Rapid Isolation and Detection of Exosomes and Associated

19 Jun 2017 - ABSTRACT: Exosomes found in the circulation are a primary source of important cancer-related RNA and protein biomarkers that are expecte...
2 downloads 0 Views 11MB Size
Download PDF
Rapid Isolation and Detection of Exosomes and Associated Biomarkers from Plasma Stuart D. Ibsen,† Jennifer Wright,† Jean M. Lewis,† Sejung Kim,‡ Seo-Yeon Ko,† Jiye Ong,† Sareh Manouchehri,§ Ankit Vyas,† Johnny Akers,∥ Clark C. Chen,∥ Bob S. Carter,∥ Sadik C. Esener,† and Michael J. Heller*,†,§ †

Department of Nanoengineering, ‡Materials Science and Engineering, §Department of Bioengineering, and ∥Moores Cancer Center, University of California San Diego, La Jolla, California 92093, United States ABSTRACT: Exosomes found in the circulation are a primary source of important cancer-related RNA and protein biomarkers that are expected to lead to early detection, liquid biopsy, and point-of-care diagnostic applications. Unfortunately, due to their small size (50−150 nm) and low density, exosomes are extremely difficult to isolate from plasma. Current isolation methods are time-consuming multistep procedures that are unlikely to translate into diagnostic applications. To address this issue, we demonstrate the ability of an alternating current electrokinetic (ACE) microarray chip device to rapidly isolate and recover glioblastoma exosomes from undiluted human plasma samples. The ACE device requires a small plasma sample (30−50 μL) and is able to concentrate the exosomes into high-field regions around the ACE microelectrodes within 15 min. A simple buffer wash removes bulk plasma materials, leaving the exosomes concentrated on the microelectrodes. The entire isolation process and on-chip fluorescence analysis is completed in less than 30 min which enables subsequent on-chip immunofluorescence detection of exosomal proteins, and provides viable mRNA for RT-PCR analysis. These results demonstrate the ability of the ACE device to streamline the process for isolation and recovery of exosomes, significantly reducing the number of processing steps and time required. KEYWORDS: dielectrophoresis, exosomes, blood plasma, cancer biomarkers, extracellular vesicles

E

other components in the blood requires major time and effort. Current state-of-the-art sample preparation kits and procedures for exosome isolation often require multiple processing steps that include extensive ultracentrifugation and incubation steps ranging from several hours to overnight.12 Other exosome recovery methods include filtration steps that fail to exclude contaminating plasma proteins.19 Ultrasound-based methods have also been developed to recover exosomes from cell culture media and packed red blood cell (pRBC) units but have not been demonstrated in undiluted plasma.20 In the case of immunoaffinity isolation methods, exosome isolation may exclude potentially important populations and the process becomes dependent upon antibody selectivity, specificity, and the affinity/binding constant.22−24 Even more recent nanoarray-based separation devices require up to 60 h of processing time to isolate exosomes from simple buffer samples.25 Additionally, many of these time-consuming multistep processes and methods can damage the exosomes8 and reduce overall collection efficiency. Thus, a major challenge exists to

xtracellular vesicles such as exosomes and other nanoscale entities released into circulation contain mRNA, miRNA, rRNA, DNA, and specific protein biomarkers that are important for early cancer detection and diagnostics and also provide information on the tissue and cell of origin.1−3 Because cell-free RNA in the blood is rapidly destroyed by endogenous RNase activity,4,5 exosomes and other extracellular vesicles (EVs) serve to encapsulate and protect RNA from RNases. Thus, vesicles are a primary source of these labile biomarkers for cancer and other disease diagnostics.2,3,6 Exosomes are small lipid/protein vesicles ranging from 50 to 150 nm in diameter7−9 that are actively secreted by both healthy and cancerous cells in vitro.10 Exosomes have also been found in vivo in various bodily fluids10 including the blood.11,12 Generally, higher levels of circulating exosomes and EVs are associated with cancer or other disease pathology.13,14 The active and passive cellular mechanisms by which RNA, DNA, and specific proteins become packaged or associated with exosomes are still not completely understood.10,15−18 Exosomes must be isolated from the blood (plasma or serum) sample before their associated RNA and protein biomarkers can be analyzed. Unfortunately, due to their small size and low buoyant density, the isolation of exosomes from © 2017 American Chemical Society

Received: January 24, 2017 Accepted: June 19, 2017 Published: July 3, 2017 6641

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

www.acsnano.org

Article

ACS Nano

In this study, we demonstrate that the ACE microarray chip device allows the preferential separation of glioblastoma exosomes from undiluted plasma based on the difference between the dielectric properties of the exosomes and the surrounding plasma. The complete process requires less than 30 min, which includes the addition of the plasma sample, AC separation, a washing step, and standard fluorescence detection of the exosomal membrane and RNA. For this initial study, exosomes and EVs collected from a cultured glioblastoma cell line were spiked into both buffer and human plasma in order to eliminate any uncertainties concerning the nature of the vesicles themselves and the final experimental results. Scanning electron microscopy (SEM) was used to verify the presence of exosomes and EVs on the ACE microelectrodes and their physical structure. On-chip (in situ) fluorescence analysis of the exosomes and EVs and associated RNA was also carried out along with immunofluorescence analysis to identify exosomespecific CD63 and TSG101 proteins. Exosomes and EVs were then eluted from the ACE chip, the RNA was released, and RTPCR was carried out to identify β-actin mRNA and glioblastoma-specific mutated EGFRvIII mRNA. ACE devices represent a powerful technology for cancer diagnostics that is particularly well-suited for the rapid isolation of exosomes and their associated RNA and proteins. It can also be applied to other EVs and nanoscale entities and their important biomarkers as well. This technology is setting the stage for rapid seamless sample-to-answer liquid biopsy, cancer patient therapy monitoring, and ultimately early disease detection.

develop a recovery method that (1) does not require dilution of the blood, plasma, or serum sample; (2) limits mechanical damage to the exosomes; (3) significantly reduces the number of processing steps and time between sample collection and exosome isolation; (4) does not depend on antibody affinity binding; and (5) maintains the viability of the exosomal RNA and protein biomarkers for subsequent detection, identification, and analysis.26,27 Alternating current electrokinetic (ACE) microarray chip devices have considerable potential for meeting the challenges of exosome isolation with subsequent analysis of the associated RNA, DNA, and protein biomarkers. In earlier work, such ACE devices were successfully used for the isolation of high molecular weight (hmw) DNA, polystyrene nanoparticles, mitochondria, and viruses from high conductance buffer solutions,28−32 undiluted blood, plasma, and serum.31,33,34 In more recent work, ACE devices were used to isolate and detect cell-free (cf) DNA biomarkers from chronic lymphocytic leukemia patient blood and plasma samples.31,33−35 ACE devices have also been used to rapidly isolate drug delivery nanoparticles, including low buoyant density liposome-based nanoparticles, from undiluted plasma.36 Such results strongly suggested that these devices might be ideal for isolating exosomes from blood, plasma, and serum samples. The dielectrophoretic (DEP) separation force generated by the ACE microarray is created by the application of an alternating current (AC) electric field.37 Nanoparticles and other nanoscale entities are attracted to the DEP high-field regions around the circular microelectrode edges, as shown in Figure 1. Cells and larger entities are pulled into the DEP low-

RESULTS AND DISCUSSION For use in this study, well-characterized glioblastoma EVs were recovered from U87-EGFRvIII cell culture media using conventional procedures. Exosomes and other EVs are difficult to separate from one another, so both exosomes and other types of EVs were present in the recovered samples. These exosomes and EVs provided known genetic and protein biomarker targets and eliminated any uncertainties concerning the origin of the vesicles. Exosomes can be differentiated from other EVs by their size range, as well as by specific protein biomarkers. Transmission electron microscopy (TEM) was used to verify size range and other morphological characteristics of the glioblastoma exosomes and EVs. Figure 2 shows

Figure 1. Schematic of the microelectrode array chip showing the cross-sectional and top views of a single electrode. Over 1000 electrodes can be in a single device. The DEP high-field regions where particles are collected are shown within the dotted lines. The darker color of the silicon dioxide layer and electrode in the top view represents the overlying transparent porous hydrogel layer.

field regions between the electrodes. Finally, small biomolecules, cations, and anions remain relatively unaffected by the DEP field. The DEP force arises from the difference between the dielectric properties of the nanoparticles and surrounding fluid (blood, plasma, serum, etc.).37 The dielectric constant determines how quickly charges will move through a material in response to a change in the external electric field. These differences cause the charges in the fluid medium and in the nanoparticles to reorient themselves at different speeds in response to the changing external electric field, creating momentary dipoles across the nanoparticles. In a nonuniform electric field, these dipoles create the forces that pull the nanoparticles into the DEP high-field regions.

Figure 2. TEM images of glioblastoma EVs that were recovered from the U87-EGFRvIII cell culture media. The EVs appear as spherical objects that ranged in size from about 50 to 150 nm.

representative images of the EVs that ranged in diameter from about 50 to 150 nm, which falls within the established size range for exosomes.17 These well-characterized glioblastoma exosomes and EVs were then spiked into normal human plasma and buffer to make reproducible samples for demonstrating the ability of ACE devices to rapidly isolate, recover, and detect 6642

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano

Figure 3. Top and cross-sectional views of an ACE microelectrode showing glioblastoma exosomes and EVs collected from 0.5× PBS buffer. (A, B, and C) Top view of a control microelectrode with no sample showing the smooth topology of the microelectrode edge, the electrode surface, and the SiO2 insulating layer. (D, E, and F) Top view image of a microelectrode after DEP collection of EVs from 0.5× PBS buffer. Many of these EVs fall within the size range for exosomes. The ACE chip was washed and then freeze-dried before SEM imaging. Individual EVs and EV clusters were seen at the electrode edge. (G) Wide-field view of an electrode that has been fractured to allow for cross-sectional imaging. This image also shows some separation and missing sections of the hydrogel layer that covers the platinum microelectrode. (H and I) Collected material at the microelectrode edge showing individual EVs and clusters of EVs, many of which fall within the size range for exosomes. (J and K) Magnified images of panels H and I showing the morphology of individual exosomes and EVs preserved by the freezedrying process. The exosomes and EVs are located in the DEP high-field regions of the ACE chip.

exosomes. Figure 3G shows an ACE chip that has been broken, allowing cross-sectional images of the microelectrodes to be made. Some of the separation and missing sections of the hydrogel layer that covers the platinum microelectrode are due to the freeze-drying process. The magnified images in Figure 3H,I provide a better view of EVs around the edges of the cross-sectioned microelectrode. Finally, the highly magnified images in Figure 3J,K provide the best views of both individual and clusters of EVs, many of which fall within the size range for exosomes. The freeze-drying process appears to preserve the original shape of the exosomes. All the exosomes and EVs appear around the microelectrode edges where the DEP highfield regions were strongest when the AC voltage was applied. In further experiments, glioblastoma exosomes and EVs and 110 nm polystyrene nanoparticles were separately spiked into normal human plasma. These samples were then loaded onto the ACE microarray chip, an AC field was applied, and isolation and collection of the exosomes and EVs were carried out. A wash buffer was then run through the ACE device followed by immediate freeze-drying or air-drying. SEM was used to analyze

exosomes and EVs and their associated RNA and protein biomarkers. For the main study, glioblastoma exosomes and EVs were first spiked into 0.5× phosphate-buffered saline (PBS) buffer, and the sample was then loaded onto the ACE microarray chip. An AC field was then applied to the chip to cause the isolation and collection of the exosomes and EVs. A wash buffer was then run through the device to remove any particles not pulled into the DEP high-field regions. The chip was then immediately freeze-dried to preserve the collected vesicles. SEM was then used to analyze the surface of the chip in order to visualize the vesicles and verify their presence on the microelectrodes. Figure 3A−C show the control microelectrode surface before the sample was applied. Figure 3D,E show the chip after collection and washing, with accumulation of exosomes and EVs around the edges of the microelectrode. Some distortion of the hydrogel layer covering the platinum microelectrodes is also observed, which is due to the freeze-drying process. The magnified image in Figure 3F clearly shows individual EVs as well as clusters of EVs that fall within the size range for 6643

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano

Figure 4. Top and cross-sectional views of the ACE microelectrode chip showing a collection of exosomes, EVs, and 110 nm nanoparticles from undiluted human plasma samples. After application of the AC field, the chips were washed and then dried for SEM imaging. (A, B, and C) Top surface images of a microelectrode that was processed using freeze-drying techniques. The hydrogel layer covering the platinum microelectrode has been ruptured. Panels B and C show materials collected from the plasma forming a rounded structure that fills the DEP high-field regions. These collected materials include plasma protein aggregates, cell-free recirculating DNA, exosomes, and EVs. (D, E, and F) Cross-sectional image of a microelectrode from an ACE chip that has been broken in half, showing collection from a sample containing 110 nm polystyrene nanoparticles in normal human plasma. The polystyrene nanoparticles have been collected along with other plasma materials in the same region as exosomes and EVs and are easier to distinguish in the collected materials than the exosomes and EVs. Frame F shows a higher magnification image view of the solid nanoparticles that have become covered and embedded in the other materials collected from the plasma. (G, H, and I) Magnified images of the microelectrode edge prepared using an air-drying technique. Here the collected layer of plasma material has collapsed into a thinner film. Individual solid polystyrene nanoparticles are even more clearly visible along the microelectrode edge in the same location that exosomes and EVs should be collected.

protruding from this thin film. As the next set of biomarker analysis experiments will clearly show, the exosomes are also among the collected materials from the plasma. In order to better track the glioblastoma exosomes and EVs through the entire ACE device collection process, they were fluorescently labeled with the red fluorescent PKH26 membrane dye before spiking them into the plasma samples. An initial bright-field image of a section of the ACE microarray chip surface shows the light-colored circular microelectrodes that are 60 μm in diameter (Figure 5A). Exosomes and EVs labeled with red PKH26 dye are visible as red fluorescent coloration in the bulk plasma sample, and some larger aggregates of exosomes and EVs are visible as red spots (Figure 5B). After the AC field was applied for 10 min at 10 Vpp and 15 kHz, the exosomes and EVs were isolated and concentrated in the DEP high-field regions around the edges of the microelectrodes where the nonuniform electric field strength was highest (Figure 5C). With the AC field still on, the exosomes and EVs remained concentrated in the DEP highfield regions, while a 1× TE buffer wash was used to remove the bulk plasma and any other materials that collected in the DEP low-field regions of the ACE chip (Figure 5D). This experiment clearly demonstrates by fluorescent imaging the collection and concentration of the exosomes and EVs along with the other plasma materials and nanoparticles into the DEP high-field regions of the ACE microarray device.

the surface of the ACE chip. Figure 4 shows SEM images of collected materials around the microelectrode edges from undiluted human plasma spiked with glioblastoma exosomes and EVs. Figure 4A−C show SEM images of the top view of the ACE chip microelectrodes after recovery from undiluted human plasma using a freeze-drying technique to preserve the collected materials. The plasma components, exosomes, and EVs form a rounded layer of aggregated material that fills the DEP highfield region around the microelectrode edges. This material collected from the plasma includes protein aggregates, cell-free circulating DNA, and cellular debris along with the exosomes and EVs. Because the exosomes and EVs are relatively soft and compressible, they were not clearly visible within the collected plasma material layer. In further studies, polystyrene nanobeads (110 nm diameter) were spiked into plasma samples in order to provide structures that have similar collection characteristics to exosomes but are more solid. These solid nanoparticles become more clearly distinguishable in cross-sectional views of the microelectrodes that were produced by breaking the ACE chip in half as shown in Figure 4D−F. Figure 4 F shows a magnified image where the solid spherical nanoparticles can be seen covered and partially embedded in other materials collected from the plasma. To better see where the nanobeads themselves were being collected, the ACE chip was air-dried instead of being freeze-dried, which caused the collected plasma material to collapse into a thinner film along the electrode edge, as shown in Figure 4G−I. The solid nanoparticles are seen 6644

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano

The RNA Select dye is membrane permeable and able to cross the vesicle membrane to stain RNA that is inside. Exosomes were allowed to incubate for 10 min with RNA Select dye before the excess dye was washed away. Figure 5E shows the composite image where the RNA stained by the RNA Select dye appears as green fluorescent areas within the red exosomes and EVs. Not all the exosomes and EVs show the presence of stained RNA, which may be due to this procedure not allowing enough time for the dye to reach and stain all the RNA that is actually present. Finally, by applying a series of low-frequency electric pulses the exosomes and EVs are repelled from the DEP high-field areas and eluted from the chip using a 1× TE wash buffer. Figure 5F shows an image taken after the exosomes and EVs were eluted from the chip, demonstrating that a clear majority of the purified exosomes and EVs can be recovered from the ACE device for further analysis. The complete process from addition of the exosome and EV spiked plasma sample to elution of the vesicles from the ACE device took only about 30 min. In further experiments, on-chip (in situ) immunofluorescence analyses were carried out to detect specific exosomal protein biomarkers after isolation and collection of the glioblastoma exosomes and EVs from undiluted plasma samples (Figure 6). Glioblastoma exosomes are known to exhibit high levels of both the transmembrane protein CD63 and the internal protein TSG101. Because TSG101 is an internal protein, some membrane permeabilization is required before it can be detected by immunofluorescence.18 Thus, the detection of CD63 and TSG101, along with the expected minimal presence of the endoplasmic reticulum-specific protein calnexin, provides a specific pattern of protein expression that allows exosomes to be differentiated from other extracellular vesicles.18 In the first of these experiments, on-chip (in situ) immunofluorescence analysis of the exosomes collected by application of the AC electric field showed high levels of CD63 on the outside of the exosomal membranes after treatment with the green fluorescent FITC-conjugated anti-CD63 antibody (Figure 6B). In control experiments using U87 cell lysate materials isolated by the same ACE procedure, no detectable fluorescence was observed after treatment with the green fluorescent FITC-conjugated antiCD63 antibody (Figure 6D), indicating that the green signal in Figure 6B was from exosomes and not from cellular debris that might have been collected from the cell culture media. No calnexin was detected on exosomes collected by the AC electric field after treatment with a calnexin-specific primary antibody and red fluorescent Alexa Fluor 594-conjugated secondary antibody (Figure 6F), indicating that cellular debris was not collected from the cell culture media along with the exosomes. Red fluorescence was strongly detected after treatment with a calnexin-specific primary antibody and red fluorescent Alexa Fluor 594-conjugated secondary antibody (Figure 6H) on the U87 cell lysate materials, indicating that the immunostaining technique was functioning properly and that the lack of red signal in Figure 6D truly shows a lack of cellular debris collection from cell culture media. Note: Figure 6A, C, E, and G are the corresponding bright-field images of the ACE microelectrodes. As shown again in Figure 7, on-chip (in situ) labeling with the green fluorescent FITC-conjugated anti-CD63 antibody demonstrates that CD63 protein can be directly detected on the isolated glioblastoma exosome surface (Figure 7C and D). However, the TSG101 protein, which is located inside the exosomes,18 required a permeabilization treatment with

Figure 5. Bright-field and fluorescence images taken at the different stages of ACE isolation and recovery of glioblastoma exosomes and EVs from undiluted human plasma. (A) Bright-field image of the ACE microarray chip surface. Each of the circles is an individual 60 μm diameter platinum microelectrode overcoated with a thin porous hydrogel layer. (B) Fluorescence image before the AC field was applied showing the red fluorescent exosomes and EVs labeled with the PKH26 dye scattered across the chip. Aggregates of the exosomes and EVs are visible as the larger red dots. (C) Fluorescence image after the AC field was applied for 10 min at 10 Vpp and 15 kHz, showing the red fluorescent exosomes and EVs now isolated and concentrated around the edges of the microelectrodes where the DEP high-field region is strongest. (D) Fluorescence image taken after the bulk plasma and other materials are removed by a 10 min wash with 1× TE buffer, showing the captured exosomes and EVs are still concentrated in the DEP highfield region on the microelectrodes. (E) Composite red and green fluorescent image of the captured exosomes and EVs after on-chip (in situ) staining of the RNA with RNA Select dye. The RNA appears as the green fluorescent areas that are present in about 25% of the visible red fluorescent exosomes and EVs. (F) Final fluorescence image shows the ACE chip after the isolated exosomes were repelled from the chip surface by applying a series of lowfrequency electric pulses and eluted by a buffer wash for further analysis.

To now demonstrate on-chip (in situ) the presence of RNA in the isolated exosomes and EVs, the green fluorescent RNAspecific dye RNA Select was included in a second wash buffer. 6645

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano

Figure 6. On-chip (in situ) selective antibody labeling of ACEisolated glioblastoma exosomes with selective labeling of ACEisolated U87 cell lysate materials as a control. (A) Bright-field image of the ACE chip microelectrodes, which corresponds to panel B. (B) Fluorescence image showing the on-chip labeling of glioblastoma exosomes when using the green FITC anti-CD63 antibody. (C) Bright-field image of ACE chip microelectrodes corresponding to the D image. (D) Negative control for the FITCconjugated anti-CD63 antibody showing no detectable labeling of U87 cell lysate materials collected by DEP. (E) Bright-field image of the microelectrodes corresponding to the F image. (F) Negative control for the calnexin-specific primary antibody (AlexaFluor 594conjugated secondary antibody) showing no detectable labeling of glioblastoma exosomes isolated by DEP. (G) Bright-field image of the ACE microelectrodes corresponding to the H image. (H) Fluorescence image showing on-chip labeling of the U87 cell lysate materials when using calnexin-specific primary antibody and AlexaFluor 594-conjugated secondary antibody.

Figure 7. On-chip (in situ) labeling of ACE-isolated glioblastoma exosomes with green fluorescent FITC-conjugated anti-CD63 antibody before and after saponin treatment to study the effects of membrane permeabilization on labeling of the internal protein TSG101 by anti-TSG101 antibodies. (A and B) Bright-field images of the circular microelectrodes. (C and D) Fluorescence images showing the labeling of the external membrane protein CD63 by the green fluorescent FITC-anti CD63 antibody with and without saponin. (E) Fluorescence image with saponin permeabilization, showing labeling of internal TSG101 protein by the anti- TSG101specific primary antibody followed by red-fluorescent AlexaFluor 594 secondary antibody. (F) Fluorescence image without saponin permeabilization showing no fluorescence labeling using the same conditions as in panel E. This indicates that the DEP collection process does not damage the exosome membrane.

device. Additionally, these experiments demonstrate the added advantage of carrying out on-chip (in situ) immunofluorescence analysis directly on the ACE microarray device. We next investigated the ability of the ACE device to isolate exosomal and EV RNA biomarkers. U87 glioblastoma exosomes and EVs contain a specific mutated mRNA for the EGFRvIII gene, which provides a further verification of their identity. Analysis of mRNA was carried out on the glioblastoma exosomes and EVs, which had been recovered from undiluted plasma after application of the AC electric field and washing steps. The purified exosomes and EVs were eluted from the ACE microarray chip using a reverse electric field and a second fluid wash. Different procedures were used to release RNA from the exosomes and EVs before RT-PCR analysis. Initially, an RNA isolation kit (Exiqon) procedure was used to release the

saponin before detection by labeling with the primary antiTSG101 antibody and red fluorescent AlexaFluor 594 secondary antibody (Figure 7E and F). Without saponin permeabilization the intact exosome membranes did not allow the antibodies to have access to the inside of the exosomes (Figure 7F). This indicates that the membranes of the collected exosomes were intact and were not damaged during the DEP collection process. Saponin was chosen because it creates small holes in the cell membrane, leaving the rest of the membrane intact.38 The TSG101 is still attached to the inside of this permeabilized membrane. The detection of the exosome specific pattern of external CD63 membrane protein and internal TSG101 protein17 further confirms the presence of glioblastoma exosomes among the collected EVs by the ACE 6646

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano mRNA from the exosomes and EVs. As shown in Figure 8A, the presence of mutated EGFRvIII mRNA in exosomes and EVs isolated from plasma by the ACE device was confirmed by RT-PCR followed by end-point PCR.

Figure 9. Results showing β-actin mRNA detection by qPCR/RTPCR analysis performed on RNA from glioblastoma exosomes and EVs isolated from plasma using the ACE device. qPCR amplification was observed for the β-actin positive control (green curve), the glioblastoma exosomes and EV positive control (blue curve), and the glioblastoma exosomes and EVs isolated by ACE from plasma (red curve). No amplification was observed for the negative control or for normal plasma without spiked exosomes and EVs (brown and purple).

Figure 8. Gel analysis results for RT-PCR and end-point PCR performed on RNA from glioblastoma exosomes and EVs isolated from plasma by the ACE device, using primers specific for the mutated EGFRvIII mRNA. (A) PCR results show the specific 181 bp amplicon for the mutated EGFRvIII mRNA was produced for the positive control, the exosome and EV control, and the exosomes and EVs isolated from plasma. The negative control is a normal plasma sample without spiked glioblastoma exosomes and EVs that went through the same DEP isolation procedure as the spiked samples. This control did not show the presence of mutated EGFRvIII mRNA. (B) RT-PCR and end-point PCR gel analysis results using heat and varying concentrations of Tween 20 surfactant to release EGFRvIII mRNA from the DEP-isolated exosomes and EVs. PCR results show the specific 181 bp amplicon for EGFRvIII mRNA was produced for the positive controls, the Tween 20 treatments (0.8−0.1%), and the heat treatment.

as well. As shown in Figure 10, endogenous exosomes can be collected from a breast cancer patient plasma sample and detected using fluorescence antibody staining for CD63. The green fluorescence signal shows the collection of exosomes that were naturally present in circulation. This also shows that cancer patients can have an exosome concentration level in circulation that allows the exosomes to be collected and detected using this method. Based on this study, Figure 11 shows a stylized representation for the ACE chip collection of different plasma components including exosomes along with their associated protein and RNA biomarkers. The blue AC field lines become more concentrated at the microelectrode edges, representing the DEP high-field regions. This preferentially attracts bionanoparticles of a certain size range and material composition into the DEP high-field regions that include exosomes, EVs, and circulating DNA nanoparticles. Larger particles, such as cells, are not attracted into the DEP high-field regions and are removed from the ACE chip along with smaller biomolecules during the washing step.

In further mRNA experiments, faster and simpler treatments were used to release RNA from exosomes and EVs, which had been isolated from plasma by the ACE device. This included heat treatment to destabilize the plasma membrane of the exosomes and EVs and the use of different concentrations of Tween 20 surfactant (0.1% to 0.8%). The RT-PCR and endpoint PCR gel analysis results showed that both the heat and Tween 20 surfactant procedures can release mRNA and did not interfere with RT-PCR analysis (Figure 8B). In addition to these procedures being faster and easier than the classical protocols, they are also more compatible with “on-chip” procedures and most certainly produce less degradation of the RNA, which is a relatively labile biomolecule. Finally, RT-PCR and qPCR procedures were used to show that β-actin mRNA, a general (housekeeping) mRNA, was also present in the exosomes and EVs isolated from plasma using the ACE devices (Figure 9). Thus, in addition to specific protein biomarkers, both a specific mRNA biomarker for mutated EGFRvIII and a general mRNA biomarker for β-actin were detected and identified in exosomes and EVs, which could be rapidly isolated directly from plasma using the ACE microarray devices. This DEP method can be used to collect not only spiked exosomes from human plasma but also endogenous exosomes

CONCLUSIONS In this study, we conclusively demonstrated that the ACE microarray chip device provides a rapid and simple three-step process to efficiently isolate and recover glioblastoma exosomes from undiluted human plasma in less than 30 min. The presence of isolated exosomes and EVs on the ACE microelectrodes was first verified by SEM; second by on-chip (in situ) fluorescence detection of the exosome and EV membranes prelabeled with PKH26 dye and encapsulated RNA with RNA Select dye; and third by on-chip (in situ) immunofluorescence detection of the exosome-specific external CD63 protein and internal TSG101 protein. The ability to 6647

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano

Figure 10. Collection and detection of endogenous CD63-positive exosomes from breast cancer patient plasma. (A) Bright-field image of the circular microelectrodes. (B) Fluorescence image showing the presence of CD63 at the electrode edges, detected using mouse anti-CD63 antibody, followed by Alexa Fluor 488 (green) anti-mouse secondary antibody. This confirms the isolation and detection of endogenous exosomes from cancer patient plasma. This result is produced from only 25 μL of plasma being loaded onto the DEP chip.

Figure 11. Schematic representation of exosome and other nanoparticle collection on the ACE device (chip) microelectrodes. The electric field lines (blue) run between individual microelectrodes on the microarray and converge onto the edges of the microelectrodes, forming the DEP high-field regions. The exosomes, EVs, nucleosomes, and aggregated protein particles collect in these high-field regions around the microelectrode edges. Any cells or larger particles in the sample (blood, plasma, serum, etc.) are concentrated into the DEP low-field areas between the microelectrodes, while the lower molecular weight biomolecules are unaffected by the DEP electric fields. A fluid wash removes any cells and the other plasma materials, while the nanosize biomarkers (exosomes, etc.) remain concentrated in the DEP high-field regions.

carry out on-chip or in situ fluorescence and immunofluorescence analysis represents a major advantage for the ACE process by providing faster detection of biomarkers and eliminating the need to remove exosomes and their associated biomarkers from the device, which could certainly contribute to biomarker loss. The detection and identification of RNA provides further verification of exosome and EV isolation by the ACE device. The RNA analysis involved eluting the isolated exosomes and EVs from the ACE chip and releasing the RNA from the vesicles for RT-PCR detection. The presence of both glioblastoma-specific mutated EGFRvIII mRNA and more general housekeeping β-actin mRNA was observed. While classical procedures for releasing RNA from exosomes and EVs were used, a simpler heat- and surfactant-based process was also developed, which was faster and more compatible with the overall ACE process. The improved process allows freshly collected plasma samples containing exosomes to be rapidly analyzed with minimal RNA degradation. Overall, the ACE

device process is significantly faster than any of the current, state-of-the-art methods that are presently being used,20−23 including nanoarray-based devices.21 Most of these devices and methods require multistep procedures that are time-consuming and expensive, making them very unlikely to translate into much-anticipated rapid diagnostic applications using exosomes and their associated protein and RNA biomarkers. Other significant advantages of the ACE device and process include (1) the use of a relatively small sample volume (30−50 μL), which is crucial for analysis of precious fetal, infant, and other rare samples; (2) no dilution of the sample is required; (3) the ACE devices and processes can be used with whole blood,29 which eliminates the need for preparing plasma or serum; and (4) under the same conditions used in this study, the ACE device is capable of isolating hmw-DNA and cf-DNA biomarkers directly from blood, plasma, and serum samples.33,34 The ACE device represents a powerful, minimally invasive technology for cancer diagnostics that is particularly 6648

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano

concentration of around 5 × 1010 particles/mL, as determined by a NanoSight particle tracking system. ACE Device Isolation and Recovery of the Exosomes. AC electrokinetic microarray chip devices were obtained from Biological Dynamics (San Diego, CA, USA). These ACE microarray chips are silicon-based devices approximately 2 cm × 3 cm in size that have 400 platinum microelectrodes (60 μm diameter), which are coated with a thin porous hydrogel layer. This hydrogel layer prevents direct contact between the plasma and the platinum electrode surface, which prevents aggressive AC electrochemistry that would otherwise occur during operation. It also prevents bubble formation through electrolysis. This helps to protect the collected biomolecules from degradation. The chip is encased in a plastic sample chamber (50 μL) that allows addition and removal of samples and wash buffers. The sample chamber also has a thin optical window covering that allows the chip surface to be monitored using an epifluorescent microscope system. The ACE device isolation and recovery of the exosomes from plasma and buffer is a three-step process that includes the following. Step 1: A 30−50 μL sample of the plasma or buffer spiked with glioblastoma exosomes and EVs was immediately placed into the ACE microarray device/chip (Biological Dynamics, San Diego, CA, USA). An AC voltage of 10 Vpp at a frequency of 15 kHz was applied to the ACE device for 10 min. This was sufficient time to collect the exosomes and EVs. Step 2: With the AC field still on, the chip was washed with a 1× TE buffer at 20 μL/min for 10 min to remove the bulk plasma. Step 3: The AC field was then switched to three pulses of 0.5 s duration at 5 Hz and 10 Vpp. This reversed the AC dielectrophoretic force and pushed the collected exosomes and EVs into the wash buffer solution, allowing them to be eluted from the chip for further analysis. The total time required from introducing the spiked plasma sample to the chip to recovering the purified exosomes and EVs was 30 min. ACE Isolation and Recovery with on-Chip (in Situ) Fluorescent Staining and Detection of Exosomes and RNA. Exosomes and EVs were concentrated at the microelectrode edges, and the excess plasma was washed away as described above. With the AC field still on, a second wash was run through the system that contained a 5000× dilution of RNA Select dye (Molecular Probes, USA) and allowed to incubate for 10 min. A fresh wash buffer without dye was then run through the chip to remove excess unbound dye. The sample was then fluorescently imaged, and a composite image was created combining the green and red images using ImageJ software (NIH). These exosomes and EVs were then collected as described in step 3 of the isolation and recovery protocol. The data shown in Figure 5 have had three replicates, all of which showed the same result, that red exosomes and EVs are pulled down to the electrode edges and are held there while the bulk plasma is washed away. They also had similar ratios of exosomes and EVs staining green for RNA. Figure 5 is a representative example of that data set. ACE Isolation of Exosomes with on-Chip (in Situ) Immunofluorescent Labeling and Detection of Proteins. Purified exosomes and EVs derived from cultured U87-EGFRvIII glioblastoma cells were prepared by differential centrifugation as described above. Whole cell lysates derived from the same cells were also provided by one of the authors (C.C.) and were used for on-chip (in situ) antibody control experiments. Purified exosomes and EVs with a final concentration of approximately (1−10) × 109 particles/mL or cell lysates were diluted into 25 μL of 0.5× PBS and then placed onto the DEP chip, at 1:4 dilution. An AC electric field was applied to the ACE device (chip) for 10 min at 18 Vpp and 15 kHz to isolate and collect the exosome nanoparticles around the microelectrode edges. With the AC field still on, the chip was washed with 200 μL of 0.5× PBS for 10 min. Following the wash step, the AC field was turned off, leaving the concentrated exosomes at the microelectrode edges. The relevant samples were then treated on-chip with 0.1% saponin in PBS for 30 min at room temperature and 5% normal goat or rabbit serum to block nonspecific sites of antibody binding, as needed. Primary antibodies were diluted into PBS, and incubated with the sample on the ACE chip for 90 min at room temperature. Following a wash with 200 μL of

well suited for the rapid isolation of exosomes, their associated RNA and proteins, and cf-DNA biomarkers. The technology is setting the stage for rapid, seamless sample-to-answer liquid biopsy, cancer patient therapy monitoring, and ultimately early disease detection.

METHODS Exosomes and EVs used in these experiments were provided by one of the authors (C.C.) and were isolated from cell culture media of cultured glioblastoma U87-EGFRvIII neuronal cells bearing an epidermal growth factor receptor mutation. These cells have been shown to naturally produce exosomes39 and excrete them into the cell culture medium. These exosomes were purified from the cells as described below and were spiked into either buffer or undiluted human plasma samples for DEP isolation, detection, and analysis. Exosome-Free Medium Preparation and Exosome Isolation from Cell Culture Medium. Exosome-depleted cell culture media was prepared by ultracentrifugation of Dulbecco’s modified Eagle’s medium supplemented with 20% fetal bovine serum (FBS) at 120000g for 18 h at 4 °C. The medium was then diluted to a final concentration of 10% FBS and used to culture cell lines as described. U87(MG)EGFRvIII cells were cultured to 60−70% confluency; then the standard culture medium was replaced with exosome-depleted medium. Cells were cultured for an additional 72 h before exosome collection from the cell-free supernatants by differential centrifugation. Conditioned medium was first centrifuged at 300g for 10 min to remove cellular debris. The supernatant was collected and further centrifuged at 2000g for 20 min. The resultant supernatant was then transferred to ultracentrifuge tubes for ultracentrifugation at 120000g for 2 h. This centrifugation procedure recovers exosomes along with other EVs. The supernatant was discarded and exosome and EV pellets were resuspended in PBS for storage at −80 °C. All centrifugation steps were performed at 4 °C. Preparation of Exosome and Nanoparticle Human Plasma and Buffer Samples. The exosomes collected from the cell culture media as described above were in a final solution of 1× PBS buffer. These were used for the buffer recovery experiments. Normal human plasma samples were purchased from ZenBio, Inc. (Research Triangle Park, NC, USA). The exosomes and EVs collected from the U87-EGFRvIII cells were then spiked into the plasma at a 1:9 dilution ratio. The final concentration of exosomes and EVs in the human plasma was 5 × 109 exosomes/mL. It has been observed that cancer patients can have up to 5.6 × 1010 exosomes/mL of plasma.14 Plasma samples spiked with 110 nm polystyrene nanoparticles (Invitrogen, USA) were prepared by first diluting the purchased stock solution of particles by 100× using 0.5× PBS buffer. This solution was then spiked into the plasma at a 1:9 dilution ratio. Fluorescent Labeling of Exosomes and EVs with PKH26 Fluorescent Dye. U87(MG)-EGFRvIII-derived exosomes and EVs were labeled with PKH26 (PKH26 red fluorescent cell linker mini kit for general cell membrane labeling, Sigma-Aldrich, St. Louis, MO, USA). Isolated exosomes and EVs were diluted to 1 mL with diluent C from the kit prior to mixing with freshly prepared PKH26 solution (4 μL of PKH26 dye was added to 1 mL of diluent C). The samples were mixed gently for 4 min before 2 mL of 1% bovine serum albumin was added to bind the excess dye. A 12 mL amount of PBS was added to the mixture, and the labeled exosomes and EVs were pelleted by ultracentrifugation at 120000g for 2 h. The supernatant was removed, and the pellets were washed with 16 mL of PBS and centrifuged at 120000g for 2 h at 4 °C. As a negative control, a PBS sample without spiked exosomes and EVs was also run through the labeling process and was found to contain minimal red fluorescence background or nanoparticles. This shows that the labeling was specific for the exosomes and EVs in the samples and that the dye had not simply precipitated and created red fluorescent particles that could be mistaken for exosomes and EVs. This means that the red fluorescence particles recovered from the cell culture media that were then spiked into the plasma were actually labeled exosomes and EVs and not just precipitated dye. The labeled exosome and EV sample had a final 6649

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano PBS, the samples were either viewed immediately (for directly conjugated fluorescent antibodies) or incubated with fluorophoreconjugated secondary antibodies for 90 min at room temperature, washed, and then viewed. The antibodies that were used included FITC-conjugated mouse anti-CD63 (ThermoFisher), goat antiTSG101 (Santa Cruz), rabbit anti-calnexin (Cell Signaling Technologies), and AlexaFluor 594- or 488-conjugated secondary antibodies (Novex, Life Technologies). Following an additional wash, the samples were viewed and imaged using an epifluorescent microscope with a 4× Nikon objective. The data shown in Figure 6A,B have been replicated dozens of times with similar results. The data shown in Figure 6E,F have been replicated three times with similar results. The data shown in Figure 7 were replicated multiple times with similar results. RT-PCR and PCR Analysis. The exosomes and EVs recovered from ACE isolation were lysed to release the RNA contents by running them through a miRCURY RNA isolation kit (Biofluids, Exiqon Life Sciences) in order to allow access for further RNA analysis. The purified RNA was then used in an RT-PCR reaction to create cDNA using random hexamers and SuperScript III reverse transcriptase (Thermo Fisher) according to the manufacturer’s instructions. After the cDNA was obtained, it was used in both qPCR reactions for detection of beta-actin and end-point PCR reactions for detection of EGFRvIII. Beta-actin primers were purchased from Life Technologies (catalog number 4331182; assay ID: Hs01060665_g1), with an amplicon length of 63 bp. The qPCR conditions were as follows: 1 hold at 50 °C for 2 min, 1 hold at 95 °C for 10 min, 50 cycles of 95 °C for 15 s, and 50 cycles of 60 °C for 1 min. We visualized the amplification using the TaqMan probe (FAM fluorescence). The EGFRvIII primers were EGFRvIII-1F and EGFRvIII-2Rb with an amplicon length of 181 bp. The EGFRvIII PCR conditions were as follows: 1 hold at 98 °C for 5 min, 50 cycles of 98 °C for 15 s, 50 cycles of 70 °C for 15 s, and 50 cycles of 72 °C for 15 s. Positive controls were cDNA created from RNA that was extracted using a standard kit from U87-EGFRvIII cells. For the exosome and EV lysis experiments Tween 20 was used. The heated exosomes and EVs were raised to 98 °C for 10 min before starting the RT-PCR protocol. The two positive controls are different starting amounts of extracted cDNA in the PCR reaction. TEM Analysis. Negative Staining of Exosomes and EVs. The carbon-coated TEM grid was treated with a glow discharge using an Emitech K350 for 1 min to make it hydrophilic. The exosome and EV dispersion collected from the cell culture media was diluted in PBS (20× v/v). A small drop of the diluted exosome and EV solution (5 μL) was placed on the treated TEM imaging grid, and the exosomes and EVs were allowed to adhere to the carbon film for 5 min. To prepare for the staining procedure, 20 drops of DI water were then applied to the grid in order to remove any unbound exosomes, EVs, and PBS. A 5 μL drop of 1% uranyl acetate solution was placed onto the grid and flicked off. This was repeated twice. The remaining 1% of the uranyl acetate solution was removed by absorbing the stain from the edge of the grid with a wedge of filter paper. The grid was air-dried for TEM imaging. TEM Parameters. The exosome and EV samples were analyzed with a transmission electron microscope (FEI Tecnai G2) using an accelerating voltage of 200 kV. SEM Analysis. Sample Drying Preparation. The ACE chips were run as described above without using the sample push-off for recovery (step 3), thereby leaving the collected material intact on the microelectrode edges. The chips were then placed in a −80 °C freezer, which quickly froze the 35 μL of water in the chip. The frozen chip was then placed in a lyophilizer overnight to remove the water, leaving the shape of the collected material preserved for SEM analysis. A second air-drying technique was used to prepare the chips for SEM analysis, where the chips were left out overnight to dry. This method of evaporation allowed the collected material to collapse into a thin film when completely dry. Both glioblastoma exosomes and EVs and 110 nm diameter polystyrene nanoparticles (Invitrogen, USA) were separately spiked into the undiluted plasma sample to help identify the collection regions around the ACE microelectrodes. The

beads did not collapse and maintained their three-dimensional shape, making them easy to identify under these conditions. SEM Parameters. Both types of dried chips were coated with iridium for 7 s. The chips were analyzed with an environmental scanning electron microscope (Phillips XL30) operated at 10 kV.

AUTHOR INFORMATION Corresponding Author

*E-mail: [email protected] ORCID

Michael J. Heller: 0000-0001-6436-0337 Notes

The authors declare the following competing financial interest(s): Michael Heller is a member of the scientific advisory board for Biological Dynamics.

ACKNOWLEDGMENTS This work was supported by Award Number W81XWH-14-20192 from the Defense Medical Research and Development Program and U.S. Army Medical Research Acquisition Activity (USAMRAA). REFERENCES (1) Zhou, H.; Xu, W.; Qian, H.; Yin, Q.; Zhu, W.; Yan, Y. Circulating RNA as A Novel Tumor Marker: An In-Vitro Study of the Origins and Characteristics of Extracellular RNA. Cancer Lett. 2008, 259, 50−60. (2) Mitchell, P. S.; Parkin, R. K.; Kroh, E. M.; Fritz, B. R.; Wyman, S. K.; Pogosova-Agadjanyan, E. L.; Peterson, A.; Noteboom, J.; O’Briant, K. C.; Allen, A. Circulating MicroRNAs as Stable Blood-Based Markers for Cancer Detection. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 10513−10518. (3) Gallo, A.; Tandon, M.; Alevizos, I.; Illei, G. G. The Majority of MicroRNAs Detectable in Serum and Saliva is Concentrated in Exosomes. PLoS One 2012, 7, e30679. (4) Tsui, N. B.; Ng, E. K.; Lo, Y. D. Stability of Endogenous and Added RNA in Blood Specimens, Serum, and Plasma. Clin. Chem. 2002, 48, 1647−1653. (5) Houseley, J.; Tollervey, D. The Many Pathways of RNA Degradation. Cell 2009, 136, 763−776. (6) Kosaka, N.; Yoshioka, Y.; Hagiwara, K.; Tominaga, N.; Ochiya, T. Functional Analysis of Exosomal MicroRNA in Cell−Cell Communication Research. In Circulating MicroRNAs; Springer: Berlin, 2013; pp 1−10. (7) Duijvesz, D.; Luider, T.; Bangma, C. H.; Jenster, G. Exosomes as Biomarker Treasure Chests for Prostate Cancer. Eur. Urol. 2011, 59, 823−831. (8) Taylor, D. D.; Shah, S. Methods of Isolating Extracellular Vesicles Impact Down-Stream Analyses of their Cargoes. Methods 2015, 87, 3− 10. (9) Vlassov, A. V.; Magdaleno, S.; Setterquist, R.; Conrad, R. Exosomes: Current Knowledge of their Composition, Biological Functions, and Diagnostic and Therapeutic Potentials. Biochim. Biophys. Acta, Gen. Subj. 2012, 1820, 940−948. (10) Simpson, R. J.; Jensen, S. S.; Lim, J. W. Proteomic Profiling of Exosomes: Current Perspectives. Proteomics 2008, 8, 4083−4099. (11) Caby, M.-P.; Lankar, D.; Vincendeau-Scherrer, C.; Raposo, G.; Bonnerot, C. Exosomal-Like Vesicles are Present in Human Blood Plasma. Int. Immunol. 2005, 17, 879−887. (12) Melo, S. A.; Luecke, L. B.; Kahlert, C.; Fernandez, A. F.; Gammon, S. T.; Kaye, J.; LeBleu, V. S.; Mittendorf, E. A.; Weitz, J.; Rahbari, N. Glypican-1 Identifies Cancer Exosomes and Detects Early Pancreatic Cancer. Nature 2015, 523, 177−182. (13) Whiteside, T. L. The Potential of Tumor-Derived Exosomes for Noninvasive Cancer Monitoring. Expert Rev. Mol. Diagn. 2015, 15, 1293−1310. 6650

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651

Article

ACS Nano

Point-of-Care Tests Using Cell-Free Nucleic Acids. Expert Rev. Mol. Diagn. 2015, 15, 1187−1200. (33) Sonnenberg, A.; Marciniak, J. Y.; Skowronski, E. A.; Manouchehri, S.; Rassenti, L.; Ghia, E. M.; Widhopf, G. F.; Kipps, T. J.; Heller, M. J. Dielectrophoretic Isolation and Detection of Cancer Related Crculating Cell Free DNA Biomarkers from Blood and Plasma. Electrophoresis 2014, 35, 1828−1836. (34) Sonnenberg, A.; Marciniak, J. Y.; Rassenti, L.; Ghia, E. M.; Skowronski, E. A.; Manouchehri, S.; McCanna, J.; Widhopf, G. F.; Kipps, T. J.; Heller, M. J. Rapid Electrokinetic Isolation of CancerRelated Circulating Cell-Free DNA Directly from Blood. Clin. Chem. 2014, 60, 500−509. (35) Manouchehri, S.; Ibsen, S.; Wright, J.; Rassenti, L.; Ghia, E.; Widhopf, G., II; Kipps, T.; Heller, M. Dielectrophoretic Recovery of DNA from Plasma for the Identification of Chronic Lymphocytic Leukemia Point Mutations. Int. J. Hematol. Oncol. 2016, 5, 27−35. (36) Ibsen, S.; Sonnenberg, A.; Schutt, C.; Mukthavaram, R.; Yeh, Y.; Ortac, I.; Manouchehri, S.; Kesari, S.; Esener, S.; Heller, M. J. Recovery of Drug Delivery Nanoparticles from Human Plasma Using an Electrokinetic Platform Technology. Small 2015, 11, 5088−5096. (37) Ramos, A.; Morgan, H.; Green, N. G.; Castellanos, A. AC Electrokinetics: A Review of Forces in Microelectrode Structures. J. Phys. D: Appl. Phys. 1998, 31, 2338−2353. (38) Seeman, P.; Cheng, D.; Iles, G. Structure of Membrane Holes in Osmotic and Saponin Hemolysis. J. Cell Biol. 1973, 56, 519−527. (39) Skog, J.; Würdinger, T.; van Rijn, S.; Meijer, D. H.; Gainche, L.; Curry, W. T.; Carter, B. S.; Krichevsky, A. M.; Breakefield, X. O. Glioblastoma Microvesicles Transport RNA and Proteins that Promote Tumour Growth and Provide Diagnostic Biomarkers. Nat. Cell Biol. 2008, 10, 1470−1476.

(14) Muller, L.; Hong, C.-S.; Stolz, D. B.; Watkins, S. C.; Whiteside, T. L. Isolation of Biologically-Active Exosomes from Human Plasma. J. Immunol. Methods 2014, 411, 55−65. (15) Mathivanan, S.; Ji, H.; Simpson, R. J. Exosomes: Extracellular Organelles Important in Intercellular Communication. J. Proteomics 2010, 73, 1907−1920. (16) Huang, X.; Yuan, T.; Tschannen, M.; Sun, Z.; Jacob, H.; Du, M.; Liang, M.; Dittmar, R. L.; Liu, Y.; Liang, M. Characterization of Human Plasma-Derived Exosomal RNAs by Deep Sequencing. BMC Genomics 2013, 14, 319. (17) Simons, M.; Raposo, G. Exosomes−Vesicular Carriers for Intercellular Communication. Curr. Opin. Cell Biol. 2009, 21, 575− 581. (18) Lötvall, J.; Hill, A. F.; Hochberg, F.; Buzás, E. I.; Di Vizio, D.; Gardiner, C.; Gho, Y. S.; Kurochkin, I. V.; Mathivanan, S.; Quesenberry, P. 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, 3.2691310.3402/jev.v3.26913 (19) Grant, R.; Ansa-Addo, E.; Stratton, D.; Antwi-Baffour, S.; Jorfi, S.; Kholia, S.; Krige, L.; Lange, S.; Inal, J. A Filtration-Based Protocol to Isolate Human Plasma Membrane-Derived Vesicles and Exosomes from Blood Plasma. J. Immunol. Methods 2011, 371, 143−151. (20) Lee, K.; Shao, H.; Weissleder, R.; Lee, H. Acoustic Purification of Extracellular Microvesicles. ACS Nano 2015, 9, 2321−2327. (21) Kalra, H.; Adda, C. G.; Liem, M.; Ang, C. S.; Mechler, A.; Simpson, R. J.; Hulett, M. D.; Mathivanan, S. Comparative Proteomics Evaluation of Plasma Exosome Isolation Techniques and Assessment of the Stability of Exosomes in Normal Human Blood Plasma. Proteomics 2013, 13, 3354−3364. (22) Im, H.; Shao, H.; Park, Y. I.; Peterson, V. M.; Castro, C. M.; Weissleder, R.; Lee, H. Label-Free Detection and Molecular Profiling of Exosomes with a Nano-Plasmonic Sensor. Nat. Biotechnol. 2014, 32, 490−495. (23) Jeong, S.; Park, J.; Pathania, D.; Castro, C. M.; Weissleder, R.; Lee, H. Integrated Magneto−Electrochemical Sensor for Exosome Analysis. ACS Nano 2016, 10, 1802−1809. (24) Zhao, Z.; Yang, Y.; Zeng, Y.; He, M. A Microfluidic ExoSearch Chip for Multiplexed Exosome Detection Towards Blood-Based Ovarian Cancer Diagnosis. Lab Chip 2016, 16, 489−496. (25) Wunsch, B. H.; Smith, J. T.; Gifford, S. M.; Wang, C.; Brink, M.; Bruce, R. L.; Austin, R. H.; Stolovitzky, G.; Astier, Y. Nanoscale Lateral Displacement Arrays for the Separation of Exosomes and Colloids Down to 20 nm. Nat. Nanotechnol. 2016, 11, 936−940. (26) Babic, A.; Wolpin, B. M. Circulating Exosomes in Pancreatic Cancer: Will They Succeed on the Long, Littered Road to Early Detection Marker? Clin. Chem. 2016, 62, 307−309. (27) Botling, J.; Edlund, K.; Segersten, U.; Tahmasebpoor, S.; Engström, M.; Sundström, M.; Malmström, P.-U.; Micke, P. Impact of Thawing on RNA Integrity and Gene Expression Analysis in Fresh Frozen Tissue. Diagn. Mol. Pathol. 2009, 18, 44−52. (28) Krishnan, R.; Dehlinger, D. A.; Gemmen, G. J.; Mifflin, R. L.; Esener, S. C.; Heller, M. J. Interaction of Nanoparticles at the DEP Microelectrode Interface Under High Conductance Conditions. Electrochem. Commun. 2009, 11, 1661−1666. (29) Sonnenberg, A.; Marciniak, J. Y.; Krishnan, R.; Heller, M. J. Dielectrophoretic Isolation of DNA and Nanoparticles from Blood. Electrophoresis 2012, 33, 2482−2490. (30) Sonnenberg, A.; Marciniak, J. Y.; McCanna, J.; Krishnan, R.; Rassenti, L.; Kipps, T. J.; Heller, M. J. Dielectrophoretic Isolation and Detection of cfc-DNA Nanoparticulate Biomarkers and Virus from Blood. Electrophoresis 2013, 34, 1076−1084. (31) Krishnan, R.; Sullivan, B. D.; Mifflin, R. L.; Esener, S. C.; Heller, M. J. Alternating Current Electrokinetic Separation and Detection of DNA Nanoparticles in High Conductance Solutions. Electrophoresis 2008, 29, 1765−1774. (32) Lewis, J. M.; Heineck, D. P.; Heller, M. J. Detecting Cancer Biomarkers in Blood: Challenges for New Molecular Diagnostic and 6651

DOI: 10.1021/acsnano.7b00549 ACS Nano 2017, 11, 6641−6651