Exosome Isolation: Cyclical Electrical Field Flow Fractionation in Low

Subscriber access provided by The University of Texas at El Paso (UTEP)

Article

Exosome Isolation: Cyclical Electrical Field Flow Fractionation in Low Ionic Strength Fluids Kevin E. Petersen, Farhad Shiri, Travis White, Gina T Bardi, Himanshu Sant, Bruce K. Gale, and Joshua L Hood Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b03146 • Publication Date (Web): 10 Oct 2018 Downloaded from http://pubs.acs.org on October 14, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

1

Exosome Isolation: Cyclical Electrical Field Flow Fractionation in Low Ionic Strength Fluids Kevin E. Petersen1, Farhad Shiri1, Travis White1, Gina T. Bardi2, Himanshu Sant1, Bruce K. Gale1*, Joshua L. Hood2* 1

University of Utah Department of Mechanical Engineering, 1495 E 100 S, Room 1550, Salt Lake City, UT 84112; 2University of Louisville, Department of Pharmacology and Toxicology & James Graham Brown Cancer Center, Kosair Charities Clinical & Translational Research Building, Room 411, 505 South Hancock Street, Louisville, KY 40202; ABSTRACT: The influence of buffer substitution and dilution effects on exosome size and electrophoretic mobility were shown for the first time. Cyclical electrical field flow fractionation (Cy-El-FFF) in various substituted fluids was applied to exosomes and other particles. Tested carrier fluids of deionized (DI) water, 1X phosphate buffered saline (PBS), 0.308 M trehalose, and 2% isopropyl alcohol (IPA) influenced Cy-El-FFF-mediated isolation of A375 melanoma exosomes. All fractograms revealed a crescent-shaped trend in retention times with increasing voltage with the maximum retention time at ~1.3V AC. A375 melanoma exosome recovery was approximately 70-80% after each buffer substitution and recovery was independent of whether the sample was substituted into 1X PBS or DI water. Exosome dilution in deionized water produced a U-shaped dependence on electrophoretic mobility. The effect of dilution using 1X PBS buffer revealed a very gradual change in electrophoretic mobility of exosomes from ~-1.6 to -0.1 µm cm/s V, as exosome concentration was decreased. This differed from the use of DI water where a large change from ~-5.5 to -0.1 µm cm/s V over the same dilution range was observed. Fractograms of separated A375 melanoma exosomes in two substituted low-ionic strength buffers were compared with synthetic particle fractograms. Overall, the ability of Cy-El-FFF to separate exosomes based on their size and charge is a highly promising, label-free approach to initially catalogue and purify exosomes subtypes for biobanking as well as to enable further exosome subtype interrogations.

Interest in exosomes has burgeoned in recent years as researchers continue to learn more about how these nanoscale vesicles are produced and released by cells. Exosomes, originally discovered as a means of removing cellular components during erythrocyte development1, are increasingly understood to play additional roles in both normal and pathological disease states. Some of these include: embryogenesis2, paracrine cellular communication3, cancer metastasis4, myocardial ischemia5, and parasite-host interactions6. Participation of exosomes in these processes is thought to be mediated in part by horizontaltransfer of RNA7. In general, exosomal nanovesicles are formed in multivesicular bodies (MVBs), or multivesicular endosomes, that subsequently fuse with the inner cellular plasma membrane resulting in exosome release into the extracellular microenvironment8. Exosomes are approximately 30 - 200 nm and can vary in content and composition. Registries such as Exocarta9-10, catalog the differences between exosome subtypes. Exosome purification methods typically consist of both an isolation and purification step11. An initial exosome isolation step such as precipitation12 (used in many commercially available kits), filtration13, or ultracentrifugation14, is followed by further purification using chromatography or by various immuno-

affinity approaches using known markers such as CD9, CD63, or CD8115. Density gradient ultracentrifugation, sieving, and size exclusion chromatography may also be used. Field flow fractionation (FFF) is a suite of techniques that has rarely been applied to exosome separation13, 16-18. FFF is a labelfree separation technique capable of separating particles based on biophysical differences that can be resolved by assessing particle interactions with any of several applied force fields (flow, density, thermal, electrical, magnetic, gravity, etc.). Briefly, in El-FFF two narrowly gapped electrodes form the top and bottom walls of a thin, flat fluid channel. Particle separation occurs within the channel because particles, exhibiting different electrophoretic mobilities in an electric field, exit the channel at different times. Cyclical-electrical-FFF (Cy-El-FFF) produces a strong, but time-changing effective field by using AC voltage. Although several size-based exosome separation techniques exist, few separation techniques exist for exosome separation based upon charge. Use of El-FFF for exosome separation would be desirable over capillary electrophoresis because El-FFF is expected to process higher sample volumes ( ~0.2– 40 mL19-20 vs. 0.03 – 18.9 nL21), is safer (~1V instead of >1000 V), interfaces well with a variety of detectors, and the sample is recovered without breaking the field.

ACS Paragon Plus Environment

Analytical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 10

2

Exosomes are known to possess negative zeta potentials22-26. They are loaded with a variety of charged molecular cargo including carbohydrates, proteins, lipids, and RNA8. The side chains of different protein-based receptors and ligands, unique to given exosome subtypes, are composed of negatively charged aspartate and glutamate amino acids, as well as positively charged lysine, arginine, or histidine amino acids. Side chain differences such as these, coupled to RNA, carbohydrate and lipid composition (phosphatidyl serine content, for example8) may contribute to pH dependent differences in overall composite exosome charge and resulting zeta potential and electrophoretic mobility. In fact, increased tumor exosome release and uptake is associated with low pH conditions as encountered in tumor microenvironments27. Exosome charge composition is expected to be exosome subpopulation specific. This expectation is supported by a recent study demonstrating that MDA-MB-231 breast cancer exosomes, isolated using anti-CD63 versus a normal IgG isotype control, generated different zeta potential profiles26. The ability of Cy-El-FFF to separate exosomes, and potentially exosome subtypes, based upon their electrophoretic mobility in low strength ionic fluids is a highly promising “label-free” biocompatible approach to gently purify and initially catalogue exosome subtypes based on biophysical properties. This effort will facilitate future biobanking of exosome subtypes, as well as enable further exosome subpopulation interrogations concerning exosome structure, biomarker potential, function, and therapeutic applications. One difficulty in using El-FFF for biological vesicle separation is that salt concentration within typical physiological vesicle buffers is high enough to reduce the strength of the effective electric field within the El-FFF channel28. Salt must be removed via buffer substitution if effective vesicle separation using El-FFF is to be achieved. Conceivably, buffer substitution may influence El-FFF dependent exosome subfractionation. Herein, we investigate the influence of various carrier fluids including deionized (DI) water, 1X PBS, 0.308 M trehalose in DI water, and 2% isopropyl alcohol (IPA) in DI water on cyclicalEl-FFF (Cy-El-FFF) fractionation of polystyrene nanoparticle standards, melanoma exosomes sourced from a common human A375 melanoma cell line, and synthetic liposome-based exosome mimetics. Experimental Section Overview Experiments were performed in 5 parts: an exosome buffer substitution study, exosome dilution study, multiangle light scattering (MALS) exosome sizing post buffer dilution or substitution, Cy-El-FFF characterization of several similarly sized particles in substituted buffers, and finally Cy-El-FFF fractionation of exosomes in two substituted buffers. In both the substitution and dilution studies, a Mobius zetasizer (Wyatt Technology Corp.) was used to measure various characteristics including conductivity, electrophoretic mobility, static light scattering, and dynamic light scattering. The dip cell with the same quartz cuvette was used for all experiments with sample fill vol-

umes of 46 µL. Samples were checked to ensure that no bubbles were present in the dip cell prior to each measurement. The exosome size (hydrodynamic radius) was measured with dynamic light scattering, and the exosome concentration was measured using the instrument’s static light scattering detection ability. Exosome buffer substitution study Purified human A375 melanoma exosomes were isolated using differential centrifugation as described previously22, 29. Exosomes were cryopreserved in 50 mM trehalose in 1X PBS buffer until use25, 30. Exosome samples were divided into equal aliquots (control and test) and subjected to a series of buffer substitution steps using a buffer substitution spin column (PALL Life Sciences Nanosep P/N OD010C34 with 10KDa filter). The control exosome sample was substituted with 0.02 µm filtered 1X PBS, whereas the test exosome sample was alternately substituted with 0.02 µm filtered deionized water, and 0.02 µm filtered 1X PBS (Table S-1, Figure S-1). Additional details are given in the supplemental material. Exosome dilution study After completing the exosome buffer substitution study, the exosomes obtained were used as the starting point for a series of controlled exosome dilutions. This concentration (after the 3rd buffer substitution step) was termed to be a concentration equal to 1.0 arbitrary units. A 1:1 dilution was termed to be an exosome concentration of 0.5 arbitrary units and so on. Dilutions were made over the course of the test over ~2 orders of magnitude. Each sample was diluted with known volumes of 0.02 µm filtered solvent: 1X PBS for the control and DI water for the test samples. After each dilution, the sample was remeasured on the Mobius. The purpose of dilution was to establish a calibration curve relating exosome concentration to static light scattering (counts per second) obtained from the Mobius. These controlled dilutions also yielded additional electrophoretic mobility and radius effects. An extrapolation from these curve-fit data allowed calculation of the relative exosome concentrations after each buffer substitution step. Additional details are given in the supplemental material. MALS exosome sizing post buffer dilution or substitution Samples were prepared using the the buffer substitution method outlined previously (Figure S-1) cognizant that the weight of the final substituted fluid was equal to the weight of the starting unsubstituted fluid. In this test however, all samples were diluted 1:10 prior to spinning down to prevent detector saturation. An Agilent 1260 Isocratic pump supplied deionized water at 0.38 mL/min. Sequentially, the flow path included a 100 nm pre-filter, Rheodyne 9725i injection port, Gilson 119 UV/Vis detector with 5 mm path length, 12 µL quartz flow cell volume, Dawn Helios II MALS instrument, and Optilab Rex refractive index detector. An injection loop of 15 µL was used and overfilled with 35-50 µL of sample to avoid any bubbles during injection. Exosomes processed four different ways were injected into the MALS detector for sizing: exosomes diluted


Page 3 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


3

with 1X PBS, exosomes substituted with 1X PBS, exosomes substituted with DI water, and exosomes substituted with 2% isopropanol in DI water. Radii were calculated using a Zimm fit method modeled to a spherical form factor to yield geometric (physical) radii as opposed to hydrodynamic radii. Cy-El-FFF of similarly sized particles in several carrier fluids A variety of control particles within the exosome size range were used to predict the necessary parameters required to separate exosomes using small voltage Cy-El-FFF. Particles employed included carboxylate modified polystyrene spheres: FS02F lot# 10597, and PC02N lot# 10582, of radius 25.5 and 100 nm respectively (Bangs Laboratories, Inc.). The particles were diluted in DI water to obtain a final concentration of 0.01% solids for the 25.5 nm radius particles, and 0.001% solids for the 100 nm radius particles. Each concentration was maintained independently in the final solution whether the particles were mixed together or injected separately. A 3rd vesicle control included a synthetic exosome produced according to established methods22. Based upon preliminary studies31, 0.38 mL/min was used as the flow rate. A square waveform, AC frequency of 7 Hz, 50% duty cycle, and a 0 V DC offset were also applied. The AC voltage was varied from 0 V to 4 V peak. All AC voltages are reported as peak voltage in this study which is half of the peak-to-peak value32. The detectors were simultaneously monitored using a variety of software packages including LabView, Astra 5.3.4.20 (Wyatt Technologies Corp.), and a Python Script. The Cy-El-FFF device was constructed the same as described previously33. Briefly, it was constructed of parallel graphite blocks including a 178 µm high, 2 cm wide, and 64 cm long channel. Electrical stimulation and current measurement was provided using Agilent devices E3642A, 33120A, and 34401A. To change El-FFF carrier fluid, the reservoir buffer fluid was changed to the new buffer. Subsequently, the channel and entire fluid path was flushed at a rate of approximately 1 mL/min for 2 hours. For sample injection, the flow rate was set to 0.38 mL/min and the detector was initially zeroed to determine the 260 nm UV baseline for pure DI water. Direct current (DC) voltage was next applied to 1 V for ~5 minutes. Sample injection then proceeded as follows. A 50 µL aliquot of sample was loaded into the injection port of a 15 µL injection loop. Immediately post sample injection, simultaneous data acquisition began using LabView, Astra, and Python. This corresponds to time = 0 in the plots. The sample then traveled through and exited the injection loop, FFF entrance tubing, and moved into the front of the Cy-El-FFF (calculated time = 31 seconds) device. The pump was then shut off at 31 seconds, and the previously applied 1 V DC voltage was allowed to relax the particles to the retention wall for 60 seconds. The AC power supply was then turned on to the pre-set frequency and voltage condition while the DC voltage was turned off. The pump was re-started at 0.38 mL/min and ran until the desired time had passed for given peaks of interest to resolve. Afterward, the AC power supply was shut off while rinse peaks passed through the system. Data was collected and saved prior to waiting at least 10 minutes before beginning the next test run.

The carrier flow fluids employed using Cy-El-FFF were DI water, 2% Isopropanol in DI water, or a 0.308 M trehalose in DI water. Since El-FFF requires the use of very low salt concentrations, each of the tested additives and concentrations were selected in order to produce a non-ionic fluid that was also isoosmolar with a standard physiological fluid. The dilute trehalose solution was also evaluated given our previous data demonstrating the use of trehalose as a membrane protectant against detrimental electrical effects during electroporation25. In each of these fluids, a mixture of 25.5 and 100 nm radius particles were injected simultaneously into the Cy-El-FFF machine for separation at various applied AC voltages. The synthetic exosome samples were diluted to 0.1X PBS final concentration prior to injection. Their pre-injection radius in 0.1X PBS, as determined using Mobius dynamic light scattering, was found to be ~78 nm. The void time is the time of the first unretained peak as it exits the channel. The retention time is denoted by the time at which the center of each retained peak detected by UV absorbance exits the channel, and the particle radius is measured with MALS. To avoid superimposition of all signal curves, an offset was added to the graphed UV signal for clarity of presentation. Cy-El-FFF of exosomes in DI water or 2% IPA in DI water Using the spin column buffer substitution technique (Figure S-1), we substituted exosome samples into either DI water or 2% IPA in DI water. The elution buffer used during fractionation matched the substituted buffer. The sample loop size was 49 µL and the amount of sample injected to overfill the loop was 100 µL. Injection time was 36 seconds, and 1.3 V AC was applied. Other El-FFF parameters were the same as described under “Cy-El-FFF characterization of similarly sized particles in several carrier fluids,” above. Similarly, synthetic exosomes were diluted in DI water resulting in a pre-injection salt concentration of 0.1X PBS. They were injected into the Cy-El-FFF system using either DI water or 2% IPA in DI water elution buffers. Results and Discussion The present study reports the effect of buffer substitution on polystyrene nanoparticle, synthetic exosomes, and natural human A375 exosome biophysical measurements including radius and electrophoretic mobility using Cy-El-FFF. Several publications showed exosomes to resist lysis under low osmotic conditions31, 34-36. As such, we sought to determine whether our A375 melanoma exosome test population would also resist hypotonic lysis, and resulting in sample loss, when substituted into DI water, then PBS and finally back to DI water. The goal was to facilitate the development of a low salt carrier fluid for CyEl-FFF to minimize electric field neutralization. Exosome loss occurred at the same rate, whether DI water or 1X PBS was used as the substitution buffer (Figure 1). Agglomeration of the particles as a result of buffer exchange is not likely a large contributor to exosome sample loss in spin columns because the loss rate was the same whether the substitution buffer was 1X PBS, or alternating with DI water. This is supported by the relation-



Page 4 of 10

4

ship between counts per second versus exosome dilution producing linear plots with R2 values near ~ >/= 0.9 (Figure S-2.). Additionally, obtained radii using direct injection into the MALS detector, were the same regardless of whether exosomes were diluted or substituted in 1X PBS, substituted in DI water or substituted in 2% IPA in DI water (Table S-2). Diminishing exosome recovery following repeated buffer substitution was also observed. This differs from what has been reported for synthetic liposomes where no appreciable loss of liposomes was observed post spin filtration37. Taken together, loss of exosome samples using spin filters likely occurs via adherence of the vesicles to the spin column, and/or spin column filter. This is consistent with a urinary exosome study, whereby exosomes were found to adhere to the membrane of a nanomembrane concentrator38.

buffer viscosity and dielectric permittivity are often used to derive a particle’s zeta potential through either the Hückel or Smoluchowski approximations. Zeta potential subsequently is a predictor of particle colloidal stability or resistance to aggregation. In recent years, exosome zeta potentials have been reported in the literature24. Salt influences zeta potential measurements by affecting the electrophoretic mobility of the particle itself, the buffer viscosity, and also the dielectric permittivity. After the substitution process, dilution of the exosomes showed a steady change in the electrophoretic mobility of the exosomes as a function of exosome concentration (Figure 2). Conductivity measurements using the Mobius indicate that the spin column removed ~99.9% of the salt during the last buffer substitution step prior to beginning the dilutions in DI water. The trace amounts of salt that were not removed during the buffer substitution process, were also be diluted at the same rate as the exosomes. The electrophoretic mobility of substituted exosomes was different in 1X PBS versus DI water. Dilution in each fluid additionally showed that the electrophoretic mobility varied with either exosome or residual salt concentration. The change in electrophoretic mobility of exosomes with dilution was shown for the first time in this work.

Figure 1. Cumulative exosome sample lost per buffer substitution was the same for 1X PBS or DI water. (see Table S-1). Exosome recovery was reduced by 20-30% per buffer substitution step using the 10 kDa spin column. The buffer substitution process further revealed that A375 exosome size remained constant in DI water. This is consistent with the findings of others concerning the resistance of exosomes to hypotonic dialysis36, resistance to desiccation prior to TEM29, 39-40, and maintenance of stability following immersion in DI water35. To the best of our knowledge, this is the first demonstration for melanoma-derived exosome resistance to hypotonic rupture. The same might hold true for other tumor exosome types. The finding is also highly relevant to Cy-El-FFF studies in that the use of DI water as an exosome carrier fluid is ideal for reducing salt neutralization of the electric force field used to fractionate exosomes. It is anticipated that the use of DI water as a carrier fluid would facilitate the ability to resolve exosome subpopulations with higher resolution in Cy-El-FFF. Exosome radii as measured by DLS in this work supports this conclusion as they remained unaffected by dilution in 1X PBS or in DI water except at the lowest exosome concentration where the radii of exosomes diluted in 1 X PBS increased slightly (Figure S-2). This increase is consistent with the effect of measuring low particle concentrations using DLS41. Knowledge of electrophoretic mobility of particles is useful to various fields of study. In El-FFF systems, electrophoretic mobility indicates a particle’s velocity in an applied electric field, which is used to determine the particle’s ultimate elution time42. Additionally, electrophoretic mobility coupled with

Figure 2. Buffer-substituted exosomes diluted in DI water or 1X PBS show a large increase in electrophoretic mobility, or a negligible change respectively. The largest mobility occurred at high exosome concentration. Residual micromolar level salt concentration may also be responsible for this trend which was extrapolated by the U-shaped curve. A plot of the exosome electrophoretic mobility with respect to salt concentration produced a U-shaped curve (Figure 2) where the highest measured negative mobility occurs at lowmid salt conditions. In high salt conditions (1X PBS), prior to dilution, all exosome particles had negative electrophoretic mobilities of ~-1 µm cm V-1 s-1. Substitution with 1X PBS gave similar mobilities. Subsequent dilution in 1X PBS showed a slight and gradual decrease to ~-0.1 µm cm V-1 s-1. For the exosomes substituted into DI water, the mobility was quite different (~-5.5 µm cm V-1 s-1) than it had originally been in PBS (~-1 µm cm V-1 s-1). Subsequent dilution with DI water caused the mobilities to gradually decline, until they eventually approached the lower negative mobility typical of the starting salty solution (~-.1 µm cm V-1 s-1). These results indicate a


Page 5 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


5

Figure 3. Offset fractograms for polystyrene (a) and synthetic exosomes (b) separated via Cy-El-FFF in 2% isopropyl alcohol show separability and size stability under substituted conditions. Labeled radii above peaks are based upon MALS (sphere model). Horizontal axis is time in minutes and vertical axis is UV absorbance in absorbance units. Injected radii of polystyrene particle mixtures and AC voltages are labeled on each fractogram (plot a). Synthetic exosomes, ~ 78 nm radius, were injected in plot b. All tests also include a 1V DC relaxation step. Arrow (b) indicates the “crescent” shape of retention times versus applied AC voltage similar to all substituted fluids (Figure S-3). maximum in electrophoretic mobilities at a low-mid salt concentration. For the exosomes measured in this study, the maximum achieved electrophoretic mobility occurred when the salt concentration was ~500 µM PBS. For a given effective electric field, there would be an optimal salt concentration (or perhaps exosome concentration) where the mobility of the exosomes is maximal on the U-shaped curve and it occurs at low-mid concentrations. For comparision, a variety of synthetic nanoparticles in literature also exhibited a U-shaped change in their mobilities relative to salt concentration43-44. Other researchers have similarly found U-shaped curves for mobilities which were dependent upon nanoparticle concentration itself independent of salt45-46. This is the first test to our knowledge that shows the dilution change in electrophoretic mobility of exosomes with regard to salt concentration or exosome concentration. Aside from influencing the electrophoretic mobility of the particles, salt concentration also greatly affects the effective field. Previous work showed that even small amounts of salt in the 15-750 µM range28 can greatly affect separations using a DC separation voltage in El-FFF. The use of cyclical AC voltages in El-FFF may overcome some of the detrimental effects of salt because AC voltages produce a higher effective field, therefore this study used AC fields to perform separations. In AC tests, synthetic exosomes possessing similar biophysical properties as exosomes22, and a mixture of two polystyrene particles in the exosome size range were used as test particles (Figure 3 & Figure S-3). Polystyrene particles are not expected to change size due to osmosis and hence provided a firm metric to which exosomes might be compared. AC separations were performed using Cy-El-FFF in various carrier fluids including DI water, 2% isopropyl alcohol in water, and 0.308 M trehalose

in DI water. The geometric size in nm measured by MALS at the peak of each distribution is written above the distribution. The polystyrene tests confirmed a size separation at each of the applied voltages tested as shown by the presence of two peaks (one for each polystyrene particle size). Interestingly, both particle types in all carrier fluids also exhibited a crescent-shaped curve (Figure 3 & Figure S-3) with regards to retention time and applied AC voltage. The longest retention times were observed for an applied voltages of ~1.3 V AC in DI water, 2% IPA, and 0.308 M trehalose (Figure 3a, & Figure S-3a,c). The longest retention time reported for synthetic exosomes occurred at ~1.1 V AC in 2% IPA (Figure 3b) and ~ 1.3V AC in DI water and 0.308 M trehalose (Figure S-3b,c). Synthetic exosomes in 2% IPA also exhibited a crescent-shaped curve (Figure 3b). These data demonstrate that Cy-El-FFF particle retention times are minimally dependent on particle type or carrier fluid types at the tested concentrations. Moreover the optimal AC voltage for particle retention in these experiments, ~1.3 V, produced easily discernible particle fractionation peaks for synthetic nanoparticles. Synthetic exosome peaks were more spread out than the 100 nm polystyrene peaks, which is indicative of a higher synthetic exosome sample polydispersity. In the trehalose solution (Figure S-3c,d), the initial relaxation added about 2-3 minutes to the “apparent” void peak retention time and also added some level of incomplete particle separation as seen with the MALS. The fractionation of polystyrene particles in 2% IPA buffer was very similar to water, except that the retention time was approximately 30 minutes (Figure 3a) for 100 nm radius instead of 39 minutes particles in water (Figure S-3a). The difference between Cy-El-FFF fractionation of polystyrene nanoparticles (Figure 4) and synthetic exosomes (Figure 5)



Page 6 of 10

6

with varying voltage might be explained by a decreased effective electric field strength due to a higher injected salt concentration with synthetic exosomes. Synthetic exosomes were injected (15 µL) in 0.1 X PBS into a Cy-El-FFF channel volume of 2.7 mL initially filled with DI water. At a DI water flow rate of 0.38 mL/min and >20 minutes of retention time, the salt concentration from the small volume injected would be very dilute after just a few minutes. However, the larger salt content in the diluted synthetic exosome sample could result in a decreased retention time relative to either: the polystyrene particles which were diluted in DI water, or the exosomes that were fully substituted into DI water prior to injection. Thus the increase in salt might reduce the effective field by degrees until its influence is negligible. While the salt concentration may be a factor, it is also likely that the polystyrene particles have a different mobility from synthetic exosome even though their size is in the same range. Overall, particles separated in 0.308M trehalose had longer retention times than particles separated in DI water or 2% isopropyl alcohol in DI water under the conditions tested. Cy-El-FFF was clearly able to separate polystyrene particles and synthetic exosome particles by size (Figure 4 & Figure 5). MALS indicated that the particles separated into a variety of sizes as a function of applied AC voltage and time. Electrically purified particles by size could be easily collected in fractions as they drip from the Cy-El-FFF device. For synthetic exosomes, the fraction was not completely separated from the void peak, but the trend to increasing size with time was still present.

Figure 4. Superimposed fractograms for polystyrene separation in DI water using Cy-El-FFF at three AC peak voltages (square wave). Geometric Radius (dots) and MALS intensity (lines, peak plots) vs. time. Red=0V, Blue=0.5V, Green=1.3V.

Figure 5. Superimposed fractograms for synthetic exosome separation in DI water using Cy-El-FFF at three AC peak volt-

ages (square wave). Geometric radius (dots) and MALS intensity (lines, peak plots) vs. time. Brown= 0V, Green=0.5V, Red=1.3V. The fractionation profile for biological exosomes and synthetic exosomes in DI water or 2% IPA in water was similar (Figure 6). Cy-El-FFF fractionation of both types of vesicles produced easy to delineate exosome fractionation peaks representing similar sized exosomes regardless of the buffer used. The alcohol had no visible effect on exosome concentration, size, or retention time for exosomes or synthetic exosomes in this study. Exosomes in their substituted buffer did fractionate differently from synthetic exosomes. Synthetic exosomes did not achieve a baseline separation from the early peaks (minimally retained particles) as the center of their distribution was at ~18 minutes. This could have been caused by their higher salt concentration. However, the portion that did get retained at 20-40 minutes did measure to be in the same size as exosomes. The substituted exosomes did achieve a baseline separation from the early peaks, and the distribution could be found between ~23 and 42 minutes. The measured radius of the exosomes appeared to be fairly flat in the middle of the exosome peak (from ~27-40 minutes) and was ~120 nm for most of the width of the peak. An analysis of the sizes of different types of particles fractionated under similar conditions allowed an inference on the nature of the electrical fractionation. The radius of the unfractionated exosomes injected directly into the MALS detector was ~120 nm (Table S-2). This same size was uniformly found for exosomes across nearly the entire (~ from 27 to 40 minutes) 13 minute long exosome peak (Figure 6a). The polystyrene mixture consists of particles that bookend the exosome size and during this same time window, separated them into their individual component sizes (Figure 3a & Figure 4). Differential retention in time requires that particles have a difference in their respective electrophoretic mobility. This electrophoretic mobility difference may either originate from the field strength, particle’s size, charge, or interaction with the elution buffer. Assuming that the exosome size and buffer composition both remained essentially the same, the difference in retention time over the 13 minute long exosome peak must be due to differences in the total charge on the exosomes.

Figure 6. a) Fractograms of exosome separations by Cy-ElFFF using low ionic strength elution fluids (DI water or 2% IPA in DI water). 90° MALS intensity (dark orange, dark green, and dark blue) and geometric radius (light orange, light green, and light blue) versus elution time. Exosomes were substituted into the elution fluid prior to injection. b) Synthetic exosomes were diluted with DI water to a final salt concentration of 0.1X PBS prior to injection. Green is in DI water. Orange and blue are in


Page 7 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


7

2% IPA in DI water. Prior to injection, exosomes rested in substituted buffers for > 24 hours. Cy-El-FFF was run at an AC peak voltage of 1.3 V (square wave) and 7 Hz. Conclusions Buffer substitution effects on exosomes for use in Cy-El-FFF were explored. Exosome sample loss on spin-filters during buffer substitution was quantified using static light scattering extrapolation. The loss rate was 20-30% per buffer substitution, and was independent of whether exosomes were substituted into 1X PBS or DI water. Both exosomes diluted in DI water and in 1X PBS had similar measured sizes but their mobilities were very different. Electrophoretic mobility dependence upon dilution was shown for the first time for exosomes and found to be U-shaped with concentration having the largest negative value near the 500 µM PBS salt concentration. For high salt (1X PBS) concentrations, the change was small and solely due to exosome concentration. Particle aggregation as a result of buffer substitution did not seem to significantly occur because exosome size remained stable across time and dilution in both 1X PBS and DI water. A characterization of similarly sized polystyrene and synthetic exosome particles revealed that Cy-El-FFF separated these similar particles into differentially sized fractions. At a single frequency, retention time of particles exhibited a crescent-shape that varied with applied voltage, typically having a maximum retention time at 1.3 V AC for all particles tested. Three carrier fluids (DI water, 2% isopropyl alcohol, and 0.308 M trehalose) each fractionated these particles under varying AC voltage conditions, and only minor differences in their fractograms were observed. Each particle could therefore be separated with a variety of buffers with different peak retention times. Exosomes are of great interest because of the multiplicity of physiological roles they play. This work showed that exosomes having the same size, may be separated using Cy-El-FFF based upon a difference in their charge. The development of Cy-ElFFF for label-free exosome separation provides another method by which these fascinating nanovesicles may fractionated, and categorized, leading to further developments in physiological understanding, biobanking, and nanomedicine applications.

ASSOCIATED CONTENT Supporting Information The Supporting Information is available free of charge on the ACS Publications website. Experimental descriptions and additional results (Suppl.PDF)

AUTHOR INFORMATION Corresponding Authors *[email protected] or [email protected] (engineering questions) or [email protected] (extracellular vesicle/nanoparticle questions)

Author Contributions

K.E. Petersen, B. K. Gale and J.L. Hood wrote and edited the manuscript. H. Sant helped secure funding that supported this work. K.E. Petersen, F. Shiri, T. White, and G.T. Bardi performed experiments and analyzed data. All authors have given approval to the final version of the manuscript.

Notes

K.E. Petersen, B. K. Gale and H. Sant are also employees of Espira, Inc. and therefore declare a financial interest in this work

ACKNOWLEDGMENT Research was supported by NIH/NIGMS (R21-GM107894) to B.K. Gale and J.L. Hood

REFERENCES 1. Hao, S.; Moyana, T.; Xiang, J., Review: cancer immunotherapy by exosome-based vaccines. Cancer biotherapy & radiopharmaceuticals 2007, 22 (5), 692-703. 2. Saadeldin, I. M.; Oh, H. J.; Lee, B. C., Embryonic– maternal cross-talk via exosomes: potential implications. Stem Cells and Cloning : Advances and Applications 2015, 8, 103-107. 3. D'Souza-Schorey, C.; Clancy, J. W., Tumor-derived microvesicles: shedding light on novel microenvironment modulators and prospective cancer biomarkers. Genes & development 2012, 26 (12), 1287-99. 4. Peinado, H.; Aleckovic, M.; Lavotshkin, S.; Matei, I.; Costa-Silva, B.; Moreno-Bueno, G.; Hergueta-Redondo, M.; Williams, C.; Garcia-Santos, G.; Ghajar, C.; Nitadori-Hoshino, A.; Hoffman, C.; Badal, K.; Garcia, B. A.; Callahan, M. K.; Yuan, J.; Martins, V. R.; Skog, J.; Kaplan, R. N.; Brady, M. S.; Wolchok, J. D.; Chapman, P. B.; Kang, Y.; Bromberg, J.; Lyden, D., Melanoma exosomes educate bone marrow progenitor cells toward a prometastatic phenotype through MET. Nature medicine 2012, 18 (6), 883-91. 5. Lai, R. C.; Arslan, F.; Lee, M. M.; Sze, N. S.; Choo, A.; Chen, T. S.; Salto-Tellez, M.; Timmers, L.; Lee, C. N.; El Oakley, R. M.; Pasterkamp, G.; de Kleijn, D. P.; Lim, S. K., Exosome secreted by MSC reduces myocardial ischemia/reperfusion injury. Stem cell research 2010, 4 (3), 214-22. 6. Szempruch, A. J.; Dennison, L.; Kieft, R.; Harrington, J. M.; Hajduk, S. L., Sending a message: extracellular vesicles of pathogenic protozoan parasites. Nat Rev Micro 2016, 14 (11), 669675. 7. Skog, J.; Wurdinger, T.; van Rijn, S.; Meijer, D. H.; Gainche, L.; Sena-Esteves, M.; Curry, W. T., Jr.; Carter, B. S.; Krichevsky, A. M.; Breakefield, X. O., Glioblastoma microvesicles transport RNA and proteins that promote tumour growth and provide diagnostic biomarkers. Nature cell biology 2008, 10 (12), 1470-6. 8. Hood, J. L.; Wickline, S. A., A systematic approach to exosome-based translational nanomedicine. Wiley interdisciplinary reviews. Nanomedicine and nanobiotechnology 2012, 4 (4), 458-67. 9. Mathivanan, S.; Fahner, C. J.; Reid, G. E.; Simpson, R. J., ExoCarta 2012: database of exosomal proteins, RNA and lipids. Nucleic acids research 2012, 40 (Database issue), D1241-4. 10. Mathivanan, S.; Simpson, R. J., ExoCarta: A compendium of exosomal proteins and RNA. Proteomics 2009, 9 (21), 4997-5000. 11. Sunkara, V.; Woo, H.-K.; Cho, Y.-K., Emerging techniques in the isolation and characterization of extracellular



Page 8 of 10

8

vesicles and their roles in cancer diagnostics and prognostics. Analyst 2016, 141 (2), 371-381. 12. Rider, M. A.; Hurwitz, S. N.; Meckes Jr, D. G., ExtraPEG: A Polyethylene Glycol-Based Method for Enrichment of Extracellular Vesicles. Scientific reports 2016, 6. 13. Kang, D.; Oh, S.; Ahn, S.-M.; Lee, B.-H.; Moon, M. H., Proteomic Analysis of Exosomes from Human Neural Stem Cells by Flow Field-Flow Fractionation and Nanoflow Liquid Chromatography - Tandem Mass Spectrometry. Journal of Proteome Research 2008, 7 (8), 3475-3480. 14. Helwa, I.; Cai, J.; Drewry, M. D.; Zimmerman, A.; Dinkins, M. B.; Khaled, M. L.; Seremwe, M.; Dismuke, W. M.; Bieberich, E.; Stamer, W. D.; Hamrick, M. W.; Liu, Y., A Comparative Study of Serum Exosome Isolation Using Differential Ultracentrifugation and Three Commercial Reagents. PLOS ONE 2017, 12 (1), e0170628. 15. Yakimchuk, K., Exosomes: isolation and characterization methods and specific markers. MATER METHODS 2015, 5, 1450. 16. Petersen, K. E.; Manangon, E.; Hood, J. L.; Wickline, S. A.; Fernandez, D. P.; Johnson, W. P.; Gale, B. K., A review of exosome separation techniques and characterization of B16-F10 mouse melanoma exosomes with AF4-UV-MALS-DLS-TEM. Analytical and bioanalytical chemistry 2014. 17. Ashames, A. A. Development of separation methods to produce uniform exosomes subpopulations using field-flow fractionation techniques. PhD, Colorado School of Mines, 2015. 18. Sitar, S.; Kejžar, A.; Pahovnik, D.; Kogej, K.; TušekŽnidarič, M.; Lenassi, M.; Žagar, E., Size characterization and quantification of exosomes by asymmetrical-flow field-flow fractionation. Analytical chemistry 2015, 87 (18), 9225-9233. 19. Kantak, A.; Srinivas, M.; Gale, B., Characterization of a microscale cyclical electrical field flow fractionation system. Lab on a chip 2006, 6 (5), 645-54. 20. Tasci, T. O.; Johnson, W. P.; Fernandez, D. P.; Manangon, E.; Gale, B. K., Biased cyclical electrical field flow fractionation for separation of sub 50 nm particles. Analytical chemistry 2013, 85 (23), 11225-32. 21. Whatley, H., Basic Principles and Modes of Capillary Electrophoresis. In Clinical and Forensic Applications of Capillary Electrophoresis, J. R. Petersen, A. A. M., Ed. Humana Press Inc.: Totowa, NJ, 2001. 22. Hood, J. L.; San, R. S.; Wickline, S. A., Exosomes released by melanoma cells prepare sentinel lymph nodes for tumor metastasis. Cancer Res 2011, 71 (11), 3792-801. 23. Kato, K.; Kobayashi, M.; Hanamura, N.; Akagi, T.; Kosaka, N.; Ochiya, T.; Ichiki, T., Electrokinetic Evaluation of Individual Exosomes by On-Chip Microcapillary Electrophoresis with Laser Dark-Field Microscopy. Jpn J Appl Phys 2013, 52 (6). 24. Weiner, B. B. Measuring the Size & Surface Charge of Exosomes, Microvesicles and Liposomes. http://www.brookhaveninstruments.com/literature/pdf/Measuring %20the%20Size%20and%20Surface%20Charge%20of%20Exoso mes.pdf (accessed 11/18/2013). 25. Hood, J. L.; Scott, M. J.; Wickline, S. A., Maximizing exosome colloidal stability following electroporation. Analytical biochemistry 2014, 448, 41-9. 26. Akagi, T.; Kato, K.; Kobayashi, M.; Kosaka, N.; Ochiya, T.; Ichiki, T., On-chip immunoelectrophoresis of extracellular vesicles released from human breast cancer cells. PloS one 2015, 10 (4), e0123603.

27. Parolini, I.; Federici, C.; Raggi, C.; Lugini, L.; Palleschi, S.; De Milito, A.; Coscia, C.; Iessi, E.; Logozzi, M.; Molinari, A.; Colone, M.; Tatti, M.; Sargiacomo, M.; Fais, S., Microenvironmental pH is a key factor for exosome traffic in tumor cells. The Journal of biological chemistry 2009, 284 (49), 3421122. 28. Ornthai, M.; Siripinyanond, A.; Gale, B. K., Effect of Ionic and Non-Ionic Carriers in Electrical Field-Flow Fractionation. Analytical chemistry 2015. 29. Hood, J. L.; Pan, H.; Lanza, G. M.; Wickline, S. A., Paracrine induction of endothelium by tumor exosomes. Laboratory investigation; a journal of technical methods and pathology 2009, 89 (11), 1317-28. 30. Bosch, S.; de Beaurepaire, L.; Allard, M.; Mosser, M.; Heichette, C.; Chretien, D.; Jegou, D.; Bach, J. M., Trehalose prevents aggregation of exosomes and cryodamage. Sci Rep 2016, 6, 36162. 31. Petersen, K. E.; Ornthai, M.; Hood, J. L.; Wickline, S. A.; Gale, B. K. In Electrical Field Flow Fractionation of Exosomes, 17th International Symposium on Field-and Flow-based Separation, Salt Lake City, October 12-16; Salt Lake City, 2014; p 78. 32. Alternating Current (AC). http://physicsnet.co.uk/alevel-physics-as-a2/current-electricity/alternating-current-ac/ (accessed 5-5-17). 33. Caldwell, K. D.; Gao, Y. S., Electrical field-flow fractionation in particle separation. 1. Monodisperse standards. Analytical chemistry 1993, 65 (13), 1764-1772. 34. Willms, E.; Johansson, H. J.; Mäger, I.; Lee, Y.; Blomberg, K. E. M.; Sadik, M.; Alaarg, A.; Smith, C. I. E.; Lehtiö, J.; El Andaloussi, S.; Wood, M. J. A.; Vader, P., Cells release subpopulations of exosomes with distinct molecular and biological properties. Scientific Reports 2016, 6, 22519. 35. Chernyshev, V. S.; Rachamadugu, R.; Tseng, Y. H.; Belnap, D. M.; Jia, Y.; Branch, K. J.; Butterfield, A. E.; Pease III, L. F.; Bernard, P. S.; Skliar, M., Size and shape characterization of hydrated and desiccated exosomes. Analytical and bioanalytical chemistry 2015, 407 (12), 3285-3301. 36. Hood, J. L., Post isolation modification of exosomes for nanomedicine applications. Nanomedicine (London, England) 2016, 11 (13), 1745-56. 37. Lane, R. E.; Korbie, D.; Anderson, W.; Vaidyanathan, R.; Trau, M., Analysis of exosome purification methods using a model liposome system and tunable-resistive pulse sensing. Scientific Reports 2015, 5, 7639. 38. Cheruvanky, A.; Zhou, H.; Pisitkun, T.; Kopp, J. B.; Knepper, M. A.; Yuen, P. S. T.; Star, R. A., Rapid isolation of urinary exosomal biomarkers using a nanomembrane ultrafiltration concentrator. Am J Physiol-Renal 2007, 292 (5), F1657-F1661. 39. Thery, C.; Clayton, A.; Amigorena, S.; Raposo, G., Isolation and Characterization of Exosomes from Cell Culture Supernatants and Biological Fluids. Current Protocols in Cell Biology 2006, 3.22.1-3.22.29. 40. Sahoo, S.; Klychko, E.; Thorne, T.; Misener, S.; Schultz, K. M.; Millay, M.; Ito, A.; Liu, T.; Kamide, C.; Agrawal, H.; Perlman, H.; Qin, G.; Kishore, R.; Losordo, D. W., Exosomes from human CD34(+) stem cells mediate their proangiogenic paracrine activity. Circ Res 2011, 109 (7), 724-8. 41. Panchal, J.; Kotarek, J.; Marszal, E.; Topp, E. M., Analyzing subvisible particles in protein drug products: a comparison of dynamic light scattering (DLS) and resonant mass measurement (RMM). The AAPS journal 2014, 16 (3), 440-451.


Page 9 of 10 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


9

42. Tasci, T. O.; Johnson, W. P.; Fernandez, D. P.; Manangon, E.; Gale, B. K., Particle Based Modeling of Electrical Field Flow Fractionation Systems. Chromatography 2015, 2 (4), 594-610. 43. Elimelech, M.; O'Melia, C. R., Effect of electrolyte type on the electrophoretic mobility of polystyrene latex colloids. Colloids and Surfaces 1990, 44, 165-178. 44. Goff, J. R.; Luner, P., Measurement of colloid mobility by laser Doppler electrophoresis: the effect of salt concentration on

particle mobility. Journal of colloid and interface science 1984, 99 (2), 468-483. 45. Medrzycka, K., The effect of particle concentration on zeta potential in extremely dilute solutions. Colloid and polymer science 1991, 269 (1), 85-90. 46. Tantra, R.; Schulze, P.; Quincey, P., Effect of nanoparticle concentration on zeta-potential measurement results and reproducibility. Particuology 2010, 8 (3), 279-285.



Page 10 of 10

10

Authors are required to submit a graphic entry for the Table of Contents (TOC) that, in conjunction with the manuscript title, should give the reader a representative idea of one of the following: A key structure, reaction, equation, concept, or theorem, etc., that is discussed in the manuscript. Consult the journal’s Instructions for Authors for TOC graphic specifications.


Exosome Isolation: Cyclical Electrical Field Flow Fractionation in Low

Recommend Documents