CBX766966 1..10 Report Strategies for enumeration of circulating microvesicles on a conventional flow cytometer: Counting beads and scatter parameters Mohammad J Alkhatatbeh1, Anoop K Enjeti2,3,4,5, Sara Baqar6,7, Elif I Ekinci6,7, Dorothy Liu6,7, Rick F Thorne3,4, and Lisa F Lincz2,3,4 Abstract Enumeration of circulating microvesicles (MVs) by conventional flow cytometry is accomplished by the addition of a known amount of counting beads and calculated from the formula: MV/ml ¼ (MV count/bead count) � final bead con- centration. We sought to optimize each variable in the equation by determining the best parameters for detecting ‘MV count’ and examining the effects of different bead preparations and concentrations on the final calculation. Three com- mercially available bead preparations (TruCount, Flow-Count and CountBright) were tested, and MV detection on a BD FACSCanto was optimized for gating by either forward scatter (FSC) or side scatter (SSC); the results were compared by calculating different subsets of MV on a series of 74 typical patient plasma samples. The relationship between the number of beads added to each test and the number of beads counted by flow cytometry remained linear over a wide range of bead concentrations (R2 � 0.997). However, TruCount beads produced the most consistent (concentration variation ¼ 3.8%) calculated numbers of plasma CD41 þ /Annexin V þ MV, which were significantly higher from that calculated using either Flow-Count or CountBright (p < 0.001). The FACSCanto was able to resolve 0.5 mm beads by FSC and 0.16 mm beads by SSC, but there were significantly more background events using SSC compared with FSC (3113 vs. 470; p ¼ 0.008). In general, sample analysis by SSC resulted in significantly higher numbers of MV (p < 0.0001) but was well correlated with enumeration by FSC for all MV subtypes (r¼ 0.62–0.89, p < 0.0001). We conclude that all counting beads provided linear results at concentrations ranging from 6 beads/ml to 100 beads/ml, but TruCount was the most consistent. Using SSC to gate MV events produced high background which negatively affected counting bead enumeration and overall MV calculations. Strategies to reduce SSC background should be employed in order to reliably use this technique. Keywords Flow cytometry, absolute counting, microvesicle, microparticles, extracellular vesicles, submicron particles, scatter Date received: 21 November 2017; accepted: 21 February 2018 1 Department of Clinical Pharmacy, Faculty of Pharmacy, Jordan University of Science and Technology, Irbid, Jordan 2 Haematology Unit, Calvary Mater Newcastle, New South Wales, Australia 3 Hunter Medical Research Institute, New Lambton, New South Wales, Australia 4 Faculty of Health and Medicine, University of Newcastle, New South Wales, Australia 5 Pathology North Hunter, NSW Health Pathology, New South Wales, Australia 6 Department of Endocrinology, Austin Health, Victoria, Australia 7 Department of Medicine, Austin Health, The University of Melbourne, Victoria, Australia Corresponding Author: Lisa F Lincz, Haematology Unit, Level 4, New Med Building, Calvary Mater Newcastle, Edith Street, Waratah, New South Wales 2298, Australia. Email: lisa.lincz@calvarymater.org.au Journal of Circulating Biomarkers Volume 7: 1–10 ª The Author(s) 2018 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1849454418766966 journals.sagepub.com/home/cbx Creative Commons Non Commercial CC BY-NC: This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 License (http://www.creativecommons.org/licenses/by-nc/4.0/) which permits non-commercial use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). http://orcid.org/0000-0002-1612-2382 http://orcid.org/0000-0002-1612-2382 mailto:lisa.lincz@calvarymater.org.au https://uk.sagepub.com/en-gb/journals-permissions https://doi.org/10.1177/1849454418766966 http://journals.sagepub.com/home/cbx https://us.sagepub.com/en-us/nam/open-access-at-sage Introduction Microvesicles (MVs) are a type of extracellular vesicle (EV) that bud off directly from the plasma membrane of activated or dying cells. 1 They are small (<1 mm) membrane-bound particles that can be identified by their cell surface markers, and they circulate throughout the body carrying biological remnants of their cells of origin. 1 MVs are regularly found in body fluids, including human plasma. 2 Their numbers and constitution have been docu- mented to change in times of pathological conditions, and as such, they are considered ideal biomarkers for diagnosis and prognosis of various disorders. 3 However, absolute MV counts vary widely between studies, sug- gesting that their detection and accurate quantification remains a challenge. Despite many new ‘nano’ technologies emerging in recent years, enumeration and analysis of circulating MV by flow cytometry continues to be the method of choice. 4 In spite of its sizing limitations, flow cytometry offers one of the few means to simultaneously detect multiple subsets of MVs and is also the most likely method to be easily adopted for clinical purposes owing to its widespread use in diagnostic pathology. Standardization remains a priority, and there have been numerous reports on how different variables can affect MV measurements. These variables include pre-analytical variables, such as blood collection and timing, needle gauge, anticoagulants, sample transport, centrifugation and storage; testing variables, such as choice of antibody and fluorochrome; and analytical variables, such as brand of flow cytometer, fluorescence compensa- tion approaches, threshold settings and gating strategies. 4–9 To this end, the International Society on Thrombosis and Haemostasis (ISTH) has been instrumental in developing consensus guidelines detailing best practice recommendations for evaluation of circulating MVs by flow cytometry. 7,10,11 More recently, ISTH has collaborated with other key organizations, the International Society of Extracellular Vesicles (ISEV) and the International Society on Advancement of Cytometry, to develop a comprehensive set of methodological guidelines for collection, isolation and measurement of EV using a range of common tech- niques. 12 In addition, ISEV published ‘minimal informa- tion for studies of EV’ in 2014, providing advice on methods and reporting of EV isolation, characterization and functional studies. 13,14 Such standardization is impera- tive to reduce the variability within and between methods in order to allow comparison between studies as well as develop diagnostic parameters for routine testing of circu- lating MVs. Although the newer generation of flow cytometers can measure volume and thus provide absolute particle counts, more conventional instruments without this capability can still be used successfully for quantitative MV detection. 15–17 An often overlooked aspect of MV enumeration using such classical flow cytometry methods involves the spiking of samples with a known amount of commercially available counting beads to enable calculation of the concentration of MV in the starting material. The general formula for this is: MV=ml ¼ðMV count= bead countÞ �ð total number of beads= test volumeÞ Hence, more variables are introduced, which can ulti- mately affect the final MV result. While all of the vari- ables in the above formula can be manipulated in each experiment, there is limited information on changes of bead concentration and its impact on total MV enumera- tion. Counting beads are distinct from the small (<1.0 mm) sizing beads available to establish MV gating parameters, and unlike the latter, there are no commercially available counting beads specific for MV enumerations; hence, much larger beads, meant for lymphocyte enumeration, are generally employed. Few manufacturers specify the final analysis volume, leaving the bead concentration up to the individual researcher. Hence, it is unknown whether the relationship remains linear over a wide range of bead concentrations or whether this differs between bead types and sizes. As for the parameters used to determine the ‘MV count’, the main testing variable is the actual flow cytometer and gating strategies employed. Enumeration and analysis of such small particles can be performed using forward scatter (FSC) or side scatter (SSC) as the main sizing parameter. The choice is usually determined by the type of instrument being used; with wide angle FSC (1–19�) machines such as Beckman Coulter (Brea, California, USA) generally per- forming better using FSC, compared with low angle FSC (1–8�) machines such as BD Biosciences (San Jose, California, USA), which typically perform better using SSC. A thorough cross-instrument evaluation has been per- formed to standardize gating parameters between the dif- ferent types. 7 However, for those cytometers that perform equally well on either parameter, it is difficult to know which provides more reliable results. Thus, the aims of this study were to (i) compare different concentrations of commercially available counting beads to establish their limitations and optimal parameters and (ii) determine whether FSC or SSC is a better sizing para- meter for enumeration of circulating MVs by flow cytome- try on a single BD FACSCanto (BD Biosciences). Materials and methods Subjects A series of 74 platelet-free plasma (PFP) samples from patients with type II diabetes were chosen as representative of a typical patient cohort for analysis. These samples were collected from patients (aged 50–75 years, body mass index (BMI) 25–35 kg/m 2 ) recruited between 2014 and 2016 from diabetes clinics at the Department of Endocrinology in Austin Health, Melbourne, Victoria, as part of a separate 2 Journal of Circulating Biomarkers study on circulating MV levels in type II diabetes. The study followed the guidelines set out within the Australian National Statement on Ethical Conduct in Human Research (2007; Updated May 2015) and was approved by the Austin Health (HREC/12/Austin/63) and Hunter New England Area Human Research Ethics and Governance Committees (SSA/15/HNE/141). All procedures were conducted in accordance with the Helsinki Declaration of 1975, as revised in 2008, and written informed consent was obtained from all participants. Blood processing Peripheral blood was collected into 3.2% sodium citrate and processed at room temperature within 2 h of collection. Whole blood was centrifuged at 400�g for 15 min to sep- arate the cellular fraction from the plasma. The latter was carefully removed, transferred to a fresh tube and further centrifuged at 2100�g for 15 min. All but the bottom 500 ml was transferred to a fresh tube and centrifuged again at 2100�g for 15 min to produce PFP. This was aliquoted, stored at �80�C and then thawed at 37�C immediately prior to analysis. Antibody staining of PFP for MV analysis Staining of MV was performed as previously described. 11,18 A 10 ml aliquot of PFP was incubated at room temperature for 30 min with various combinations of antibodies conju- gated to phycoerythrin (PE), fluorescein isothiocyanate (FITC), allophycocyanin (APC) or PE-cyanine (PE-Cy5): CD41-PE (clone PL2-49, Biocytex, Marseille, France; pla- telet marker), CD42b-FITC (clone HIP1, BD Pharmingen, San Diego, California, USA; platelet marker), CD235a- APC (clone GA-R2 (HIR2), BD Pharmingen; erythrocyte marker), CD105-PE (clone 1G2, Beckman Coulter; endothelial marker), CD31-PE (clone WM59, BD Phar- mingen; against endothelial marker PECAM-1), CD62e PE-Cy5 (clone 68-5H11, BD Pharmingen, against activated endothelial marker E-selectin) and Annexin V-APC (eBioscience, San Diego, California, USA; phosphatidyl- serine). All assays were diluted to a final volume of 500 ml in phosphate-buffered saline (without Ca 2þ and Mg 2þ ) or calcium-rich binding buffer (for those stained with Annexin V), with the addition of a known quantity of counting beads and 15 mM D-Phe-Pro-Arg-chloromethylk- etone (PPAK) to inhibit clumping. Counting beads Three different popular brands of counting beads were compared: TruCount (BD Biosciences; size not specified), Flow-Count (Beckman Coulter; 10 mm diameter) and CountBright (Molecular Probes, Eugene, Oregon, USA; 7 mm diameter). Each bead preparation was diluted into the final test volume at the indicated concentration by adding exact quantities based on the individual lot concentration provided by the manufacturer. The plasma source, concen- tration and acquiring time on the flow cytometer were kept constant. For experiments comparing FSC with SSC, CountBright beads were used at a final concentration of 50 beads/ml. Analysis of MV by flow cytometry All flow cytometry analyses were performed on a standard configuration BD FACSCanto (BD Biosciences) equipped with two lasers (488 nm and 640 nm). The FACS flow pressure was set to 3.0 lbf/in 2 and the low flow rate adjusted to a factor of 0.61 (decreased from the original factory settings of 4.5 lbf/in 2 and 0.75, respectively) to improve resolution at smaller sizes. Analysis of MV was performed as previously described 11,18 and according to guidelines established by the ISTH Vascular Biology Scientific Standardization Committee on the standardization of plate- let microparticle enumeration by flow cytometry incorpor- ating modifications suggested for the BD FACSCanto (BD Biosciences). 10 The cytometer was calibrated for FSC res- olution using Megamix sizing beads (a blend of 2:1:1 of 0.5, 0.9 and 3 mm diameter fluorescent beads) or for SSC resolution using Megamix-Plus SSC (a mixture of 0.16 mm, 0.20 mm, 0.24 mm and 0.5 mm beads) both purchased from Biocytex. Voltages were set at FSC ¼ 570 V and SSC ¼ 390 V for FSC detection or FSC ¼ 350 V and SSC ¼ 631 V for SSC detection. The lower MV detection limits were set according to the manufacturer’s instructions, with thresh- olds of FSC ¼ 200/SSC ¼ 200 employed for FSC enumera- tion and SSC ¼ 3200 for SSC gating. Fluorescent voltages were set to 654 V for FITC, 485 V for PE, 544 V for PE- Cy5 and 400 V for APC (with the exception of Annexin V- APC detected at 500 V). Counting beads were detected on PerCP-Cy5.5 at 290 V using FSC gating and 549 V using SSC gating. Events were collected for 60 s (bead experi- ments) or 120 s (patient samples; to enable adequate num- ber of counting bead events 19 ) at low flow rate prior to analysis using FACS Diva software (BD Biosciences). The absolute number of MV in each plasma sample was calcu- lated using the formula: MV/ml ¼ (MV count/bead count) � (total # beads/test volume). Statistical analysis Data for continuous variables are expressed as mean + standard deviation or median (interquartile range) where appropriate. Variables that were not normally distributed were analysed using non-parametric tests. Differences in mean levels of multiple normally distributed continuous variables were assessed using one-way analysis of variance (ANOVA) with post hoc Scheffe test for multiple compar- isons. Mann–Whitney U test and Wilcoxon matched-pair tests were used to detect differences between medians for individual and paired data, respectively. Correlations Alkhatatbeh et al. 3 between continuous variables were assessed by Pearson’s product-moment or Spearman’s rho (r) where appropriate. All calculations were performed with Statistica v10.0 (StatSoft, Tulsa, Oklahoma, USA) or STATA v11 (StataCorp LLC, College Station, Texas, USA) using two-tailed tests, and p values <0.05 were considered statistically significant. Results Comparison of counting beads The manufacturer’s method was followed as closely as possible to prepare a series of identical plasma samples containing varying concentrations of three different brands of counting beads: TruCount (BD Biosciences), Flow- Count (Beckman Coulter) and CountBright (Molecular Probes), respectively. All tests were performed in triplicate. Figure 1 shows that for all three brands of counting beads tested, the relationship between the number of beads added to each test and the number of beads counted by flow cytometry remained linear over a wide range of bead concentrations (up to 200 beads/ml), as indicated by all correlation coefficients (R 2 ¼ 0.999, 0.997, 0.997, respectively), which were close to 1.0. Importantly, these relationships were maintained even at the lowest bead concentrations of 6.26 beads/ml (equivalent to adding just 3125 beads to a 500 ml test). However, the slope (m) of the lines fitted for the CountBright (m ¼ 1.54) and Flow-Count (m ¼ 1.51) beads was slightly higher than that of the TruCount beads (m¼1.22). Thus, the absolute number of TruCount bead events became significantly different from that of the CountBright and Flow-Count at concentrations greater than 50 beads/ml (p < 0.05). We next sought to determine whether the number of raw MV events counted was stable in the presence of different levels of counting beads. As indicated by the dotted lines in Figure 1, the number of CD41 þ /Annexin V þ raw events was on average 428 + 42 and was not significantly differ- ent in tubes with added counting beads versus tubes without counting beads (data not shown). This was with the excep- tion of tubes containing 200 beads/ml of Flow-Count beads that derived significantly less MV events than expected (p < 0.001), indicating that spiking with high amounts of Flow-Count beads interferes with MV detection. The numbers of bead and CD41 þ /Annexin V þ events were then used to calculate the final concentration of CD41 þ /Annexin V þ MVs in each sample for the three brands of counting beads at different concentrations (Table 1). With the exception of the highest concentration Figure 1. Correlation between the number of bead events counted versus added to each test and MV events counted using different brands of absolute counting beads. Solid lines and filled icons represent bead events, whereas dotted lines and outlined icons represent corresponding raw CD41þ/Annexin Vþ MV events detected by flow cytometry for each bead dilution. *p � 0.05 for TruCount versus CountBright and/or Flow-Count bead events; **p � 0.001 for Flow-Count versus TruCount and/or CountBright MV events. MV: microvesicle. 4 Journal of Circulating Biomarkers of Flow-Count beads (this data point was omitted from the overall analysis), the results were consistent within each bead manufacturer, showing no significant difference between the calculated values at all bead concentrations (within manufacturer ANOVA p values ¼ 0.685, 0.417, 0.479, respectively). TruCount tubes gave the highest over- all consistency, with a concentration variation of 3.80% compared to 8.86% and 7.50% for Flow-Count and CountBright, respectively. The number of calculated MV events was highest using TruCount beads (3768 + 143 MV/ml), and this was significantly different from that cal- culated using Flow-Count (3058 + 271 MV/ml, p < 0.001) and CountBright (2886 + 217 MV/ml; p < 0.001). Comparison of FSC versus SSC for gating MV by flow cytometry In an attempt to optimize the number of raw MV events detected by flow cytometry, we compared two alternate gating strategies on our FACSCanto, one using FSC and the other using SSC as the main sizing parameter. Fluor- escent beads of known diameters selected to cover a major part of the theoretical MV size range (0.1–1.0 mm) were used to determine resolution aptitude and establish appro- priate MV gates. As the relative position of biological MVs and beads in SSC is different from that in FSC, reference beads of sizes specifically designed for each parameter were used. As shown in Figure 2, the FACSCanto was equally capable of adequately resolving the respective bead mixtures by either FSC or SSC. Threshold parameters were set to exclude as much background as possible, leaving the MV gates set to capture all events below the 0.5 mm bead cloud using FSC and all events between the 0.2 and 0.5 mm bead limits detected using SSC. These MV gates have been shown to be equivalent in order to allow inter-platform comparisons of MV counts. 19 The respective MV gates were used to detect six dif- ferent MV subsets in a series of 74 patient plasma sam- ples. The number and type of bead were kept constant (CountBright beads were added to all samples at a final concentration of 50 beads/ml), and events were collected from the same tube for 120 s on each gating parameter. Table 2 presents the number of raw events detected using the respective MV gates as well as the number of CountBright beads counted. These amounts were deter- mined in the absence of added patient plasma at the beginning of each run (n ¼ 9) in order to establish the amount of background electronic noise detected using either parameter. This consistently showed significantly more background events in the MV gate using SSC com- pared to FSC (3113, 2098–23,860 vs. 470, 404–3994; p ¼ 0.008), but the number of bead events detected remained equivalent (786, 733–804 vs. 822, 730–865; p ¼ 0.374). With the addition of individual plasma sam- ples, events detected in the MV gate were significantly higher using SSC compared to FSC (119,640, 84,320– 180,233 vs. 12,476, 7530–27,211; p < 0.00001) with an increase that was disproportionate and could not be explained by the initial higher background noise events. In addition, the number of beads counted was signifi- cantly reduced in the presence of plasma using SSC (730, 697–771 vs. 822, 730–865 without plasma; p ¼ 0.008) and when plasma containing samples were measured on SSC compared to FSC (730, 697–771 vs. 766, 738–793; p < 0.0001). In contrast, the number of beads enumerated remained stable when FSC was used to count beads in samples with or without additional plasma (766, 738–793 vs. 786, 733–804; p ¼ 0.520). This is further illustrated in Figure 3, which shows a strong negative correlation between the number of CountBright beads counted and the number of events in the MV gate when SSC but not FSC is used as the main gating parameter. Table 1. Calculated number of CD41 þ /Annexin V þ MVs using different manufacturer’s brands of counting beads at different concentrations. TruCountTM Flow-Count CountBrightTM Final bead concentration per ml Mean + stdev %CV Mean + stdev %CV Mean + stdev %CV 200 3397 + 211 6.2 32 + 16 50.1 2621 +49 1.9 100 3912 + 449 11.5 2591 + 613 23.7 2841 + 159 5.6 50 3581 + 203 5.7 3086 + 249 8.1 2653 + 153 5.8 25 3745 + 204 5.4 3154 + 244 7.7 3001 + 171 5.7 12.5 3910 + 638 16.3 3290 + 452 13.7 3147 + 870 27.6 6.26 3693 + 632 17.1 3167 + 565 17.8 3051 + 217 7.1 mean + stdev 3768 + 143 10.4 3058 + 271a 14.21a 2886 + 217 9.0 %CV 3.80 8.86a 7.50 p values 0.685 0.417 a 0.479 <0.001 0.724 Between brands <0.001 CV: coefficient of variation; MV: microvesicle; stdev: standard deviation. a Omitting 200 beads/ml results. Alkhatatbeh et al. 5 Not surprisingly, calculation of plasma MV concentra- tions using data derived from SSC analysis resulted in sig- nificantly higher absolute amounts of most MV subsets compared to analysis using FSC (p < 0.0001; Figure 4(a)). However, the results for each MV subset were indi- vidually well correlated between the two methodologies (r¼ 0.619–0.992; p < 0.0001; Figure 4(b)). Discussion Absolute MV counts vary widely between studies, with circulating platelet MV levels ranging from hundreds to thousands even in control populations. 20–22 Although much attention has been given to the many pre-analytical and methodical variables that can result in such discrepancies, 23 few have addressed the addition of the all-important Table 2. Comparison of raw MV events and beads counted in the presence or absence of plasma using FSC or SSC as the flow cytometry sizing parameter. Number of events in MV gate a Number of beads counted a Without plasma (n ¼ 9) With plasma (n ¼ 74) Without plasma (n ¼ 9) With plasma (n ¼ 74) p value FSC 470 (444–656) 12,476 (7530–27,211) 786 (733–804) 766 (738–793) 0.520 SSC 3113 (2497–4727) 119,640 (84,320–180,233) 822 (730–865) 730 (697–771) 0.008 p value 0.008 <0.00001 0.374 <0.00001 MV: microvesicle. aValues are presented as median (interquartile range). Figure 2. Flow cytometry resolution of sizing beads and MV gate settings using FSC versus SSC as the main size parameter. (a) The histogram in the top panel shows resolution of 0.5 mm (orange) and 0.9 mm (yellow) Megamix beads by FSC. The same beads are depicted in the dot plot below where the 0.9 mm bead cloud is used to set the MV gate. (b) The histogram in the top panel shows resolution of 0.16 mm (pink), 0.20 mm (blue), 0.24 mm (green) and 0.5 mm (red) Megamix-Plus SSC beads using FITC as the main parameter. The same beads are depicted in the dot plot below using SSC as the threshold to eliminate the 0.16 mm (pink) beads and use the 0.5 and 0.2 mm bead clouds to set the MV gate. The gate for capturing the counting beads is depicted in blue on both dot plots. 6 Journal of Circulating Biomarkers counting beads that enable such calculations. Herein, we compared three different popular brands of fluorescent absolute counting beads: TruCount by BD Biosciences, which is supplied as a lyophilized pellet in individual flow cytometry tubes; Flow-Count (Beckman Coulter) and CountBright (Molecular Probes), both of which are sup- plied as slurries. All are brightly fluorescent and intended to be used for determining absolute counts of leucocytes in blood. Our results show that although the number of beads counted by the flow cytometer remained linear over a wide range of concentrations for all three brands tested, the TruCount beads gave the most accurate enumerations with the least variation in MV calculated levels. This is perhaps due to the lyophilized format of the beads that may deter clumping and thus minimize the potential for pipetting error that would be more common with slurries. However, the single-use tubes are the most expensive of the three preparations, contain a set amount of beads and require the presence of proteins (such as from plasma or serum) for proper performance, which may limit the utility of such tubes for other, such as purified, MV preparations. Although all manufacturers specify a certain amount of beads to use per test (equating to 50,000–100,000 beads), only BD provides a recommended final test volume of 520 ml (100 beads/ml) of their TruCount beads, while CountBright manufacturers warn to maintain a final vol- ume of at least 300 ml per test (143 beads/ml). Despite this, both brands performed well at concentrations of up to 200 beads/ml. Of concern was the finding that high levels of Flow-Count beads (� 200 beads/ml) significantly inter- fered with the detection of MV particles. The underlying reasons for inhibition are unclear, and we cannot discount factors other than the beads themselves, such as proprie- tary stabilizers added by the manufacturer. Nevertheless, our findings are instructive to delineate the concentration parameters where Flow-Count beads can be used to measure MV. We did not count 1000 bead events as suggested by the manufacturers of Flow-Count and CountBright because our MV enumeration protocol has always been based on that recommended by the ISTH standardization papers. The first of these employed 30 ml of Flow-Count beads in a final volume of 580 ml (approximately 50 beads/ml) and a timed collection of events for 60 s at low flow rate. 10 On our instrument this allows for counts of 350–400 beads, but obviously this varies between laboratories and will be highly dependent on the fluidics pressure of individual machines. We, therefore, have adopted the latest recommendation to increase the collection time to 120 s if the number of beads counted is <500. 19 However, the results presented herein suggest that this will not make any significant difference to the calculated MV results for major populations that are readily detectable (i.e. approximately 400 positive events detected in the MV gate). Perhaps a better guide would be to collect a minimum number of MV events of interest. Original attempts to standardize flow cytometry analysis of MV using FSC as the main sizing parameter proved that reproducible platelet-derived (CD41 þ ) MV counts could be obtained across many different laboratories worldwide. 10 However, this success was not always shared by labora- tories using BD instruments, which demonstrated discre- pancies between location of sizing beads and that of biological particles on the FSC parameter, thought to be due to the relatively lower solid angle used to collect FSC signals on these instruments. This could be ameliorated by removing the upper gate limit set by the Megamix beads, a recommendation adopted for the current study. 10 However, the same authors found a more reproducible solution in using SSC as the main sizing parameter, with the use of different sized reference beads for FSC versus SSC being the most critical element for standardization across the dif- ferent platforms. 7,19 Our BD instrument produced less variability in MV subset calculations using FSC. Although we found higher Figure 3. Correlation between the number of beads counted and the number of MV events detected by SSC. Scatterplots show relationship between the number of beads counted and the number of MV events detected in the MV gate when using (a) FSC or (b) SSC as the main detection parameter. Alkhatatbeh et al. 7 background events detected by the SSC channel compared to FSC, these were not above the recommended back- ground noise threshold ratio of 1.0 (calculated as the num- ber of events per second/maximal number of events per second acceptable by the instrument ¼ [3113/60]/4000 ¼ 0.012 for SSC on our FACSCanto) and consistent, if not better, then most instruments surveyed by Cointe et al. 19 However, the significantly increased number of MV events detected in plasma suggests that SSC may be a more sen- sitive parameter for small particles, resulting in much higher absolute counts for the majority of MV subsets. This phenomenon was not observed across different instruments in the ISTH multicentre workshop, with similar counts recorded between instruments using SSC or FSC as the preferred sizing parameter. 19 However, the results were highly variable, with standard deviations of up to 50% of the mean for identical samples measured on different instruments. Hence, only a paired study design would be able to address the difference between results measured by different scatter parameters on individual flow cytometers. The utility of employing polystyrene beads to establish sizing gates for biological material has been the source of much contention. It is well known that polystyrene has a much higher refractive index, resulting in light scattering properties much different from plasma membranes. 24,25 A 400 nm polystyrene microsphere has been shown to pro- duce the same forward light scatter as a 1 mm lipid or cellular vesicle. 24 The Megamix gating strategy originally Figure 4. Results of MV subsets calculated from data using FSC or SSC as the main detection parameter. (a) Bar graph illustrating differences in concentrations of MV subsets when calculated from data obtained using FSC versus SSC as the main detection parameter. Bars represent median and interquartile ranges. (b) Correlation of individual MV subset concentrations when calculated from data obtained using FSC versus SSC as the main detection parameter. Scatter graphs are presented on log scales. *p < 0.0001. MV: microvesicle; FSC: forward scatter; SSC: side scatter. 8 Journal of Circulating Biomarkers established by the Scientific Standardization Committee set the upper size detection limit using 900 nm beads, but this has been estimated in actuality to gate biological vesicles measuring 800–2400 nm in diameter. 25 Such discrepancies have led to proposals by us and others that triggering on fluorescence may provide a much more useful approach. 6,26,27 However, plasma contains many different sizes and shapes of particles, some as small as 30 nm, 28,29 and current technology in flow cytometry remains biased towards detection of only the largest and brightest particles, with many events destined to be lost in the instrument ‘noise’. 23 Much more sensitive detection and sizing meth- ods of nanoparticle tracking analysis and/or resistive pulse sensing can provide more accurate measurement of EV concentrations and have confirmed that total plasma EV is highly underestimated by flow cytometry. 28,30,31 The current study is limited by the few different types of counting beads assayed and the use of a single flow cytometer. Smaller sized counting beads such as the 5.2 mM CytoCount (DAKO, Agilent Pathology Solu- tions, Santa Clara, California, USA) are becoming more popular and would have made a welcome comparison to the larger Flow-Count and CountBright beads used here. Similarly, all the analyses were done on a single flow cytometer, rendering it the equivalent of a technological ‘case study’. It would be of interest to compare our results to other FACSCanto machines as well as other newer instruments with integrated cell counting. None- theless, we have highlighted the importance of bead selection, concentration and background minimization for MV analysis by flow cytometry. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article. Funding The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This study was funded in part by a Hunter Medical Research Institute (HMRI) Special Project Grant, supported by Lions District 201 N3 Diabetes Foundation, awarded to RFT and LFL. Maintenance for the BD FACSCanto was supplied by the Hunter Cancer Research Alliance. MJA was supported by Jordan University of Science and Technology. ORCID iD Lisa F Lincz http://orcid.org/0000-0002-1612-2382 References 1. Raposo G and Stoorvogel W. Extracellular vesicles: exo- somes, microvesicles, and friends. J Cell Biol 2013; 200: 373–383. 2. George J, Thoi L, McManus L, et al. Isolation of human platelet membrane microparticles from plasma and serum. Blood 1982; 60: 834–840. 3. Enjeti AK, Lincz LF and Seldon M. Microparticles in health and disease. Semin Thromb Hemost 2008; 34: 683–691. 4. Chandler WL. Measurement of microvesicle levels in human blood using flow cytometry. Cytometry B Clin Cytom 2016; 90: 326–336. 5. van Ierssel SH, Van Craenenbroeck EM, Conraads VM, et al. Flow cytometric detection of endothelial microparticles (EMP): effects of centrifugation and storage alter with the phenotype studied. Thromb Res 2010; 125: 332–339. 6. Arraud N, Gounou C, Turpin D, et al. Fluorescence trigger- ing: a general strategy for enumerating and phenotyping extracellular vesicles by flow cytometry. Cytometry A 2016; 89: 184–195. 7. Poncelet P, Robert S, Bouriche T, et al. Standardized count- ing of circulating platelet microparticles using currently available flow cytometers and scatter-based triggering: For- ward or side scatter? Cytometry A 2016; 89: 148–158. 8. Enjeti AK, Lincz LF and Seldon M. Detection and measure- ment of microparticles: an evolving research tool for vascular biology. Semin Thromb Hemost 2007; 33: 771–779. 9. Poncelet P, Robert S, Bailly N, et al. Tips and tricks for flow cytometry-based analysis and counting of microparticles. Transfus Apher Sci 2015; 53: 110–126. 10. Lacroix R, Robert S, Poncelet P, et al. Standardization of platelet-derived microparticle enumeration by flow cytome- try with calibrated beads: results of the International Society on Thrombosis and Haemostasis SSC Collaborative work- shop. J Thromb Haemost 2010; 8: 2571–2574. 11. Robert S, Poncelet P, Lacroix R, et al. Standardization of platelet-derived microparticle counting using calibrated beads and a Cytomics FC500 routine flow cytometer: a first step towards multicenter studies? J Thromb Haemost 2009; 7: 190–197. 12. Coumans FAW, Brisson AR, Buzas EI, et al. Methodological guidelines to study extracellular vesicles. Circ Res 2017; 120: 1632–1648. 13. Witwer KW, Soekmadji C, Hill AF, et al. Updating the MISEV minimal requirements for extracellular vesicle stud- ies: building bridges to reproducibility. J Extracell Vesicles 2017; 6: 1396823. 14. Lotvall J, Hill AF, Hochberg F, 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. 15. Inglis H, Norris P and Danesh A. Techniques for the analysis of extracellular vesicles using flow cytometry. J Vis Exp 2015; 97: e52484. 16. Marcoux G, Duchez AC, Cloutier N, et al. Revealing the diversity of extracellular vesicles using high-dimensional flow cytometry analyses. Sci Rep 2016; 6: 35928. 17. Nielsen MH, Beck-Nielsen H, Andersen MN, et al. A flow cytometric method for characterization of circulating cell-derived microparticles in plasma. J Extracell Vesicles 2014; 3: 20795. Alkhatatbeh et al. 9 http://orcid.org/0000-0002-1612-2382 http://orcid.org/0000-0002-1612-2382 http://orcid.org/0000-0002-1612-2382 18. Alkhatatbeh MJ, Mhaidat NM, Enjeti AK, et al. The putative diabetic plasma marker, soluble CD36, is non-cleaved, non- soluble and entirely associated with microparticles. J Thromb Haemost 2011; 9: 844–851. 19. Cointe S, Judicone C, Robert S, et al. Standardization of microparticle enumeration across different flow cytometry platforms: results of a multicenter collaborative workshop. J Thromb Haemost 2017; 15: 187–193. 20. Helal O, Defoort C, Robert S, et al. Increased levels of micro- particles originating from endothelial cells, platelets and ery- throcytes in subjects with metabolic syndrome: relationship with oxidative stress. Nutr Metab Cardiovasc Dis 2011; 21: 665–671. 21. Alkhatatbeh MJ, Enjeti AK, Acharya S, et al. The origin of circulating CD36 in type 2 diabetes. Nutr Diabetes 2013; 3: e59. 22. Chirinos JA, Heresi GA, Velasquez H, et al. Elevation of endothelial microparticles, platelets, and leukocyte activation in patients with venous thromboembolism. J Am Coll Cardiol 2005; 45: 1467–1471. 23. Nolan JP and Jones JC. Detection of platelet vesicles by flow cytometry. Platelets 2017; 28: 256–262. 24. Chandler WL, Yeung W and Tait JF. A new microparticle size calibration standard for use in measuring smaller micro- particles using a new flow cytometer. J Thromb Haemost 2011; 9: 1216–1224. 25. van der Pol E, van Gemert MJ, Sturk A, et al. Single vs. swarm detection of microparticles and exosomes by flow cytometry. J Thromb Haemost 2012; 10: 919–930. 26. Arraud N, Gounou C, Linares R, et al. A simple flow cyto- metry method improves the detection of phosphatidylserine- exposing extracellular vesicles. J Thromb Haemost 2015; 13: 237–247. 27. Enjeti AK, Lincz L and Seldon M. Bio-maleimide as a gen- eric stain for detection and quantitation of microparticles. Int J Lab Hematol 2008; 30: 196–199. 28. van der Pol E, Coumans FA, Grootemaat AE, et al. Particle size distribution of exosomes and microvesicles determined by transmission electron microscopy, flow cytometry, nano- particle tracking analysis, and resistive pulse sensing. J Thromb Haemost 2014; 12: 1182–1192. 29. Arraud N, Linares R, Tan S, et al. Extracellular vesicles from blood plasma: determination of their morphology, size, pheno- type and concentration. J Thromb Haemost 2014; 12: 614–627. 30. Enjeti AK, Ariyarajah A, D’Crus A, et al. Correlative analysis of nanoparticle tracking, flow cytometric and functional mea- surements for circulating microvesicles in normal subjects. Thromb Res 2016; 145: 18–23. 31. Enjeti AK, Ariyarajah A, Warwick E, et al. Challenges in analysis of circulating extracellular vesicles in human plasma using nanotracking and tunable resistive pulse sensing. J Nanomed Nanotechnol 2017; 8: 5. 10 Journal of Circulating Biomarkers << /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /None /Binding /Left /CalGrayProfile (Gray Gamma 2.2) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Warning /CompatibilityLevel 1.4 /CompressObjects /Off /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages false /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /DetectCurves 0.1000 /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedOpenType false /ParseICCProfilesInComments true /EmbedJobOptions true /DSCReportingLevel 0 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams true /MaxSubsetPct 100 /Optimize true /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveDICMYKValues true /PreserveEPSInfo true /PreserveFlatness false /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Remove /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true ] /AntiAliasColorImages false /CropColorImages false /ColorImageMinResolution 266 /ColorImageMinResolutionPolicy /OK /DownsampleColorImages true /ColorImageDownsampleType /Average /ColorImageResolution 175 /ColorImageDepth -1 /ColorImageMinDownsampleDepth 1 /ColorImageDownsampleThreshold 1.50286 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.40 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.76 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /CropGrayImages false /GrayImageMinResolution 266 /GrayImageMinResolutionPolicy /OK /DownsampleGrayImages true /GrayImageDownsampleType /Average /GrayImageResolution 175 /GrayImageDepth -1 /GrayImageMinDownsampleDepth 2 /GrayImageDownsampleThreshold 1.50286 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.40 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.76 /HSamples [2 1 1 2] /VSamples [2 1 1 2] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /CropMonoImages false /MonoImageMinResolution 900 /MonoImageMinResolutionPolicy /OK /DownsampleMonoImages true /MonoImageDownsampleType /Average /MonoImageResolution 175 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50286 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /CheckCompliance [ /None ] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox false /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile (U.S. Web Coated \050SWOP\051 v2) /PDFXOutputConditionIdentifier (CGATS TR 001) /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown /CreateJDFFile false /Description << /ENU >> /Namespace [ (Adobe) (Common) (1.0) ] /OtherNamespaces [ << /AsReaderSpreads false /CropImagesToFrames true /ErrorControl /WarnAndContinue /FlattenerIgnoreSpreadOverrides false /IncludeGuidesGrids false /IncludeNonPrinting false /IncludeSlug false /Namespace [ (Adobe) (InDesign) (4.0) ] /OmitPlacedBitmaps false /OmitPlacedEPS false /OmitPlacedPDF false /SimulateOverprint /Legacy >> << /AllowImageBreaks true /AllowTableBreaks true /ExpandPage false /HonorBaseURL true /HonorRolloverEffect false /IgnoreHTMLPageBreaks false /IncludeHeaderFooter false /MarginOffset [ 0 0 0 0 ] /MetadataAuthor () /MetadataKeywords () /MetadataSubject () /MetadataTitle () /MetricPageSize [ 0 0 ] /MetricUnit /inch /MobileCompatible 0 /Namespace [ (Adobe) (GoLive) (8.0) ] /OpenZoomToHTMLFontSize false /PageOrientation /Portrait /RemoveBackground false /ShrinkContent true /TreatColorsAs /MainMonitorColors /UseEmbeddedProfiles false /UseHTMLTitleAsMetadata true >> << /AddBleedMarks false /AddColorBars false /AddCropMarks false /AddPageInfo false /AddRegMarks false /BleedOffset [ 9 9 9 9 ] /ConvertColors /ConvertToRGB /DestinationProfileName (sRGB IEC61966-2.1) /DestinationProfileSelector /UseName /Downsample16BitImages true /FlattenerPreset << /ClipComplexRegions true /ConvertStrokesToOutlines false /ConvertTextToOutlines false /GradientResolution 300 /LineArtTextResolution 1200 /PresetName ([High Resolution]) /PresetSelector /HighResolution /RasterVectorBalance 1 >> /FormElements true /GenerateStructure false /IncludeBookmarks false /IncludeHyperlinks false /IncludeInteractive false /IncludeLayers false /IncludeProfiles true /MarksOffset 9 /MarksWeight 0.125000 /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe) (CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /DocumentCMYK /PageMarksFile /RomanDefault /PreserveEditing true /UntaggedCMYKHandling /UseDocumentProfile /UntaggedRGBHandling /UseDocumentProfile /UseDocumentBleed false >> ] /SyntheticBoldness 1.000000 >> setdistillerparams << /HWResolution [288 288] /PageSize [612.000 792.000] >> setpagedevice