The release and uptake of nano-sized extracellular vesicles (EV) is a highly conserved means of intercellular communication. The molecular composition of EV, and thereby their signaling function to target cells, is regulated by cellular activation and differentiation stimuli. EV are regarded as snapshots of cells and are, therefore, in the limelight as biomarkers for disease. Although research on EV-associated RNA has predominantly focused on microRNAs, the transcriptome of EV consists of multiple classes of small non-coding RNAs with potential gene-regulatory functions. It is not known whether environmental cues imposed on cells induce specific changes in a broad range of EV-associated RNA classes. Here, we investigated whether immune-activating or -suppressing stimuli imposed on primary dendritic cells affected the release of various small non- coding RNAs via EV. The small RNA transcriptomes of highly pure EV populations free from ribonucleoprotein particles were analyzed by RNA sequencing and RT-qPCR. Immune stimulus-specific changes were found in the miRNA, snoRNA, and Y-RNA content of EV from dendritic cells, whereas tRNA and snRNA levels were much less affected. Only part of the changes in EV-RNA content reflected changes in cellular RNA, which urges caution in interpreting EV as snapshots of cells. By comprehensive analysis of RNA obtained from highly purified EV, we demonstrate that multiple RNA classes contribute to genetic messages conveyed via EV. The identification of multiple RNA classes that display cell stimulation-dependent association with EV is the prelude to unraveling the function and biomarker potential of these EV-RNAs. Keywords Immune activation · Immune suppression · Small RNA sequencing · Biomarker · Extracellular RNA Abbreviations Electronic supplementary material The online version of this BSA Bovine serum albumin article (https ://doi.org/10.1007/s0001 8-018-2842-8) contains DC Dendritic cell supplementary material, which is available to authorized users. EV Extracellular vesicles LPS Lipopolysaccharide * Esther N. M. Nolte-‘t Hoen miRNA/miR MicroRNA firstname.lastname@example.org miscRNA Miscellaneous RNA Department of Biochemistry and Cell Biology, Faculty ncRNA Non-coding RNA of Veterinary Medicine, Utrecht University, Utrecht, RNP Ribonucleoprotein (complex) The Netherlands rRNA Ribosomal RNA Department of Human Genetics, Leiden University Medical snRNA Small nuclear RNA Center, Leiden, The Netherlands snoRNA Small nucleolar RNA Leiden Genome Technology Center, Leiden University TLR Toll-like receptor Medical Center, Leiden, The Netherlands tRNA Transfer RNA School of Biological Sciences, Centre for Immunity, Infection and Evolution, Institute of Immunology and Infection Research, University of Edinburgh, Edinburgh, UK Centre for Biomolecular and Molecular Informatics, Radboud Institute for Molecular Life Sciences, Radboud University Medical Center Nijmegen, Nijmegen, The Netherlands Vol.:(0123456789) 1 3 T. A. P. Driedonks et al. functions of these RNAs [19–21]. It has been shown that Introduction changes in the activation status of the EV-producing cell lead to changes in the EV-miRNA composition [22–24]. Extracellular vesicles (EV) released by cells are consid- Moreover, miRNAs in EV were shown to be functionally ered as important mediators of intercellular communica- transferred between cells, leading to repression of genes tion [1, 2]. These 50–200 nm-sized vesicles are released in target cells [25, 26]. It is, therefore, thought that differ - by a broad range of cells, and have been detected in a ences in the miRNA content of EV underlie distinct func- wide range of body fluids, such as milk, plasma, urine, and tional effects of EV released by differentially stimulated semen [1, 3]. The collective term ‘EV’ refers to a hetero- cells [22–24]. In addition, changes in EV-miRNA levels in geneous population of secreted vesicles that are formed via plasma or serum have been associated with diseases such different pathways. Exosomes are formed as intraluminal as cancer, rheumatoid arthritis, and Alzheimer’s disease vesicles (ILVs) inside multi-vesicular endosomes (MVE) [27–29]. and are released upon fusion of the MVE with the plasma Despite the interest in miRNAs, a larger part of EV- membrane, whereas microvesicles directly bud off from associated RNA consists of other ncRNA classes. We and the plasma membrane . These vesicle populations over- others previously showed that EV released by cultured cells lap in size and molecular composition, which currently and EV present in a variety of body fluids contain many hampers their discrimination based on biophysical or bio- other small non-coding RNA species (20–300 nt), such as chemical parameters [4–6]. tRNA, snoRNA, snRNA, Y-RNA, Vault RNA, and SRP- EV are multi-component entities that transfer proteins, RNA (also named 7SL) [23, 30–35]. Some of these non- lipids, and RNA between cells. The RNA associated with coding RNAs are relatively enriched in EV, suggesting that EV mainly consists of small non-coding RNA species these RNAs are specic fi ally shuttled into EV for release into (ncRNA), which are thought to be of major importance the extracellular milieu. Inside cells, the above-mentioned in altering molecular processes in the recipient cell [3, 4, non-coding RNAs are known to play a role in basic cellular 7, 8]. In addition, the RNA composition of EV can give processes such as translation, RNA splicing, and RNA qual- information about the (patho)physiological status of the ity surveillance. Recent studies showed that full-length and EV-producing cell. Since EV can be easily obtained from fragmented forms of these non-coding RNAs can addition- body fluids, such EV-associated RNAs may be used as ally be involved in other processes including immunological non-invasive biomarkers for the early detection of dis- signaling, gene regulation, and guiding of other RNAs and ease [9, 10]. The molecular composition of EV is often nucleases [36–41]. However, it is not known whether activa- assumed to be a ‘snapshot’ of the producing cell, from tion stimuli imposed on the EV-producing cell ae ff ct the lev - which the tissue type or mutational status of the parental els of these ncRNA classes in EV, similar to EV-associated cell can be deduced [9, 11, 12]. miRNAs. Evaluation of cell stimulation-dependent changes Although the EV-RNA field receives major attention, in EV-associated ncRNA classes is a first step in uncovering with quickly rising numbers of publications on this topic, their function in EV-mediated signaling processes and their the technical difficulties to obtain pure EV remain under - potential as biomarkers indicative of the activation status exposed. It is important to note that a substantial amount of cells. of extracellular RNA is not contained in EV, but is associ- Here, we studied how changes in the activation/differen- ated with ribonucleoprotein particles (RNPs), including tiation status of the parental cell affect the small non-coding argonaute 2 (AGO2), or lipoprotein complexes [13–15]. RNA content of EV by employing a primary immune cell These contaminating structures have been shown to co- model, two different types of cell stimulation, and methods isolate together with EV via commonly used isolation pro- yielding highly pure EV populations. For our analysis, we cedures, and may, therefore, greatly affect the outcomes used primary dendritic cells (DC), which are master regu- of experiments [4, 5, 16]. The currently most reliable lators of immune responses and have been shown to com- method to obtain pure EV is by high-speed ultracentrifu- municate with various other immune cells via EV [2, 8, 42]. gation followed by density centrifugation in which EV are In addition, the use of DC-derived EV has been proposed separated from contaminating structures based on differ - as a strategy for cancer immune(chemo)therapy [43–46]. ences in buoyant density . Subsequently, analyzing the Depending on external stimuli, DC can be differentiated presence of common EV markers (such as CD9, CD63, into functionally different phenotypes that either activate or and CD81) and absence of contaminant proteins should down-regulate immune responses . These DC, therefore, be performed to assess the purity of EV preparations used present a suitable model to investigate how different external in the study [17, 18]. stimuli affect incorporation of a broad range of RNA classes MiRNAs are the most intensely studied type of small into EV. We compared cellular and EV-associated RNA lev- RNA in EV, likely due to the interest in gene-regulatory els in control, immune-stimulating, or immune-suppressing 1 3 Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… DC conditions using RNA deep sequencing and RT-qPCR. incubator. Experiments were approved by the institutional Our data indicate that differentiation signals imposed on ethical animal committee at Utrecht University (Utrecht, The dendritic cells affect the EV-associated levels of particular Netherlands). RNA classes, such as miRNA, snoRNA, and Y-RNA, while other non-coding RNA types remain largely unchanged. Flow cytometry Moreover, only a minor part of the stimulation-induced changes in EV-RNA content reflects changes in cellular Day 12 bone marrow DC (BMDC) were collected and RNA levels. This study exemplifies how comprehensive labeled for 30 min in PBS + 1% BSA (Bovine Serum Albu- analysis of RNA obtained from highly purified EV yields min, cat. K45-001, GE Healthcare Bio-Sciences, Pasch- candidate small ncRNAs beyond miRNAs for further explo- ing, Austria) containing anti-CD11c-APC (eBioscience, ration as biomarkers or functional entities within EV. clone N418, 1:400), anti-MHCII-FITC (eBioscience, clone M5/114.15.2, 1:1500), and anti-CD40-PE (BD Biosciences, clone 3/23, 1:200), or anti-CD86-PE (eBioscience, clone Materials and methods GL1, 1:200) or anti-CD274-PE (PD-L1, eBioscience, clone M1H5, 1:400). As control, corresponding isotype control Cell culture antibodies (eBioscience, San Diego, CA) were used. Sur- face labeling was analyzed using a FACS Calibur Flow Complete culture medium was prepared by supplementing Cytometer (BD Biosciences, San Jose, CA). Data analysis Iscove’s Modified Dulbecco’s Medium (IMDM, Lonza, was performed using FCS Express V3 (DeNovo Software, Verviers, Belgium) with 2 mM Ultraglutamine (Lonza, Glendale, CA). Verviers, Belgium), 10% fetal calf serum (FCS, GE Health- care Bio-Sciences, Pasching, Austria), 100 IU/ml penicil- Fluorescent labeling and purification of EV lin and 100 μg/ml streptomycin (Gibco, Paisley, United Kingdom), and 50 µM β-mercaptoethanol. The GM-CSF Conditioned cell supernatants were pooled to volumes of producing NIH 3T3 cell line (R1) was grown in complete 100–140 ml per condition and were subjected to differential medium. Primary bone marrow cells were grown in com- centrifugation as described previously . In brief: super- plete medium supplemented with 30% conditioned medium natant was sequentially centrifuged 2 × 200g for 10 min, from R1 cells. To prepare EV-depleted medium, a mixture of 2 × 500g for 10 min, and 1 × 10,000g for 30 min. Next, EV 150 ml conditioned R1 culture supernatant and 42.5 ml FCS were pelleted by ultracentrifugation at 100,000g for 65 min (end concentration 30% FCS) was depleted of EV by over- using an SW28 rotor (k-factor 334.2) (Beckman Coulter, night centrifugation at 100,000g in an SW28 rotor (k-factor Brea, CA). For EV quantification by high-resolution flow 334.2) (Beckman Coulter, Brea, CA). The EV-depleted cytometry, pellets were resuspended in 20 µl PBS + 0.2% supernatant was carefully pipetted from the tubes, leaving BSA (cleared from aggregates by overnight ultracentrifuga- 5 ml in the tubes to prevent disturbance of the pellet. This tion at 100,000g) and labeled with 1.5 µl PKH67 (Sigma, supernatant was filtered through a 0.22 μm bottle top filter St. Louis, MO) in 100 µl Diluent C per pellet. For EV-RNA (Millipore, Billerica, MA) after which additional IMDM, isolation, 100,000 g pellets were resuspended in 50 µl ultraglutamine, antibiotics, and β-mercaptoethanol were PBS + 0.2% BSA. PKH67-labeled samples or samples for added to prepare complete medium as indicated above. The RNA isolation were mixed with 1.5 ml 2.5 M sucrose, and efficiency of EV depletion from culture medium, as regu - overlaid with a linear sucrose gradient (2.0–0.4 M sucrose in larly assessed by high-resolution flow cytometric analy - PBS). Gradients were centrifuged 15–18 h at 192,000g in a sis, is on average ~ 90%. Primary bone marrow cells were SW40 rotor (k-factor 144.5) (Beckman Coulter, Brea, CA). flushed from the femur and tibia of 8–12-week-old C57bl/6 PKH67-labeled EV were used for high-resolution flow cyto- mice and differentiated to dendritic cells according to Lutz metric analysis (see below). Alternatively, the fractions with et al. . To generate tolerogenic DC, cells were treated densities of 1.12–1.18 g/ml (as measured by refractometry) from day 2 onward with 10 nM 1α,25-dihydroxyvitaminD3 were pooled, diluted six times in PBS + 0.2% EV-depleted (Sigma, St Louis, MO). To generate immunogenic DC, 1 µg/ BSA, and centrifuged again at 192,000g for 65 min in a ml lipopolysaccharide (LPS, cat. L-2630, Sigma, St. Louis, SW40 rotor (k-factor 144.5) prior to EV-RNA isolation. MO) was added on day 12. On day 13, non-adherent and semi-adherent cells were recovered and cultured at 3 × 10 RNA isolation cells/dish for 20 h in EV-depleted culture medium. Cell via- bility, as determined by Trypan blue exclusion, did not differ Small RNA was isolated from EV pellets and from 1 × 10 between treatments and was above 90% for all cultures. All cells using the miRNeasy Micro kit according to the small cells were maintained at 37 °C and 5% C O in a humidified RNA enrichment protocol provided by the manufacturer 1 3 T. A. P. Driedonks et al. (Qiagen, Hilden, Germany). RNA yield and size profile were Sequences with a minimal length of 17 bp after adapter assessed using the Agilent 2100 Bioanalyzer with Pico 6000 trimming were retained to ensure high quality reads (cut- RNA chips (Agilent Technologies, Waldbronn, Germany). adapt -q 30 -u -51 -U -51 -n1 --minimum-length 17 -a A G A T CG G AA G A G C A C A CG T CT G AA C -A G A T CG T CGG A CT G TA G AA CT C T G A A). R eads w er e mapped Preparation of small RNA sequencing libraries to the mouse genome mm10 using gsnap (v2014-12-23; gsnap --novelsplicing 1 --npaths 3 --format sam). Count 50 ng cellular small RNA and 2 ng EV-derived small RNA tables were prepared using Htseq (v0.6.1p1; htseq-count was treated with DNase [Turbo DNA-free kit (Life Tech- --minaqual 0 --format bam --order pos --stranded yes). For nologies, Carlsbad, CA)] according to the manufacturer’s miRNA count tables, miRbase v20 annotation was used. instructions. The DNase-treated RNA was pelleted using Gene annotation for snoRNA, snRNA, and miscRNA was Pellet Paint (Merck, Darmstadt, Germany) according to the retrieved from Biomart (Ensembl 84). tRNA and rRNA manufacturer’s instructions, reconstituted in milliQ (MQ) chromosomal positions were retrieved from the UCSC and subsequently subjected to ribosomal RNA depletion mm10 repeat mask track. Reproducibility of triplicate using RiboZero Gold kit (Human/Mouse/Rat) (Illumina, experiments (performed on different days) was assessed San Diego, CA) according to the manufacturer’s instruc- by evaluating Pearson correlation values (after logarithmic tions. This is required to deplete residual rRNA from cel- transformation of the small RNA counts). lular RNA samples and, for comparability, we subjected EV- Differential abundance was determined separately for RNA samples to the same procedure. Hereafter, RNA was each ncRNA class, and separately for EV and cells, using pelleted with Pellet Paint and reconstituted into 6 µl MQ. the edgeR package in R . Data were normalized using cDNA libraries were prepared using the NebNext smallRNA the TMM method (weighted trimmed mean of M values). library prep kit for Illumina (New England Biolabs, Ipswich, Estimating the common, trended (overexpression values), MA), according to the manufacturer’s instructions but with and tagwise dispersion, a generalized linear model, con- the following adaptations: 3′ adapter-ligation was carried taining the effect of cell stimulation and the effect of the out overnight at 16 °C; PCR amplification was done using day-to-day variation between triplicate experiments, was Kapa HiFi Readymix 2× PCR mastermix (Kapa Biosystems, fit. The log-likelihood ratio test was used to evaluate dif- Wilmington, MA) using barcoded primers and the follow- ferential expression between treatments relative to control. ing PCR programme: 2 min at 95 °C, 15 cycles of 20 s at p values were adjusted for multiple testing using Benja- 98 °C, 30 s at 62 °C, 15 s at 70 °C, and a final elongation mini and Hochberg’s false discovery rate (FDR). Average step of 5 min at 70 °C. cDNA was purified using magnetic fold-change over three independent experiments and stand- AMPure XP beads (Beckman Coulter, Brea, CA) and quan- ard deviation were plotted. Analysis of RNA fragments tified using Agilent 2100 Bioanalyzer and DNA HiSensi- was done using the UCSC genome browser and Integrated tivity chips (Agilent Technologies, Waldbronn, Germany). Genome Viewer . Adapter dimers (126 nt in size) were removed by running the cDNA on 6% TBE gels (Life Technologies, Carlsbad, CA) for 60 min at 145 V, after which cDNA products of Quantitative real‑time PCR 15–300 nt (+ 126 nt adapters) were cut from the gel and purified. Subsequently, all libraries were pooled at equimo- cDNA was generated from cellular or EV-derived small lar ratios and run on a 4–12% TBE gel (Life Technologies, RNA using the miScript RT2 kit (Qiagen, Hilden, Ger- Carlsbad, CA). Size fractions of 15–25 nt (includes miR- many). An equivalent of 20 pg RNA was used per qPCR NAs), 25–60 nt, 60–80 nt (includes tRNAs), and 80–275 nt reaction and mixed with 100 nM primers (Isogen Life Sci- (each + 126 nt adapters) were cut from the gel, and purified ences, De Meern, The Netherlands) and 4 µl SYBR Green and pooled in a volumetric ratio of 2:2:3:3. Sequencing was Sensimix (Bioline Reagents Ltd., United Kingdom) in an done using 150 bp paired-end reads on an Illumina HiSeq 8 µl reaction. No-RT-controls confirmed the absence of 4000 machine (Illumina, San Diego, CA) at ServiceXS (Lei- genomic DNA and non-specific amplification. den, The Netherlands). Cycling conditions were 95 °C for 10 min followed by 50 cycles of 95 °C for 10 s, 57 °C for 30 s, and 72 °C for RNAseq data analysis 20 s. All PCR reactions were performed on the Bio-Rad iQ5 Multicolor Real-Time PCR Detection System (Bio- All data was processed as paired-end reads. Data qual- Rad, Hercules, CA). Quantification cycle (Cq) values were ity was checked with FastQC and reads were processed determined using Bio-Rad CFX software using automatic with cutadapt (version 1.8) to remove low-quality reads, baseline settings. Thresholds were set in the linear phase clip the reads to 100 bp, and remove adapter sequences. of the amplification curve. 1 3 Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… High‑resolution flow cytometric analysis of EV Nanoparticle tracking analysis (NTA) High-resolution flow cytometric analysis of PKH67- EV purified by density gradient ultracentrifugation as labeled EV was performed using a BD Influx flow cytom- described above were resuspended in 50 µl PBS + 0.4% eter (BD Biosciences, San Jose, CA) with an optimized EV-depleted BSA. Samples were diluted 20 times in PBS configuration, as previously described [49, 52]. In brief, (confirmed to be particle-free when analyzed with the same we applied threshold triggering on fluorescence derived settings as used for the EV samples) before measurement from PKH67-labeled EV passing the first laser. Forward on a Nanosight NS500 instrument (Malvern, Worchester- scatter (FSC) was detected with a collection angle of shire, UK). Data acquisition and processing were performed 15°–25° (reduced wide-angle FSC). Fluorescent 100- using NTA software 3.1. Each sample was recorded for 5 and 200-nm polystyrene beads (FluoSpheres, Invitrogen, times 30 s, 25 frames per second, camera level 14, detection Carlsbad, CA) were used to calibrate the fluorescence and threshold 5. rw-FSC settings. Sucrose gradient fractions containing PKH67-labeled EV were diluted 25× in PBS and vor- Electron microscopy texed just before measurement. This dilution factor was sufficient to avoid ‘coincidence’ (multiple EV arriving EV and RNP were separated by density gradient centrifuga- at the measuring spot at the same time), thereby allowing tion as described above, and resuspended in 15 µl PBS. 3 µl accurate quantitative comparison of EV numbers in dif- aliquots of either EV or RNP were added on top of carbon- ferent conditions. Moreover, samples were measured at coated 300 mesh, copper grids, and incubated for 2 min at maximally 10,000 events per second, which is far below room temperature. The grids were then washed two times the limit in the electronic pulse processing speed of the in 50 µl PBS and stained with 2% uranyl acetate. Imaging BD Inf lux . was performed on a T20 electron microscope (FEI) operated at 200 keV. Images were recorded on a CCD Eagle camera (FEI). Western blotting Northern blotting Cell pellets were lyzed in PBS + 1% Nonidet-P40 with protein inhibitor cocktail (Roche, Basel, Switzerland) for Cellular or EV-derived small RNA (between 100 ng and 15 min on ice. Nuclei were spun down at 16,000 g for 1 ng per lane, as indicated) was denatured in gel loading 15 min at 4 °C, supernatant was used for Western blotting. buffer II (Invitrogen) for 2 min at 70 °C and snap-cooled on Cell lysates and EV were denatured in SDS-sample buffer ice before loading on a denaturing 15% PAGE gel (National at 100 °C for 3 min, and separated using 12% SDS-PAGE Diagnostics, Nottingham, UK) as described previously . gels, after which proteins were transferred onto Immo- RNA was transferred onto a Hybond-N Nylon membrane bilon-P 0.45 μm PVDF membranes (Millipore, Cork, Ire- (Amersham Pharmacia Biotech, GE Healthcare Bio-Sci- land). After blocking for 1–2 h in blocking buffer (0.5% ences, Pasching, Austria) and chemically crosslinked with Cold Fish Skin Gelatin (Sigma-Aldrich, St. Louis, CA) in 1-ethyl-3-(3-dimethylaminopropyl) carbodimide (Sigma) at PBS + 0.05% Tween-20), blots were incubated overnight 55 °C for 2 h . P-labeled DNA oligo probes perfectly at 4 °C with primary antibodies [anti-mouse-CD9 (eBio- complementary to each RNA species were hybridized over- science, clone KMC8, 1: 1000), anti-mouse-CD63 (MBL, night at 42 °C in PerfectHyb (Sigma) solution. Blots were clone D263-3, 1:1000), anti-mouse–galectin-3 (eBiosci- analyzed by phosphorimaging using a Typhoon Scanner (GE ence, clone M3/38, 1:500), anti-MHCII-p55 (GenScript, Healthcare). Piscataway, NJ, custom Ab raised against MHCII bèta chain peptide sequence RSQKGPRGPPPAGLLQC, List of DNA oligo probes 1:5000), or anti-mouse-beta-actin (ThermoScientific, polyclonal PA1-16889, 1:5000)] in blocking buffer, and RT‑qPCR probes washed and incubated for 1–2 h with HRP-coupled sec- ondary antibodies (Dako, cat P0450 and P0448, 1:5000). The following forward primers were used in RT-qPCR (5′ ECL solution (ThermoScientific, SuperSignal West Dura to 3′) in combination with the miScript universal reverse Extended Duration Substrate, cat. 34075) was used for primer. detection on a Chemidoc imager (Bio-Rad, Hercules, CA). Images were analyzed by the Image Lab software miR-146a-5p TGA GAA CTG AAT TCC ATG GGT (Bio-Rad, Hercules, CA). miR-155-5p GGG TTA ATG CTA ATT GTG ATA G miR-9-5p GGG TCT TTG GTT ATC TAG C 1 3 T. A. P. Driedonks et al. miR-10a-5p TAC CCT GTA GAT CCG AAT TTG TG as p < 0.05. Statistical tests on data not derived from RNA miR-27b-3p TTC ACA GTG GCT AAG TTC TGC sequencing were done in SPSS (v24, IBM). miR-378-5p ACT GGA CTT GGA GTC AGA AG mY1 GTT ATC TCA ATT GAT TGT TCA CAG Availability of data and materials TC mY3 GGC TGG TCC GAG TGC AGT GG Raw data and count tables are deposited in the GEO data- RNU1 CCA TGA TCA CGA AGG TGG TTT base under accession number GSE105151. Written details RNU6 CTC GCT TCG GCA GCACA on experimental procedures have been submitted to the EV- snord65 TAG TGG TGA GCC TAT GGT TTT TRACK knowledgebase (EV-TRACK ID: EV170030) . snord68 AGT ACT TTT GAA CCC TTT TCCA Northern oligo’s Results Functionally distinct subsets of mouse bone marrow- Y1_5p TTG AGA TAA CTC ACT ACC TTC derived DCs were generated by exposure of cells to LPS, GGA CCA GCC thereby inducing immune-stimulatory DC (LPS-DC), or Y1_3p GTC AAG TGC AGT AGT GAG to 1α,25-dihydroxyvitaminD3, thereby inducing immune- AAG suppressive DC (VitD3-DC) . Untreated DC were used 5′-tRNA-Glu_CTC CCG AAT CCT AAC CAC TAG ACC as control cells. The different DC subtypes displayed the ACC AG expected differences in expression of the activation mark - 5′-tRNA-Gly_GTT GCA TTG GTG GTT CAG TGG TAG ers, with LPS-DC showing increased levels of MHC class AAT TCT CGC C II and CD86 expression, while VitD3-DC showed increased PD-L1/CD86 ratios  and resistance to LPS activation Statistical analyses  (Supplementary Fig. 1A–F). We isolated EV from cell culture supernatant of LPS-DC, VitD3-DC, and con- Differences between the numbers of EV released by LPS- trol DC using differential (ultra)centrifugation followed and VitD3-stimulated DC versus control DC were analyzed by density gradient separation, as previously published by by one-way ANOVA with Dunnett’s two-sided post hoc test. our group (Supplementary Fig. 2A and ). Transmission Similar statistical testing was performed on differences in electron microscopic analysis indicated that low-density RNA yield. Differences in RNA biotypes in EV and cells fractions contained 100–150 nm-sized EV, whereas no EV were analyzed by two-tailed t test. Differences in fold- were observed in high-density fractions enriched in protein changes of individual RNAs as measured in RT-qPCR were complexes (Supplementary Fig. 2B). Our in-house devel- analyzed by one-way ANOVA. Significance was defined oped high-resolution flow cytometric method was used for EV count RNA yield ab n.s. Fig. 1 LPS- and VitD3-stimulated DCs release different numbers experiments, one-way ANOVA with Dunnett’s two-sided post hoc of extracellular vesicles. a EV derived from equal numbers of con- test, *p < 0.05. b Bioanalyzer-based quantification of RNA isolated trol, and LPS- and VitD3-treated DC were fluorescently labeled from EV derived from equal numbers of cells purified by density gra- using PKH67, separated from RNPs and free dye by sucrose density dient centrifugation. Small RNA yields were normalized to the con- centrifugation, and quantified by high-resolution flow cytometry. trol DC condition. Indicated are the mean ± SD values of n = 4 inde- Indicated are the number of fluorescent events detected in 30 s, nor - pendent experiments (one-way ANOVA with Dunnett’s two-sided malized to the control condition (mean ± SD of n = 3 independent post hoc test, *p < 0.05) 1 3 Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… high-throughput EV analysis at the single particle level [49, Vault RNA (142 nt) (Supplementary Fig. 3). More than 13 52]. The light scatter patterns induced by EV from differ - million reads were obtained for each library (Supplemen- ently stimulated DC were highly similar (Supplementary tary Table 1). Pearson coefficients for biological replicate Fig. 2C). However, we detected differences in the number samples were ≥ 0.85 for EV, and ≥ 0.95 for cells (Supple- of released EV, with LPS-DC releasing more, and VitD3- mentary Fig. 4). As expected based on the previous studies DC releasing less EV than non-treated control DC (Fig. 1a). [23, 30–32, 34], we detected the presence of various small These quantitative differences were confirmed by nanopar - RNA classes in EV (miRNA, snoRNA, snRNA, tRNA, mis- ticle tracking analysis (NTA) (Supplementary Fig. 2D). In cRNA, and rRNA). Some RNA types were enriched in EV, addition, the NTA data indicated that the majority of EV such as Y-RNA, Vault RNA, and 7SL RNA classified as observed in all conditions were in the 100–200 nm size ‘miscRNA’, while other RNA types, such as snoRNA, were range. Western blot analysis showed that the DC-derived relatively less abundant in EV compared to cells (Fig. 2). EV contained variable levels of common EV proteins such Since we depleted ribosomal RNA before preparation of as CD9, MHCII, CD63, and Galectin-3 , whereas abun- the sequencing libraries, apparent enrichment of rRNA in dant cellular proteins such as beta-actin were not detected EV is probably caused by differences in depletion efficiency (Supplementary Fig. 2E). These data demonstrate that VitD3 between cells and EV. MiRNA and snRNA abundance was and LPS treatment of DC differentially affect the number similar between cells and EV. Stimulation of DC with LPS and protein content of EV released by these cells. or VitD3 did not lead to major changes in the distribution of Next, we analyzed how LPS- or VitD3 treatment of DC sequencing reads over different RNA classes in EV and cells affected the RNA content of EV released by these cells. It (Supplementary Fig. 5A–B). is important to note that extracellular RNA can be associ- Next, we assessed whether, within each of the different ated with either EV or ribonucleoprotein complexes, which classes, the levels of individual RNAs in EV changed as a may sediment at similar centrifugal force . Using density result of cell stimulation. Hereto, read counts were normal- gradient-based purification, we separated EV from con- ized per RNA class, after which the fold changes in reads taminating ribonucleoprotein complexes. Depending on the differentiation status of the parental DC, the EV-free ab c ribonucleoprotein fraction contained between 26 and 55% of total extracellular RNA released from cells (Supplemen- tary Fig. 2F–G). The total amount of EV-associated RNA released by equal numbers of control, LPS-, and VitD3- treated DC was different and reflected the differences in EV numbers (Fig. 1a, b). To assess whether DC stimulation altered the RNA composition of EV, we prepared sequenc- ing libraries of EV-associated and cellular small RNA from control-, LPS-, and VitD3-DC cultures (n = 3 biological replicates). Sequencing libraries were prepared using an adapter-ligation-based method routinely used in miRNA de f profiling . Such a method has been frequently applied in studies to demonstrate the presence of miRNA and other small non-coding RNA types in EV (e.g., [23, 24, 31, 32, 61]). All libraries contained RNAs with lengths ranging between 20 and 300 nucleotides. Since short RNA mole- cules have a selective advantage during PCR amplification and pre-amplification on the sequencing flow-cell, leading to an apparent over-representation of short read lengths, we enriched the libraries for longer RNAs to ensure suf- Fig. 2 Cells and EV differ in the relative abundance of different small ficient coverage of these sequences. To this end, cellular RNA classes. a–f The relative abundance of the indicated ncRNA and EV-associated cDNA libraries were split into different classes in cells versus EV was assessed by small RNAseq analysis size fractions of 15–25 nt (includes miRNAs), 25–60 nt, of cellular and EV-associated RNA. Total read count for each RNA 60–80 nt (includes tRNAs), and 80–275 nt. Subsequently, class was normalized to total library size. To compare the distribution of ncRNA classes between cells and EV, normalized RNA class read we enriched for longer RNA molecules by pooling these counts in EV were scaled to the normalized RNA class read counts size fractions in a volumetric ratio of 2:2:3:3 (for details, see in cellular RNA (set to 1). Indicated are mean values ± SD of n = 3 “Materials and methods”). This, indeed, resulted in a better independent experiments (two-tailed t test on normalized CPM val- coverage of 100–300 nt mid-size RNAs, such as full-length ues, *p < 0.05) 1 3 T. A. P. Driedonks et al. Fig. 3 LPS- versus VitD3-induced changes in EV-associated RNA LPS-EV relative to control-EV versus log2fold change in VitD3-EV classes. EV-associated RNA from control, LPS, and VitD3 condi- relative to control-EV. Cutoffs for log2fold changes larger or smaller than 1 are indicated with dashed lines (red = LPS; green = VitD3), so tions was isolated and analyzed by RNA sequencing. Read counts all data points beyond these lines are differentially expressed with for individual RNAs were normalized to the total read counts of log2FC > 1. Dot size represents the normalized abundance (logCPM) each RNA class. a–e LPS- or VitD3-induced fold changes and cor- of individual RNAs. RNAs that changed with non-adjusted p < 0.05 responding p values were calculated relative to the control condition are highlighted in black with edgeR GLM method. Data are expressed as log2fold change in per million (RPM) of individual RNAs in LPS or VitD3 LysTTT) (Supplementary Fig. 6A). Although relatively high over control conditions were calculated. The scatter plots in tRNA read counts were observed in the EV-associated RNA Fig. 3a–e display the log-fold changes induced by LPS ver- pool (Supplementary Fig. 5B), the levels of these six abun- sus VitD3 treatment for the indicated RNA classes (dashed dant tRNAs remained stable (log2FC < 1) in response to cel- lines correspond to a log2fold change of 1). Dot size indi- lular stimulation (Supplementary Fig. 6B). Besides tRNAs, cates the average (n = 3) abundance of a transcript in EV the highly purified EV populations used in this study also and RNAs with non-adjusted p < 0.05 are highlighted in contained substantial levels of snRNA, which is an RNA black. The scatter plots of these data indicated that LPS- type known to be mainly confined to the nucleus. Although and VitD3-treatment affected the levels of miRNA, snoRNA, cell stimulation seemed to induce changes in EV-associated snRNA, and miscRNA in EV (Fig. 3a–e). Some of the stim- snRNA levels (Fig. 3c), many of the individual data points ulation-induced changes were similar for LPS and VitD3 mapped to multicopy genes with highly similar sequences conditions, while other RNAs show opposing alterations, or (> 95% sequence identity) corresponding to known snRNAs. change only in one of the stimulation conditions. Cumulative analysis of these multicopy genes indicated that Of the tested RNA classes, tRNA levels in EV showed 96% of snRNA reads mapped to four different snRNAs (U1, the lowest rate of change due to LPS or VitD3 stimulation. U2, U5, and U6) (Fig. 4a), of which the levels in EV did 97% of tRNA reads mapped to six different tRNA isoaccep- not significantly differ between stimulation versus control tors (GluCTC, GlyGCC, LysCTT, GluTTC, GluCTG, and conditions (Fig. 4b). Thus, the levels of snRNAs and tRNAs, 1 3 Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… Fig. 4 EV from LPS- and VitD3-stimulated DC display stable lev- abundant snRNAs, constituting 96% of snRNA reads in EV-RNA, are els of abundant snRNAs. EV-associated RNA from control, LPS, expressed as percentage of the total snRNA read count in EV. b Data and VitD3 conditions was isolated and analyzed by RNA sequenc- from a are expressed as fold change in the indicated snRNA levels of ing. Read counts for individual RNAs were normalized to the total VitD3- or LPS-EV relative to control-EV (mean ± SD of n = 3 experi- read counts of each RNA class. a Read counts for the top-four most ments; in one-way ANOVA, no significant differences were observed) which are commonly detected RNA types in EV, remained significance (Fig. 5b). Together, these data indicate that constant upon different immune stimuli imposed on the EV- opposing stimuli imposed on the same type of parental cell producing cells. induce the release of functionally different sets of miRNAs In contrast to tRNAs and snRNAs, the EV-associated via EV. levels of three other RNA types tested here, i.e., miRNAs, Besides changes in miRNA levels, the levels of several snoRNAs, and miscRNAs, changed in response to the differ - EV-associated snoRNAs were found to differ between LPS ent stimulate imposed on the DC. We aimed to identify the and VitD3 conditions (Fig. 5c), with three H/ACA box RNAs differentially present in EV released by LPS- versus snoRNAs (snora2b, snora32, and snora55) displaying FDR VitD3-treated DC, since such RNAs may underlie distinct levels < 0.05. The overall low abundance of this RNA class effects of EV on recipient cells. in EV (Fig. 2c) hampered reliable quantification of snoR - Within the miRNA group, we observed the highest num- NAs in EV by RT-qPCR (data not shown). Nevertheless, ber of RNAs that differed between the LPS and VitD3 groups our data are a first indication that EV-associated snoRNA with a fold change > 2 and with FDR values < 0.05 (Fig. 5a, levels change with the activation status of parental cells. red dots). Interestingly, for the majority (15 out of 20) of the Although none of the RNAs in the miscRNA group reached top miRNAs that differed significantly between VitD3 and FDR levels < 0.05, analysis by RT-qPCR indicated that LPS conditions with lowest FDR values, DC functions have the levels of Y3-RNA (Rny3) and an additional member been described previously (Table 1 and [62–73]). of the Y-RNA family, Y1-RNA (Rny1), significantly dif - MiRNAs known to be involved in the pro-inflammatory fered between LPS and VitD3 conditions (Fig. 5d, e). In function of DC were enriched in LPS-induced EV, whereas cells, the conserved family of Y-RNAs is known to bind to EV from VitD3-treated DC were enriched in miRNAs that several different proteins that play a role in RNA quality dampen immune-stimulatory signaling cascades in DC. We control [75, 76]. Members of the small non-coding Y-RNA validated three LPS-enriched and three VitD3-enriched family have frequently been detected in EV from multiple RNAs by RT-qPCR. The use of stable reference transcripts different cell types and different body fluids [30– 32, 34, 35, in RT-qPCR analysis is important to ensure reliable normali- 77–81]. We here provide evidence that Y-RNA levels in EV zation between different samples. However, the identity and can be regulated by stimuli imposed on the EV-producing general applicability of reference transcripts stably present cell. The reduced Y-RNA levels in EV from LPS-stimulated in EV are understudied. Here, we used our sequencing data DC compared to EV from VitD3-stimulated DC (Fig. 5e) set to select four non-coding RNAs that were present in EV indicate that the presence of Y-RNAs in EV may be indica- at comparable levels across all conditions, and confirmed tive for the immune function of the parental cells. Overall, their stability by RT-qPCR (Supplementary Fig. 7A–B). our data indicate that immune cell stimuli cause changes We normalized our qPCR data to the geometric mean of in the EV-associated levels of specific miRNAs, snoRNAs these four genes to minimize errors caused by technical and Y-RNAs. and/or biological variation . Using this normalization So far, we compared the abundance of specific RNA strategy, RT-qPCR analysis confirmed the sequencing data classes in EV by assessing the total numbers of reads for the tested miRNAs, although miR146-5p did not reach mapping to non-coding RNA genes. However, also the 1 3 T. A. P. Driedonks et al. a miRNA b VitD3-enriched LPS-enriched snoRNA VitD3-enriched LPS-enriched miscRNA de VitD3-enriched LPS-enriched fragmentation profile of EV-associated RNAs may be fluids [30–32, 34, 35, 77, 78, 81]. Especially, Y-RNA functionally important and indicative for the status of the fragments (19–35 nt) have gained attention, because they EV-producing cell. Fragmented forms of RNA have been were found to associate with Argonaute, suggesting scope described in EV from a wide range of cells and biological for gene-regulatory functions [82, 83], and because their 1 3 Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… ◂Fig. 5 EV from LPS-/VitD3-stimulated DC display differences in in the RNA pool of the parental cells and vice versa. the levels of several miRNAs, snoRNAs, and Y-RNAs. EV-asso- After normalization within each RNA class, fold-changes ciated RNA from control, LPS, and VitD3 conditions was isolated in EV were plotted against the fold changes in cells for and analyzed by RNA sequencing. Read counts for individual RNAs miRNA, miscRNA, and snoRNA (Fig. 7a), with dashed were normalized to the total read counts of each RNA class. LPS- or VitD3-induced fold changes and corresponding p values were calcu- lines indicating log2FC of 1 or − 1. Dot size indicates the lated relative to the control condition. a Volcano plots of EV-associ- average (n = 3) abundance of a transcript in EV, and dot ated miRNAs in LPS versus VitD3 conditions. Thresholds for twofold coloring is used to indicate RNAs that were significantly change and non-adjusted p < 0.05 are indicated. Data points represent changed in cells, EV, or both. First, we tested whether average values of n = 3 biological replicates. Significant changes are indicated with different colours. Grey: non-significant, blue: non- changes in EV-associated RNAs reflected stimulation- adjusted p value < 0.05, and red: FDR < 0.05. b RT-qPCR validation induced changes in cells. RNAs of which the levels in EV of six miRNAs showing differential abundance between LPS-EV significantly changed (non-adjusted p values < 0.05) upon and VitD3-EV. Data are expressed as log2fold change in LPS- and LPS or VitD3-treatment were categorized as ‘reflective VitD3-EV compared to control-EV. Indicated are the mean ± SD val- ues of n = 4 independent experiments, one-way ANOVA, *p < 0.05. RNAs’ (p < 0.05 in both cells and EV), ‘EV-responsive Volcano plots of EV-associated snoRNAs (c) and miscRNAs (d) in RNAs’ (p < 0.05 in EV, p > 0.05 in cells), or ‘reciprocal LPS versus VitD3 conditions. Thresholds for twofold change and RNAs’ (opposite changes in EV and cells with p < 0.05) non-adjusted p < 0.05 are indicated. Data points represent average (Supplementary Table 2). Of all RNAs that significantly values of n = 3 biological replicates. Significant changes are indi- cated in different colours. Grey: non-significant, blue: non-adjusted changed in EV (set to 100%), a minor percentage of the p value < 0.05, red: FDR < 0.05. e RT-qPCR validation of Y3 and its miRNAs and snoRNAs reflected the up/downregulation family member Y1 showing differential abundance between LPS-EV in cells or were reciprocally regulated, while most of and VitD3-EV compared to control-EV. Indicated are the mean ± SD the RNAs changed only in EV, but not (significantly) in values of n = 4 independent experiments, one-way ANOVA, *p < 0.05 cells (Fig. 7b–d). Conversely, we aimed to assess which proportion of LPS- or VitD3-induced changes in cellu- presence in plasma has been linked to disease [78, 84]. We lar RNA was reflected in the EV-RNA content. RNAs examined our data to determine whether reads mapping that changed in cells upon cell stimulation (non-adjusted to Y-RNAs included these fragments in the purified DC- p values < 0.05) were, therefore, classified as ‘reflec- derived EV. Based on RNAseq data, 5′ and 3′ fragments of tive’ (p < 0.05 in both cells and EV), ‘cell-responsive’ Y1-RNA were present in EV and such fragments seemed (p < 0.05 in cells, p > 0.05 in EV), ‘reciprocal’ (opposite more abundant than the full-length form of this Y-RNA changes in EV and cells with p < 0.05), or ‘not in EV’ (Fig. 6a). To confirm the full-length or fragmented nature of (p < 0.05 in cells, but not detected in EV). Comparable the Y-RNA molecules, we performed Northern blot analysis. to what we observed in the above analysis of EV-RNA, Using probes recognizing the 5′ and 3′ end of Y1-RNA, we only a minor percentage of changes in cellular miRNAs found that both 5′ -and 3′ fragments and full-length Y-RNA and snoRNAs was reflected in EV. The majority of the could be detected in both cellular and EV-RNA (Fig. 6b–d). RNAs that changed in cells did not significantly change Strikingly, the relative amount of full-length Y1-RNA in EV (Fig. 7e–g), and 10–20% of the RNAs that changed detected by Northern blot was much higher than expected in cells were not detected in EV. We selected three reflec- based on the sequencing coverage plots. This indicates that tive and three EV-responsive RNAs (Fig. 7h) for valida- RNAseq-based detection of the full-length form of Y1-RNA tion by RT-qPCR. For the reflective miRNAs miR-155, was highly inefficient. This is likely due to the 5′ -triphos- -9, and -10a, we confirmed that the stimulation-induced phate group of full-length Y-RNAs or their strong secondary fold-change in EV-associated and cellular RNA was highly structure, which may hamper sequencing adaptor ligation similar. For the EV-responsive miRNA miR-146a and the and thereby efficient amplification and sequencing of these two Y-RNAs, significant cell stimulation-induced changes RNAs. Interestingly, Northern blot analysis also indicated were observed in EV, while cellular levels did not change that the levels of full-length Y1-RNA in LPS-EV were significantly compared to the control conditions. Impor - reduced, which was in accordance to the RT-qPCR analysis tantly, these data indicate that part of the cell stimulation- for full-length Y1 (Fig. 5e), while the levels of fragmented induced changes in EV-RNA content match the changes Y1-RNA in EV remained relatively stable (Fig. 6d). These observed in cellular RNA, but that cell stimulation also data urge caution in classifying fragmented and full-length induces changes in RNA levels that are only observed in forms of Y-RNA based on RNA sequencing data. cells or in EV. Changes in the RNA content of EV have been suggested Overall, the results obtained for our primary DC cul- to directly reflect changes in the RNA levels of the paren- tures show that the molecular messages enclosed in EV tal cell. To test this hypothesis, we investigated whether are composed of multiple RNA classes that are specifically all of the stimulation-induced changes in miRNAs, snoR- shuttled into EV dependent on the activation status of the NAs, and miscRNAs observed in EV reflected changes 1 3 T. A. P. Driedonks et al. Table 1 MicroRNAs enriched in LPS- or VitD3-EV with known functions in DC Identifier Enriched in FDR Function in DC References mmu-miR-155-5p LPS-EV 4.04E−12 Master regulator in DC maturation  mmu-miR-708-3p VitD3-EV 2.21E−09 Downregulated in mature/activated DC  mmu-miR-10a-5p VitD3-EV 2.31E−07 Inhibits DC activation and Th1/Th17 cell immune responses  mmu-miR-146a-5p LPS-EV 4.53E−06 Down regulates IL-12p70, IL-6, and TNF-α production by DC  mmu-miR-9-5p LPS-EV 1.86E−05 Regulatory circuitry controlling monocyte activation by LPS  mmu-miR-223-5p VitD3-EV 6.82E−05 Repression of pro-inflammatory cytokine release by DC  mmu-miR-378a-3p VitD3-EV 8.24E−05 Upregulated in VitD3-treated DC  mmu-miR-203-3p VitD3-EV 0.000328 Upregulated in tolerogenic DC  mmu-miR-199a-3p VitD3-EV 0.000484 Upregulated in tolerogenic DC  mmu-miR-27b-5p VitD3-EV 0.000578 Suppression of inflammatory cytokine production via NF-κB  mmu-miR-7a-5p LPS-EV 0.000703 Upregulated in LPS/IFNg stimulated DC  mmu-miR-126a-3p VitD3-EV 0.000946 Reduces the responsiveness of DCs to TLR7/9 ligands  mmu-miR-708-5p VitD3-EV 0.000946 Suppresses NF-κB signaling  mmu-miR-181b-3p VitD3-EV 0.001923 Inhibition of CD40 and MHCII expression  mmu-miR-27a-5p VitD3-EV 0.002511 Suppression of inflammatory cytokine production  EV-associated RNA from control, LPS, and VitD3 conditions was isolated and analyzed by RNA sequencing. Read counts for individual RNAs were normalized to the total read counts of each RNA class. LPS- or VitD3-induced fold-changes and corresponding p values were calculated relative to the control condition with edgeR GLM method. We created a top 20 list of miRNAs with the lowest FDR values. Indicated are miR- NAs from this top 20 for which DC-related functions have been reported 5’fragment 3’fragment 3’ Y1-RNA 5’ b cd Y1_5p Y1_3p Y1-5p cell EV cell EV CL V 150 150 FL FL FL 100 100 90 90 40 40 5p 5p 3p Fig. 6 Full-length versus fragmented forms of Y1-RNA in cells and was loaded per lane. c Northern blot detection of full-length (FL) and EV validated by Northern blot. a Representative coverage plot of 3′fragments (3p) of Y1-RNA using Y1-3p probe. 10 ng of small RNA Y1-RNA as observed in EV-RNA seq data (sequencing coverage was loaded per lane. d Northern blot detection of full-length and depth 3981) visualized in the UCSC genome browser. b RNA iso- 5′fragments of Y1-RNA in EV-RNA from differently stimulated DC lated from control, LPS, and VitD3-treated cells and their EV were (C = control, L = LPS, V = VitD3). 10 ng of small RNA was loaded analyzed by Northern blot for the detection of full-length (FL) and per lane. Data are representative for n = 2 independent experiments 5′fragments (5p) of Y1-RNA using Y1-5p probe. 10 ng of small RNA 1 3 Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… cell. Moreover, the stimulus-induced changes in the small larger and highly modified/structured RNAs, for which the RNA content of cells and EV only partly overlap. sequencing efficiency may be compromised, are enriched in EV compared to cellular RNA , which may have caused the percentages of alignment to be lower in EV Discussion compared to cells. For our studies, we selected a com- mercial adapter-ligation-based method that has been fre- The data presented here broaden our view on the plasticity quently used in EV-RNA sequencing studies [23, 24, 31, of the small RNA content of EV in relation to changes in 32, 61]. While optimized for miRNA analysis, it should be the activation status of the EV-producing cell. Whereas noted that these library preparation methods show bias in most of the previous studies focused on the miRNA con- the efficiency with which sequences with base modifica- tent of EV, we show that cell status-dependent changes in tions, terminal triphosphate groups, and strong secondary the RNA composition of EV also extend to other small structures are ligated and sequenced . As an example, RNA classes, which may, therefore, also contribute to the we here provide Northern blot-based evidence that the specific genetic messages conveyed by EV to recipient frequently reported predominance of Y-RNA fragments cells. The immune cell model which we employed, i.e., DC over full-length Y-RNA in EV [30, 32, 35, 90] represents that are well known for their strong and diverse functional a sequencing artefact. We observed a similar discrepancy responsiveness to external stimuli, served two purposes. between sequencing read coverage and Northern blot anal- First, the cell system illustrates that diverse stimuli can ysis for tRNAs (Supplementary Fig. 8). While the triphos- induce the release of EV with variable levels of multiple phate group on the 5′-end  or their strong secondary RNA types, and that only some of the stimulation-induced structure may hamper efficient sequencing of full-length changes in EV-associated RNAs reflect changes in the Y-RNA, difficulties with sequencing of full-length tRNAs cellular RNA. Second, the data extend our knowledge on are likely due to the inability of the reverse transcriptase the central role of DC in raising and regulating immune enzyme used during sequencing library preparation to responses. The identification of multiple EV-RNA types read through the highly modified tRNA structure . that are specifically associated with either the immune- This is corroborated by recent sequencing studies deploy- activating or immune-suppressing status of DC is the prel- ing reverse transcriptases that are insensitive to second- ude to unraveling the function of these RNAs in EV. ary RNA structures [93, 94]. Together, these findings urge So far, RNA sequencing studies in which multiple caution in the interpretation of RNA fragmentation based RNA classes were identified in EV employed EV isolation on RNA sequencing data alone. Moreover, we reduced the methods known to co-isolate extracellular protein–RNA effect of sequencing biases on the assessment of quantita- complexes (RNP) that overlap in size with EV, e.g., ultra- tive differences in EV-RNA content by determining the centrifugation and precipitation-based methods [4, 24, fold changes in identical transcripts between different 30, 32, 34, 80, 85, 86]. The results of our present study conditions. indicate that up to 55% of total extracellular RNA pre- Our experimental approach allowed comprehensive sent in ultracentrifugation pellets of cell culture superna- analysis of changes in EV-RNA classes induced by diverse tant can be associated with RNPs. Buoyant density-based immune-relevant stimuli. The most prominent stimulation- separation of EV from RNP, as employed in this study, induced changes in EV-associated RNA levels were found allowed us to specifically identify RNA species that cells for miRNA, Y-RNA, and snoRNAs (Fig. 3). Although differentially sort into EV upon exogenous stimulation. detected in EV from multiple sources [23, 31, 33–35], the In RNA sequencing libraries prepared from these highly presence of snoRNAs in EV has been largely understud- pure EV, we observed that the percentage of alignment to ied. Here, we provide evidence that the levels of EV-associ- the mouse genome was lower than in cellular RNA librar- ated snoRNAs can be regulated by stimuli imposed on the ies. This is likely caused by the (ultra)low-input quantity EV-producing cell. Since changes in the cellular levels of of RNA in our EV-sequencing libraries, which amplifies snoRNA have been associated with diseases [36, 95, 96], our the contribution of lab-derived contaminant RNAs, e.g., present data point to EV-associated snoRNAs as potential from commercial nucleic acid extraction kits or sample functional entities in EV and interesting candidate biomark- cross-contamination [87, 88]. The percentage of align- ers indicative of the cellular activation status. ment may also be dependent on the size range of RNAs Numerous studies already demonstrated that miRNA selected for sequencing. Whereas, in most published EV- levels in EV change upon disease induction or cell stim- RNA sequencing studies, RNAs in the 20–40 nt size range ulation [22–24]. Unique aspects of our present RNAseq were sequenced, thereby strongly selecting for microR- study are the use of two different stimuli for comparative NAs, we here sequenced RNAs of a much larger size range analysis of RNA levels in highly purified EV and the par - (20–300 nt). We previously showed that several of these allel assessment of cellular and EV-associated RNA. The 1 3 T. A. P. Driedonks et al. mm iRNA iscRNA snoRNA legend log2FC in EV log2FC in EV log2FC in EV mm iRNA iscRNA snoRNA log2FC in EV log2FC in EV log2FC in EV bc d LPS VitD3 e fg LPS VitD3 reflective EV-responsive 1 3 VitD3 LPS log2FC in cells log2FC in cells log2FC in cells log2FC in cells log2FC in cells log2FC in cells Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… ◂Fig. 7 LPS-/VitD3-induced changes in cellular versus EV-associated to immune-activating or immune-suppressive functions of RNA levels. Cellular and EV-associated RNA from control, LPS, differentially stimulated DC. and VitD3 conditions was isolated and analyzed by RNA sequenc- Not only immune-suppressive miRNAs but also two types ing. Read counts for individual RNAs were normalized to the total of Y-RNA were found to be specifically excluded from EV read counts of each RNA class. LPS- or VitD3-induced fold changes and corresponding p values were calculated relative to the control released by LPS-stimulated DC (Fig. 5e). This raises the condition for cellular and EV-associated RNA. a Fold changes in question whether Y-RNAs contribute to immune-related EV-RNAs were plotted against the fold changes in cellular RNA in functions of EV. Until now, different types of Y-RNA (mY1 response to LPS (top graphs) and VitD3 (lower graphs) for miRNA and mY3 in mice and hY1, hY3, hY4, and hY5 in humans) (left), miscRNA (middle), and snoRNA (right). Dashed lines indicate log2FC larger or smaller than 1, so all data points beyond these lines have been detected in EV from a multitude of different cell are differentially expressed with log2FC > 1. Colored dots indicate types and in different body fluids [30– 32, 34, 35, 77–81]. transcripts that changed with non-adjusted p values < 0.05 in cells The extracellular presence of Y4- and Y5-RNA and frag- (blue), in EV (red), or in both cells and EV (purple), grey dots indi- ments thereof has mostly been associated to cancer [32, 81, cate unchanged transcripts. Dot size represents the normalized abun- dance (logCPM) of individual RNAs, mean of n = 3 experiments. b–d 84, 102] and coronary artery disease [78, 103]. Immune- Transcripts showing significant changes in stimulated versus control related functions of Y-RNAs include their capacity to trig- EV (p value < 0.05) were selected, and fold changes in EV versus cor- ger TLR signaling. Interestingly, Y-RNAs may vary in their responding cells were compared. Indicated are percentages of tran- specificity for different TLRs, with Y3-RNA triggering scripts categorized as ‘reflective’ (p value < 0.05 in EV and cells), ‘EV-responsive’ (p value < 0.05 in EV but not in cells), and ‘recipro- predominantly TLR3, while Y1-, Y3-, and Y4-RNA trig- cal’ (p value < 0.05 in cells and EV, but with opposite fold changes). ger TLR7 . On the contrary, the reduction of Y1- and e–g Analogous to b–d, transcripts showing significant changes in Y3-RNA levels in EV released by immune-stimulatory DC, cells (p value < 0.05) were selected, and fold changes in cells were shown in this study, suggests an immune downregulatory compared with those in EV. Indicated are percentages of transcripts categorized as ‘reflective’, ‘reciprocal’, cell-responsive (p value < 0.05 role for these EV-associated Y-RNAs. In line with this, high in cells but not in EV), or ‘not in EV’ (transcripts only found in cells). levels of Y-RNAs have been reported in immune-suppressive h RT-qPCR validation of six genes that were found to be reflective EV released by the parasite Heligmosomoides polygyrus and (left panels) or EV-responsive (right panels) on EV-associated RNA in seminal fluid EV for which immunosuppressive effects and cellular RNA from control, LPS- or VitD3-treated DC. Fold changes in EV and in cells were calculated relative to the control con- have been described [35, 105–107]. Overall, the observed dition in cells or EV. N = 4, one-way ANOVA, *p < 0.05 variable presence of Y-RNA in EV from immune cells has raised further interest in unraveling the immune-related enrichment of endotoxin-responsive miRNAs, such as miR- function and biomarker potential of these RNAs in EV. 155 and miR-146a [20, 97, 98], that we observed in EV With regard to the biomarker potential of EV-associ- from LPS-stimulated DC concurs with previously published ated RNAs, our data also urge caution in interpreting EV semi-quantitative microarray-based data on the miRNA as snapshots of the cell from which they arise. We found content of EV from stimulated DC [22, 26]. Both miR-155 that only a subset of changes in EV-RNA reflected changes and miR-146a are known to be endotoxin-responsive, but that occurred at the cellular level. The highest proportion of miR-155 has an immune-activating role, while miR-146a is reflective RNAs was observed within the category of miR - involved in dampening of immune responses and could act NAs and included prominent immune-related RNAs such as as a molecular brake on inflammation . It is thought that miR-155, miR-9, and miR-10-5p (Fig. 7h). For a substantial the coordinated action of these two is important in regula- number of EV-associated miRNAs and both Y-RNA types, tion of immune response . Differences in the miR-155/ however, stimulation-induced changes were observed in EV, miR-146a ratios that we observed in LPS-EV and VitD3 EV but not in cells. This could suggest that EV-levels of reflec- (Fig. 5b) may, therefore, lead to different effects of these EV tive RNAs are mainly regulated by transcription, whereas on immune activation. In addition, we here demonstrated levels of EV-responsive RNAs are mainly determined by that a different (i.e., immune downregulatory) type of stimu- shuttling rate. For RNAs known to primarily reside in the lus imposed on DC led to both a reduction of EV-associated nucleus (such as snRNAs and snoRNAs), subtle changes levels of these immune-activating miRNAs and an enrich- in the cytoplasmic pool of these RNAs may be overshad- ment in miRNAs implicated in the suppression of immune owed by the much larger nuclear pool. Comparison of EV- responses via modulation of multiple pathways. Examples associated levels with cytoplasmic instead of total cellular include dampening of TLR signaling by miR-27a and miR- RNA levels may, therefore, provide a better insight in shut- 126a [70, 73], and downregulation of (regulators of) PI3K/ tling of these nuclear RNAs into EV. Interestingly, we also Akt and NF-κB by miR-378 , miR-708  and miR- observed that many of the stimulation-induced changes in 27b-3p , processes that have been implicated in the cellular RNA levels were not reflected in the RNA levels of suppression of pro-inflammatory cytokine release. Such dif- EV released by these cells. This strengthens the hypothesis ferences in the EV-miRNA content may therefore contribute that cells release EV with selective sets of RNAs into the extracellular milieu. From a biomarker perspective, these 1 3 T. A. P. Driedonks et al. Framework Programme [FP/2007-2013]/ERC Grant Agreement num- data implicate that, although the presence of specific EV- ber  to [ENMNtH]; COST Action BM1202 Microvesicles and RNAs may correlate with disease, the EV transcriptome is Exosomes in Health and Disease Short-Term Scientific Mission Grant not necessarily predictive for RNA levels in the EV-pro- [191116-083388] to [TAPD]; and the Human Frontiers Science Pro- ducing cell. gram [RGY0069] to [AHB]. We observed two classes of RNA, i.e., snRNA and tRNA, of which the EV-associated levels remained stable Compliance with ethical standards after differential stimulation of DC. These RNA classes Conflict of interest The authors declare that they have no conflict of have been commonly detected in small RNAseq studies of interest. EV from multiple cell types and body fluids, but factors impacting their incorporation into EV had until now not Open Access This article is distributed under the terms of the Crea- been investigated. Although these data need confirmation tive Commons Attribution 4.0 International License (http://creat iveco in different cell types with various external stimuli, our mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- tion, and reproduction in any medium, provided you give appropriate data may be a first indication that tRNAs and snRNAs credit to the original author(s) and the source, provide a link to the are constitutive components of EV. Speculatively, these Creative Commons license, and indicate if changes were made. RNAs may be required for basal scaffolding functions in EV biogenesis, or may aid the functional transfer of gene-regulatory RNAs. Based on their stable association with EV, these types of RNAs also have potential to be References used as EV-reference RNAs for RT-qPCR normalization, since comparing the EV transcriptome of differentially 1. Colombo M, Raposo G, Théry C (2014) Biogenesis, secretion, stimulated cells is complicated by potential differences and intercellular interactions of exosomes and other extracel- lular vesicles. Annu Rev Cell Dev Biol 30:255–289. https ://doi. in the number of EV, RNA yield, and RNA composition. org/10.1146/annur ev-cellb io-10151 2-12232 6 Importantly, the use of well-known stably expressed cel- 2. van der Grein SG, Nolte-’t Hoen ENM (2014) “Small Talk” in lular RNAs is not necessarily valid for normalization of the innate immune system via RNA-containing extracellular EV qPCR data due to potential differences in RNA shut- vesicles. Front Immunol 5:1–8. https ://doi.org/10.3389/fimmu .2014.00542 tling between tested conditions. Therefore, normalization 3. Théry C, Ostrowski M, Segura E (2009) Membrane vesicles as of RT-qPCR data to reference genes which are stable in conveyors of immune responses. Nat Rev Immunol 9:581–593. EV is a reliable way to quantitate differences in individual https ://doi.org/10.1038/nri25 67 RNAs. Here, we show how RNA sequencing and RT-qPCR 4. Mateescu B, Kowal EJK, van Balkom BWM et al (2017) Obsta- cles and opportunities in the functional analysis of extracellu- validation can be used to identify candidate EV-reference lar vesicle RNA—an ISEV position paper. J Extracell Vesicles genes which are constitutively expressed between differ - 6:1286095. https ://doi.org/10.1080/20013 078.2017.12860 95 ently stimulated cells. 5. Van Deun J, Mestdagh P, Agostinis P et al (2017) EV-TRACK: In conclusion, this study indicates that external stim- transparent reporting and centralizing knowledge in extracel- lular vesicle research. Nat Methods 14:228–232. https ://doi. uli imposed on cells can lead to changes in the levels of org/10.1038/nmeth .4185 some EV-associated RNA classes, while leaving others 6. Tkach M, Kowal J, The C (2017) Why the need and how to unchanged. In our DC-model, two different immune- approach the functional diversity of extracellular vesicles. Philos related cell stimuli led to disparate changes in the type Trans R Soc B Biol Sci. https ://doi.org/10.1098/rstb.2016.0479 7. Camussi G, Deregibus MC, Bruno S et al (2010) Exosomes/ and quantity of miRNAs, snoRNAs, and Y-RNAs in EV. microvesicles as a mechanism of cell-to-cell communication. Only part of the changes in RNA contents of EV reflected Kidney Int 78:838–848. https ://doi.org/10.1038/ki.2010.278 changes in the cellular RNA, which urges caution in inter- 8. Robbins PD, Morelli AE (2014) Regulation of immune responses preting EV as snapshots of cells. These findings show that by extracellular vesicles. Nat Rev Immunol 14:195–208. https:// doi.org/10.1038/nri36 22 differences in multiple structurally distinct RNA classes 9. Redzic JS, Balaj L, van der Vos KE, Breakefield XO (2014) shape the EV transcriptome. These RNA types may con- Extracellular RNA mediates and marks cancer progression. tribute to the functional properties of EV in cellular com- Semin Cancer Biol 28:14–23. https ://doi.org/10.1016/j.semca munication, and may be further explored as candidate ncer.2014.04.010 10. Revenfeld ALS, Bæk R, Nielsen MH et al (2014) Diagnostic EV-RNA-based indicators for the immune status of the and prognostic potential of extracellular vesicles in peripheral EV-producing cell. blood. Clin Ther 36:830–846. https ://doi.or g/10.1016/j.clint hera.2014.05.008 Acknowledgements The authors would like to thank Dr. G. J. A. 11. Kahlert C, Melo SA, Protopopov A et al (2014) Identification Arkesteijn for the assistance with high-resolution flow cytometric of doublestranded genomic dna spanning all chromosomes with analysis of EV, Dr. Z. Zeev Ben Mordehai for the assistance with mutated KRAS and P53 DNA in the serum exosomes of patients transmission electron microscopy, and Drs. S. L. Wolin, S. Sim, and with pancreatic cancer. J Biol Chem 289:3869–3875. https://doi. D. Staněk for the valuable discussions. This work was supported by org/10.1074/jbc.C113.53226 7 European Research Council under the European Union’s Seventh 1 3 Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… 12. Raab-Traub N, Dittmer DP (2017) Viral effects on the content 30. Nolte-’t Hoen ENM, Buermans HPJ, Waasdorp M et al (2012) and function of extracellular vesicles. Nat Rev Microbiol 15:559– Deep sequencing of RNA from immune cell-derived vesicles 572. https ://doi.org/10.1038/nrmic ro.2017.60 uncovers the selective incorporation of small non-coding RNA 13. Arroyo JD, Chevillet JR, Kroh EM et al (2011) Argonaute2 biotypes with potential regulatory functions. Nucleic Acids Res complexes carry a population of circulating microRNAs inde- 40:9272–9285. https ://doi.org/10.1093/nar/gks65 8 pendent of vesicles in human plasma. Proc Natl Acad Sci USA 31. van Balkom BWM, Eisele AS, Pegtel DM et al (2015) Quantita- 108:5003–5008. https ://doi.org/10.1073/pnas.10190 55108 tive and qualitative analysis of small RNAs in human endothelial 14. Turchinovich A, Weiz L, Langheinz A, Burwinkel B (2011) cells and exosomes provides insights into localized RNA pro- Characterization of extracellular circulating microRNA. Nucleic cessing, degradation and sorting. J Extracell Vesicles 4:26760. Acids Res 39:7223–7233. https ://doi.org/10.1093/nar/gkr25 4https ://doi.org/10.3402/jev.v4.26760 15. Vickers KC, Palmisano BT, Shoucri BM et al (2011) MicroR- 32. Tosar JP, Gambaro F, Sanguinetti J et al (2015) Assessment of NAs are transported in plasma and delivered to recipient cells by small RNA sorting into different extracellular fractions revealed high-density lipoproteins. Nat Cell Biol 13:423–433. https://doi. by high-throughput sequencing of breast cell lines. Nucleic Acids org/10.1038/ncb22 10 Res. https ://doi.org/10.1093/nar/gkv43 2 16. Van Deun J, Mestdagh P, Sormunen R et al (2014) The impact 33. Yeri A, Courtright A, Reiman R et al (2017) Total extracellular of disparate isolation methods for extracellular vesicles on small RNA profiles from plasma, saliva, and urine of healthy downstream RNA profiling. J Extracell Vesicles 3:1–14. https subjects. Sci Rep 7:44061. https ://doi.org/10.1038/srep4 4061 ://doi.org/10.3402/jev.v3.24858 34. Lunavat TR, Cheng L, Kim D-K et al (2015) Small RNA deep 17. Lötvall J (2014) Minimal experimental requirements for defi- sequencing discriminates subsets of extracellular vesicles nition of extracellular vesicles and their functions: a position released by melanoma cells—evidence of unique microRNA statement from the. Int Soc Extracell Vesicles. 1:1–6 cargos. RNA Biol. https: //doi.org/10.1080/154762 86.2015.10569 18. Vergauwen G, Dhondt B, Van Deun J et al (2017) Confounding 75 factors of ultrafiltration and protein analysis in extracellular 35. Vojtech L, Woo S, Hughes S et al (2014) Exosomes in human vesicle research. Sci Rep 7:2704. https://doi.or g/10.1038/s4159 semen carry a distinctive repertoire of small non-coding RNAs 8-017-02599 -y with potential regulatory functions. Nucleic Acids Res 42:7290– 19. O’Connell RM, Taganov KD, Boldin MP et al (2007) Micro- 7304. https ://doi.org/10.1093/nar/gku34 7 RNA-155 is induced during the macrophage inflammatory 36. Youssef OA, Safran SA, Nakamura T et al (2015) Potential role response. Proc Natl Acad Sci USA 104:1604–1609. https :// for snoRNAs in PKR activation during metabolic stress. Proc doi.org/10.1073/pnas.06107 31104 Natl Acad Sci USA 112:5023–5028. https ://doi.org/10.1073/ 20. Smyth LA, Boardman DA, Tung SL et al (2015) MicroRNAs pnas.14240 44112 affect dendritic cell function and phenotype. Immunology 37. Michel CI, Holley CL, Scruggs BS et al (2011) Small nucleo- 144:197–205. https ://doi.org/10.1111/imm.12390 lar RNAs U32a, U33 and U35a are critical mediators of meta- 21. Mehta A, Baltimore D (2016) MicroRNAs as regulatory ele- bolic stress. Cell Metab 14:33–44. https ://doi.or g/10.1016/j. ments in immune system logic. Nat Rev Immunol 16:279–294. cmet.2011.04.009.Small https ://doi.org/10.1038/nri.2016.40 38. Stepanov GA, Filippova JA, Komissarov AB et al (2015) Regula- 22. Montecalvo A, Larregina AT, Shufesky WJ et al (2012) Mech- tory role of small nucleolar RNAs in human diseases. Biomed anism of transfer of functional microRNAs between mouse Res Int 2015:206849. https ://doi.org/10.1155/2015/20684 9 dendritic cells via exosomes. Blood 119:756–766. https ://doi. 39. Haussecker D, Huang Y, Lau A et al (2010) Human tRNA- org/10.1182/blood -2011-02-33800 4 derived small RNAs in the global regulation of RNA silencing. 23. Bellingham SA, Coleman BM, Hill AF (2012) Small RNA RNA 16:673–695. https ://doi.org/10.1261/rna.20008 10.also deep sequencing reveals a distinct miRNA signature released 40. Ninomiya S, Kawano M, Abe T et al (2015) Potential small guide in exosomes from prion-infected neuronal cells. Nucleic Acids RNAs for tRNase ZL from human plasma, peripheral blood mon- Res 40:10937–10949. https ://doi.org/10.1093/nar/gks83 2 onuclear cells, and cultured cell lines. PLoS One 10:e0118631. 24. Cha DJ, Franklin JL, Dou Y et al (2015) KRAS-dependent https ://doi.org/10.1371/journ al.pone.01186 31 sorting of miRNA to exosomes. Elife 4:1–22. https ://doi. 41. Nabet BY, Qiu Y, Shabason JE et al (2017) Exosome RNA org/10.7554/eLife .07197 unshielding couples stromal activation to pattern recognition 25. Valadi H, Ekström K, Bossios A et al (2007) Exosome-medi- receptor signaling in cancer. Cell 170(352–366):e13. https://doi. ated transfer of mRNAs and microRNAs is a novel mechanism org/10.1016/j.cell.2017.06.031 of genetic exchange between cells. Nat Cell Biol 9:654–659. 42. Tkach M, Thery C (2016) Communication by extracellu- https ://doi.org/10.1038/ncb15 96 lar vesicles: where we are and where to go. Cell. https ://doi. 26. Alexander M, Hu R, Runtsch MC et al (2015) Exosome- org/10.1016/j.cell.2016.01.043 delivered microRNAs modulate the inflammatory response 43. Zitvogel L, Regnault A, Lozier A et al (1998) Eradication of to endotoxin. Nat Commun 6:7321. https ://doi.org/10.1038/ established murine tumors using a novel cell-free vaccine: den- ncomm s8321 dritic cell-derived exosomes. Nat Med 4:594–600. https ://doi. 27. Taylor DD, Gercel-Taylor C (2008) MicroRNA signatures of org/10.1038/nm059 8-594 tumor-derived exosomes as diagnostic biomarkers of ovarian 44. Pitt JM, André F, Amigorena S et al (2016) Dendritic cell-derived cancer. Gynecol Oncol 110:13–21. h ttps ://doi.or g/10.1016/j. exosomes for cancer therapy. J Clin Investig 126:1224–1232. ygyno .2008.04.033https ://doi.org/10.1172/JCI81 137 28. Murata K, Yoshitomi H, Tanida S et al (2010) Plasma and syno- 45. Wu T, Qi Y, Zhang D et al (2017) Bone marrow dendritic cells vial fluid microRNAs as potential biomarkers of rheumatoid derived microvesicles for combinational immunochemotherapy arthritis and osteoarthritis. Arthritis Res Ther 12:R86. https :// against tumor. Adv Funct Mater 27:1–15. https : //doi.or g/10.1002/ doi.org/10.1186/ar301 3adfm.20170 3191 29. Cheng L, Doecke JD, Sharples RA et al (2014) Prognostic serum 46. Tao SC, Guo SC, Zhang CQ (2018) Modularized extracellular miRNA biomarkers associated with Alzheimer’s disease shows vesicles: the dawn of prospective personalized and precision medicine. Adv Sci. https ://doi.org/10.1002/advs.20170 0449 concordance with neuropsychological and neuroimaging assess- ment. Mol Psychiatry. https ://doi.org/10.1038/mp.2014.127 1 3 T. A. P. Driedonks et al. 47. Raker VK, Domogalla MP, Steinbrink K (2015) Tolerogenic den- 64. Wu W, He C, Liu C et al (2015) miR-10a inhibits dendritic cell dritic cells for regulatory T cell induction in man. Front Immunol activation and Th1/Th17 cell immune responses in IBD. Gut 6:1–11. https ://doi.org/10.3389/fimmu .2015.00569 64:1755–1764. https ://doi.org/10.1136/gutjn l-2014-30798 0 48. Lutz MB, Kukutsch N, Ogilvie AL et al (1999) An advanced 65. Park H, Huang X, Lu C et al (2015) MicroRNA-146a and micro- culture method for generating large quantities of highly pure RNA-146b regulate human dendritic cell apoptosis and cytokine dendritic cells from mouse bone marrow. J Immunol Methods production by targeting TRAF6 and IRAK1 proteins. J Biol 223:77–92. https ://doi.org/10.1016/S0022 -1759(98)00204 -X Chem 290:2831–2841. https://doi.or g/10.1074/jbc.M114.591420 49. van der Vlist EJ, Nolte-’t Hoen ENM, Stoorvogel W et al (2012) 66. Bazzoni F, Rossato M, Fabbri M et al (2009) Induction and regu- Fluorescent labeling of nano-sized vesicles released by cells and latory function of miR-9 in human monocytes and neutrophils subsequent quantitative and qualitative analysis by high-resolu- exposed to proinflammatory signals. Proc Natl Acad Sci USA tion flow cytometry. Nat Protoc 7:1311–1326 106:5282–5287. https ://doi.org/10.1073/pnas.08109 09106 50. Robinson MD, McCarthy DJ, Smyth GK (2009) edgeR: a bio- 67. Zhou H, Xiao J, Wu N et al (2015) MicroRNA-223 regulates conductor package for differential expression analysis of digital the differentiation and function of intestinal dendritic cells and gene expression data. Bioinformatics 26:139–140. https ://doi. macrophages by targeting C/EBP-beta. Cell Rep 13:1149–1160. org/10.1093/bioin forma tics/btp61 6https ://doi.org/10.1016/j.celre p.2015.09.073 51. Robinson JT, Thorvaldsdóttir H, Winckler W et al (2011) Inte- 68. Pedersen AW, Holmstrøm K, Jensen SS et al (2009) Phenotypic grative genomics viewer. Nat Biotechnol 29:24–26. https://doi. and functional markers for 1alpha,25-dihydroxyvitamin D(3)- org/10.1038/nbt01 11-24 modified regulatory dendritic cells. Clin Exp Immunol 157:48– 52. Nolte-’t Hoen ENM, van der Vlist EJ, Aalberts M et al (2012) 59. https ://doi.org/10.1111/j.1365-2249.2009.03961 .x Quantitative and qualitative flow cytometric analysis of nano- 69. Stumpfova Z, Hezova R, Meli AC et al (2014) MicroRNA profil- sized cell-derived membrane vesicles. Nanomedicine 8:712– ing of activated and tolerogenic human dendritic cells. Mediat 720. https ://doi.org/10.1016/j.nano.2011.09.006 Inflamm 2014:259689. https ://doi.org/10.1155/2014/25968 9 53. Groot Kormelink T, Arkesteijn GJA, Nauwelaers FA et al 70. Agudo J, Ruzo A, Tung N et al (2013) The miR-126-VEGFR2 (2016) Prerequisites for the analysis and sorting of extracel- axis controls the innate response to pathogen-associated nucleic lular vesicle subpopulations by high-resolution flow cytom- acids. Nat Immunol 15:54–62. https ://doi.org/10.1038/ni.2767 etry. Cytom Part A 89:135–147. https ://doi.or g/10.1002/ 71. Baer C, Oakes CC, Ruppert AS et al (2015) Epigenetic silencing cyto.a.22644 of miR-708 enhances NF-kB signaling in chronic lymphocytic 54. Libri V, Helwak A, Miesen P et al (2012) Murine cytomegalo- leukemia. Int J Cancer 137:1352–1361. https ://doi.org/10.1002/ virus encodes a miR-27 inhibitor disguised as a target. Proc Natl ijc.29491 Acad Sci USA 109:279–284. https://doi.or g/10.1073/pnas.11142 72. Lin J, Xia J, Tu CZ et al (2017) H9N2 avian influenza virus 04109 protein PB1 enhances the immune responses of bone marrow- 55. Pall GS, Hamilton AJ (2008) Improved northern blot method for derived dendritic cells by down-regulating miR375. Front Micro- enhanced detection of small RNA. Nat Protoc 3:1077–1084. https biol 8:1–10. https ://doi.org/10.3389/fmicb .2017.00287 ://doi.org/10.1038/nprot .2008.67 73. Lv Y, Ou-yang A, Fu L (2017) MicroRNA-27a negatively modu- 56. Piemonti L, Monti P, Sironi M et al (2000) Vitamin D3 affects lates the inflammatory response in lipopolysaccharide-stimulated differentiation, maturation, and function of human monocyte- microglia by targeting TLR4 and IRAK4. Cell Mol Neurobiol derived dendritic cells. J Immunol 164:4443–4451 37:195–210. https ://doi.org/10.1007/s1057 1-016-0361-4 57. Unger WWJ, Laban S, Kleijwegt FS et al (2009) Induction of 74. Vandesompele J, De Preter K, Pattyn F et al (2002) Accurate nor- Treg by monocyte-derived DC modulated by vitamin D3 or malization of real-time quantitative RT-PCR data by geometric dexamethasone: differential role for PD-L1. Eur J Immunol averaging of multiple internal control genes. Genome Biol. https 39:3147–3159. https ://doi.org/10.1002/eji.20083 9103://doi.org/10.1186/gb-2002-3-7-resea rch00 34 58. Penna G, Adorini L (2000) 1,25-dihydroxyvitamin D3 inhibits 75. Stein AJ, Fuchs G, Fu C et al (2005) Structural insights into RNA differentiation, maturation, activation, and survival of dendritic quality control: the Ro autoantigen binds misfolded RNAs via cells leading to impaired alloreactive T cell activation. J Immunol its central cavity. Cell 121:529–539. https ://doi.org/10.1016/j. 164:2405–2411. https ://doi.org/10.4049/jimmu nol.164.5.2405 cell.2005.03.009 59. Thery C, Boussac M, Veron P et al (2001) Proteomic analysis 76. Chen X, Smith JD, Shi H et al (2003) The Ro autoantigen binds of dendritic cell-derived exosomes: a secreted subcellular com- misfolded U2 small nuclear RNAs and assists mammalian cell partment distinct from apoptotic vesicles. J Immunol 166:7309– survival after UV irradiation. Curr Biol 13:2206–2211. https :// 7318. https ://doi.org/10.4049/jimmu nol.166.12.7309 doi.org/10.1016/j.cub.2003.11.028 60. ’t Hoen PAC, Friedländer MR, Almlöf J et al (2013) Reproduc- 77. Dhahbi JM, Spindler SR, Atamna H et al (2013) 5′-YRNA frag- ibility of high-throughput mRNA and small RNA sequencing ments derived by processing of transcripts from specific YRNA across laboratories. Nat Biotechnol 31:1015–1022. https ://doi. genes and pseudogenes are abundant in human serum and org/10.1038/nbt.2702 plasma. Physiol Genom 45:990–998. https ://doi.org/10.1152/ 61. Koppers-Lalic D, Hackenberg M, Bijnsdorp IV et al (2014) Non-physi olgen omics .00129 .2013 templated nucleotide additions distinguish the small RNA com- 78. Repetto E, Lichtenstein L, Hizir Z et al (2015) RNY-derived position in cells from exosomes. Cell Rep 8:1649–1658. https :// small RNAs as a signature of coronary artery disease. BMC Med doi.org/10.1016/j.celre p.2014.08.027 13:259. https ://doi.org/10.1186/s1291 6-015-0489-y 62. Dueck A, Eichner A, Sixt M, Meister G (2014) A miR-155-de- 79. Kaudewitz D, Skroblin P, Bender LH et al (2015) Association of pendent microRNA hierarchy in dendritic cell maturation and microRNAs and YRNAs with platelet function. Circ Res. https macrophage activation. FEBS Lett 588:632–640. https ://doi.://doi.org/10.1161/CIRCR ESAHA .114.30566 3 org/10.1016/j.febsl et.2014.01.009 80. Cambier L, de Couto G, Ibrahim A et al (2017) Y RNA fragment 63. Ma Z-X, Tan X, Shen Y et al (2015) MicroRNA expression pro- in extracellular vesicles confers cardioprotection via modulation file of mature dendritic cell in chronic rhinosinusitis. Inflamm of IL-10 expression and secretion. EMBO Mol Med. https://doi. Res 64:885–893. https ://doi.org/10.1007/s0001 1-015-0870-5org/10.15252 /emmm.20160 6924 1 3 Immune stimuli shape the small non-coding transcriptome of extracellular vesicles released… 81. Haderk F, Schulz R, Iskar M et al (2017) Tumor-derived 95. Gee HE, Buffa FM, Camps C et al (2011) The small-nucleolar exosomes modulate PD-L1 expression in monocytes. Sci Immu- RNAs commonly used for microRNA normalisation correlate nol 2:1–12. https ://doi.org/10.1126/sciim munol .aah55 09 with tumor pathology and prognosis. Br J Cancer 104:1168– 82. Meiri E, Levy A, Benjamin H et al (2010) Discovery of micro- 1177. https ://doi.org/10.1038/sj.bjc.66060 76 RNAs and other small RNAs in solid tumors. Nucleic Acids Res 96. Martens-Uzunova ES, Olvedy M, Jenster G (2013) Beyond 38:6234–6246. https ://doi.org/10.1093/nar/gkq37 6 microRNA—novel RNAs derived from small non-coding RNA 83. Thomson DW, Pillman KA, Anderson ML et al (2015) Assessing and their implication in cancer. Cancer Lett 340:201–211. https the gene regulatory properties of argonaute-bound small RNAs ://doi.org/10.1016/j.canle t.2012.11.058 of diverse genomic origin. Nucleic Acids Res 43:470–481. https 97. Baltimore D, Boldin MP, O’Connell RM et al (2008) MicroR- ://doi.org/10.1093/nar/gku12 42 NAs: new regulators of immune cell development and function. 84. Dhahbi JM, Spindler SR, Atamna H et al (2014) Deep sequenc- Nat Immunol 9:839–845. https ://doi.org/10.1038/ni.f.209 ing of serum small RNAs identifies patterns of 5′ tRNA half 98. O’Connell RM, Kahn D, Gibson WSJ et al (2010) Micro- and YRNA fragment expression associated with breast cancer. RNA-155 promotes autoimmune inflammation by enhancing Biomark Cancer 6:37–47. https ://doi.org/10.4137/BIC.S2076 inflammatory T cell development. Immunity 33:607–619. https 4.RECEI VED://doi.org/10.1016/j.immun i.2010.09.009 85. Li CCY, Eaton SA, Young PE et al (2013) Glioma microvesicles 99. Boldin MP, Taganov KD, Rao DS et al (2011) miR-146a is a carry selectively packaged coding and non-coding RNAs which significant brake on autoimmunity, myeloproliferation, and can- alter gene expression in recipient cells. RNA Biol 10:1333–1344. cer in mice. J Exp Med 208:1189–1201. https://doi.or g/10.1084/ https ://doi.org/10.4161/rna.25281 jem.20101 823 86. Fiskaa T, Knutsen E, Nikolaisen MA et al (2016) Distinct small 100. Ruckerl D, Jenkins SJ, Laqtom NN et al (2012) Induction of rna signatures in extracellular vesicles derived from breast cancer IL-4R alpha-dependent microRNAs identifies PI3K/Akt signal- cell lines. PLoS One 11:e0161824. https:/ /doi.org/10.1371/journ ing as essential for IL-4-driven murine macrophage proliferation al.pone.01618 24 in vivo. Blood 120:2307–2317. https ://doi.org/10.1182/blood 87. Tosar JP, Rovira C, Naya H, Cayota A (2014) Mining of public -2012-02-40825 2.The sequencing databases supports a non-dietary origin for putative 101. Lee J-J, Drakaki A, Illiopoulos D, Struhl K (2012) MiR-27b tar- foreign miRNAs: underestimated effects of contamination in gets PPARy to inhibit growth, tumor progression, and the inflam- NGS. RNA 20:754–757. https://doi.or g/10.1261/rna.04426 3.114 matory response in neuroblastoma cells. Oncogene 31:3818– 88. Tosar JP, Cayota A, Eitan E et al (2017) Ribonucleic artefacts: 3825. https ://doi.org/10.1146/annur ev-immun ol-03271 3-12024 are some extracellular RNA discoveries driven by cell culture 0.Micro glia medium components? J Extracell Vesicles 6:1–10. https ://doi. 102. Chakrabortty SK, Prakash A, Nechooshtan GAL et al (2015) org/10.1080/20013 078.2016.12728 32 Extracellular vesicle-mediated transfer of processed and func- 89. Raabe CA, Tang T-H, Brosius J, Rozhdestvensky TS (2014) tional RNY5 RNA. RNA. https ://doi.or g/10.1261/r na.05362 Biases in small RNA deep sequencing data. Nucleic Acids Res 9.115 42:1414–1426. https ://doi.org/10.1093/nar/gkt10 21 103. Hizir Z, Bottini S, Grandjean V et al (2017) RNY (YRNA)- 90. Wei Z, Batagov AO, Schinelli S et al (2017) Coding and noncod- derived small RNAs regulate cell death and inflammation in ing landscape of extracellular RNA released by human glioma monocytes/macrophages. Cell Death Dis 8:e2530. https ://doi. stem cells. Nat Commun 8:1145. https ://doi.org/10.1038/s4146 org/10.1038/cddis .2016.429 7-017-01196 -x 104. Greidinger EL, Zang Y, Martinez L et al (2007) Differential tis- 91. Hendrick JP, Wolin SL, Rinke J et al (1981) Ro small cytoplasmic sue targeting of autoimmunity manifestations by autoantigen- ribonucleoproteins are a subclass of La ribonucleoproteins: fur- associated Y RNAs. Arthritis Rheumatol 56:1589–1597. https ther characterization of the Ro and La small ribonucleoproteins ://doi.org/10.1002/art.22601 from uninfected mammalian cells. Mol Cell Biol 1:1138–1149. 105. Skibinski G, Kelly R, Harkiss D, James K (1992) Immunosup- https ://doi.org/10.1128/MCB.1.12.1138.Updat ed pression by human seminal plasma—extracellular organelles 92. Motorin Y, Muller S, Behn-Ansmant I, Branlant C (2007) Iden- (prostasomes) modulate activity of phagocytic cells. Am J tification of modified residues in RNAs by reverse transcrip- Reprod Immunol 28:97–103 tion-based methods. Methods enzymol 425:21–53. https ://doi. 106. Tarazona R, Delgado E, Guarnizo MC et al (2011) Human org/10.1016/S0076 -6879(07)25002 -5 prostasomes express CD48 and interfere with NK cell function. 93. Qin Y, Yao J, Wu DC et al (2015) High-throughput sequencing Immunobiology 216:41–46. https ://doi.or g/10.1016/j.imbio of human plasma RNA by using thermostable group II intron .2010.03.002 reverse transcriptases. RNA 22:1–18. https ://doi.org/10.1261/ 107. Buck AH, Coakley G, Simbari F et al (2014) Exosomes secreted rna.05480 9.115.3 by nematode parasites transfer small RNAs to mammalian cells 94. Shurtleff MJ, Yao J, Qin Y et al (2017) Broad role for YBX1 in and modulate innate immunity. Nat Commun 5:5488. https://doi. defining the small noncoding RNA composition of exosomes. org/10.1038/ncomm s6488 Proc Natl Acad Sci 114:E8987–E8995. https ://doi.org/10.1073/ pnas.17121 08114 1 3
Cellular and Molecular Life Sciences – Springer Journals
Published: May 28, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera