Subscriber access provided by Nottingham Trent University
Communication
A rapid lipid-based approach for normalization of quantum dot-detected biomarker expression on extracellular vesicles in complex biological samples Meryl Rodrigues, Nicole Richards, Bo Ning, Christopher J Lyon, and Tony Y Hu Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.9b02232 • Publication Date (Web): 18 Jul 2019 Downloaded from pubs.acs.org on July 18, 2019
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 26 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
Nano Letters
A rapid lipid-based approach for normalization of quantum dot-detected biomarker expression on extracellular vesicles in complex biological samples Meryl Rodrigues1, 2, Nicole Richards1, Bo Ning3, Christopher J. Lyon1, and Tony Y. Hu 4* 1
Virginia G. Piper Biodesign Center for Personalized Diagnostics, Arizona State University
Biodesign Institute, Tempe, Arizona, 85287, USA 2
School of Biological and Health Systems Engineering, Arizona State University, Tempe,
Arizona, 85287, USA 3 Center for Molecular Design and Biomimetics, The Biodesign Institute, Arizona State University,
Tempe, Arizona, 85281, USA 4
Center for Cellular and Molecular Diagnostics, Department of Biochemistry and Molecular
Biology, School of Medicine, Tulane University, New Orleans, Louisiana, 70112, USA *
* Corresponding author email:
[email protected] phone: 504-605-8004
1 ACS Paragon Plus Environment
Nano Letters 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
Abstract: Extracellular vesicles (EVs) are of considerable interest as tumor biomarkers, since tumor-derived EVs contain a broad array of information about tumor pathophysiology. However, current EV assays cannot distinguish between EV biomarker differences resulting from altered abundance of a target EV population with stable biomarker expression, altered biomarker expression in a stable target EV population, or effects arising from changes in both parameters. We now describe a rapid nanoparticle- and dye-based fluorescent immunoassay that can distinguish among these possibilities by normalizing EV biomarker level(s) to EV abundance. In this approach, EVs are captured from complex samples (e.g. serum), stained with a lipophilic dye and hybridized with antibody-conjugated quantum dot probes for specific EV surface biomarkers. EV dye signal is used to quantify EV abundance and normalize EV surface biomarker expression levels. EVs from malignant and nonmalignant pancreatic cell lines exhibited similar staining, and probe-to-dye ratios did not change with EV abundance, allowing direct analysis of normalized EV biomarker expression without a separate EV quantification step. This EV biomarker normalization approach markedly improved the ability of serum levels of two pancreatic cancer biomarkers, EV EpCAM and EV EphA2, to discriminate pancreatic cancer patients from nonmalignant control subjects. The streamlined workflow and robust results of this assay are suitable for rapid translation to clinical applications and its modular design permits it to be rapidly adapted to quantitate other EV biomarkers by the simple expedient of swapping the antibody-conjugated quantum dot probes for those that recognize a different disease-specific EV biomarker.
Keywords: Extracellular vesicles, exosomes, quantum dots, normalization, cancer-associated markers, protein expression.
2 ACS Paragon Plus Environment
Page 2 of 26
Page 3 of 26 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
Nano Letters
Liquid biopsy samples (e.g. plasma/serum, saliva, and urine) are of great interest as a means to rapidly diagnose different diseases, including cancer, and evaluate their responses to treatment, since it is not always feasible to obtain biopsies or repeat biopsies from affected sites. Liquid biopsy methods have influenced the development of several therapies that require biomarker assays 1, 2. Circulating EVs are an excellent source of biomarkers for such assays since EVs are abundantly secreted by most cells (> 1010 vesicles / mL of peripheral blood) 3, particularly infected or malignant cells, and can carry factors that reflect the phenotype of their parental cells. 4-7.
However, current EV assays do not differentiate between altered secretion of EVs carrying a
biomarker of interest and altered expression of this biomarker within an EV population under different pathological conditions and in response to physiological changes 8. Studies have revealed that circulating EV levels are altered in individuals with, or animal models of, Parkinson’s disease, insulin resistance, atherosclerosis, hypertension, following a stroke or myocardial infarction, and during normal pregnancy
9-12.
EV release can also be affected by several common tumor
microenvironment features (e.g. hypoxia, acidification, and heparanase overexpression) 13-16. Several studies have analyzed how altered levels of biomarkers present in EVs provide valuable information regarding the physiological state of their parent cells to evaluate disease prognosis and inform treatment decisions
17.
For example, increased EV surface expression of
specific integrins is associated with preferential metastasis to the lung (integrin α6β4 and α6β1) and liver (integrin αvβ5), and does not reflect cellular expression levels of these factors
18.
EV
biomarker associated with tumor development and metastasis can, however, also be influenced by the patient’s physiologic state or treatment regimen. One study has reported that interferon-γ (IFN-γ) enhances PD-L1 expression on EVs of melanoma cells, that there is a correlation between circulating IFN-γ and EV PD-L1 levels and tumor size, that EV PD-L1 expression varies during 3 ACS Paragon Plus Environment
Nano Letters 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 4 of 26
anti-PD-1 therapy, and that the degree of this increase during early treatment could differentiate responsive and non-responsive tumors 19. Notably, this study analyzed the number of circulating EVs, which is not done for most EV biomarkers, and thus could distinguish between a biomarker difference resulting from a difference in EV expression level and EV abundance. Another study reported using a λ-DNA microfluidic platform to segregate EV subpopulations, using aptamer probes to assess changes in individual EV surface expression of HER2 and EpCAM, and differentiate patients with II breast cancer from non-malignant controls
20.However,
both these
studies were uncommon in that they assessed the relative instead of absolute expression of their EV biomarkers. Circulating levels of a disease-associated EV biomarker can reflect both the release rate of disease-associated EVs from the target tissue and their relative expression of the target biomarker. The ability to distinguish such differences may allow clinicians to detect and better analyze disease biomarkers that may otherwise be masked by altered physiologic conditions. This difference is not addressed by standard EV analyses unless a study protocol also quantifies the number of target EVs in an analysis sample. This is not practical in most clinical setting, since nanoparticle tracking analysis (NTA)
21
or tunable resistive pulse sensing (TRPS)
22
instruments used to quantify EV
diameters and numbers require purified EV samples, and EV isolation procedure required to generate such samples are laborious and yield variable results not suitable for clinical assays. Measurement of EV RNA, DNA, or protein can be used as a surrogate for relative EV abundance in highly defined and homogeneous EV samples 17, 23, 24, such as those obtained from cell culture supernatants. Such approaches are not feasible with complex biological samples as altered abundance of EVs with different cargo compositions can skew these normalizations. By contrast, EV lipid bilayers represent a much more static feature 25. EV lipid content reflects EV size and 4 ACS Paragon Plus Environment
Page 5 of 26 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
Nano Letters
abundance and mean EV lipid content should significantly change only in response to a marked change in the size of the EVs present in a sample. Since lipophilic dyes have been employed in EV localization, uptake and transfer studies 26-29, we hypothesized that quantifying the signal from EVs stained with such a dye could serve as a surrogate marker of EV abundance to normalize the mean expression of target EV biomarkers. This would represent a significant improvement over current assays that attempt to evaluate the relative expression of target biomarkers on EVs, which require purified EV samples for analysis. To test this hypothesis, we designed an assay where EVs are immobilized on the surface of a 96-well plate with an antibody specific for the common EV membrane protein CD81
30.
Captured EV are then stained with the lipophilic dye DiO (3,3′-dihexadecyloxacarbocyanine perchlorate) to provide a quantitative readout reflecting the total number of EVs in the sample. EVs are then hybridized with quantum dot-labeled antibody probes specific to candidate EV biomarkers, and the probe-to-dye signal ratio is measured to determine the mean EV expression of these biomarkers (Fig. 1). This approach circumvents the need for separate EV purification and quantification steps, which are time-consuming, low-throughput, require expensive and specialized equipment, and are difficult to optimize, rendering them unsuitable for clinical translation 21, 31.
5 ACS Paragon Plus Environment
Nano Letters 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
Figure 1: Schematic of the quantum dot assay. EVs captured by an antibody to an EV-specific surface are stained with the lipophilic fluorescent dye DiO and then hybridized with antibodyconjugated quantum dot probes specific for biomarker targets on the EV membrane (e.g., EpCAM and EphA2). DiO signal from the captured EVs functions as a surrogate marker of EV abundance and allows direct normalization of quantum dot probe signal to permit quantification of mean biomarker levels in a captured population without the need for an independent EV isolation and quantitation procedure. This allows direct comparison of relative EV biomarker levels among different cohorts for disease diagnosis (e.g., cancer patients vs. healthy subjects with suspected malignancy).
To analyze the feasibility of this approach, EVs from malignant (PANC-1) and nonmalignant (HPNE) pancreatic cell lines were examined to determine the correlation between their 6 ACS Paragon Plus Environment
Page 6 of 26
Page 7 of 26 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
Nano Letters
abundance and dye signal. PANC-1 and HPNE cells exhibited similar growth (Fig. 2A), but significantly different secretion rates (Fig. 2B-C) of EVs with similar size distributions and mean diameters (Fig. S1). Spiking known concentrations of these EVs into EV-free serum produced progressive DiO signal changes that could differentiate 2-fold sequential EV dilutions (Fig. 2D), where all but the lowest PANC-1 sample significantly differed from its next dilution sample. Samples spiked with equal numbers of HPNE or PANC-1 EVs also revealed strong linear correlation, with a slope approximating one (Fig. 2E), implying that any potential differences in membrane lipid composition or EV size distributions between these samples did not substantially affect their staining. DiO was utilized for this analysis since it requires a one-step labeling process, unlike other fluorescent lipophilic dyes that have been used to stain EVs. For example, PKH dyes require extended incubation in an iso-osmotic mannitol solution, which can reduce EV yields and increase the variability of downstream EV analyses 29. HPNE and PANC-1 EVs stained with DiO also demonstrated similar fluorescent intensity on confocal microscope images, which was distinct from the signal of a quantum dot probe bound to these EVs (Fig. 2F).
7 ACS Paragon Plus Environment
Nano Letters 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
Figure 2. DiO similarly stains EVs isolated from PANC-1 and HPNE cells. (A) PANC-1 and HPNE cells exhibit similar growth rates, but PANC-1 cells secrete more (B) total EVs and (C) EVs per cell than HPNE cells. (** p
