Abstract
Objectives
Small extracellular vesicles (EVs) contain various signaling molecules, thus playing a crucial role in cell-to-cell communication and emerging as a promising source of biomarkers. However, the lack of standardized procedures impedes their translation to clinical practice. Thus, we compared different approaches for high-throughput analysis of small EVs transcriptome.
Methods
Small EVs were isolated from 150 μL of serum. Quality and quantity were assessed by dynamic light scattering, transmission electron microscopy, and Western blot. Comparison of RNA extraction efficiency was performed, and expression of selected genes was analyzed by RT-qPCR. Whole transcriptome analysis was done using microarrays.
Results
Obtained data confirmed the suitability of size exclusion chromatography for isolation of small EVs. Analyses of gene expression showed the best results in case of samples isolated by Monarch Total RNA Miniprep Kit. Totally, 7,182 transcripts were identified to be deregulated between colorectal cancer patients and healthy controls. The majority of them were non-coding RNAs with more than 70 % being lncRNAs, while protein-coding genes represented the second most common gene biotype.
Conclusions
We have optimized the protocol for isolation of small EVs and their RNA from low volume of sera and confirmed the suitability of Clariom D Pico Assays for transcriptome profiling.
Introduction
Over the last years, secreted membrane-enclosed vesicles are being intensively studied. The best described are exosomes, small extracellular vesicles (EVs) of endocytic origin with size ranging from 30 to 150 nm [1]. Initially, they were considered to be an unfunctional garbage bags. However, it was proved that they contain variety of molecules such as nucleic acids, lipids, proteins, or different metabolites [2], facilitate the transport of these intracellular components, and thus play a crucial role in cell-to-cell communication [3]. Importantly, recent studies have demonstrated their active release into the peripheral circulation by cancerous cells [4], thus they have emerged as a promising source of novel non-invasive biomarkers for early cancer diagnosis, prognosis and therapy response monitoring [5].
Despite the growing number of studies describing circulating small EVs as a promising liquid biopsy, only few biomarkers associated with EVs have entered the clinical trials [6, 7] due to the lack of standardized procedures. Although the ultracentrifugation is the most widely used method [8], it is time-consuming, requires the right equipment, and usually higher sample volume. Moreover, shearing forces may cause the disruption of vesicles and its efficiency is reduced when high viscosity fluids such as serum or plasma are used [9]. Therefore, numerous commercial kits have been developed to improve and facilitate isolation of EVs. These methods are based on precipitation, ultrafiltration, size-exclusion chromatography (SEC), or immunoassays [10]. Compared to the centrifugation, the structure of vesicles is usually not changed, low volumes of sample are required, different eluting solutions may be used, and in case of immunoassays, highly pure fraction of small EVs may be acquired. Nevertheless, precipitation reagents inhibit recovery of intact EVs and polymers could interfere with downstream analysis and influence the biological activities [11, 12]. Similarly, ultrafiltration methods have been shown to result in low purity and adhesion of EVs to the filtration membrane, while SEC requires longer running time and vesicles are eluted in a larger volume, thus additional concentration step is necessary. Finally, immunoassays are expensive and usually used for low volumes of highly concentrated samples. In addition, non-specific binding was observed, and the yield is relatively small to be sufficient for majority of further analysis [13].
Once small EVs have been isolated, their comprehensive characterization must be performed. According to the International Society for Extracellular Vesicles (ISEV), single vesicles should be characterized by at least two different technologies such as transmission electron microscopy (TEM), atomic force microscopy (AFM), nanoparticle-tracking analysis (NTA), resistive pulse sensing (RPS), or dynamic light scattering (DLS). In addition, the amount of three or more proteins from different categories that are expected to be present in small EVs should be assessed by Western blot (WB), flow cytometry, or global proteomic analysis [14]. Importantly, the level of other proteins not expected to be enriched in vesicles of endosomal origin should be determined in order to reveal the co-isolation of EVs of different origin [14].
Further, RNA extraction together with the measurements of its quality and quantity is a crucial step before the high-throughput screening of RNA content of EVs. Currently, a number of conventional RNA extraction methods are available. However, previous studies showed that different isolation methods give extensive variation in RNA yield, pattern, and purity [15, 16]. Thus, it is necessary to choose the best approach according to the aims of the research. Due to the low levels of RNA in EVs, its quality and quantity is usually assessed using Agilent 2100 Bioanalyzer together with an RNA 6000 Pico Kit [16]. Finally, identification of small EVs RNA-based biomarkers may be performed using several different methods, including RT-qPCR, digital PCR or high-throughput approaches such as microarrays or next-generation sequencing (NGS). While microarray procedure is simple and requires lower sample volume, it is prone to DNA contamination and enable detection of only known RNA sequences. On the contrary, NGS is time-consuming and its results have to be validated, but detection of different types of RNA may be performed in the same run independently of previous sequence information and together with existing isoforms [13, 17].
To date, several studies have compared the effect of different techniques for isolation of small EVs and their RNA on efficiency, reproducibility, and downstream applications [16, 18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28]. Nevertheless, none of these papers have performed in depth analyses of all important parameters. Thus, the aim of this study was to compare several techniques widely recommended for the isolation of small EVs, their characterization, and subsequent RNA isolation together with quality and quantity assessments. Further, we hypothesize that expression profiles of small EVs could differentiate between CRC patients and healthy controls. Thus, Human Clariom D Pico Assays have been tested in order to confirm its suitability for transcriptome profiling of small EVs isolated from minimal volume of blood serum as this product should be able to generate robust expression profiles data from as little as 100 pg of total RNA. To our knowledge, this is the first study comparing and optimizing different methodological approaches for credible analysis of small EVs-enriched RNAs from low volume of sera using microarrays.
Materials and methods
Patients’ samples and study design
Optimization of small EVs isolation as well as RNA was performed using serum samples of healthy controls. Six serum samples of histopathologically verified CRC patients (three men, three women; three C18, three C20; 2× stage I, 1× stage II, 2× stage III, 1× stage IV; median age 63 years) and six samples of healthy controls (three men, three women; median age 68 years) were used for the expression profiling by Clariom D Pico Assays (Applied Biosystems, Waltham, MA, USA). All patients underwent the resection from 2016 through 2020 at Masaryk Memorial Cancer Institute (MMCI, Brno, Czech Republic) and samples were collected prior the surgery. Serum samples from healthy controls were collected at the Cancer Prevention Center between 2018 and 2021 (MMCI, Brno, Czech Republic); these donors had no prior diagnosis of any malignancy. All serum samples have been stored at −80 °C before the exosome isolation.
Isolation of small extracellular vesicles
Isolation of small EVs was performed using miRcury Exosome Isolation Kit – Serum and Plasma (Exiqon, Vedbaek, Denmark), Plasma/Serum Exosome Purification Mini Kit (Norgen Biotek, Thorold, Canada), and qEVsingle/35 nm columns (SP6, iZON Science Ltd., Oxford, UK). All samples were centrifuged at 1,500×g for 10 min at 4 °C and then at 10,000×g for 20 min at 4 °C. Small EVs were isolated according to the manufacturer’s recommendations. In case of SEC, individual fractions were collected (each 200 μL). Subsequently, fractions six and seven were combined. All samples were treated with 10 μL of proteinase K (20 mg/mL) and 5 μL of RNase Cocktail Enzyme Mix (20,000 U/mL) at 37 °C for 10 min. Finally, 4 μL of SUPERaseIn RNase Inhibitor (20 U/μL) were added and mixed well (all Invitrogen, Waltham, MA, USA).
Dynamic light scattering
Fifty microliters of the sample were placed in low volume quartz batch cuvette ZEN2112 (minimal volume 12 μL of sample for back scatter, 20 μL for side scatter, and 45 μL for forward scatter measurements) and measured using multi-angled DLS technique (MADLS®), Zetasizer Ultra at a constant temperature of 25 °C. The data were collected at three angles, 173°, 90°, and 13° and evaluated using ZS Xplorer software version 2.50 (all Malvern Panalytical Ltd, UK).
Transmission electron microscopy
Four microliters of the sample were applied to freshly plasma-cleaned TEM grids with thin carbon layer and stained with 2 % uranyl acetate. The grids were loaded to Talos F200C TEM (Thermo Fisher Scientific, Brno, Czech Republic). The microscope was operated at 200 kV. The images were collected on Ceta-16 M CMOS camera at the 36,000× nominal magnification with the underfocus of 2–4 µm.
Protein analysis by Western blot
All samples were concentrated to 80 μL using Concentrator plus 5305 Vacuum Centrifuge (Eppendorf, Hamburg, Germany). Protein quantification was performed using the Pierce Micro BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) and 8 µg of proteins were loaded per lane. Proteins were resolved by 10 % SDS-PAGE gel and wet-transferred to Immobilion-P PVDF membrane (EMD Millipore, Burlington, MA, USA). The signals were visualized using Clarity Western ECL Substrate (Bio-Rad, Hercules, CA, USA) and detected by the UVITEC chemiluminescence imager (UVITEC Cambridge, UK). A list of antibodies is summarized in Supplementary Table S1.
Isolation of RNA
Three different kits were compared including MagMAX mirVana Total RNA Isolation Kit (Applied Biosystems, Waltham, MA, USA), Plasma/Serum RNA Purification Mini Kit (Norgen Biotek, Thorold, Canada), and Monarch® Total RNA Miniprep Kit (New England Biolabs, Ipswich, MA, USA). Total RNA including small RNAs was isolated according to the manufacturer’s recommendations, treated with DNase I, and eluted with 50 μL of DEPC-treated water.
Reverse transcription and RT-qPCR
Complementary DNA was synthesized using 5 μL of total RNA and the High-Capacity cDNA Reverse Transcription Kit according to the manufacturer’s recommendations. A preamplification step was performed mixing 2.5 μL of cDNA with pooled TaqMan assays (0.2×) and TaqMan PreAmp Master Mix using 14 cycles of preamplification. Quantitative PCR was carried out using specific probes (Supplementary Table S2), 2.5 μL of preampliefied cDNA diluted 20× and TaqMan Gene Expression Master Mix, fluorescence was measured by QuantStudio 12K Flex Real-Time PCR system (all Applied Biosystems, Waltham, MA, USA).
Microarray analysis
Microarray analysis was performed using Clariom D Pico Assay, human (Applied Biosystems, Waltham, MA, USA), according to the manufacturer’s recommendations. The pre-IVT amplification consisted of 12 cycles. Arrays were stained using an Affymetrix GeneChip Fluidics Station 450 according to the FS450_0001 protocol and scanned with an Affymetrix GeneChip Scanner 3000 7G.
Statistical analysis
Quality control check was done using the ‘arrayQualityMetrics’ [29]. To have a full control over the process of gene annotation, we created our own CDF using probe sequences from BRAINARRAY CDF [30] and the GENCODE (v40) database as a target. The 25 bp long probe sequences were aligned using BLASTn [31]. Promiscuous probes mapping to multiple different genes were removed. Only probe sets of more than four probes were allowed. Previously published script [32] was applied to build our own custom CDF and then the ‘pdInfoBuilder’ was used to create the R package. The differential expression analysis was done employing the ‘oligo’ and ‘limma’ packages. The expression values were normalized by RMA algorithm and p-values were adjusted by Benjamini–Hochberg.
Results
Identification of the best approach for isolation of small extracellular vesicles
To choose the best approach for the isolation of small EVs from limited volume of human blood serum, we compared qEVsingle/35 nm columns from iZON, Plasma/Serum Exosome Purification Mini Kit from Norgen Biotek, and miRCURY Exosome Serum/Plasma Kit from Exiqon. In all cases, small EVs were isolated from 150 μL of serum and their quality and quantity was assessed using DLS and TEM. In case of SEC, nine different fractions (F1–F9) were collected and separately analyzed to identify fractions with the highest amount and quality of vesicles. In addition, protein content-based characterization of gained particles was performed by Western blot according to ISEV recommendations [33] to demonstrate the degree of purity of isolates.
TEM of samples isolated by kits from Norgen Biotek and Exiqon showed a considerable number of smaller particles as well as proteins (Figure 1B and C). DLS analysis corresponded to the results of TEM and showed high polydispersity index (PDI) values for these samples indicating the presence of particles with very broad size range (Table 1). In case of SEC, TEM and DLS confirmed the presence of proteins and their aggregates and almost no small EVs in fractions F1–F5, which are the part of the column initial void volume (average size of eluted particles was bigger than 300 nm). Fractions F6 and F7 contained mostly the particles with the expected size of small EVs (around 100 nm), while F8 and F9 contained a mix of particles with various sizes, but smaller vesicles predominated (Figure 2A). DLS measurements proved the highest concentration of particles in F8 and F9, however, with smaller sizes, while the vesicles in F6 and F7 contained the particles between 50 and 150 nm (Supplementary Table S3). According to the results of WB, the highest levels of small EVs markers CD81 and Syntenin were detected in F7 with lower levels in F6 and F8. Nevertheless, the amount of ApoB increased significantly from F6 to F9, thus the combination of only F6 and F7 was chosen for further experiments to minimize the contamination by lipoproteins (Figure 2B). Comparison of different isolation methods proved the expression of CD63 and TSG101 in all analyzed samples, while CD81 was detected only in samples isolated by SEC and Syntenin was not present in the samples isolated by the kit from Exiqon. Concerning the lipoprotein contamination, the purest isolates were gained using Norgen procedure, while samples isolated by Exiqon kit or SEC contained detectable amount of ApoB (Figure 2C). These data confirm the suitability of qEVsingle/35 nm columns for isolation of small EVs from low volume of sera as this approach enable to get high number of vesicles with corresponding size, shape, and sufficient purity although the lipoproteins are co-isolated to some extent.
Sample | Z ± SD, nm | Z-average ± SD, nm | PDI ± SD | PDI-average ± SD | c ± SD, particles/mL | c-average ± SD, particles/mL |
---|---|---|---|---|---|---|
iZON 1a | 72 ± 1 | 70 ± 4 | 0.268 ± 0.005 | 0.234 ± 0.032 | 9.3 × 1012 ± 2.8 × 1012 | 3.4 × 1013 ± 4.1 × 1013 |
iZON 2a | 66 ± 0 | 0.205 ± 0.010 | 8.1 × 1013 ± 8.5 × 1013 | |||
iZON 3a | 72 ± 12 | 0.229 ± 0.034 | 1.2 × 1013 ± 2.5 × 1012 | |||
Norgen 1b | 25 ± 0 | 26 ± 6 | 0.465 ± 0.018 | 0.446 ± 0.023 | 1.0 × 1018 ± 3.3 × 1017 | 5.9 × 1017 ± 3.8 × 1017 |
Norgen 2b | 32 ± 1 | 0.452 ± 0.015 | 2.9 × 1017 ± 1.8 × 1017 | |||
Norgen 3b | 20 ± 1 | 0.420 ± 0.003 | 4.5 × 1017 ± 7.7 × 1016 | |||
Exiqon 1c | 51 ± 2 | 49 ± 13 | 0.477 ± 0.009 | 0.452 ± 0.023 | 9.3 × 1016 ± 1.3 × 1017 | 5.2 × 1016 ± 4.1 × 1016 |
Exiqon 2c | 60 ± 1 | 0.447 ± 0.009 | 1.0 × 1016 ± 5.5 × 1015 | |||
Exiqon 3c | 35 ± 1 | 0.432 ± 0.038 | 5.3 × 1016 ± 6.4 × 1016 |
-
SD, standard deviation; PDI, polydispersity index; c, concentration. a iZON – qEVsingle/35 nm columns from iZON (fraction 6 + 7), bNorgen – Plasma/Serum Exosome Purification Mini Kit from Norgen Biotek, cExiqon – miRCURY Exosome Serum/Plasma Kit from Exiqon.
Identification of optimal procedure for the RNA isolation from small extracellular vesicles
To find the best method for the extraction of RNA from isolated small EVs, we compared three different commercial kits – MagMAX mirVana Total RNA Isolation Kit (Applied Biosystems), Exosomal RNA Isolation Kit (Norgen Biotek), and Monarch® Total RNA Miniprep Kit (New England BioLabs). As the aim of the study was to further analyze the expression of lncRNAs as well as protein-coding genes by Clariom D Pico Assays, we assessed the efficiency of RNA extraction kits by RT-qPCR analyses of selected lncRNAs and protein-coding genes previously described to be present in small EVs [34], [35], [36], [37]. In case of all genes, the lowest expression was observed in samples isolated by the kit from Norgen Biotek. The detection of all targets by two other kits was comparable, only in case of lncRNA GAS5 the detection rate was higher when isolation was performed by Monarch kit compared to MagMAX kit (Table 2). To further evaluate the suitability of RNA isolated by the selected kits for high-throughput expression profiling, we compared the microarray results from two independent samples of healthy controls. Interestingly, the global gene expression was higher in case of samples isolated by Norgen Biotek kit compared to the MagMAX isolation kit and comparable to the results of arrays using samples isolated by Monarch kit (Supplementary Figure S1). Based on these data, we decided to use Monarch Total RNA Miniprep Kit for further isolation of RNA as the detection rate of all genes was the highest in case of RT-qPCR as well as in microarray analyses.
Sample | 18S [CT ± SD] | GAPDH [CT ± SD] | B2M [CT ± SD] | MALAT1 [CT ± SD] | ZFAS1 [CT ± SD] | GAS5 [CT ± SD] |
MagMAX 1a | 19.13 ± 0.10 | 27.83 ± 0.04 | 19.19 ± 0.25 | 26.66 ± 0.02 | 24.01 ± 0.04 | ND |
MagMAX 2a | 18.72 ± 0.07 | 28.18 ± 0.07 | 19.26 ± 0.25 | 26.59 ± 0.13 | 24.57 ± 0.08 | ND |
MagMAX 3a | 18.11 ± 0.03 | 27.69 ± 0.05 | 18.14 ± 0.07 | 25.17 ± 0.02 | 23.83 ± 0.09 | ND |
Monarch 1b | 19.38 ± 0.07 | 27.54 ± 0.15 | 20.28 ± 0.07 | 26.47 ± 0.08 | 25.80 ± 0.08 | 25.92 ± 0.25 |
Monarch 2b | 19.22 ± 0.04 | 28.17 ± 0.16 | 19.92 ± 0.08 | 27.10 ± 0.17 | 25.80 ± 0.15 | ND |
Monarch 3b | 18.46 ± 0.19 | 25.01 ± 0.05 | 19.59 ± 0.00 | 24.81 ± 0.04 | 26.77 ± 0.09 | 27.91 ± 0.10 |
Norgen Biotek 1c | 22.03 ± 0.18 | 29.87 ± 0.01 | 21.20 ± 0.11 | 26.48 ± 0.07 | 25.49 ± 0.00 | ND |
Norgen Biotek 2c | 23.05 ± 0.19 | 30.35 ± 0.25 | 21.27 ± 0.25 | 28.46 ± 0.14 | 28.87 ± 0.12 | ND |
Norgen Biotek 3c | 23.86 ± 0.16 | 32.79 ± 0.25 | 24.32 ± 0.25 | 30.17 ± 0.25 | 29.43 ± 0.25 | ND |
-
CT, treshold cycle; SD, standard deviation; B2M, beta-2-microglobulin; ND, non-detected. aMagMAX – MagMAX™ mirVana™ Total RNA Isolation Kit (Applied Biosystems, Waltham, MA, USA), bMonarch – Monarch® Total RNA Miniprep Kit (New England BioLabs, Ipswich, MA, USA), cNorgen Biotek – Exosomal RNA Isolation Kit (Norgen Biotek, Thorold, Canada).
Expression profiling of serum-derived small extracellular vesicles using Clariom D Pico Assays
Based on the results of the optimization phase of the study, we decided to use qEV single/35 nm columns from iZON for exosome isolation from sera, while RNA was isolated using the Monarch Total RNA Miniprep Kit. To assess the suitability of such isolated RNA for high-throughput screening of lncRNAs as well as protein-coding genes, we have determined expression profiles of serum-derived small EVs in six samples of CRC patients (arrays 1–6) and six samples of healthy controls (arrays 7–12) using Clariom D Pico Assays as this technology should be able to analyze more than 540,000 transcripts from as little as 100 pg of total RNA. In all cases, a yield of cRNA was higher than 62 μg after 12 cycles of PCR, which is in accordance with kit expectations. Further, 20 μg of cRNA were used for subsequent synthesis of ds-cDNA, which should result in yields higher than 6.6 μg. This step was also successfully completed as all samples provided yields higher than 12 μg of ds-cDNA. Additionally, the efficiency of hybridization was verified using hybridization controls added to the samples during the procedure and all samples passed successfully. Moreover, quality control from probe level model was performed using relative log expression (RLE) and normalized unscaled standard error (NUSE) metrics to assess the quality of individual arrays. As expected, RLE values were between −0.06 and 0.04, while NUSE values were between 1.00 and 1.02 indicating the similar quality of analyzed chips (Figure 3A and B). Finally, quality control of raw data was performed using “arrayQualityMetrics” to get feedback on the whole experimental procedure and identify potential outliers. Totally, three different approaches were used including between array comparison (Figure 4A and B), array intensity distribution (Figure 4C and D), and individual array quality (Figure 4E and F). Based on the results of Kolmogorov–Smirnov test, two samples of healthy controls (arrays 8 and 9) were considered to be outliers as their K a values were higher than determined treshold (Figure 4C). Nevertheless, the other used methods did not mark these samples as outliers. As the outlier detection is a poorly defined question and performed methods serve only as hints, which are intended to be followed up manually, we decided not to exclude these two arrays from further analyses.
Comparing CRC samples and healthy controls, 7,182 transcripts were identified to be significantly deregulated between these two groups of samples (adjusted p<0.05). The majority of them were non-coding RNAs with more than 70 % being lncRNAs, while protein-coding genes represented the second most common gene biotype (Table 3). The hierarchical clustering showed that CRC samples may be distinguished from healthy controls based on the different expression of all significant genes (Figure 5A), protein-coding genes (Figure 5B), or lncRNAs separately (Figure 5C) (adjusted p<0.05). Altogether, our results confirm the suitability of Clariom D Pico Assays for expression profiling of small EVs isolated from low volume of sera.
Gene biotype | Class of RNA | Total number of transcripts | Percentage from all identified genes, % | Percentage from particular gene biotype, % |
---|---|---|---|---|
Non-coding RNAs | 3,137 | 44.0 % | – | |
|
||||
lncRNAs | 2,309 | 32.0 % | 74.0 % | |
miscRNA | 304 | 4.5 % | 10.0 % | |
miRNA | 269 | 4.0 % | 9.0 % | |
snRNA | 166 | 2.5 % | 5.0 % | |
snoRNA | 80 | 1.0 % | 2.0 % | |
scaRNA | 3 | 0.0 % | 0.0 % | |
rRNA | 3 | 0.0 % | 0.0 % | |
ribozymes | 2 | 0.0 % | 0.0 % | |
mt_tRNA | 1 | 0.0 % | 0.0 % | |
|
||||
Protein-coding | 2,659 | 37.0 % | – | |
Pseudogenes | 1,285 | 18.0 % | – | |
|
||||
Processed | 845 | 12.0 % | 66.0 % | |
Unprocessed | 352 | 5.0 % | 27.0 % | |
rRNA | 52 | 1.0 % | 4.0 % | |
Unitary | 33 | 0.0 % | 3.0 % | |
Polymorphic | 3 | 0.0 % | 0.0 % | |
|
||||
Others | 101 | 1.0 % | – |
-
lncRNA, long non-coding RNA; miscRNA, miscellaneous RNA; miRNA, microRNA; snRNA, small nuclear RNA; snoRNA, small nucleolar RNA; scaRNA, small cajal body-specific RNA; rRNA, ribosomal RNA; mt_tRNA, mitochondrial transfer RNA.
Discussion
EVs have been intensively studied over the past decade, especially due to their abundance in different body fluids as well as their ability to transport various biomolecules. Thus, they could serve as novel promising targets of translational cancer research. However, despite a huge efforts, their clinical application is still highly restrained due to the lack of standardized methods for EVs isolation and characterization as well as non-existence of validated protocols for high-throughput screening of novel biomarkers. The major obstacles are the complexity of samples, overlapping biochemical and physiochemical characteristics between different EVs subtypes, the diversity of vesicles, and currently limited understanding of their biogenesis and release. Therefore, numerous isolation techniques based on different principles introduce significant variations in the concentration, size, and purity of gained EVs [38], [39], [40].
In the present study, we have compared a performance of three different methodical approaches commonly used for the isolation of small EVs from human blood serum to get the purest fraction of small EVs in sufficient quantity. As the volume of human samples in clinical settings is often limited, we decided to optimize the isolation procedure using only 150 μL of serum. Based on the results of TEM and DLS, we can conclude that EVs isolated by kits from Norgen Biotek and Exiqon were characterized by high PDI values indicating the presence of particles with very broad size range and all samples contained a considerable number of smaller particles. In case of SEC, F6 and F7 were proved to contain mostly the particles with the expected size of small EVs, although the detectable amount of ApoB was also confirmed by WB. These data confirm the suitability of qEVsingle/35 nm columns for small EVs isolation from low volume of sera as this approach enable to get high number of vesicles with proper size, shape, and sufficient purity although the lipoproteins are co-isolated to some extent.
Extraction of RNA is a crucial step for gene expression studies. As we were working with a low amount of starting material, we struggled for maximum yields. Previously, Eldh et al. [15] compared seven different approaches for the extraction of RNA from EVs including kits from four companies specialized in RNA extractions. Interestingly, spin column-based methods showed higher yields compared to phenol-based methods including combined phenol- and column-based techniques. This could be caused by a more rigid membrane of EVs enriched for cholesterol and sphingomyelin leading to incomplete lysis by phenol. Further, Bioanalyzer electrophoretograms of RNA patterns proved a significant difference in the relative presence of small and long RNAs depending on the isolation method. While column-based techniques isolated RNA with a broad size distribution, methods involving phenol were confirmed to be more efficient for isolation of small RNAs. Thus, to find the best approach for the extraction of RNA from our small EVs, we compared three different commercial kits not involving phenol. We planned to compare the kits’ effectiveness by RNA concentration measurements, however, it was too low to be reliably quantified. Thus, we assessed the efficiency by RT-qPCR analyses of selected genes as well as by comparing the results from two independent samples analyzed directly by selected microarrays. The best results were obtained in case of samples isolated by Monarch Total RNA Miniprep Kit, thus we decided to use this approach for all samples designed to transcriptome profiling.
The suitability of isolated RNA for high-throughput screening of lncRNAs as well as protein-coding genes was assessed using six samples from CRC patients and six samples from healthy controls. So far, several different studies investigated the expression profiles of small EVs using microarray technology [41], [42], [43]. Nevertheless, no one has tested the appropriateness of Clariom D Pico Assays for transcriptome profiling of serum-derived small EVs in CRC so far. Despite the low concentration of isolated RNA, we gained high yields of ds-cDNA and all samples passed the quality as well as hybridization control. The annotation of specific genes emerged to be rather complicated task due to the lack of updates in original annotation files and the use of probes derived from outdated transcripts that are no longer present in current databases. Thus, we were forced to create our own CDFs using probe sequences from BRAINARRAY project and GENCODE database as a target. Differential expression analysis revealed more than 7,000 transcripts significantly deregulated between compared groups including a high portion of lncRNAs and protein-coding genes. Importantly, further analyses confirmed our initial hypothesis that expression profiles of small EVs are significantly different between CRC patients and healthy controls, thus selected RNA molecules could serve as promising non-invasive diagnostic biomarkers in the future.
To summarize, we have achieved our objective of protocol optimization. Importantly, our results confirm the suitability of Clariom D Pico Assays for transcriptome profiling of small EVs isolated from low volume of sera indicating the possible utilization in clinical settings. Nevertheless, the use of expansion cohorts is necessary and robust validation by independent method should be performed. In addition, there is an increasing evidence that EVs are highly heterogenous consisting of several subpopulations with its own biological role and wide spectrum of effects on recipient cells [44], [45], [46], [47]. Thus, it is crucial to develop novel methods enabling precise separation of particular subpopulations as well as to identify reliable markers characteristic for distinct EVs subtypes. Finally, studies focusing on EVs isolation directly from cancer tissues are highly desirable [48, 49] in order to better understand the role of these vesicles in the tumor microenvironment and to perform comprehensive molecular characterization of disease-related EVs.
Funding source: Ministerstvo Zdravotnictví-Ceské Republiky
Award Identifier / Grant number: NU20-03-00127
Acknowledgments
The authors thanks to the Cryo-electron microscopy core facility (CEMCOF), CEITEC Masaryk University for the help with TEM imaging of small EVs.
-
Research ethics: The study was approved by the local Ethics Board at Masaryk Memorial Cancer Institute (ID 2018/1671/MOU) and conforms to recognized standards of Helsinki Declaration.
-
Informed consent: Informed consent was obtained from all individuals included in this study.
-
Author contributions: Petra Vychytilova-Faltejskova: Conceptualization, Methodology, Validation, Writing – Original Draft, Visualization. Sara Vilmanova: Methodology, Investigation, Writing – Original Draft. Lucie Pifkova: Methodology, Investigation. Tina Catela Ivković: Conceptualization, Methodology, Formal Analysis, Data Curation. Marie Mᶏdrzyk: Methodology, Formal Analysis, Investigation, Writing – Review and Editing, Visualization. Robin Jugas: Software, Formal Analysis, Resources, Data Curation. Tana Machackova: Conceptualization, Methodology, Validation. Jan Kotoucek: Methodology, Investigation. Milana Sachlova: Resources, Funding acquisition. Lucia Bohovicova: Resources, Writing – Review and Editing. Teodor Staněk: Resources, Data Curation. Jana Halamkova: Resources, Data Curation, Project administration. Ondrej Slaby: Supervision, Funding acquisition. All authors have accepted responsibility for the entire content of this manuscript and approved its submission.
-
Competing interests: The authors state no conflict of interest.
-
Research funding: This work was financially supported by Ministry of Health of the Czech Republic, grant No. NU20-03-00127, by the project National Institute for Cancer Research (Programme EXCELES, ID Project No. LX22NPO5102) – Funded by the European Union – Next Generation EU, and by the project BBMRI-CZ no. LM2018125. All rights reserved.
References
1. Yáñez-Mó, M, Siljander, PR-M, Andreu, Z, Zavec, AB, Borràs, FE, Buzas, EI, et al.. Biological properties of extracellular vesicles and their physiological functions. J Extracell Vesicles 2015;4:27066. https://doi.org/10.3402/jev.v4.27066.Search in Google Scholar PubMed PubMed Central
2. Zhang, Y, Liu, Y, Liu, H, Tang, WH. Exosomes: biogenesis, biologic function and clinical potential. Cell Biosci 2019;9:19. https://doi.org/10.1186/s13578-019-0282-2.Search in Google Scholar PubMed PubMed Central
3. Mittelbrunn, M, Sánchez-Madrid, F. Intercellular communication: diverse structures for exchange of genetic information. Nat Rev Mol Cell Biol 2012;13:328–35. https://doi.org/10.1038/nrm3335.Search in Google Scholar PubMed PubMed Central
4. Atay, S, Godwin, AK. Tumor-derived exosomes. Commun Integr Biol 2014;7:e28231. https://doi.org/10.4161/cib.28231.Search in Google Scholar PubMed PubMed Central
5. Chung, I-M, Rajakumar, G, Venkidasamy, B, Subramanian, U, Thiruvengadam, M. Exosomes: current use and future applications. Clin Chim Acta 2020;500:226–32. https://doi.org/10.1016/j.cca.2019.10.022.Search in Google Scholar PubMed
6. Bao, Q, Gong, L, Wang, J, Wen, J, Shen, Y, Zhang, W. Extracellular vesicle RNA sequencing reveals dramatic transcriptomic alterations between metastatic and primary osteosarcoma in a liquid biopsy approach. Ann Surg Oncol 2018;25:2642–51. https://doi.org/10.1245/s10434-018-6642-z.Search in Google Scholar PubMed
7. Krug, AK, Enderle, D, Karlovich, C, Priewasser, T, Bentink, S, Spiel, A, et al.. Improved EGFR mutation detection using combined exosomal RNA and circulating tumor DNA in NSCLC patient plasma. Ann Oncol 2018;29:700–6. https://doi.org/10.1093/annonc/mdx765.Search in Google Scholar PubMed PubMed Central
8. Royo, F, Théry, C, Falcón-Pérez, JM, Nieuwland, R, Witwer, KW. Methods for separation and characterization of extracellular vesicles: results of a worldwide survey performed by the ISEV rigor and standardization subcommittee. Cells 2020;9:1955. https://doi.org/10.3390/cells9091955.Search in Google Scholar PubMed PubMed Central
9. 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. https://doi.org/10.3389/fphys.2012.00162.Search in Google Scholar PubMed PubMed Central
10. Kurian, TK, Banik, S, Gopal, D, Chakrabarti, S, Mazumder, N. Elucidating methods for isolation and quantification of exosomes: a Review. Mol Biotechnol 2021;63:249–66. https://doi.org/10.1007/s12033-021-00300-3.Search in Google Scholar PubMed PubMed Central
11. Gámez-Valero, A, Monguió-Tortajada, M, Carreras-Planella, L, Franquesa, Mla, Beyer, K, Borràs, FE. Size-Exclusion Chromatography-based isolation minimally alters Extracellular Vesicles’ characteristics compared to precipitating agents. Sci Rep 2016;6:33641. https://doi.org/10.1038/srep33641.Search in Google Scholar PubMed PubMed Central
12. Paolini, L, Zendrini, A, Di Noto, G, Busatto, S, Lottini, E, Radeghieri, A, et al.. Residual matrix from different separation techniques impacts exosome biological activity. Sci Rep 2016;6:23550. https://doi.org/10.1038/srep23550.Search in Google Scholar PubMed PubMed Central
13. Tellez-Gabriel, M, Heymann, D. Exosomal lncRNAs: the newest promising liquid biopsy. Cancer Drug Resist 2019;2:1002–17. https://doi.org/10.20517/cdr.2019.69.Search in Google Scholar PubMed PubMed Central
14. Lötvall, J, Hill, AF, Hochberg, F, Buzás, EI, Vizio, DD, Gardiner, C, et al.. Minimal experimental requirements for definition of extracellular vesicles and their functions: a position statement from the International Society for Extracellular Vesicles. J Extracell Vesicles 2014;3:26913. https://doi.org/10.3402/jev.v3.26913.Search in Google Scholar PubMed PubMed Central
15. Eldh, M, Lötvall, J, Malmhäll, C, Ekström, K. Importance of RNA isolation methods for analysis of exosomal RNA: evaluation of different methods. Mol Immunol 2012;50:278–86. https://doi.org/10.1016/j.molimm.2012.02.001.Search in Google Scholar PubMed
16. Tang, Y-T, Huang, Y-Y, Zheng, L, Qin, S-H, Xu, X-P, An, T-X, et al.. Comparison of isolation methods of exosomes and exosomal RNA from cell culture medium and serum. Int J Mol Med 2017;40:834–44. https://doi.org/10.3892/ijmm.2017.3080.Search in Google Scholar PubMed PubMed Central
17. Wu, Y, Wang, Y, Wei, M, Han, X, Xu, T, Cui, M. Advances in the study of exosomal lncRNAs in tumors and the selection of research methods. Biomed Pharmacother 2020;123:109716. https://doi.org/10.1016/j.biopha.2019.109716.Search in Google Scholar PubMed
18. An, M, Wu, J, Zhu, J, Lubman, DM. Comparison of an optimized ultracentrifugation method versus size-exclusion chromatography for isolation of exosomes from human serum. J Proteome Res 2018;17:3599–605. https://doi.org/10.1021/acs.jproteome.8b00479.Search in Google Scholar PubMed PubMed Central
19. Andreu, Z, Rivas, E, Sanguino-Pascual, A, Lamana, A, Marazuela, M, González-Alvaro, I, et al.. Comparative analysis of EV isolation procedures for miRNAs detection in serum samples. J Extracell Vesicles 2016;5:31655. https://doi.org/10.3402/jev.v5.31655.Search in Google Scholar PubMed PubMed Central
20. Baranyai, T, Herczeg, K, Onódi, Z, Voszka, I, Módos, K, Marton, N, et al.. Isolation of exosomes from blood plasma: qualitative and quantitative comparison of ultracentrifugation and size exclusion chromatography methods. PLoS One 2015;10:e0145686. https://doi.org/10.1371/journal.pone.0145686.Search in Google Scholar PubMed PubMed Central
21. Helwa, I, Cai, J, Drewry, MD, Zimmerman, A, Dinkins, MB, Khaled, ML, et al.. A comparative study of serum exosome isolation using differential ultracentrifugation and three commercial reagents. PLoS One 2017;12:e0170628. https://doi.org/10.1371/journal.pone.0170628.Search in Google Scholar PubMed PubMed Central
22. Crossland, RE, Norden, J, Bibby, LA, Davis, J, Dickinson, AM. Evaluation of optimal extracellular vesicle small RNA isolation and qRT-PCR normalisation for serum and urine. J Immunol Methods 2016;429:39–49. https://doi.org/10.1016/j.jim.2015.12.011.Search in Google Scholar PubMed
23. Lobb, RJ, Becker, M, Wen Wen, S, Wong, CSF, Wiegmans, AP, Leimgruber, A, et al.. Optimized exosome isolation protocol for cell culture supernatant and human plasma. J Extracell Vesicles 2015;4:27031. https://doi.org/10.3402/jev.v4.27031.Search in Google Scholar PubMed PubMed Central
24. Patel, GK, Khan, MA, Zubair, H, Srivastava, SK, Khushman, M, Singh, S, et al.. Comparative analysis of exosome isolation methods using culture supernatant for optimum yield, purity and downstream applications. Sci Rep 2019;9:5335. https://doi.org/10.1038/s41598-019-41800-2.Search in Google Scholar PubMed PubMed Central
25. Prendergast, EN, de Souza Fonseca, MA, Dezem, FS, Lester, J, Karlan, BY, Noushmehr, H, et al.. Optimizing exosomal RNA isolation for RNA-Seq analyses of archival sera specimens. PLoS One 2018;13:e0196913. https://doi.org/10.1371/journal.pone.0196913.Search in Google Scholar PubMed PubMed Central
26. Rekker, K, Saare, M, Roost, AM, Kubo, A-L, Zarovni, N, Chiesi, A, et al.. Comparison of serum exosome isolation methods for microRNA profiling. Clin Biochem 2014;47:135–8. https://doi.org/10.1016/j.clinbiochem.2013.10.020.Search in Google Scholar PubMed
27. Stranska, R, Gysbrechts, L, Wouters, J, Vermeersch, P, Bloch, K, Dierickx, D, et al.. Comparison of membrane affinity-based method with size-exclusion chromatography for isolation of exosome-like vesicles from human plasma. J Transl Med 2018;16:1. https://doi.org/10.1186/s12967-017-1374-6.Search in Google Scholar PubMed PubMed Central
28. Serrano-Pertierra, E, Oliveira-Rodríguez, M, Rivas, M, Oliva, P, Villafani, J, Navarro, A, et al.. Characterization of plasma-derived extracellular vesicles isolated by different methods: a comparison study. Bioengineering 2019;6:8. https://doi.org/10.3390/bioengineering6010008.Search in Google Scholar PubMed PubMed Central
29. Kauffmann, A, Gentleman, R, Huber, W. arrayQualityMetrics—a bioconductor package for quality assessment of microarray data. Bioinformatics 2009;25:415–6. https://doi.org/10.1093/bioinformatics/btn647.Search in Google Scholar PubMed PubMed Central
30. Liu, G, Loraine, AE, Shigeta, R, Cline, M, Cheng, J, Valmeekam, V, et al.. NetAffx: Affymetrix probesets and annotations. Nucleic Acids Res 2003;31:82–6. https://doi.org/10.1093/nar/gkg121.Search in Google Scholar PubMed PubMed Central
31. Johnson, M, Zaretskaya, I, Raytselis, Y, Merezhuk, Y, McGinnis, S, Madden, TL. NCBI blast: a better web interface. Nucleic Acids Res 2008;36:W5–9. https://doi.org/10.1093/nar/gkn201.Search in Google Scholar PubMed PubMed Central
32. Todoerti, K, Ronchetti, D, Manzoni, M, Taiana, E, Neri, A, Agnelli, L. Bioinformatics pipeline to analyze lncRNA arrays. Methods Mol Biol 2021;2348:45–53. https://doi.org/10.1007/978-1-0716-1581-2_3.Search in Google Scholar PubMed
33. Théry, C, Witwer, KW, Aikawa, E, Alcaraz, MJ, Anderson, JD, Andriantsitohaina, R, et al.. Minimal information for studies of extracellular vesicles 2018 (MISEV2018): a position statement of the International Society for Extracellular Vesicles and update of the MISEV2014 guidelines. J Extracell Vesicles 2018;7:1535750. https://doi.org/10.1080/20013078.2018.1535750.Search in Google Scholar PubMed PubMed Central
34. Ng, CT, Azwar, S, Yip, WK, Sham, SYZ, Jabar, MF, Sahak, NH, et al.. Isolation and identification of long non-coding RNAs in exosomes derived from the serum of colorectal carcinoma patients. Biology 2021;10:918. https://doi.org/10.3390/biology10090918.Search in Google Scholar PubMed PubMed Central
35. Zeringer, E, Li, M, Barta, T, Schageman, J, Pedersen, KW, Neurauter, A, et al.. Methods for the extraction and RNA profiling of exosomes. World J Methodol 2013;3:11–8. https://doi.org/10.5662/wjm.v3.i1.11.Search in Google Scholar PubMed PubMed Central
36. Xu, J, Xiao, Y, Liu, B, Pan, S, Liu, Q, Shan, Y, et al.. Exosomal MALAT1 sponges miR-26a/26b to promote the invasion and metastasis of colorectal cancer via FUT4 enhanced fucosylation and PI3K/Akt pathway. J Exp Clin Cancer Res 2020;39:54. https://doi.org/10.1186/s13046-020-01562-6.Search in Google Scholar PubMed PubMed Central
37. Singh, AD, Patnam, S, Koyyada, R, Samal, R, Alvi, SB, Satyanaryana, G, et al.. Identifying stable reference genes in polyethene glycol precipitated urinary extracellular vesicles for RT-qPCR-based gene expression studies in renal graft dysfunction patients. Transpl Immunol 2022;75:101715. https://doi.org/10.1016/j.trim.2022.101715.Search in Google Scholar PubMed
38. Ludwig, N, Whiteside, TL, Reichert, TE. Challenges in exosome isolation and analysis in health and disease. Int J Mol Sci 2019;20:4684. https://doi.org/10.3390/ijms20194684.Search in Google Scholar PubMed PubMed Central
39. Yang, D, Zhang, W, Zhang, H, Zhang, F, Chen, L, Ma, L, et al.. Progress, opportunity, and perspective on exosome isolation – efforts for efficient exosome-based theranostics. Theranostics 2020;10:3684. https://doi.org/10.7150/thno.41580.Search in Google Scholar PubMed PubMed Central
40. Doyle, LM, Wang, MZ. Overview of extracellular vesicles, their origin, composition, purpose, and methods for exosome isolation and analysis. Cells 2019;8:727. https://doi.org/10.3390/cells8070727.Search in Google Scholar PubMed PubMed Central
41. Budden, CF, Gearing, LJ, Kaiser, R, Standke, L, Hertzog, PJ, Latz, E. Inflammasome-induced extracellular vesicles harbour distinct RNA signatures and alter bystander macrophage responses. J Extracell Vesicles 2021;10:e12127. https://doi.org/10.1002/jev2.12127.Search in Google Scholar PubMed PubMed Central
42. Luan, Y, Li, X, Luan, Y, Zhao, R, Li, Y, Liu, L, et al.. Circulating lncRNA UCA1 promotes malignancy of colorectal cancer via the miR-143/MYO6 Axis. Mol Ther Nucleic Acids 2019;19:790–803. https://doi.org/10.1016/j.omtn.2019.12.009.Search in Google Scholar PubMed PubMed Central
43. Hu, D, Zhan, Y, Zhu, K, Bai, M, Han, J, Si, Y, et al.. Plasma exosomal long non-coding RNAs serve as biomarkers for early detection of colorectal cancer. Chem Pharm Bull 2018;51:2704–15. https://doi.org/10.1159/000495961.Search in Google Scholar PubMed
44. Singh, K, Nalabotala, R, Koo, KM, Bose, S, Nayak, R, Shiddiky, MJA. Separation of distinct exosome subpopulations: isolation and characterization approaches and their associated challenges. Analyst 2021;146:3731–49. https://doi.org/10.1039/d1an00024a.Search in Google Scholar PubMed
45. Tkach, M, Kowal, J, Théry, C. Why the need and how to approach the functional diversity of extracellular vesicles. Philos Trans R Soc Lond B Biol Sci 2018;373:20160479. https://doi.org/10.1098/rstb.2016.0479.Search in Google Scholar PubMed PubMed Central
46. Raghavan, KS, Francescone, R, Franco-Barraza, J, Gardiner, JC, Vendramini-Costa, DB, Luong, T, et al.. NetrinG1+ cancer-associated fibroblasts generate unique extracellular vesicles that support the survival of pancreatic cancer cells under nutritional stress. Cancer Res Commun 2022;2:1017–36. https://doi.org/10.1158/2767-9764.crc-21-0147.Search in Google Scholar PubMed PubMed Central
47. Mimmi, S, Zimbo, AM, Rotundo, S, Cione, E, Nisticò, N, Aloisio, A, et al.. SARS CoV-2 spike protein-guided exosome isolation facilitates detection of potential miRNA biomarkers in COVID-19 infections. Clin Chem Lab Med 2023;61:1518–24. https://doi.org/10.1515/cclm-2022-1286.Search in Google Scholar PubMed
48. Crescitelli, R, Lässer, C, Lötvall, J. Isolation and characterization of extracellular vesicle subpopulations from tissues. Nat Protoc 2021;16:1548–80. https://doi.org/10.1038/s41596-020-00466-1.Search in Google Scholar PubMed
49. Maisano, D, Mimmi, S, Dattilo, V, Marino, F, Gentile, M, Vecchio, E, et al.. A novel phage display based platform for exosome diversity characterization. Nanoscale 2022;14:2998–3003. https://doi.org/10.1039/d1nr06804k.Search in Google Scholar PubMed
Supplementary Material
This article contains supplementary material (https://doi.org/10.1515/cclm-2023-0610).
© 2023 the author(s), published by De Gruyter, Berlin/Boston
This work is licensed under the Creative Commons Attribution 4.0 International License.