pubs.acs.org/ac Article Electrochemical Genosensing of Overexpressed GAPDH Transcripts in Breast Cancer Exosomes Arnau Pallares-Rusiñol, Silio Lima Moura, Merce ̀ Martí, and Maria Isabel Pividori* Cite This: https://doi.org/10.1021/acs.analchem.2c04773 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. Downloaded via 158.109.212.84 on January 23, 2023 at 14:48:44 (UTC). ABSTRACT: Exosomes are receiving highlighted attention as new biomarkers for the detection of cancer since they are profusely released by tumor cells in different biological fluids. In this paper, the exosomes are preconcentrated from the serum by immunomagnetic separation (IMS) based on a CD326 receptor as a specific epithelial cancer-related biomarker and detected by glyceraldehyde-3-phosphate dehydrogenase (GAPDH) transcripts. Following the lysis of the captured exosomes, the released GAPDH transcripts are amplified by reverse transcription polymerase chain reaction (RT-PCR) with a double-tagging set of primers on poly(dT)-modified-MPs to increase the sensitivity. The double-tagged amplicon is then quantified by electrochemical genosensing. The IMS/double-tagging RT-PCR/electrochemical genosensing approach is first demonstrated for the sensitive detection of exosomes derived from MCF7 breast cancer cells and compared with CTCs in terms of the analytical performance, showing an LOD of 4 × 102 exosomes μL−1. The genosensor was applied to human samples by immunocapturing the exosomes directly from serum from breast cancer patients and showed a higher electrochemical signal (3.3-fold, p < 0.05), when compared with healthy controls, suggesting an overexpression of GAPDH on serum-derived exosomes from breast cancer patients. The detection of GAPDH transcripts is performed from only 1.0 mL of human serum using specific magnetic particles, improving the analytical simplification and avoiding ultracentrifugation steps, demonstrating to be a promising strategy for minimal invasive liquid biopsy. ■ INTRODUCTION Breast cancer is a highly lethal malignancy and the most cancer-related biomarkers (e.g., BRCA1 ELISA kit, from MBS; CA15-3 ELISA kit, from Abcam; or BCAR ELISA kit, from commonly diagnosed cancer among women, with an estimated Biogen). Besides, other techniques aim for the detection of over 2 million new cases in 2020.1 Most of the currently circulating tumor cells (CTCs),5 considered one of the most available technologies for breast cancer diagnosis are based on significant breast cancer-related biomarkers. CellSearch6,7 is imaging techniques.2 In high-income countries, breast cancer is the first and the only CTC-based assay commercially available detected at stages I and II in 70% of women, while 20−50% in and approved by the US Food and Drug Administration low-income countries. Moreover, the time delay that exists (FDA). CellSearch enriches CTCs using magnetic particles between diagnosis and treatment is about 4 to 6 weeks in high- containing antibodies against the Epithelial Cell Adhesion income countries, but it can be as long as 8 months in low- and middle-income countries.3 The use of biomarkers related with breast cancer in liquid biopsies for early identification of Received: October 28, 2022 individuals could potentially bridge this gap in low-resource Accepted: January 10, 2023 settings, by reducing the technical requirements and opera- tional costs. The routine clinical diagnosis in liquid biopsies is based on enzyme-linked immunosorbent assay (ELISA) methodologies,4 existing many commercial kits targeting breast © XXXX The Authors. Published by American Chemical Society https://doi.org/10.1021/acs.analchem.2c04773 A Anal. Chem. XXXX, XXX, XXX−XXX Analytical Chemistry pubs.acs.org/ac Article Molecule (EpCAM) (also known as CD326). EpCAM is a analyzer. A magneto-actuated graphite-epoxy composite (m- cell-surface glycoprotein that is known to be highly expressed GEC) electrode as the working electrode (geometric area = 0.5 in epithelial carcinomas, including breast cancer and prostate cm2), Ag/AgCl/KCl(sat.) as the reference electrode, a disc cancer.8 However, the clinical use of CTCs is limited by their platinum counter electrode (geometric area = 3.0 cm2), and a scarcity in the peripheral blood (1 CTC/105−6 blood cells).9 standard 20-mL one compartment three-electrode cell was Exosomes10 (30−200 nm in diameter) are receiving used in all experiments. The detailed preparation of the m- highlighted attention as new biomarkers for the detection of GEC electrodes has been extensively described by Pividori and cancer in early stages.11,12 Exosomes are intercellular shuttle- co-workers.16−19 like vesicles with molecular cargo as mRNA, microRNA, DNA, Chemicals and Biochemicals. Tosyl-activated magnetic lipids, and proteins.13 Most cell types, including normal and particles (MPs) (Dynabeads M450 Tosylactivated, ref. 14013), tumor cells, release exosomes in many different biological MPs modified with EpCAM antibody (antiCD326-MPs, fluids such as blood, plasma, serum, or urine, among others.13 Dynabeads Epithelial Enrich, ref. 16102), MPs modified with It is known that a single cell can release many exosomes per poly(dT) (polydT-MPs, Dynabeads Oligo(dT)25, ref. 61002), hour into the extracellular space,14 at an increased rate by MPs modified with streptavidin (strep-MPs, Dynabeads tumor cells. The high number of exosomes that can be released MyOne Streptavidin T1, ref. 65601), mouse monoclonal by a single tumor cell reveals the strong potential application of antibody antiCD81 (ref. 10630D), and BCA protein assay kit exosomes as an alternative biomarker for early diagnostics, (ref. 23225) were purchased from Thermo Fisher Scientific overcoming the most challenging limitation that presents CTC (MA, US). Mouse monoclonal antibody antiCD326 or assays: their very low concentration in blood. Exosomes can EpCAM (ref. ab7504) and a goat anti-mouse IgG H&L potentially be used to detect the presence of tumor cells and (Cy5) (antimouse-Cy5, ref. ab97037) were purchased from deposits in the early stage of growth with a simple and Abcam (Cambridge, GB). Antidigoxigenin-horseradish perox- minimally invasive procedure such as liquid biopsy. idase Fab fragments (antiDIG-HRP, ref. 11207733910) were In this work, it is described an electrochemical genosensor purchased from Roche Diagnostics (Basel, CH). for the detection of glyceraldehyde-3-phosphate dehydrogen- The primers for the double-tagging PCR were selected for ase (GAPDH) gene transcripts, which we found to be the specific amplification of GAPDH and were purchased from overexpressed in breast cancer cells and exosome-derived Sigma-Aldrich (Merck KGaA, DE). The sequence for the human serum in breast cancer patients. The approach is based digoxigenin-modified forward primer (DIG-Fw) was 5′-[DIG] on immunomagnetic separation (IMS) of the exosomes using CTTCTTTTGCGTCGCCAG, while the sequence for the CD326 cancer-related biomarker, followed by amplification by biotin-modified reverse primer (BIO-Rev) was 5′-[BIO] double-tagging reverse transcription PCR of the GAPDH AGCCCCAGCCTTCTCCA. The primers were also checked transcripts on poly(dT)-MPs. The integration of PCR with in terms of secondary structures, to avoid hairpins, self, or cross electrochemical genosensing was previously reported,15 as well dimers. All solutions, described in S1 (Supplementary Data), as the further detection of the labeled DNA amplicons from were prepared with Ultrapure water (Millipore System, double-tagging PCR16,17 and quadruple-tagging multiplex resistivity 18.2 MΩ cm) and solutions used in RNA PCR.18,19 This approach is first optimized with breast cancer preparation were RNase-free by treatment with 0.1% DEPC. cell line MCF7 cells and exosomes obtained from a cell culture Cell Culturing, Exosome Isolation, and Purification supernatant. Only 1.0 mL of human serum is used to from the MCF7 Cell Line. The exosomes were obtained from specifically capture the exosomes on MPs which can be easily the MCF7 cell line (ATCC, ref. HTB-22), and the culture integrated on the biosensing device, improving the analytical conditions are detailed in S2 (Supplementary Data). The simplification by avoiding ultracentrifugation. The electro- MCF7 cells were used as a model of breast cancer. The chemical genosensing approach allows the quantitative exosomes were purified from the culture supernatant by measurement of transcripts with high sensitivity, robustness, differential ultracentrifugation according to Théry et al.20 with and simplicity. To the best of the authors’ knowledge, this is minor changes. Exosomes were resuspended in Tris 1× buffer the first study on the expression of GAPDH genes in exosomes (pH 7.4, 0.22 μm sterile-filtered) and stored at −80 °C. All the from breast cancer patients. experimental data are provided in S2 (Supplementary Data). ■ EXPERIMENTAL SECTION Instrumentation. Nanoparticle tracking analysis (NTA) Characterization of the Exosomes Derived from MCF7 Breast Cancer Cell Lines. The size distribution and concentration of exosomes were measured by nanoparticle was performed using the NanoSight LM10-HS system tracking analysis (NTA). The morphology was analyzed by (NanoSight Ltd., Malvern, GB). The cryogenic transmission cryogenic transmission electron microscopy (Cryo-TEM). The electron microscopy (TEM) images were collected by a Jeol total protein concentration of exosomes samples was estimated JEM 2011 (JEOL USA Inc., MA, US) microscope. Flow by the BCA protein assay kit (Pierce BCA protein assay kit, ref. cytometry was performed using BD FACSCANTO II (BD 23227, Thermo Fisher Scientific). Biosciences, NJ, US) equipment. Mean fluorescence intensity To set up the technical approach, the expression study of (MFI) and beads count data were obtained by FlowJo analysis CD81, a tetraspanin general marker for exosomes, and CD326, software (FlowJo LLC, BD Biosciences) of every sample- a cancer-related epithelial receptor, on the MCF7 cell line and reading file. The confocal images were collected on the their derived exosomes was carried out by flow cytometry. In microscope Leica, TCS SP5 (Leica Microsystems, DE). the case of the cells, the indirect labeling was performed by the SimpliAmp Thermal Cycler (Applied Biosystems, US) was incubation of specific antibodies antiCD81 and antiCD326, used for the double-tagging reverse transcription polymerase followed by labeling with the antimouse-Cy5 antibody (a far- chain reaction (RT-PCR) amplification. All electrochemical red-fluorescent dye, excitation 647 nm, emission 665 nm). The experiments were performed using an Autolab PGSTAT10 labeled cells were resuspended in Tris 1× buffer solution (Metrohm AG, CH) potentiostat/galvanostat electrochemical containing 0.5% BSA solution. B https://doi.org/10.1021/acs.analchem.2c04773 Anal. Chem. XXXX, XXX, XXX−XXX Analytical Chemistry pubs.acs.org/ac Article In order to compare the expression on the exosomes, the antiCDX antibody (CDX being either CD81 or CD326; Figure same procedure of labeling was performed, but in this case, the 1, panel A1). Then, they were lysed and the released exosomes were first immobilized on the surface of MPs due to messenger RNAs (Figure 1, panel A2) were captured by their size and the resolution of the technique. To achieve that, polydT-MPs based on the poly(A) tail followed by reverse exosomes were covalently immobilized on tosyl-activated MPs, transcription to obtain cDNA (Figure 1, panel B1). After that, as detailed in S3 and Figure S1, panel A (Supplementary the cDNA was amplified by double-tagging PCR on the Data). Then, indirect labeling was performed first incubating magnetic beads (Figure 1, panel B2), using a double-tagging with antiCD81 or antiCD326, followed by incubation with the set of primers specific for GAPDH. During PCR, the cDNA is antimouse-Cy5 antibody. In parallel, the same batch of cells not only amplified but also labeled at the same time with and exosomes analyzed by flow cytometry was subjected to biotin/digoxigenin (BIO/DIG) tags. Finally, the electro- confocal microscopy imaging for the study of the binding chemical magneto-genosensing was performed on streptavi- pattern of antibodies. In the case of cells, nuclear DNA was din-magnetic particles as a support, based on the BIO tag stained with Hoechst dye (a blue-fluorescent dye, emission through biotin−streptavidin interactions. The DIG tag was wavelength 490 nm) before labeling with antibodies. Further used for labeling with the antiDIG-HRP conjugate. The experimental details and incubations are described in S4 electrochemical readout of the double-tagged amplicons was (Supplementary Data). based on peroxidase (HRP) enzymes as electrochemical Immunomagnetic Separation, Double-Tagging Re- reporters and performed on m-GEC electrodes, as previously verse Transcription PCR of GAPDH Transcripts, and reported18 (Figure 1, panel C). The experimental details are Electrochemical Genosensing. The procedure was eval- described in the next sections and further detailed in S5 uated on cells and exosomes derived from the MCF7 breast (Supplementary Data). cancer cell line as a model. Briefly, it consists of (i) Immunomagnetic Separation of the Cells and Exosomes. immunomagnetic separation of the cells/exosomes, (ii) Cells (100 μL) (at different concentrations ranging from 50 to double-tagging reverse transcription PCR of GAPDH tran- 5000 cells mL−1) or exosomes (from 100 to 4.0 × 104 scripts, and (iii) electrochemical genosensing. After the exosomes μL−1) were incubated with 1 × 106 antiCDX-MPs optimization, the approach was used for the evaluation of (CDX being either CD81 or CD326) for 30 min at 25 °C exosomes in human serum from breast cancer individuals. while shaking, followed by washing with Tris 1× buffer This approach sequentially combines three different types of containing 0.5% BSA. The content of the preconcentrated cells magnetic separations, as depicted in Figure 1. First, the method or exosomes on antiCDX-MPs was released by resuspending involves the cells or exosome preconcentration based on the them on 1 mL of lysis/binding buffer and disrupted using a specific separation with magnetic particles modified with the syringe. Double-Tagging RT-PCR on Magnetic Beads. The mRNA extraction and purification on polydT-MPs based on the polyA tail of the transcripts was performed, followed by reverse transcription on polydT-MPs (Figure 1, panel B1). The lysate was incubated with 15 μL of polydT-MPs (75 μg, equivalent to 7.5 × 107 MPs) for 15 min under gentle shaking at 25 °C, washed three times, and stored in ice. In order to obtain the cDNA, the retrotranscription (RT) was carried out on poly(dT)-MPs with Moloney Murine Leukemia Virus (M- MLV) reverse transcriptase. The RNA-poly(dT)-MPs were incubated with 10 nmol of dNTPs mix for 5 min at 65 °C and cooled on ice for 1 min. After that, a mix containing 200 nmol of DTT, 40 U of RNaseOUT inhibitor, and 1× First Strand Buffer was added and incubated at 37 °C for 2 min. Finally, 200 U of M-MLV reverse transcriptase were added and incubated for 50 min at 37 °C and 15 min at 70 °C for inactivating the reaction. The cDNA was stored at −21 °C until use. The double-tagging polymerase chain reaction (PCR) was performed in 15 μL of the reaction mixture containing the cDNA (Figure 1, panel B2). Each reaction mixture contained 7.5 pmol of each primer (DIG-Fw and BIO-Rev), 3.75 nmol of each dNTPs, and 3 U of Taq polymerase. The reaction was carried out in a buffer with 7.5 mmol L−1 Tris buffer (pH 9.0), 5.0 mmol L−1 KCl, 2.0 mmol L−1 (NH4)2SO4, and 0.2 mmol L−1 MgCl2 as a cofactor of the enzyme. The reaction mixture was exposed to an initial step at 95 °C for 3 min followed by 32 cycles of 95 °C for 30 s, 61 °C for 30 s, 72 °C for 30 s, and a Figure 1. Schematic representation for the detection of GAPDH expression by the immunomagnetic separation of exosomes (panel last step of 7 min at 72 °C. Negative controls for both the RT A1) and lysis (panel A2); mRNA extraction with poly(dT)-MPs and and PCR were performed as above, except adding mRNA or reverse transcription (panel B1), and double-tagging PCR (panel B2); cDNA, respectively. and electrochemical magneto genosensing with amperometric readout The performance of the double-tagging RT-PCR amplifica- (panel C). Created with BioRender.com. tion was checked with 2% agarose gel electrophoresis in TAE C https://doi.org/10.1021/acs.analchem.2c04773 Anal. Chem. XXXX, XXX, XXX−XXX Analytical Chemistry pubs.acs.org/ac Article Figure 2. (A) Characterization by NTA of purified exosomes derived from the MCF7 breast cancer cell line. (B) Cryo-TEM images (i) and (ii) of purified exosomes at an acceleration voltage of 200 kV. (C) Confocal microscopy images and (D) flow cytometry study for the (i) MCF7 breast cancer cell line and (ii) their exosomes covalently immobilized on MPs. For confocal microscopy, DNA appears blue, magnetic particles in green color, while the exosome protein membrane in red color. For flow cytometry, in all cases, the negative controls (obtained with incubation with the secondary antimouse-Cy5) are shown in blue, while the positive CD326 and CD81 samples obtained with the incubation with the antiCD326 and antiCD81 primary antibodies, are shown in yellow and red, respectively. The negative control obtained with the exosomes-modified MPs without any labeling is shown in gray in panel D (ii). buffer containing 1× GelRed dye. The DNA bands were Electrochemical Magneto-Genosensing. Briefly, after visualized by UV transillumination. A single DNA band was the double-tagging RT-PCR, the BIO-tag was used for the obtained in all samples sized around 371 bp. To confirm that immobilization of the amplicons on streptavidin-magnetic GAPDH was amplified, all bands were cut from the gel and particles through the high affinity biotin−streptavidin inter- purified with the GeneJET kit, and DNA sequencing was action, while the DIG tag allowed the labeling by the antiDIG- performed. HRP, in one 15 min step. The procedure comprised, as RNA Integrity Analysis and DNA Sequencing. For the described in Figure 1, panel C: (a) the immobilization and integrity analysis, the RNA from breast cancer cells and preconcentration of the tagged amplicons on 7 × 107 strep- exosomes was extracted using the total exosome RNA and the MPs and (b) the incubation with the electrochemical reporters protein isolation kit (Thermo Fisher Scientific, ref. 4478545) in one step for 15 min at RT, with 10 μL (130 mU) of and analyzed with the Agilent RNA 6000 Nano Kit (ref. 5067- antiDIG-HRP. Two washing steps with 500 μL of Tris 1× 1511, Agilent) by Genomics Bioinformatics Service (Institute buffer for 2 min at RT were performed. After the incubation or of Biotechnology and Biomedicine, UAB, ES). The DNA washing step, a magnetic separator was positioned under the sequences of the PCR amplicons were obtained with an ABI tubes until pellet formation on the tube side wall, followed by Prism 3130XL Genetic Analyzer by the same GBS and were supernatant separation; (c) magnetic actuation on the m-GEC; analyzed using Chromas v 2.6.6 (Technelysium Pty Ltd., and (d) amperometric readout using applying a potential of Brisbane, QLD, AU) and Clustal Omega21 software to check −100 mV (vs Ag/AgClsat), under enzyme saturation conditions the chromatograms and the alignment of both sequences. in ePBS buffer, upon the addition of hydroquinone and D https://doi.org/10.1021/acs.analchem.2c04773 Anal. Chem. XXXX, XXX, XXX−XXX Analytical Chemistry pubs.acs.org/ac Article Figure 3. Electrochemical genosensing of GAPDH transcripts from (A) MCF7 cells ranging from 50 to 5000 cells mL−1 and (B) their exosomes ranging from 100 to 4.0 × 104 exosomes μL−1, according to NTA counting. In all cases, the cells and exosomes were lysed preconcentrated by IMS using antiCD81-MPs (▲) and antiCD326-MPs (●), followed by double-tagging RT-PCR on poly(dT)-MPs. The error bars show the standard deviation for n = 3. hydrogen peroxide. All experimental steps are described in with U.S. Department of Health and Human Services detail in S7 (Supplementary Data). guidelines for level 2 laboratory Biosafety.22 The steady-state cathodic amperometric current (Icat, in μA) was used for the electrochemical signal plotted in all the figures. Different parameters of the electrochemical genosens- ■ RESULTS AND DISCUSSION Characterization of the Exosomes Derived from ing, such as the washing step time, the incubation time with the MCF7 Breast Cancer Cell Line. An estimation of the size electrochemical reporter, the concentration of strep-MPs, and diameter distribution and the concentration of purified the electrochemical reporter, and finally, the procedure in one exosomes derived from MCF7 breast cancer cell lines was or two steps for electrochemical genosensing were previously performed by nanoparticle tracking analysis (NTA). Figure 2, optimized by our group.18 panel A, shows that the size diameter distribution of exosomes Electrochemical Magneto-Genosensing of Tran- ranges from 30 up to 210 nm, which is represented by scripts from Exosomes of Breast Cancer Patients. exosomes with 90, 120, 150, and 195 nm in diameter, in Blood samples from healthy donors (n = 10, 5 men and 5 accordance with the expected size range for exosomes.23 women, mean age 30/SD = 5) and breast cancer donors (n = Further information of size diameter distribution was obtained 10, stage IV, all women, mean age 50/SD = 6) were obtained by Cryo-TEM. Micrographs of the exosome sample show well- already anonymized from the Hospital del Mar (Barcelona, shape exosomal vesicles with closed circular lipid bilayers ES) and pooled for further separation of the exosomes. The (Figure 2, panel B) around 110 nm in diameter. Cryo-TEM work was carried out in accordance with the principles of micrographs also reveal the presence of some exosome voluntariness and confidentiality. The samples were treated as aggregates. described in S7 (Supplementary Data). In this instance, the Confocal microscopy demonstrated qualitatively that the IMS of the exosomes from 1 mL of pooled and anonymized CD81 and CD326 membrane receptors are well-expressed in undiluted human serum (healthy and breast cancer patients) the MCF7 breast cancer cell line (Figure 2, panel C(i)), as well was directly performed on magnetic particles modified with the as their expression on MCF7-derived exosomes covalently epithelial biomarker CD326 (antiCD326-MPs). The IMS immobilized on magnetic particles (Figure 2, panel C(ii)). The involved the following steps: (i) IMS of the exosomes with intense green color of the magnetic particles was due to antiCD326-MPs (containing 2 × 106 MPs per tube) and 1.0 autofluorescence around 580 nm.24 Negligible nonspecific mL of human serum were simultaneously incubated for 30 min adsorption was observed (Figure 2, panel C(ii), negative). The with gentle shaking at 25 °C, followed by washing with Tris 1× CD81 tetraspanin was also shown with strong labeling in buffer containing 0.5% BSA. Then, the exosome-coated exosomes, although a poor labeling pattern was achieved for antiC326-MPs were resuspended with 100 μL of Tris 1× the CD326 biomarker. buffer, stored on ice, and immediately used for RNA Quantitative patterns of the MCF7 cell line were also extraction. All further steps were performed as described in studied by flow cytometry analyzing the expression to CD81 the Experimental Section, including double-tagging reverse and CD326, as shown in Figure 2, panel D. The negative transcription PCR of GAPDH transcripts and electrochemical control in which the signal appears onto the left side confirms genosensing. The complete assay protocols, as well as the that there is a negligible (<0.1%) nonspecific reaction with the preparation of human serum from blood, are provided in S7 secondary antibody (antimouse-Cy5 antibody) with the MCF7 (Supplementary Data). cells (Figure 2, panel D(i)). As expected, the percentage Statistical Analysis. The statistical analyses and calcu- marker expression to CD81 and CD326 biomarkers was high lations were performed using GraphPad Prism 8 (San Diego, as >95% for MCF7 cells (Figure 2, panel D(i)). The same CA, US) while plots were represented using Origin Pro 2017 CD81 and CD326 biomarkers on exosomes derived from the (Northampton, MA, US). MCF7 breast cancer cell line were also studied by flow Safety Considerations. All studies were performed in a cytometry. As expected, exosomes covalently immobilized on biosafety cabinet, and all material decontaminated by magnetic particles (exosomes-MPs) highly expressed CD81, autoclaving or disinfected before discarding in accordance but CD326 showed a low expression pattern with this model E https://doi.org/10.1021/acs.analchem.2c04773 Anal. Chem. XXXX, XXX, XXX−XXX Analytical Chemistry pubs.acs.org/ac Article (<5%) (Figure 2, panel D(ii)), in accordance with the results positive selection of CD326 exosomes by IMS and obtained by confocal microscopy. Tetraspanins are the most preconcentration, followed by double-tagging end-point RT- frequently identified proteins in exosomes and are considered PCR at 32 cycles for GAPDH, which improves the sensitivity classical markers. Comparing the expression levels of cells and of the approach. IMS of exosomes improves analytical their derived exosomes, and according to many studies,25−27 simplification, avoiding ultracentrifugation or other separation cell-membrane biomarkers are not always identically expressed steps and have the advantage of a specific capture of exosomes in the cells as well as in their derived exosomes. These data by epithelial breast cancer biomarker, which is currently used suggest that the exosomal molecular profile needs to be in most of the CTC-enrichment methods such as CellSearch.7 carefully assessed to achieve a better experimental approach Since the number of exosomes in plasma ranges from 105 to design. 109 exosomes μL−1,30 the LOD for double-tagging RT-PCR Regarding the genetic material in MCF7 cells and exosomes, based on GAPDH transcripts coupled with electrochemical it was characterized by RNA integrity analysis, obtaining genosensing was feasible and reliable to detect and quantify significantly different patterns, as shown in S6 (Supplementary cancer-related exosomes. Data). In this approach, the exosomes are specifically isolated and Immunomagnetic Separation, Double-Tagging Re- preconcentrated by IMS, while the double-tagging RT-PCR is verse Transcription PCR of GAPDH Transcripts, and used as a strategy to amplify the signal and thus improve the Electrochemical Genosensing. First, all steps from the LOD by an order of 2, compared to previous studies (from 1 × proposed IMS/double-tagging RT-PCR/electrochemical gen- 105 to 4 × 102 exosomes μL−1).31,32 For this reason, in this osensing detection method were tested and optimized with approach, a common and ubiquitous transcript based on MCF7 cells and purified exosomes, as described in S5 GAPDH was selected for the double-tagging amplification,33 (Supplementary Data). The DNA sequences of the PCR while the previous studies were based on a second labeled amplicons were also obtained. The genome sequence for the antibody in a sandwich immunosensing format27,31 or the Homo sapiens mRNA GAPDH transcript was identified by intrinsic alkaline phosphatase enzyme activity in exosomes.32 using BLAST software.28 Further details are provided in S6 The main shortcoming is that in this approach, the unique and Figure S3 (Supplementary Data). source of specificity is provided by the antibody on the MPs The calibration plots for the detection of GAPDH during the IMS. transcripts from MCF7 cells and its derived exosomes are The LOD obtained in this work was better in analytical comparatively shown in Figure 3. Two different immunomag- performance than in fluorescence,34 electrochemical,35 and netic separation (IMS) approaches were tested. First, IMS of surface-enhanced Raman scattering36 devices, and comparable MCF7 cells by using antiCDX-MPs (where CDX is any of to other reported approaches, such as rolling circle CD81 or CD326 biomarkers), followed by the double-tagging amplification37 and microfluidic graphene oxide-based38 RT-PCR and electrochemical genosensing. Thus, different detection. concentrations of MCF7 cells ranging from 50 to 5000 cells Electrochemical Magneto-Genosensing of Tran- mL−1 were evaluated for the calibration plot. The electro- scripts from Exosomes of Breast Cancer Patients. The chemical responses were fitted using nonlinear regression performance of the double-tagging RT-PCR on MPs and (four-parameter logistic equation, GraphPad prism software) electrochemical genosensing was evaluated in serum-derived (Figure 3, panel A). The limit of detection (LOD) of 45 cells exosomes from healthy and breast cancer patients. The mL−1 (r2 = 0.996) and 67 cells mL−1 (r2 = 0.998) was reached procedure and the results are depicted in Figure 4. All for cells immunocaptured by using CD81 and CD326 experimental parameters are described in S7 (Supplementary biomarkers, respectively. Although the strategy was able to Data). First, the GAPDH expression was evaluated in purified clearly detect cells by GAPDH transcript amplification and exosomes (without preconcentration on MPs) derived from improved analytical simplification, these LOD values are not healthy controls and breast cancer patients (Figure 4, panel A), suitable for applications in breast cancer diagnosis, since the normalized per micrograms of exosomes (BCA protein assay clinical CTC count assay approved by the U.S Food and Drug results are detailed in S7, Supplementary Data). In this Administration (FDA) must be smaller than 5 cells per 7.5 approach, the specific IMS and as such the positive selection of mL−1.7 At this point, considering the performance of the CD326 exosomes (Figure 4, panel B) were replaced by a proposed genosensor, the study was further focused on cancer- nonspecific physical isolation (ultracentrifugation at 100,000 × related exosomes. These extracellular vesicles are considered as g). In this case, the approach is based on amplification and new biomarkers for the detection of cancer in early stages, detection through a nonspecific GAPDH biomarker. Then, the since it is related to cell-to-cell communication and increased double-tagging RT-PCR for the specific GAPDH transcripts in cancer cells.11,29 on poly(dT)-MPs was performed using 0.33 μg per assay of The IMS of exosomes (ranged from 100 to 4.0 × 104 serum-derived exosomes from both groups of samples, exosomes μL−1) derived from the MCF7 cell line was followed by subsequent electrochemical genosensing detection. performed on antiCD81-MPs and/or antiCD326-MPs fol- The results of this analysis are shown in Figure 4, panel A. The lowed by the double-tagging RT-PCR for the specific GAPDH results suggested that breast cancer patients overexpress transcripts on poly(dT)-MPs and subsequent electrochemical GAPDH in total exosomes (6.7-fold) and can be well genosensing. The electrochemical responses were fitted using discriminated from healthy individuals. nonlinear regression (four-parameter logistic equation, Graph- Next, in order to achieve the analytical simplification, IMS of Pad prism software) (Figure 3, panel B). The LOD of 415 exosomes directly from human serum by using antiCD326- exosomes μL−1 (r2 = 0.991) and 1225 exosomes μL−1 (r2 = MPs was performed, followed by RNA extraction and PCR on 0.980) were reached using CD81 and CD326 biomarkers, poly(dT)-MPs of primer specific for GAPDH transcript respectively. Although the expression of CD326 is much lower labeled with DIG/BIO tags and subsequent electrochemical than that of CD81, this effect is minimized by using the genosensing. In this case, the approach is based on the specific F https://doi.org/10.1021/acs.analchem.2c04773 Anal. Chem. XXXX, XXX, XXX−XXX Analytical Chemistry pubs.acs.org/ac Article exosomes from breast cancer patients. This is in accordance with the highly expressed GAPDH in breast cancer cells.42 ■ CONCLUSIONS Early diagnosis of breast cancer by standard techniques remains a difficult task due to the low specificity, availability, and high cost, added to the lack of specific symptoms in the early stage and the small size of the primary tumor. The study of novel biomarkers including exosomes are currently under intense investigation. Here, a double-tagging RT-PCR on magnetic beads and electrochemical genosensing demonstrated high sensitivity and specificity for GAPDH gene expression based on specific epithelial CD326 cancer-related exosomes, being able to detect the transcripts produced by as low as 1225 Figure 4. Panel A shows the control of the purified total exosome exosomes μL−1. population obtained by ultracentrifugation (100,000 × g) normalized Although further studies should be done with samples in according to the protein content (0.33 μg per assay). Panel B. Electrochemical genosensing of CD326+ exosomes from 1 mL of cell- early stages of cancer, our data clearly suggest the GAPDH free undiluted human serum (centrifuged at 10,000 × g) based on expression in total exosomes from human serum from healthy immunomagnetic separation with antiCD326-MP and further and breast cancer patients, which revealed that the GAPDH GAPDH transcripts detection. The whole procedure is also shown gene is overexpressed by 6.7-fold in breast cancer women at in Figure 1. In all cases, serum-derived exosomes from healthy stage IV, when compared to the healthy controls. Also, CD326 controls (n = 10, pooled) and breast cancer (n = 10, pooled) patients were processed. The error bars show the standard deviation for n = 3. (+) exosomes specifically immunocaptured expressed the The raw amperograms are also shown in Figure S4 (Supplementary GAPDH gene as high as 3.3-fold, when compared to the Data). Created with BioRender.com. healthy controls. Interestingly, the positive selection of exosomes expressing CD326 by using immunomagnetic separation provides an effective strategy to separate exosomes capture of exosomes by the CD326 epithelial cancer-related with higher efficiency than ultracentrifugation, achieving a mild biomarker and further detection through a nonspecific and specific approach for the isolation and preconcentration a GAPDH mRNA biomarker. The electrochemical genosensor subpopulation of exosomes. This strategy is amenable with is performed in 1 mL of human serum only pretreated by a point-of-care diagnosis since MPs can be easily integrated in short centrifugation pulse at 10,000 × g (to eliminate any different IVD platforms, including biosensing devices. remaining cells or particulate debris), followed by IMS with In conclusion, although further clinical validation should be antiCD326-MP and electrochemical magneto-genosensing, as performed with a higher number of samples at early stages, the depicted in Figure 1. The results are presented in Figure 4, panel B. To confirm significant increase of the GAPDH transcript content in the significance of the differences in the value for the healthy exosomes from patients compared to healthy individuals control and breast cancer patient samples, a one-tailed p-test envisages its role as a putative biomarker for breast cancer (Hi > Ho) at a 95% significance level was performed, Hi being diagnostics and monitoring of metastatic disease. This work hypothesis and Ho the null hypothesis (as detailed in S7, thus shows a promising strategy for being implemented at Supporting Data). Notably, a significant overexpression (3.3- primary health care in low-resource settings based on a fold) of GAPDH on the immunocaptured CD326 positive minimally invasive liquid biopsy. exosomes in breast cancer samples when compared with serum-derived exosomes from healthy individuals. The signal obtained from the CD326-positive exosomes from healthy individuals is probably due to some biomarkers ■ ASSOCIATED CONTENT * Supporting Information sı such as CD326 that may also exist on the surface of the The Supporting Information is available free of charge at nontumorigenic cell-derived exosomes.39 Thus, it is expected https://pubs.acs.org/doi/10.1021/acs.analchem.2c04773. that exosomes derived from healthy individuals also contain low amounts of CD326 biomarkers in exosomes, but at Chemicals and biochemicals; buffers and solutions; cell increased levels in cancer-related exosomes from various culturing, exosome isolation, and purification from carcinomas.40 Magnetic particles used for exosome separation MCF7 cell lines; immobilization of exosomes and avoid the amplification of free mRNA that can be present in antibodies on magnetic particles; Immobilization of the serum samples. As mentioned in previous publications,41 exosomes on magnetic particles; covalent immobiliza- the expression of GAPDH may vary in several situations, given tion; characterization of the exosomes derived from the that it is a multifunctional protein involved in more than ten MCF7 breast cancer cell line; electrochemical magneto- functions in mammalian cells. Although GAPDH has been genosensing; RNA integrity analysis and DNA sequenc- considered as a housekeeping gene in several studies for gene ing; immunomagnetic separation of the exosomes from expression normalization, its expression can vary in diseases in undiluted human serum; immunomagnetic separation of which the metabolic state of the cells is altered.41 To the best the exosomes from undiluted human serum; detection of of the authors’ knowledge, this is the first study that reports an GAPDH transcripts from purified exosomes; and overexpression of the GAPDH gene on serum-derived double-tagging RT-PCR on magnetic beads (PDF) G https://doi.org/10.1021/acs.analchem.2c04773 Anal. Chem. XXXX, XXX, XXX−XXX Analytical Chemistry pubs.acs.org/ac Article ■ AUTHOR INFORMATION Corresponding Author (7) Allard, W. J.; Matera, J.; Miller, M. C.; Repollet, M.; Connelly, M. C.; Rao, C.; Tibbe, A. G. J.; Uhr, J. W.; Terstappen, L. W. M. M. Clin. Cancer Res. 2004, 10, 6897−6904. Maria Isabel Pividori − Grup de Sensors i Biosensors, (8) de Wit, S.; Manicone, M.; Rossi, E.; Lampignano, R.; Yang, L.; Departament de Química and Biosensing and Bioanalysis Zill, B.; Rengel-Puertas, A.; Ouhlen, M.; Crespo, M.; Berghuis, A. M. Group, Institute of Biotechnology and Biomedicine, S.; Andree, K. C.; Vidotto, R.; Trapp, E. K.; Tzschaschel, M.; Universitat Autònoma de Barcelona, Bellaterra 08193, Colomba, E.; Fowler, G.; Flohr, P.; Rescigno, P.; Fontes, M. S.; Spain; orcid.org/0000-0002-5266-7873; Phone: +34 93 Zamarchi, R.; Fehm, T.; Neubauer, H.; Rack, B.; Alunni-Fabbroni, 581 2806; Email: isabel.pividori@uab.cat; Fax: +34 93 M.; Farace, F.; de Bono, J.; IJzerman, M. J.; Terstappen, L. W. M. M. 581 2379 Oncotarget 2018, 9, 35705−35716. (9) Ross, B. A. A.; Cooper, B. W.; Lazarus, H. M.; Mackay, W.; Authors Moss, T. J.; Ciobanu, N.; Tallman, M. S.; Kennedy, M. J.; Davidson, Arnau Pallares-Rusiñol − Grup de Sensors i Biosensors, N. E.; Sweet, D. Blood 1993, 82, 2605−2610. Departament de Química and Biosensing and Bioanalysis (10) Johnstone, R. M.; Adam, M.; Hammond, J. R.; Orr, L.; Turbide, Group, Institute of Biotechnology and Biomedicine, C. J. Biol. Chem. 1987, 262, 9412−9420. Universitat Autònoma de Barcelona, Bellaterra 08193, Spain (11) Mathieu, M.; Martin-Jaular, L.; Lavieu, G.; Théry, C. Nat. Cell Silio Lima Moura − Grup de Sensors i Biosensors, Biol. 2019, 21, 9−17. (12) Harding, C.; Heuser, J.; Stahl, P. Eur. J. Cell Biol. 1984, 35, Departament de Química, Universitat Autònoma de 256−263. Barcelona, Bellaterra 08193, Spain (13) Samanta, S.; Rajasingh, S.; Drosos, N.; Zhou, Z.; Dawn, B.; Mercè Martí − Biosensing and Bioanalysis Group, Institute of Rajasingh, J. Acta Pharmacol. Sin. 2018, 39, 501−513. Biotechnology and Biomedicine, Universitat Autònoma de (14) Davidson, S. M.; Riquelme, J. A.; Zheng, Y.; Vicencio, J. M.; Barcelona, Bellaterra 08193, Spain Lavandero, S.; Yellon, D. M. Sci. Rep. 2018, 8, 15885. Complete contact information is available at: (15) Pividori, M. I.; Merkoçi, A.; Barbé, J.; Alegret, S. Electroanalysis https://pubs.acs.org/10.1021/acs.analchem.2c04773 2003, 15, 1815−1823. (16) Lermo, A.; Zacco, E.; Barak, J.; Delwiche, M.; Campoy, S.; Barbé, J.; Alegret, S.; Pividori, M. I. Biosens. Bioelectron. 2008, 23, Author Contributions 1805−1811. The manuscript was written through contributions of all (17) Liébana, S.; Lermo, A.; Campoy, S.; Barbé, J.; Alegret, S.; authors. All authors have given approval to the final version of Pividori, M. I. Anal. Chem. 2009, 81, 5812−5820. the manuscript. A.P.-R. and S.L.M. contributed equally. (18) Carinelli, S.; Xufré, C.; Martí, M.; Pividori, M. I. Biosens. Funding Bioelectron. 2018, 117, 183−190. This work was funded by the Ministry of Science and (19) Ben Aissa, A.; Jara, J. J.; Sebastián, R. M.; Vallribera, A.; Innovation (PID2019-106625RB-I00/AEI/10.13039/ Campoy, S.; Pividori, M. I. Biosens. Bioelectron. 2017, 88, 265−272. 501100011033) and from Generalitat de Catalunya (2017- (20) Théry, C.; Amigorena, S.; Raposo, G.; Clayton, A. Curr. Protoc. Cell Biol. 2006, 30, Unit 3.22. SGR-220). Also, the Ministry of Universities (Grant FPU16/ (21) Chojnacki, S.; Cowley, A.; Lee, J.; Foix, A.; Lopez, R. Nucleic 01579) is gratefully acknowledged. Acids Res. 2017, 45, W550−W553. Notes (22) Hamaguchi, J. R.; Tobey, R. A.; Pines, J.; Crissman, H. A.; The authors declare no competing financial interest. Hunter, T.; Bradbury, E. M. J. Cell Biol. 1992, 117, 1041−1053. ■ ACKNOWLEDGMENTS C. Garcia Martín is gratefully acknowledged for her (23) Chernyshev, V. S.; Rachamadugu, R.; Tseng, Y. H.; Belnap, D. M.; Jia, Y.; Branch, K. J.; Butterfield, A. E.; Pease, L. F.; Bernard, P. S.; Skliar, M. Anal. Bioanal. Chem. 2015, 407, 3285−3301. (24) Agrawal, A.; Sathe, T.; Nie, S. J. Agric. Food Chem. 2007, 55, contribution. Also, ICTS “NANBIOSIS” NTA analysis service 3778−3782. of Institut de Ciència dels Materials de Barcelona and Service (25) Rupp, A.-K.; Rupp, C.; Keller, S.; Brase, J. C.; Ehehalt, R.; of Microscopy of Universitat Autònoma de Barcelona are Fogel, M.; Moldenhauer, G.; Marmé, F.; Sültmann, H.; Altevogt, P. gratefully acknowledged. Gynecol. Oncol. 2011, 122, 437−446. ■ CTC ABBREVIATIONS circulating tumor cell (26) Stoeck, A.; Keller, S.; Riedle, S.; Sanderson, M. P.; Runz, S.; Le Naour, F.; Gutwein, P.; Ludwig, A.; Rubinstein, E.; Altevogt, P. Biochem. J. 2006, 393, 609−618. EpCAM epithelial cell adhesion molecule (27) Moura, S. L.; Martín, C. G.; Martí, M.; Pividori, M. I. Talanta 2020, 211, No. 120657. GAPDH glyceraldehyde-3-phosphate dehydrogenase (28) Agarwala, R.; Barrett, T.; Beck, J.; et al. Nucleic Acids Res. 2018, IMS immunomagnetic separation 46, D8−D13. RT-PCR reverse transcription polymerase chain reaction. (29) Théry, C.; Witwer, K. W.; Aikawa, E.; et al. J. Extracell. Vesicles ■ REFERENCES (1) Sung, H.; Ferlay, J.; Siegel, R. L.; Laversanne, M.; 2018, 7, No. 1535750. (30) Johnsen, K. B.; Gudbergsson, J. M.; Andresen, T. L.; Simonsen, J. B. Biochim. Biophys. Acta - Rev. Cancer 2019, 1871, 109−116. Soerjomataram, I.; Jemal, A.; Bray, F. CA Cancer J. Clin. 2021, 71, (31) Moura, S. L.; Martín, C. G.; Martí, M.; Pividori, M. I. Biosens. 209−249. Bioelectron. 2020, 150, No. 111882. (2) Weissleder, R.; Pittet, M. J. Nature 2008, 452, 580−589. (32) Moura, S. L.; Pallarès-Rusiñol, A.; Sappia, L.; Martí, M.; (3) Unger-Saldaña, K. World J. Clin. Oncol. 2014, 5, 465−477. Pividori, M. I. Biosens. Bioelectron. 2022, 198, No. 113826. (4) Japp, N. C.; Souchek, J. J.; Sasson, A. R.; Hollingsworth, M. A.; (33) Lermo, A.; Campoy, S.; Barbé, J.; Hernández, S.; Alegret, S.; Batra, S. K.; Junker, W. M. J. Immunol. Res. 2021, 2021, 1. Pividori, M. I. Biosens. Bioelectron. 2007, 22, 2010−2017. (5) Poudineh, M.; Sargent, E. H.; Pantel, K.; Kelley, S. O. Nat. (34) Zhao, Z.; Yang, Y.; Zeng, Y.; He, M. Lab Chip 2016, 16, 489− Biomed. Eng. 2018, 2, 72−84. 496. (6) Miller, M. C.; Doyle, G. V.; Terstappen, L. W. M. M. J. Oncol. (35) Zhou, Q.; Rahimian, A.; Son, K.; Shin, D. S.; Patel, T.; Revzin, 2010, 2010, 1. A. Methods 2016, 97, 88−93. H https://doi.org/10.1021/acs.analchem.2c04773 Anal. Chem. XXXX, XXX, XXX−XXX Analytical Chemistry pubs.acs.org/ac Article (36) Zong, S.; Wang, L.; Chen, C.; Lu, J.; Zhu, D.; Zhang, Y.; Wang, Z.; Cui, Y. Anal. Methods 2016, 8, 5001−5008. (37) Huang, L.; Wang, D.-B.; Singh, N.; Yang, F.; Gu, N.; Zhang, X.- E. Nanoscale 2018, 10, 20289−20295. (38) Zhang, P.; He, M.; Zeng, Y. Lab Chip 2016, 16, 3033−3042. (39) Schmelzer, E.; Reid, L. M. Front. Biosci. 2008, 13, 3096−3100. (40) Tamkovich, S. N.; Yunusova, N. V.; Somov, A. K.; Afanas’ev, S. G.; Kakurina, G. V.; Kolegova, E. S.; Tugutova, E. A.; Laktionov, P. P.; Kondakova, I. V. Biochem. Moscow Suppl. B Biomed. Chem. 2018, 12, 151−155. (41) Sirover, M. A. Biochim. Biophys. Acta - Gen. Subj. 2011, 1810, 741−751. (42) Révillion, F.; Pawlowski, V.; Hornez, L.; Peyrat, J.-P. Eur. J. Cancer 2000, 36, 1038−1042. I https://doi.org/10.1021/acs.analchem.2c04773 Anal. Chem. XXXX, XXX, XXX−XXX