- All RT-qPCRs were carried out in triplicate with ABI products tration and tested the effect of a twofold primer dilution. Results and were performed on the MX3000p thermal cycler system from showed that reducing the final concentration of primers only had Stratagene with the following conditions: one denaturing step at a minor impact on the detection threshold, which increased from 95◦ C for 10 min, followed by 40 cycles consisting of denaturing less than 0.5 cycles (data not shown). Thus, primer was used at at 95◦ C for 15 s and annealing and elongation at 60◦ C for 60 s, a final concentration of 0.5× for all further analyzes. Concern- followed by an inactivation step of 10 min at 99.9◦ C. qPCR target ing the RT of plasma miRNAs, due to the low plasma miRNA sequences are provided in Table 1. expression levels, a fixed volume of 3 μl of the 80 μl plasma RNA eluate was used as input in the reaction mix described above. RT reaction conditions were as follows: 30 min at 16◦ C to anneal Table 1 | List of TaqMan® miRNAs and snoRNAs used in this study. primers, 30 min at 42◦ C for the extension of primers on miRNA and the synthesis of the first cDNA strand, 5 min at 85◦ C to stop Name Assay ID Target sequence the reaction. cDNA was then stored at −20◦ C until use. hsa-miR-1 002222 UGGAAUGUAAAGAAGUAUGUAU Multiple miRNA reverse transcription hsa-miR-15a 000389 UAGCAGCACAUAAUGGUUUGUG miRNAs were reverse-transcribed using the TaqMan® microRNA hsa-miR-16 000391 UAGCAGCACGUAAAUAUUGGCG RT kit (#4366596) and the associated miRNA-specific stem-loop hsa-miR-20a 000580 UAAAGUGCUUAUAGUGCAGGUAG primers (TaqMan® microRNA assay kit, #4427975) with some hsa-miR-21 000397 UAGCUUAUCAGACUGAUGUUGA modifications. A customized RT primer pool was prepared by pooling all miRNA-specific stem-loop primers of interest. In brief, hsa-miR-126 002228 UCGUACCGUGAGUAAUAAUGCG miRNA-specific primers were pooled and diluted in 1× Tris-EDTA hsa-miR-133a 002246 UUUGGUCCCCUUCAACCAGCUG (TE) buffer to obtain a final dilution of 0.05× each. 6 μl of this hsa-miR-146a 000468 UGAGAACUGAAUUCCAUGGGUU mixture were added to the reaction mix containing 0.3 μl 100 mM hsa-miR-210 000512 CUGUGCGUGUGACAGCGGCUGA dNTP, 3 μl enzyme (50 U/μl), 1.5 μl 10× RT buffer, 0.19 μl hsa-miR-221 000524 AGCUACAUUGUCUGCUGGGUUUC RNase inhibitor (20 U/μl) and 50 ng of muscle total RNA or 3 μl of plasma RNA. A final volume of 15 μl was reverse-transcribed hsa-miR-222 002276 AGCUACAUCUGGCUACUGGGU with the following conditions: 30 min at 16◦ C to anneal primers, hsa-miR-454 002323 UAGUGCAAUAUUGCUUAUAGGGU 30 min at 42◦ C for the extension phase, 5 min at 85◦ C to stop the hsa-miR-486 001278 UCCUGUACUGAGCUGCCCCGAG reaction. cDNA was then stored at −20◦ C. hsa-miR-494 002365 UGAAACAUACACGGGAAACCUC RNU44 001094 CCUGGAUGAUGAUAGCAAAUGCUGACUGAA PRE-AMPLIFICATION CAUGAAGGUCUUAAUUAGCUCUAACUGACU In order to increase the amount of cDNA and to improve the sensitivity of the TaqMan® qPCR reaction, a PA step was performed RNU48 001006 GAUGACCCCAGGUAACUCUGAGUGUGUCG on plasma cDNAs. PA PCR conditions consisted in 10 min at 95◦ C, CUGAUGCCAUCACCGCAGCGCUCUGACC www.frontiersin.org November 2014 | Volume 5 | Article 413 | 3 Le Carré et al. Multiplexing reverse transcription for microRNAs quantification Table 2 | Improvement of sensitivity through pre-amplification (PA) step on individual and pooled cDNA. Individual cDNA Pooled cDNAs w/o PA (CT) With PA (CT) Differencies w/o PA (CT) With PA (CT) Differencies miR-16 25.83 17.72 8.1 28.73 20.64 8.1 miR-20a 30.77 21.82 8.9 30.96 23.67 7.3 miR-146a 31.40 23.21 8.2 34.57 26.35 8.2 miR-454 32.53 24.89 7.6 35.05 26.66 8.4 RT-qPCR with individual cDNA based on RNA dilution and not cDNA. A standard curve including Real-time PCR reactions were modified for a smaller final volume 5 RNA dilution points (5, 10, 50, 100, and 200 ng of total RNA) of 15 μl per well, using the same reagent proportions as recom- was established for the miRNAs of interest present in muscle tissue. mended by the manufacturer. In each well, 1 μl of muscle cDNA or This represents a dilution of cDNA ranging from 0.067 to 2.6 ng. 1.5 μl of diluted PA plasma cDNA was added to the reaction mix The low amount of RNA in plasma prevents standard RNA quan- containing 0.75× TaqMan® Small RNA Assay (20×; #4427975) tification via optical density measurement. Thus, a fixed volume of and 0.75× TaqMan® Universal PCR MasterMix (2×), no UNG plasma RNA was used to perform a standard curve ranging from 1 (#4440040). Plates were mixed by hand and briefly centrifuged to 4 μl of plasma RNA. All dilution points were reverse-transcribed before being loaded onto the qPCR machine. separately before being amplified for each miRNA. All tested miR- NAs were successfully amplified in both muscle and plasma. The RT-qPCR with multiple cDNAs efficiency and the coefficient of determination (R2 ) were deter- Real-time PCR reactions with multiple cDNAs were performed in a mined from the standard curve for each miRNA present in the RT 20 μl final volume. A reaction mix containing 0.2 μl of multiplexed primer pool. The R2 of qPCR reactions on muscle samples were muscle cDNAs or 0.2 μl of pre-amplified plasma cDNAs, 7.5 μl higher than 0.993 (Figure 2A) with efficiency ranged from 94.9 to 2× TaqMan® Universal PCR MasterMix, no UNG (#4440040) 166.6%. The R2 of qPCR reactions on plasma samples were higher and 11.3 μl DEPC-water was loaded in each well and 1 μl of 1× TaqMan® Small RNA Assay (20×; #4427975) was added. The used MasterMix concentration is 25% lower than recommended by ABI. Previous experiments from our lab demonstrated that such reduction of the MasterMix concentration did not influence detec- tion thresholds. Plates were mixed by hand and briefly centrifuged before being loaded onto the qPCR machine. RESULTS PRE-AMPLIFICATION STEP As expected, the PA step significantly reduced CT detection thresh- old in both individual and pooled experiments. Table 2 shows the CT values for the two conditions aforementioned and the dif- ferences in CT for the four miRNAs selected. In our hands we observed an average of 8.2 and 8.0 cycles difference for individual and pooled cDNA, respectively. SENSITIVITY AND SPECIFICITY The sensitivity of miRNA quantification in muscle and plasma using the customized TaqMan® miRNA Assay protocol was eval- uated by a dose-response curve for each miRNA used in the RT primer pool. The RT primer pool for muscle analysis consisted in miR-1, miR-15a, miR-16, miR-21, miR-126, miR-133a, miR-210, miR-221, miR-222, RNU44, and RNU48. In the same manner, the RT primer pool for plasma analysis consisted in miR-16, miR-20a, miR-21, miR-126, miR-133a, miR-146a, miR-210, and miR-454. The protocol described in this manuscript is based on a fixed vol- ume of RT product to be used in the PCR reaction. Therefore, FIGURE 2 | (A) Standard curves from 11 miRNAs expressed in human muscle tissue. (B) Standard curves from eight microRNAs (miRNAs) increasing the cDNA input requires to increase the RNA amount expressed in human plasma samples. used in the RT reaction. This explains why the standard curve was Frontiers in Genetics | Non-Coding RNA November 2014 | Volume 5 | Article 413 | 4 Le Carré et al. Multiplexing reverse transcription for microRNAs quantification than 0.960 (Figure 2B) with efficiency ranged from 88.5 to 119.1% miRNAs (miR-1, miR-15a, miR-16, miR-21, miR-126, miR- (Table 3). These results demonstrate a good linearity of the dif- 133a, miR-210, miR-221, miR-222, RNU44, and RNU48) ferent assays and the fact that the first point of the standard curve were simultaneously reverse-transcribed and then four miRNAs is as low as 5 ng RNA (corresponding to predicted cDNA inputs (miR-1, miR-16, miR-21, and RNU44) were amplified indi- of 0.067 ng for the multiplex method) indicates a high sensitivity vidually by qPCR. Similarly, one plasma sample was divided of miRNAs amplification. The efficiency of these standard curves in three aliquots of 400 μl each and RNA was extracted as does not only reflect the specificity of qPCR primers but also the described. For each plasma sample, eight miRNAs (miR-16, efficiency of the RT to convert RNA into cDNA. Indeed, standard miR-20a, miR-21, miR-126, miR-133a, miR-146a, miR-210, curves do not result from a serial dilution of a single RT product, and miR-454) were simultaneously reverse-transcribed and then but from a serial dilution of a unique RNA sample, thus five RT pre-amplified. Four plasma miRNAs (miR-16, miR-20a, miR- products. This explains why some miRNAs have an excellent lin- 146a, and miR-454) were then analyzed individually by qPCR. earity despite of a reduced or increased efficiency. To verify that The coefficient of variation (CV%) was calculated to deter- the efficiency of miR-21 (166.6%) in muscle was not related to the mine the intra-assay variability. The CVs ranged from 0.27 to multiplexing method, we completed a standard curve following 0.38% and from 0.21 to 0.45% for muscle and plasma sam- an individual RT of miR-21. Results showed an efficiency largely ples, respectively, indicating a high reproducibility of the assay superior to 100% as well (142%, R2 = 0.95), suggesting that the (Table 4). efficiency of miR-21 in the multiplexing protocol is not related to the method. Together, these results revealed that the presence ACCURACY of multiple miRNA-specific stem-loop primers in the same RT To assess the accuracy of the RT-qPCR system described reaction mix did not alter or inhibit the RT reaction by non- herein, we show that the inter-subject expression variabil- specific interactions. Moreover, the presence of multiple cDNA ity for one specific miRNA is similar in the individual and species in the qPCR reaction mix did not influence the speci- pooled method. Therefore, we extracted RNA from four mus- ficity of the amplification. Furthermore, a no template control cle and four plasma samples. Eight miRNAs used for the (NTC) was run for each miRNA to rule out cross contaminations repeatability experiment in muscle (miR-1, miR-16, miR-21, of reagents or surfaces. No amplification curves were observed for and RNU44) and plasma (miR-16, miR-20a, miR-146a, and any NTC, while all 60 assays were successfully amplified suggesting miR-454) were reverse-transcribed in each sample, once indi- the absence of non-specific interactions. vidually and once pooled in the same RT. After the amplifi- cation of these cDNAs, we observed that the individual cycle REPEATABILITY threshold (CT) values differed depending on the RT approach Intra-assay variability was tested to assess the repeatability chosen. The use of reagents at different volumes and con- of the assay. Three independent RTs were performed from centrations in the RT reaction explains these variations. It the same muscle RNA sample. For each aliquot, 11 muscle is however expected that the ratio CT (pooled cDNAs)/CT (individual cDNA) remains stable. The coefficient of varia- tion (CV%) of this ratio was calculated for each reaction. The Table 3 | R 2 and efficiency of the standard curves obtained with the modified TaqMan® Small RNA Assay. inter-assay CVs ranged from 0.5 to 2.3% and from 1.2 to 3.0% for muscle and plasma samples, respectively (Table 5), Muscle Plasma indicating that the CT value of the miRNAs amplified with R2 Efficiency R2 Efficiency mir-1 0.999 117.9 miR-16 0.986 100.0 Table 4 | Intra-assay repeatability of the modified TaqMan® Small RNA Assay. miR-15a 0.993 110.1 miR-20a 0.988 107.0 miR-16 0.999 133.5 miR-21 0.960 110.3 Aliquot 1 Aliquot 2 Aliquot 3 miR-21 0.994 166.6 miR-126 0.993 94.4 CT CT CT Average CT SD CV % miR-126 0.999 94.9 miR-133a 0.996 88.5 Muscle miR-133a 1.000 99.0 miR-146a 0.967 109.4 miR-1 25.93 25.81 25.93 25.89 0.07 0.27 miR-210 0.996 101.0 miR-454 0.994 119.1 miR-16 26.58 26.41 26.53 26.51 0.09 0.33 miR-221 0.993 133.2 miR-486 0.998 96.6 miR-21 34.94 35.09 35.18 35.07 0.12 0.34 miR-222 0.996 115.7 RNU44 33.98 33.88 34.13 34.00 0.13 0.38 RNU44 0.999 100.6 Plasma RNU48 0.998 101.3 miR-16 21.94 21.77 21.78 21.83 0.10 0.45 R2 value is the coefficient of determination of the standard curves and the miR-20a 26.45 26.24 26.26 26.32 0.12 0.44 efficiency is the rate at which the polymerase chain reaction (PCR) amplicon miR-146a 26.69 26.65 26.58 26.64 0.06 0.21 is generated. The R2 and the efficiency were calculated by the MxPro qPCR Software. miR-454 28.79 28.64 28.75 28.73 0.08 0.28 www.frontiersin.org November 2014 | Volume 5 | Article 413 | 5 Le Carré et al. Multiplexing reverse transcription for microRNAs quantification Table 5 | Coefficient of variation (CV%) of the ratio CT (pooled cDNAs)/CT (individual cDNA). Ratio CT pooled/individual method Sample 1 Sample 2 Sample 3 Sample 4 Average SD CV % Muscle miR-1 1.43 1.38 1.42 1.42 1.41 0.02 1.8 miR-16 1.20 1.19 1.18 1.16 1.18 0.02 1.3 miR-21 1.36 1.38 1.33 1.31 1.35 0.03 2.3 RNU44 1.29 1.29 1.29 1.28 1.29 0.01 0.5 Plasma miR-16 1.24 1.19 1.28 1.22 1.23 0.04 3.0 miR-20a 1.27 1.25 1.29 1.26 1.26 0.02 1.6 miR-146a 1.15 1.19 1.20 1.19 1.18 0.02 1.7 miR-454 1.12 1.12 1.14 1.14 1.13 0.01 1.2 the pool method accurately reflects the expected levels of ACKNOWLEDGMENTS expression. The authors are grateful to Raphaël Faiss for providing with muscle biopsies and plasma samples. Séverine Lamon is supported by an CONCLUSION Alfred Deakin Postdoctoral Fellowship from Deakin University. The individual analysis of multiple miRNAs is time and reagent consuming. Here, we describe an approach allow- REFERENCES ing to simultaneously measure multiple miRNAs in muscle Aoi, W., Ichikawa, H., Mune, K., Tanimura, Y., Mizushima, K., Naito, Y., et al. (2013). Muscle-enriched microRNA miR-486 decreases in circulation in and plasma samples. To our knowledge this method was response to exercise in young men. Front. Physiol. 4:80. doi: 10.3389/fphys.2013. never described in the literature whereas ABI published on 00080 their web site a protocol for creating custom RT pools using Baggish, A. L., Hale, A., Weiner, R. B., Lewis, G. D., Systrom, D., Wang, F., et al. TaqMan® microRNA Assays. In this way, we validate and (2011). Dynamic regulation of circulating microRNA during acute exhaustive bring minor adaptations to a technique allowing to multi- exercise and sustained aerobic exercise training. J. Physiol. 589, 3983–3994. doi: 10.1113/jphysiol.2011.213363. plex the RT and PA steps. Validation results show that this Bartel, D. P., Lee, R., and Feinbaum, R. (2004). MicroRNAs: Genomics, Biogenesis, customized method is not only sensitive and highly specific Mechanism, and Function Genomics: the miRNA Genes. Cell 116, 281–297. doi: but also repeatable and accurate. In addition, the adapta- 10.1016/S0092-8674(04)00045-5 tions brought to the original protocol allow to save expensive Buckingham, M., and Rigby, P. W. J. (2014). Gene regulatory networks and tran- reagents, such as stem-loop primers, MasterMix and PA mix, scriptional mechanisms that control myogenesis. Dev. Cell 28, 225–238. doi: 10.1016/j.devcel.2013.12.020 as well as time by eliminating the PA step for every tissue but Bueno, M. J., de Castro, I. P., and Malumbres, M. (2008). Control of cell prolif- plasma. eration pathways by microRNAs. Cell Cycle 7, 3143–3148. doi: 10.4161/cc.7.2 Lao et al. (2006) previously described an elegant and effi- 0.6833 cient multiplexing method based on TaqMan® probes that Bye, A., Rø sjø, H., Aspenes, S. T., Condorelli, G., Omland, T., and Wisløff, U. (2013). Circulating microRNAs and aerobic fitness–the HUNT-Study. PLoS ONE allows the simultaneous amplification and analysis of up to 8:e57496. doi: 10.1371/journal.pone.0057496 190 miRNAs. However, this method is restricted to the Caporali, A., and Emanueli, C. (2011). MicroRNA regulation in angiogenesis. Vascul. use of a pre-designed RT primer pool and therefore does Pharmacol. 55, 79–86. doi: 10.1016/j.vph.2011.06.006 not allow the multiplex analysis of novel miRNAs that may Chan, S. Y., Zhang, Y.-Y., Hemann, C., Mahoney, C. E., Zweier, J. L., and Loscalzo, not be included in this pool. In contrast, our protocol J. (2009). MicroRNA-210 controls mitochondrial metabolism during hypoxia by repressing the iron-sulfur cluster assembly proteins ISCU1/2. Cell Metab. 10, allows to create a RT primer pool unique to the miRNAs 273–284. doi: 10.1016/j.cmet.2009.08.015 of interest. Moreover, whereas Lao et al. (2006) results must Chen, C., Ridzon, D. A., Broomer, A. J., Zhou, Z., Lee, D. H., Nguyen, J. T., et al. be treated as microarray results and therefore require fur- (2005). Real-time quantification of microRNAs by stem-loop RT-PCR. Nucleic ther confirmation by single PCR, the method we describe Acids Res. 33:e179. doi: 10.1093/nar/gni178 Chen, X., Ba, Y., Ma, L., Cai, X., Yin, Y., Wang, K., et al. (2008). Characterization here does not require this step. Our method is therefore of microRNAs in serum: a novel class of biomarkers for diagnosis of cancer and of high relevance to simultaneously analyze a limited num- other diseases. Cell Res. 18, 997–1006. doi: 10.1038/cr.2008.282 ber of miRNAs, selected from a previous microarray screen- Dai, R., and Ahmed, S. A. (2011). MicroRNA, a new paradigm for understanding ing or from the literature, on a large number of sam- immunoregulation, inflammation, and autoimmune diseases. Transl. Res. 157, ples. Altogether, it will facilitate the identification and mea- 163–179. doi: 10.1016/j.trsl.2011.01.007 Davidsen, P. K., Gallagher, I. J., Hartman, J. W., Tarnopolsky, M. A., Dela, F., Helge, sure of specific miRNAs biomarkers and greatly contribute J. W., et al. (2011). High responders to resistance exercise training demonstrate to the understanding of how miRNAs regulate biological differential regulation of skeletal muscle microRNA expression. J. Appl. Physiol. processes. (1985) 110, 309–17. doi: 10.1152/japplphysiol.00901.2010 Frontiers in Genetics | Non-Coding RNA November 2014 | Volume 5 | Article 413 | 6 Le Carré et al. Multiplexing reverse transcription for microRNAs quantification Faiss, R., Léger, B., Vesin, J.-M., Fournier, P.-E., Eggel, Y., Dériaz, O., et al. (2013). Rasheed, S. A. K., Teo, C. R., Beillard, E. J., Voorhoeve, P. M., and Casey, P. J. (2013). Significant molecular and systemic adaptations after repeated sprint training in MicroRNA-182 and microRNA-200a control G-protein subunit α-13 (GNA13) hypoxia. PLoS ONE 8:e56522. doi: 10.1371/journal.pone.0056522 expression and cell invasion synergistically in prostate cancer cells. J. Biol. Chem. Gidron, Y., Zwaan, M., De Quint, K., and Ocker, M. (2010). Influence of stress 288, 7986–95. doi: 10.1074/jbc.M112.437749. and health-behaviour on miRNA expression. Mol. Med. Rep. 3, 455–457. doi: Sood, P., Krek, A., Zavolan, M., Macino, G., and Rajewsky, N. (2006). Cell-type- 10.3892/mmr specific signatures of microRNAs on target mRNA expression. Proc. Natl. Acad. Krol, J., Loedige, I., and Filipowicz, W. (2010). The widespread regulation of Sci. U.S.A. 103, 2746–2751. doi: 10.1073/pnas.0511045103 microRNA biogenesis, function and decay. Nat. Rev. Genet. 11, 597–610. doi: Subramanian, S., and Steer, C. J. (2010). MicroRNAs as gatekeepers of apoptosis. 10.1038/nrg2843 J. Cell. Physiol. 223, 289–98. doi: 10.1002/jcp.22066 Lao, K., Xu, N. L., Yeung, V., Chen, C., Livak, K. J., and Straus, N. A. (2006). Wang, J., and Cui, Q. (2012). Specific roles of microRNAs in their interactions Multiplexing RT-PCR for the detection of multiple miRNA species in small sam- with environmental factors. J. Nucleic Acids 2012:978384. doi: 10.1155/2012/ ples. Biochem. Biophys. Res. Commun. 343, 85–89. doi: 10.1016/j.bbrc.2006. 978384 02.106 Zacharewicz, E., Lamon, S., and Russell, A. P. (2013). MicroRNAs in skeletal muscle Lawrie, C. H., Saunders, N. J., Soneji, S., Palazzo, S., Dunlop, H. M., Cooper, C. D. O., and their regulation with exercise, ageing, and disease. Front. Physiol. 4:266. doi: et al. (2008). MicroRNA expression in lymphocyte development and malignancy. 10.3389/fphys.2013.00266 Leukemia 22, 1440–1446. doi: 10.1038/sj.leu.2405083 Zampetaki, A., Willeit, P., Drozdov, I., Kiechl, S., and Mayr, M. (2012a). Profiling of Léger, B., Cartoni, R., Praz, M., Lamon, S., Dériaz, O., Crettenand, A., et al. circulating microRNAs: from single biomarkers to re-wired networks. Cardiovasc. (2006). Akt signalling through GSK-3beta, mTOR and Foxo1 is involved in Res. 93, 555–562. doi: 10.1093/cvr/cvr266 human skeletal muscle hypertrophy and atrophy. J. Physiol. 576, 923–933. doi: Zampetaki, A., Willeit, P., Tilling, L., Drozdov, I., Prokopi, M., Renard, J.-M., et al. 10.1113/jphysiol.2006.116715 (2012b). Prospective study on circulating microRNAs and risk of myocardial Lim, L. P., Lau, N. C., Garrett-Engele, P., Grimson, A., Schelter, J. M., Castle, infarction. J. Am. Coll. Cardiol. 60, 290–9. doi: 10.1016/j.jacc.2012.03.056 J., et al. (2005). Microarray analysis shows that some microRNAs downregu- late large numbers of target mRNAs. Nature 433, 769–73. doi: 10.1038/nature Conflict of Interest Statement: The authors declare that the research was conducted 03315 in the absence of any commercial or financial relationships that could be construed Mei, Q., Li, X., Meng, Y., Wu, Z., Guo, M., Zhao, Y., et al. (2012). A facile and specific as a potential conflict of interest. assay for quantifying microRNA by an optimized RT-qPCR approach. PLoS ONE 7:e46890. doi: 10.1371/journal.pone.0046890 Received: 09 September 2014; accepted: 06 November 2014; published online: 26 Mitchell, P. S., Parkin, R. K., Kroh, E. M., Fritz, B. R., Wyman, S. K., Pogosova- November 2014. Agadjanyan, E. L., et al. (2008). Circulating microRNAs as stable blood-based Citation: Le Carré J, Lamon S and Léger B (2014) Validation of a multiplex reverse markers for cancer detection. Proc. Natl. Acad. Sci. U.S.A. 105, 10513–10518. doi: transcription and pre-amplification method using TaqMan® MicroRNA assays. Front. 10.1073/pnas.0804549105 Genet. 5:413. doi: 10.3389/fgene.2014.00413 Nielsen, S., Åkerström, T., Rinnov, A., Yfanti, C., Scheele, C., Pedersen, B. K., This article was submitted to Non-Coding RNA, a section of the journal Frontiers in et al. (2014). The miRNA plasma signature in response to acute aerobic exer- Genetics. cise and endurance training. PLoS ONE 9:e87308. doi: 10.1371/journal.pone.00 Copyright © 2014 Le Carré, Lamon and Léger. This is an open-access article distributed 87308 under the terms of the Creative Commons Attribution License (CC BY). The use, dis- Protocol for Creating Custom RT, and Preamplification Pools using TaqMan® tribution or reproduction in other forums is permitted, provided the original author(s) MicroRNA Assays User Bulletin (Pub. no. 4465407 Rev. C) – cms_094060. pdf or licensor are credited and that the original publication in this journal is cited, in (2014). Available at: https://tools.lifetechnologies.com/content/sfs/manuals/cms_ accordance with accepted academic practice. No use, distribution or reproduction is 094060.pdf [accessed September 1, 2014]. permitted which does not comply with these terms. www.frontiersin.org November 2014 | Volume 5 | Article 413 | 7