CLB-08532; No. of pages: 4; 4C: Clinical Biochemistry xxx (2013) xxx–xxx
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Short Communication
Comparison of serum exosome isolation methods for microRNA profiling Kadri Rekker a,b,⁎, Merli Saare a,b,c, Anne Mari Roost a, Anna-Liisa Kubo d, Natasa Zarovni d, Antonio Chiesi d, Andres Salumets a,b,c, Maire Peters a,b a
Competence Centre on Reproductive Medicine and Biology, Tiigi 61b, 50410 Tartu, Estonia Department of Obstetrics and Gynecology, University of Tartu, L. Puusepa 8, 51014 Tartu, Estonia Institute of Bio- and Translational Medicine, University of Tartu, Ravila 19, 50411 Tartu, Estonia d HansaBioMed Ltd, Akadeemia tee 15, 12618 Tallinn, Estonia b c
a r t i c l e
i n f o
Article history: Received 27 August 2013 Received in revised form 15 October 2013 Accepted 21 October 2013 Available online xxxx Keywords: Exosomes miRNA Extracellular vesicles Ultracentrifugation ExoQuick
a b s t r a c t Objectives: Exosomes are small membrane bound vesicles secreted by most cell types. Exosomes contain various functional proteins, mRNAs and microRNAs (miRNAs) that could be used for diagnostic and therapeutic purposes. Currently, a standard method for serum exosome isolation is differential ultracentrifugation, but a search for alternative, less time-consuming and labour extensive exosomal isolation method for use in clinical settings is ongoing. The effect of serum exosome isolation method on obtained miRNA profile is not yet clear. The aim of this study was to determine to which extent selected exosome isolation methods influence the serum exosomal miRNA profile. Design and methods: Exosomes were isolated from blood serum of healthy individuals by ultracentrifugation and ExoQuick Precipitation methods. The expression profile of 375 miRNAs was determined by real time PCR using Exiqon miRCURY LNA™ microRNA Human panel I assays. Results: Although a strong correlation of exosomal miRNA profiles was observed between the two isolation methods, distinct clusters of miRNA levels between the used methods were identified. The detected levels of two miRNAs, miR-92a and miR-486-5p, were significantly influenced by the exosome isolation method used. Conclusions: Both exosome isolation methods are suitable for serum exosomal miRNA profiling. Differences found in miRNA patterns between the two methods indicate that the observed exosomal miRNA profile is slightly affected by the extracellular vesicle isolation method. © 2013 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction Cells release various types of membrane vesicles into the extracellular space that differ in origin, size, morphology and content [1]. The most studied of them are exosomes, small (40–100 nm) vesicles of endocytic origin. Exosomes have been isolated from a number of body fluids, including blood plasma and serum, breast milk, saliva and urine (reviewed in [2]). These extracellular vesicles contain various functional proteins, microRNAs (miRNAs) and mRNAs [3] which have a potential to be used as biomarkers for numerous diseases. Previous studies have demonstrated that exosomes are actively released by cancerous cells into the peripheral circulation [4] and biomarker studies highlight the possibility to use exosomal protein content and RNA profiles in cancer diagnosis [5,6]. Although circulating exosomes and their content can serve as biomarkers for various diseases, the standard ultracentrifugation based exosome isolation methods established for research purposes are Abbreviations: UC, ultracentrifugation; EQ, ExoQuick Precipitation Solution. ⁎ Corresponding author at: Competence Centre on Reproductive Medicine and Biology, Tiigi 61b, 50410 Tartu, Estonia. E-mail address: [email protected] (K. Rekker).
time-consuming and labour intensive and may not be suitable for use in clinical laboratories. Recently, alternative methods for exosome isolation with reduced hands-on time and without the requirement for specialised expensive equipment have been developed. For instance, precipitation based methods have been used for biomarker discovery studies [7] and a microfiltration based method for the isolation of urinary exosomes has been introduced [8]. To our knowledge, a comparative study of blood serum exosome isolation methods for miRNA profiling has not been conducted. Therefore, the aim of our study was to determine whether the exosomal miRNA profile in serum samples depends on exosome isolation method. In this study, miRNA content from exosomes obtained by a “gold standard” ultracentrifugation based method and ExoQuick Precipitation Solution was compared.
Material and methods The study was approved by the Ethics Review Committee on Human Research of the University of Tartu (Tartu, Estonia) and informed consent was obtained from all participants. A detailed description of all materials and methods used can be found in Supplementary data.
0009-9120/$ – see front matter © 2013 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clinbiochem.2013.10.020
Please cite this article as: Rekker K, et al, Comparison of serum exosome isolation methods for microRNA profiling, Clin Biochem (2013), http:// dx.doi.org/10.1016/j.clinbiochem.2013.10.020
2
K. Rekker et al. / Clinical Biochemistry xxx (2013) xxx–xxx
Sample collection
Data analysis
Peripheral blood samples from five healthy women (age 24.8 ± 4.0 years (mean± SD); body mass index 22.0± 1.4kg/m2) were collected into serum collection tubes. Blood samples were centrifuged at 1600 ×g for 10 min for serum collection followed by an additional centrifugation step at 16,000 ×g for 10 min and filtration through 0.2 μm pore size syringe filter (Sartorius GmbH, Göttingen, Germany). In order to avoid miRNA expression differences due to inter-individual variations, the same serum samples were used for both exosome isolation methods.
Data analysis was performed with GenEx 5.3 Software (MultiD Analyses AB, Sweden). Pearson's correlation and two-tailed paired t-test were used for the comparison of exosome isolation methods. Significance threshold was set to a fold change ≥ 2 with a Bonferroni adjusted p-value ≤ 0.00048 (106 miRNA comparisons).
Exosome isolation Ultracentrifugation (UC) Exosomal fraction from 1 mL of serum was purified by four successive centrifugation steps. At first, centrifugation at 300 ×g for 10 min was performed, serum was diluted with sterile PBS in 1:1 ratio and centrifuged at 10,000×g for 30min followed by ultracentrifugation for 2h at 200,000 ×g (Optima MAX-XP, Beckman Coulter, rotor TLA 100.3). The crude exosome pellet was washed once in a large volume of PBS, filtered through a 0.2 μm syringe filter and centrifuged at 200,000 ×g for 1 h, then re-suspended in PBS. All centrifugations were performed at 4 °C. ExoQuick Precipitation (EQ) Exosomal fraction from 1 mL of serum was isolated by ExoQuick™ Exosome Precipitation Solution according to the manufacturer's recommendations (System Biosciences Inc., Mountain View, CA, USA). Briefly, 1/4 volume of ExoQuick Solution was added to serum and samples were refrigerated at 4 °C overnight. The mixture was centrifuged at 1500 ×g for 30min and supernatant was removed by aspiration. Pelleted fraction was re-suspended in nuclease-free water. RNA isolation RNA was extracted from exosomal fractions using miRNeasy Mini kit (Qiagen, Hilden, Germany) with the final elution volume of 50 μL. RNase A and Proteinase K treatment In separate experiments, exosomal fractions isolated with EQ and UC methods were treated with RNase A before and after RNA extraction. The presence of miRNAs in RNase A treated samples was determined by quantitative real-time PCR (qRT-PCR) using TaqMan® microRNA assays (Applied Biosystems, Foster City, CA, USA) hsa-let-7a-5p (assay ID 000377), hsa-miR-16-5p (000391) and hsa-miR-21-5p (000397). In addition, Proteinase K treatment of serum samples prior to exosomal isolation with EQ and UC methods was conducted. The presence of miRNAs in Proteinase K treated and untreated samples was assessed with TaqMan® microRNA assay hsa-miR-92a (000431). miRNA profiling Serum exosomal RNA samples isolated with UC (n=5) or EQ (n=5) methods from healthy women were used for miRNA profiling by real time PCR using miRCURY LNA™ microRNA Human panel I assays (Exiqon, Vedbaek, Denmark) following the manufacturer's recommendations. Briefly, cDNA synthesis was performed using the Universal cDNA Synthesis Kit (Exiqon). For each reaction 4 μL of exosomal RNA was used. Reverse transcription reaction products were combined with SYBR® Green master mix (Exiqon) and loaded onto ready-to-use Human Panel assays. Amplification was performed on a 7900HT thermocycler (Applied Biosystems) using cycling parameters recommended by Exiqon.
Results To evaluate whether the miRNAs extracted from the pelleted fraction collected with EQ and UC methods are confined inside vesicles, experiments with RNase A treatment were performed. All three miRNAs (hsa-let-7a, hsa-miR-16 and hsa-miR-21) analysed with TaqMan® assays were detected after the RNase A treatment of exosomal pellets isolated with both methods, indicating that analysed miRNAs are confined inside vesicles. miRNAs were not detected from samples that were treated with RNase A after the RNA isolation from the pellets. The expression profiles of 375 miRNAs in serum exosome samples isolated with UC (n = 5) and EQ (n = 5) methods were determined by qRTPCR using Exiqon Human panel I assays. The average numbers of detected exosomal miRNAs above the detection limit (Ct b 35) were 128 (range 100 to 160) and 93 (range 87 to 108) for EQ and UC methods, respectively. According to raw Ct values, the average miRNA detection threshold levels were 1.7 cycles lower in EQ samples, indicating that the total exosomal miRNA quantity isolated from an equal amount of serum is higher in EQ processed samples. The most abundant miRNAs detected by qRT-PCR were similar between the isolation methods (Table 1). A strong correlation was observed between the two methods (Fig. 1A, Pearson r = 0.95, p b 0.05). No miRNAs were uniquely dependent on exosome isolation method. However, miR-92a and miR-486-5p showed statistically different levels between the two methods (fold change 2.9 and 4.2, respectively; p-value b 0.00048). In addition, the clustering analysis of 17 miRNAs with the largest differences between the study groups (according to t-test, unadjusted p-value ≤ 0.05 and fold change ≥ 2) revealed a clearly distinct profile for both exosome isolation methods (Fig. 1B). To assess whether the differences in miR-92a levels could be related to possible co-precipitation of protein-bound miRNAs in isolated exosomal samples, proteinase K treatment was performed. miR-92a levels were determined from protease treated and untreated samples isolated with EQ and UC methods and a slight decrease in miR-92a levels of protease treated samples was observed in both cases. This Table 1 Top 15 serum exosomal miRNAs detected by Exiqon assays. miRNAs are listed according to abundance. Exosome isolation method ExoQuick Precipitation (EQ)
Ultracentrifugation (UC)
miRNA ID
Average Ct value
miRNA ID
Average Ct value
hsa-miR-451 hsa-miR-92aa hsa-miR-16 hsa-miR-486-5pa hsa-miR-19b hsa-miR-223 hsa-miR-20a hsa-miR-15a hsa-miR-93 hsa-miR-126 hsa-miR-144 hsa-miR-106a hsa-miR-21 hsa-miR-320a hsa-let-7b
22.0 25.0 25.5 25.5 25.9 27.2 27.2 27.3 27.5 27.7 27.7 27.8 27.8 27.9 28.0
hsa-miR-451 hsa-miR-16 hsa-miR-223 hsa-miR-19b hsa-miR-92aa hsa-miR-144 hsa-miR-20a hsa-miR-21 hsa-miR-486-5pa hsa-miR-15a hsa-miR-126 hsa-miR-142-3p hsa-miR-106a hsa-miR-93 hsa-let-7b
24.2 27.3 27.7 28.0 28.2 28.5 28.9 29.1 29.3 29.4 29.4 29.6 29.6 29.7 29.9
a miRNA level differences were statistically significant between the isolation methods (fold change ≥ 2, p-value ≤ 0.00048; Bonferroni significant).
Please cite this article as: Rekker K, et al, Comparison of serum exosome isolation methods for microRNA profiling, Clin Biochem (2013), http:// dx.doi.org/10.1016/j.clinbiochem.2013.10.020
K. Rekker et al. / Clinical Biochemistry xxx (2013) xxx–xxx
Fig. 1. Comparison of miRNA profiles between exosomal isolation methods. [A] Correlation analysis of miRNA levels between the exosome isolation methods. Pearson correlation coefficient (r) is reported for the comparison. [B] Cluster analysis of serum exosomal miRNAs. Red indicates miRNAs with higher and green with lower levels than average. EQ — ExoQuick Precipitation; UC — ultracentrifugation.
result indicates a small protein-bound miRNA contamination in exosome samples, but the decrease in detected miR-92a levels was similar between the two exosomal isolation methods.
3
methods (Supplementary file 2) leading to lower efficiency of exosome recovery with UC method. Both exosome isolation methods proved to be suitable for miRNA profiling as around 100 exosomal miRNAs were detected from a relatively small amount of blood serum. The overall exosomal miRNA profiles were similar regardless of the isolation method used, however, the miRNA levels showed a slight variation between the two methods. Hierarchical clustering of 17 miRNAs generated a distinguishable pattern where the samples were grouped according to the used exosome isolation method indicating that the observed exosomal miRNA profile is somewhat dependent upon the extracellular vesicle isolation method. We must consider that both methods used in the present study do not ensure stringent criteria for exosome specific isolation, being based on their physical features, rather than on their unique molecular composition. The purity of material isolated with ultracentrifugation is inferior compared to other methods such as sucrose gradient or immunocapture based isolation [10], however combined with differential centrifugation and microfiltration steps ultracentrifugation yields pellets that are consistently enriched with fairly intact exosome like vesicles, being only slightly contaminated by other membrane vesicles or fragments, while containing large amount of protein complexes [11]. Additional disadvantage of this method regards poor recovery of exosomes from highly viscous biofluids such as plasma and serum [12,13]. ExoQuick on the other hand precipitates very efficiently most of the vesicles and protein–protein, lipoprotein or protein–RNA complexes present in the serum regardless of their origin. This can also explain differences in particular miRNA species detected in this study in differently processed samples where miR-486-5p and miR-92a showed lower relative expression in ultracentrifugation isolated samples compared to ExoQuick precipitated samples. Arroyo et al. showed that miR-92a and miR-4865p are mostly bound to ribonucleoprotein complexes rather than confined inside vesicles [14]. However, in our study the experiments with protease treatment unveiled a similar slight decrease in miR-92a levels from pellets isolated with both methods suggesting that the detected fluctuations were probably not because of different levels of proteinbound miRNAs but rather due to the differences in the composition of extracellular vesicles isolated with EQ and UC methods. ExoQuick can be considered an efficient method for exosomal RNA analysis while being unsuitable for biochemical and immunodetection analysis of protein content of the sample [9]. Possibility of amplification of low abundant and specific target downstream to RNA isolation can compensate for a scarce specificity of an exosome isolation step. However, such unspecific precipitation method may prove to be confounding if specific disease markers are sought in the serum, as these are likely to be associated to specific vesicles sets, having either distinct tissue/cell-type identity and/or subcellular origin, that cannot be separated using neither method addressed in this study. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiochem.2013.10.020.
Discussion In the present study, miRNA profile isolated from blood serum exosome samples with ExoQuick™ Exosome Precipitation Solution was compared to conventionally used ultracentrifugation method. To our knowledge, although the overall exosomal miRNA yields obtained from serum have been addressed [4,7] this is the first study where serum exosomal miRNA content is compared between the exosome isolation methods. Alvarez et al. have studied protein, miRNA and mRNA levels using different methods for the isolation of urinary exosomes [9], where the highest exosome yield and RNA quantities were obtained by using a modified ExoQuick based method, while for protein analysis ultracentrifugation methods were found to be more suitable. Larger panel of miRNAs was investigated in our study, still, our results are concordant showing that similar to urine samples, higher miRNA yield in serum exosomes was detected by ExoQuick method. This could be explained with technical differences between the exosome isolation
Acknowledgements This research was funded by grant SF0180044s09 from the Estonian Ministry of Education and Research, by Enterprise Estonia, grant no. EU30020 and EXO-ID EU29269, Eurostars EU41564 NOTED project. The authors are very grateful to the volunteers for their participation in this study. Also, the authors thank Steven Smit from Estonian Genome Centre for technical assistance and Katrin Kepp for collecting the samples. References [1] Mathivanan S, Ji H, Simpson RJ. Exosomes: extracellular organelles important in intercellular communication. J Proteomics 2010;73:1907–20. [2] Raposo G, Stoorvogel W. Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol 2013;200:373–83.
Please cite this article as: Rekker K, et al, Comparison of serum exosome isolation methods for microRNA profiling, Clin Biochem (2013), http:// dx.doi.org/10.1016/j.clinbiochem.2013.10.020
4
K. Rekker et al. / Clinical Biochemistry xxx (2013) xxx–xxx
[3] Valadi H, Ekstrom K, Bossios A, Sjostrand M, Lee JJ, Lotvall JO. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 2007;9:654–9. [4] Taylor DD, Gercel-Taylor C. MicroRNA signatures of tumor-derived exosomes as diagnostic biomarkers of ovarian cancer. Gynecol Oncol 2008;110:13–21. [5] Tanaka Y, Kamohara H, Kinoshita K, Kurashige J, Ishimoto T, Iwatsuki M, et al. Clinical impact of serum exosomal microRNA-21 as a clinical biomarker in human esophageal squamous cell carcinoma. Cancer 2013;119:1159–67. [6] Liang B, Peng P, Chen S, Li L, Zhang M, Cao D, et al. Characterization and proteomic analysis of ovarian cancer-derived exosomes. J Proteomics 2013;80C:171–82. [7] Murakami Y, Toyoda H, Tanahashi T, Tanaka J, Kumada T, Yoshioka Y, et al. Comprehensive miRNA expression analysis in peripheral blood can diagnose liver disease. PLoS One 2012;7:e48366. [8] Merchant ML, Powell DW, Wilkey DW, Cummins TD, Deegens JK, Rood IM, et al. Microfiltration isolation of human urinary exosomes for characterization by MS. Proteomics Clin Appl 2010;4:84–96. [9] Alvarez ML, Khosroheidari M, Kanchi Ravi R, DiStefano JK. Comparison of protein, microRNA, and mRNA yields using different methods of urinary
[10]
[11]
[12] [13]
[14]
exosome isolation for the discovery of kidney disease biomarkers. Kidney Int 2012;82:1024–32. Tauro BJ, Greening DW, Mathias RA, Ji H, Mathivanan S, Scott AM, et al. Comparison of ultracentrifugation, density gradient separation, and immunoaffinity capture methods for isolating human colon cancer cell line LIM1863-derived exosomes. Methods 2012;56:293–304. Gyorgy B, Modos K, Pallinger E, Paloczi K, Pasztoi M, Misjak P, et al. Detection and isolation of cell-derived microparticles are compromised by protein complexes resulting from shared biophysical parameters. Blood 2011;117: e39–48. Taylor DD, Zacharias W, Gercel-Taylor C. Exosome isolation for proteomic analyses and RNA profiling. Methods Mol Biol 2011;728:235–46. Momen-Heravi F, Balaj L, Alian S, Trachtenberg AJ, Hochberg FH, Skog J, et al. Impact of biofluid viscosity on size and sedimentation efficiency of the isolated microvesicles. Front Physiol 2012;3:162. Arroyo JD, Chevillet JR, Kroh EM, Ruf IK, Pritchard CC, Gibson DF, et al. Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma. Proc Natl Acad Sci U S A 2011;108:5003–8.
Please cite this article as: Rekker K, et al, Comparison of serum exosome isolation methods for microRNA profiling, Clin Biochem (2013), http:// dx.doi.org/10.1016/j.clinbiochem.2013.10.020
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
Short Communication
Comparison of serum exosome isolation methods for microRNA profiling Kadri Rekker a,b,⁎, Merli Saare a,b,c, Anne Mari Roost a, Anna-Liisa Kubo d, Natasa Zarovni d, Antonio Chiesi d, Andres Salumets a,b,c, Maire Peters a,b a
Competence Centre on Reproductive Medicine and Biology, Tiigi 61b, 50410 Tartu, Estonia Department of Obstetrics and Gynecology, University of Tartu, L. Puusepa 8, 51014 Tartu, Estonia Institute of Bio- and Translational Medicine, University of Tartu, Ravila 19, 50411 Tartu, Estonia d HansaBioMed Ltd, Akadeemia tee 15, 12618 Tallinn, Estonia b c
a r t i c l e
i n f o
Article history: Received 27 August 2013 Received in revised form 15 October 2013 Accepted 21 October 2013 Available online xxxx Keywords: Exosomes miRNA Extracellular vesicles Ultracentrifugation ExoQuick
a b s t r a c t Objectives: Exosomes are small membrane bound vesicles secreted by most cell types. Exosomes contain various functional proteins, mRNAs and microRNAs (miRNAs) that could be used for diagnostic and therapeutic purposes. Currently, a standard method for serum exosome isolation is differential ultracentrifugation, but a search for alternative, less time-consuming and labour extensive exosomal isolation method for use in clinical settings is ongoing. The effect of serum exosome isolation method on obtained miRNA profile is not yet clear. The aim of this study was to determine to which extent selected exosome isolation methods influence the serum exosomal miRNA profile. Design and methods: Exosomes were isolated from blood serum of healthy individuals by ultracentrifugation and ExoQuick Precipitation methods. The expression profile of 375 miRNAs was determined by real time PCR using Exiqon miRCURY LNA™ microRNA Human panel I assays. Results: Although a strong correlation of exosomal miRNA profiles was observed between the two isolation methods, distinct clusters of miRNA levels between the used methods were identified. The detected levels of two miRNAs, miR-92a and miR-486-5p, were significantly influenced by the exosome isolation method used. Conclusions: Both exosome isolation methods are suitable for serum exosomal miRNA profiling. Differences found in miRNA patterns between the two methods indicate that the observed exosomal miRNA profile is slightly affected by the extracellular vesicle isolation method. © 2013 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Introduction Cells release various types of membrane vesicles into the extracellular space that differ in origin, size, morphology and content [1]. The most studied of them are exosomes, small (40–100 nm) vesicles of endocytic origin. Exosomes have been isolated from a number of body fluids, including blood plasma and serum, breast milk, saliva and urine (reviewed in [2]). These extracellular vesicles contain various functional proteins, microRNAs (miRNAs) and mRNAs [3] which have a potential to be used as biomarkers for numerous diseases. Previous studies have demonstrated that exosomes are actively released by cancerous cells into the peripheral circulation [4] and biomarker studies highlight the possibility to use exosomal protein content and RNA profiles in cancer diagnosis [5,6]. Although circulating exosomes and their content can serve as biomarkers for various diseases, the standard ultracentrifugation based exosome isolation methods established for research purposes are Abbreviations: UC, ultracentrifugation; EQ, ExoQuick Precipitation Solution. ⁎ Corresponding author at: Competence Centre on Reproductive Medicine and Biology, Tiigi 61b, 50410 Tartu, Estonia. E-mail address: [email protected] (K. Rekker).
time-consuming and labour intensive and may not be suitable for use in clinical laboratories. Recently, alternative methods for exosome isolation with reduced hands-on time and without the requirement for specialised expensive equipment have been developed. For instance, precipitation based methods have been used for biomarker discovery studies [7] and a microfiltration based method for the isolation of urinary exosomes has been introduced [8]. To our knowledge, a comparative study of blood serum exosome isolation methods for miRNA profiling has not been conducted. Therefore, the aim of our study was to determine whether the exosomal miRNA profile in serum samples depends on exosome isolation method. In this study, miRNA content from exosomes obtained by a “gold standard” ultracentrifugation based method and ExoQuick Precipitation Solution was compared.
Material and methods The study was approved by the Ethics Review Committee on Human Research of the University of Tartu (Tartu, Estonia) and informed consent was obtained from all participants. A detailed description of all materials and methods used can be found in Supplementary data.
0009-9120/$ – see front matter © 2013 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.clinbiochem.2013.10.020
Please cite this article as: Rekker K, et al, Comparison of serum exosome isolation methods for microRNA profiling, Clin Biochem (2013), http:// dx.doi.org/10.1016/j.clinbiochem.2013.10.020
2
K. Rekker et al. / Clinical Biochemistry xxx (2013) xxx–xxx
Sample collection
Data analysis
Peripheral blood samples from five healthy women (age 24.8 ± 4.0 years (mean± SD); body mass index 22.0± 1.4kg/m2) were collected into serum collection tubes. Blood samples were centrifuged at 1600 ×g for 10 min for serum collection followed by an additional centrifugation step at 16,000 ×g for 10 min and filtration through 0.2 μm pore size syringe filter (Sartorius GmbH, Göttingen, Germany). In order to avoid miRNA expression differences due to inter-individual variations, the same serum samples were used for both exosome isolation methods.
Data analysis was performed with GenEx 5.3 Software (MultiD Analyses AB, Sweden). Pearson's correlation and two-tailed paired t-test were used for the comparison of exosome isolation methods. Significance threshold was set to a fold change ≥ 2 with a Bonferroni adjusted p-value ≤ 0.00048 (106 miRNA comparisons).
Exosome isolation Ultracentrifugation (UC) Exosomal fraction from 1 mL of serum was purified by four successive centrifugation steps. At first, centrifugation at 300 ×g for 10 min was performed, serum was diluted with sterile PBS in 1:1 ratio and centrifuged at 10,000×g for 30min followed by ultracentrifugation for 2h at 200,000 ×g (Optima MAX-XP, Beckman Coulter, rotor TLA 100.3). The crude exosome pellet was washed once in a large volume of PBS, filtered through a 0.2 μm syringe filter and centrifuged at 200,000 ×g for 1 h, then re-suspended in PBS. All centrifugations were performed at 4 °C. ExoQuick Precipitation (EQ) Exosomal fraction from 1 mL of serum was isolated by ExoQuick™ Exosome Precipitation Solution according to the manufacturer's recommendations (System Biosciences Inc., Mountain View, CA, USA). Briefly, 1/4 volume of ExoQuick Solution was added to serum and samples were refrigerated at 4 °C overnight. The mixture was centrifuged at 1500 ×g for 30min and supernatant was removed by aspiration. Pelleted fraction was re-suspended in nuclease-free water. RNA isolation RNA was extracted from exosomal fractions using miRNeasy Mini kit (Qiagen, Hilden, Germany) with the final elution volume of 50 μL. RNase A and Proteinase K treatment In separate experiments, exosomal fractions isolated with EQ and UC methods were treated with RNase A before and after RNA extraction. The presence of miRNAs in RNase A treated samples was determined by quantitative real-time PCR (qRT-PCR) using TaqMan® microRNA assays (Applied Biosystems, Foster City, CA, USA) hsa-let-7a-5p (assay ID 000377), hsa-miR-16-5p (000391) and hsa-miR-21-5p (000397). In addition, Proteinase K treatment of serum samples prior to exosomal isolation with EQ and UC methods was conducted. The presence of miRNAs in Proteinase K treated and untreated samples was assessed with TaqMan® microRNA assay hsa-miR-92a (000431). miRNA profiling Serum exosomal RNA samples isolated with UC (n=5) or EQ (n=5) methods from healthy women were used for miRNA profiling by real time PCR using miRCURY LNA™ microRNA Human panel I assays (Exiqon, Vedbaek, Denmark) following the manufacturer's recommendations. Briefly, cDNA synthesis was performed using the Universal cDNA Synthesis Kit (Exiqon). For each reaction 4 μL of exosomal RNA was used. Reverse transcription reaction products were combined with SYBR® Green master mix (Exiqon) and loaded onto ready-to-use Human Panel assays. Amplification was performed on a 7900HT thermocycler (Applied Biosystems) using cycling parameters recommended by Exiqon.
Results To evaluate whether the miRNAs extracted from the pelleted fraction collected with EQ and UC methods are confined inside vesicles, experiments with RNase A treatment were performed. All three miRNAs (hsa-let-7a, hsa-miR-16 and hsa-miR-21) analysed with TaqMan® assays were detected after the RNase A treatment of exosomal pellets isolated with both methods, indicating that analysed miRNAs are confined inside vesicles. miRNAs were not detected from samples that were treated with RNase A after the RNA isolation from the pellets. The expression profiles of 375 miRNAs in serum exosome samples isolated with UC (n = 5) and EQ (n = 5) methods were determined by qRTPCR using Exiqon Human panel I assays. The average numbers of detected exosomal miRNAs above the detection limit (Ct b 35) were 128 (range 100 to 160) and 93 (range 87 to 108) for EQ and UC methods, respectively. According to raw Ct values, the average miRNA detection threshold levels were 1.7 cycles lower in EQ samples, indicating that the total exosomal miRNA quantity isolated from an equal amount of serum is higher in EQ processed samples. The most abundant miRNAs detected by qRT-PCR were similar between the isolation methods (Table 1). A strong correlation was observed between the two methods (Fig. 1A, Pearson r = 0.95, p b 0.05). No miRNAs were uniquely dependent on exosome isolation method. However, miR-92a and miR-486-5p showed statistically different levels between the two methods (fold change 2.9 and 4.2, respectively; p-value b 0.00048). In addition, the clustering analysis of 17 miRNAs with the largest differences between the study groups (according to t-test, unadjusted p-value ≤ 0.05 and fold change ≥ 2) revealed a clearly distinct profile for both exosome isolation methods (Fig. 1B). To assess whether the differences in miR-92a levels could be related to possible co-precipitation of protein-bound miRNAs in isolated exosomal samples, proteinase K treatment was performed. miR-92a levels were determined from protease treated and untreated samples isolated with EQ and UC methods and a slight decrease in miR-92a levels of protease treated samples was observed in both cases. This Table 1 Top 15 serum exosomal miRNAs detected by Exiqon assays. miRNAs are listed according to abundance. Exosome isolation method ExoQuick Precipitation (EQ)
Ultracentrifugation (UC)
miRNA ID
Average Ct value
miRNA ID
Average Ct value
hsa-miR-451 hsa-miR-92aa hsa-miR-16 hsa-miR-486-5pa hsa-miR-19b hsa-miR-223 hsa-miR-20a hsa-miR-15a hsa-miR-93 hsa-miR-126 hsa-miR-144 hsa-miR-106a hsa-miR-21 hsa-miR-320a hsa-let-7b
22.0 25.0 25.5 25.5 25.9 27.2 27.2 27.3 27.5 27.7 27.7 27.8 27.8 27.9 28.0
hsa-miR-451 hsa-miR-16 hsa-miR-223 hsa-miR-19b hsa-miR-92aa hsa-miR-144 hsa-miR-20a hsa-miR-21 hsa-miR-486-5pa hsa-miR-15a hsa-miR-126 hsa-miR-142-3p hsa-miR-106a hsa-miR-93 hsa-let-7b
24.2 27.3 27.7 28.0 28.2 28.5 28.9 29.1 29.3 29.4 29.4 29.6 29.6 29.7 29.9
a miRNA level differences were statistically significant between the isolation methods (fold change ≥ 2, p-value ≤ 0.00048; Bonferroni significant).
Please cite this article as: Rekker K, et al, Comparison of serum exosome isolation methods for microRNA profiling, Clin Biochem (2013), http:// dx.doi.org/10.1016/j.clinbiochem.2013.10.020
K. Rekker et al. / Clinical Biochemistry xxx (2013) xxx–xxx
Fig. 1. Comparison of miRNA profiles between exosomal isolation methods. [A] Correlation analysis of miRNA levels between the exosome isolation methods. Pearson correlation coefficient (r) is reported for the comparison. [B] Cluster analysis of serum exosomal miRNAs. Red indicates miRNAs with higher and green with lower levels than average. EQ — ExoQuick Precipitation; UC — ultracentrifugation.
result indicates a small protein-bound miRNA contamination in exosome samples, but the decrease in detected miR-92a levels was similar between the two exosomal isolation methods.
3
methods (Supplementary file 2) leading to lower efficiency of exosome recovery with UC method. Both exosome isolation methods proved to be suitable for miRNA profiling as around 100 exosomal miRNAs were detected from a relatively small amount of blood serum. The overall exosomal miRNA profiles were similar regardless of the isolation method used, however, the miRNA levels showed a slight variation between the two methods. Hierarchical clustering of 17 miRNAs generated a distinguishable pattern where the samples were grouped according to the used exosome isolation method indicating that the observed exosomal miRNA profile is somewhat dependent upon the extracellular vesicle isolation method. We must consider that both methods used in the present study do not ensure stringent criteria for exosome specific isolation, being based on their physical features, rather than on their unique molecular composition. The purity of material isolated with ultracentrifugation is inferior compared to other methods such as sucrose gradient or immunocapture based isolation [10], however combined with differential centrifugation and microfiltration steps ultracentrifugation yields pellets that are consistently enriched with fairly intact exosome like vesicles, being only slightly contaminated by other membrane vesicles or fragments, while containing large amount of protein complexes [11]. Additional disadvantage of this method regards poor recovery of exosomes from highly viscous biofluids such as plasma and serum [12,13]. ExoQuick on the other hand precipitates very efficiently most of the vesicles and protein–protein, lipoprotein or protein–RNA complexes present in the serum regardless of their origin. This can also explain differences in particular miRNA species detected in this study in differently processed samples where miR-486-5p and miR-92a showed lower relative expression in ultracentrifugation isolated samples compared to ExoQuick precipitated samples. Arroyo et al. showed that miR-92a and miR-4865p are mostly bound to ribonucleoprotein complexes rather than confined inside vesicles [14]. However, in our study the experiments with protease treatment unveiled a similar slight decrease in miR-92a levels from pellets isolated with both methods suggesting that the detected fluctuations were probably not because of different levels of proteinbound miRNAs but rather due to the differences in the composition of extracellular vesicles isolated with EQ and UC methods. ExoQuick can be considered an efficient method for exosomal RNA analysis while being unsuitable for biochemical and immunodetection analysis of protein content of the sample [9]. Possibility of amplification of low abundant and specific target downstream to RNA isolation can compensate for a scarce specificity of an exosome isolation step. However, such unspecific precipitation method may prove to be confounding if specific disease markers are sought in the serum, as these are likely to be associated to specific vesicles sets, having either distinct tissue/cell-type identity and/or subcellular origin, that cannot be separated using neither method addressed in this study. Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.clinbiochem.2013.10.020.
Discussion In the present study, miRNA profile isolated from blood serum exosome samples with ExoQuick™ Exosome Precipitation Solution was compared to conventionally used ultracentrifugation method. To our knowledge, although the overall exosomal miRNA yields obtained from serum have been addressed [4,7] this is the first study where serum exosomal miRNA content is compared between the exosome isolation methods. Alvarez et al. have studied protein, miRNA and mRNA levels using different methods for the isolation of urinary exosomes [9], where the highest exosome yield and RNA quantities were obtained by using a modified ExoQuick based method, while for protein analysis ultracentrifugation methods were found to be more suitable. Larger panel of miRNAs was investigated in our study, still, our results are concordant showing that similar to urine samples, higher miRNA yield in serum exosomes was detected by ExoQuick method. This could be explained with technical differences between the exosome isolation
Acknowledgements This research was funded by grant SF0180044s09 from the Estonian Ministry of Education and Research, by Enterprise Estonia, grant no. EU30020 and EXO-ID EU29269, Eurostars EU41564 NOTED project. The authors are very grateful to the volunteers for their participation in this study. Also, the authors thank Steven Smit from Estonian Genome Centre for technical assistance and Katrin Kepp for collecting the samples. References [1] Mathivanan S, Ji H, Simpson RJ. Exosomes: extracellular organelles important in intercellular communication. J Proteomics 2010;73:1907–20. [2] Raposo G, Stoorvogel W. Extracellular vesicles: exosomes, microvesicles, and friends. J Cell Biol 2013;200:373–83.
Please cite this article as: Rekker K, et al, Comparison of serum exosome isolation methods for microRNA profiling, Clin Biochem (2013), http:// dx.doi.org/10.1016/j.clinbiochem.2013.10.020
4
K. Rekker et al. / Clinical Biochemistry xxx (2013) xxx–xxx
[3] Valadi H, Ekstrom K, Bossios A, Sjostrand M, Lee JJ, Lotvall JO. Exosome-mediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells. Nat Cell Biol 2007;9:654–9. [4] Taylor DD, Gercel-Taylor C. MicroRNA signatures of tumor-derived exosomes as diagnostic biomarkers of ovarian cancer. Gynecol Oncol 2008;110:13–21. [5] Tanaka Y, Kamohara H, Kinoshita K, Kurashige J, Ishimoto T, Iwatsuki M, et al. Clinical impact of serum exosomal microRNA-21 as a clinical biomarker in human esophageal squamous cell carcinoma. Cancer 2013;119:1159–67. [6] Liang B, Peng P, Chen S, Li L, Zhang M, Cao D, et al. Characterization and proteomic analysis of ovarian cancer-derived exosomes. J Proteomics 2013;80C:171–82. [7] Murakami Y, Toyoda H, Tanahashi T, Tanaka J, Kumada T, Yoshioka Y, et al. Comprehensive miRNA expression analysis in peripheral blood can diagnose liver disease. PLoS One 2012;7:e48366. [8] Merchant ML, Powell DW, Wilkey DW, Cummins TD, Deegens JK, Rood IM, et al. Microfiltration isolation of human urinary exosomes for characterization by MS. Proteomics Clin Appl 2010;4:84–96. [9] Alvarez ML, Khosroheidari M, Kanchi Ravi R, DiStefano JK. Comparison of protein, microRNA, and mRNA yields using different methods of urinary
[10]
[11]
[12] [13]
[14]
exosome isolation for the discovery of kidney disease biomarkers. Kidney Int 2012;82:1024–32. Tauro BJ, Greening DW, Mathias RA, Ji H, Mathivanan S, Scott AM, et al. Comparison of ultracentrifugation, density gradient separation, and immunoaffinity capture methods for isolating human colon cancer cell line LIM1863-derived exosomes. Methods 2012;56:293–304. Gyorgy B, Modos K, Pallinger E, Paloczi K, Pasztoi M, Misjak P, et al. Detection and isolation of cell-derived microparticles are compromised by protein complexes resulting from shared biophysical parameters. Blood 2011;117: e39–48. Taylor DD, Zacharias W, Gercel-Taylor C. Exosome isolation for proteomic analyses and RNA profiling. Methods Mol Biol 2011;728:235–46. Momen-Heravi F, Balaj L, Alian S, Trachtenberg AJ, Hochberg FH, Skog J, et al. Impact of biofluid viscosity on size and sedimentation efficiency of the isolated microvesicles. Front Physiol 2012;3:162. Arroyo JD, Chevillet JR, Kroh EM, Ruf IK, Pritchard CC, Gibson DF, et al. Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma. Proc Natl Acad Sci U S A 2011;108:5003–8.
Please cite this article as: Rekker K, et al, Comparison of serum exosome isolation methods for microRNA profiling, Clin Biochem (2013), http:// dx.doi.org/10.1016/j.clinbiochem.2013.10.020
