Correlation between cancerous exosomes and protein markers based

2 hours ago - Exosomes, which are nano-vesicles secreted by cells, are promising biomarkers for cancer diagnosis and prognosis, based on their specifi...
0 downloads 0 Views 874KB Size
Download PDF
Subscriber access provided by University of Sunderland

Article

Correlation between cancerous exosomes and protein markers based on surface-enhanced Raman spectroscopy (SERS) and principal component analysis (PCA) Hyunku Shin, Hyesun Jeong, Jaena Park, Sunghoi Hong, and Yeonho Choi ACS Sens., Just Accepted Manuscript • DOI: 10.1021/acssensors.8b01047 • Publication Date (Web): 01 Nov 2018 Downloaded from http://pubs.acs.org on November 2, 2018

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

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

Page 1 of 9 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

ACS Sensors

Correlation between cancerous exosomes and protein markers based on surface-enhanced Raman spectroscopy (SERS) and principal component analysis (PCA) Hyunku Shin1, Hyesun Jeong2, Jaena Park1, Sunghoi Hong2,3, and Yeonho Choi1,4,* 1Department

of Bio-convergence Engineering, Korea University, Seoul, 02841, South Korea. of Integrated Biomedical and Life Science, Korea University, Seoul, 02841, South. 3School of Biosystem and Biomedical Science, Korea University, Seoul, 02841, South Korea. 4School of Biomedical Engineering, Korea University, Seoul, 02841, South Korea. KEYWORDS. exosome, lung cancer, protein, surface enhanced Raman spectroscopy, principal component analysis 2Department

ABSTRACT: Exosomes, which are nano-vesicles secreted by cells, are promising biomarkers for cancer diagnosis and prognosis, based on their specific surface protein compositions. Here, we demonstrate the correlation of non-small cell lung cancer (NSCLC) cell-derived exosomes and potential protein markers by unique Raman scattering profiles and principal component analysis (PCA) for cancer diagnosis. Based on surface enhanced Raman scattering (SERS) signals of exosomes from normal and NSCLC cells, we extracted Raman patterns of cancerous exosomes by PCA and clarify specific patterns as unique peaks through quantitative analysis with ratiometric mixtures of cancerous and normal exosomes. The unique peaks correlated well with cancerous exosome ratio (R2 > 90%) as the unique Raman band of NSCLC exosome. To examine the origin of the unique peaks, we compared these unique peaks with characteristic Raman bands of several exosomal protein markers (CD9, CD81, EpCAM, and EGFR). EGFR had 1.97-fold similarity in Raman profiles than other markers, and it showed dominant expression against the cancerous exosomes in an immunoblotting result. We expect that these results will contribute to studies on exosomal surface protein markers for diagnosis of cancers.

Lung cancer has a high mortality rate.1 Radiological imaging and biopsy are the most common approaches utilized to screen for lung cancer, but they cannot be conducted repeatedly because of excessive radiation exposure, substantial cost, and discomfort to the patient.2 Repetitive and non-invasive liquid biopsies to detect lung cancer biomarkers in body fluids may be an effective method for early diagnosis and prognostic prediction. Early diagnosis and prognosis can improve the survival rate of patients.3,4 Many researchers have reported several specific biomarkers containing molecular information regarding tumors, such as circulating tumor cells, cell-free circulating tumor DNA, and extracellular vesicles. These markers are expected to be practical targets for the screening of in vivo cancer development. Exosomes are promising biomarkers for cancer diagnosis. Exosomes are 30–200 nm vesicles secreted by all living cells.5 They possess the molecular and genetic characteristics of their original cell, such as proteins, lipids, and RNA.6 Many body fluids, such as blood, ascites, saliva, and urine, contain abundant exosomes.7 Exosomes play important roles in cancer pathology, indicating their usefulness in the early diagnosis of cancer.8 Therefore, many studies have examined the molecular information in exosomes as biomarkers for the diagnosis and prognosis of cancer.9,10

The surfaces of cancerous exosomes can contain membrane proteins from the original cell. Therefore, the surface proteins of cancerous exosomes are of great interest to cancer diagnostics researchers. Although the unique surface proteins on cancerous exosomes remain unclear, many researchers have suggested that certain proteins are useful as cancer diagnostic and prognostic markers.11-13 Therefore, various techniques, such as immunoblotting, proteomics, enzyme-linked immunosorbent assays, zeta potential analysis, electron microscopy, fluorescence, and surface plasmon resonance, have been utilized to investigate these surface protein markers and have significantly advanced the study of cancerous exosomes.14-20 Raman spectroscopy is a suitable tool for investigating the protein markers of exosomes. Raman spectroscopy facilitates the sensing of molecular fingerprints as the vibrational and rotational modes of chemical bonding structures through sharp and narrow spectral peaks.21-25 However, the typical signal intensity of spontaneous Raman spectroscopy is extremely weak. Surface-enhanced Raman spectroscopy (SERS) is a representative method for the amplification of subtle signal intensities based on strong electromagnetic (EM) fields generated in metallic nanogaps. Signals in close proximity to the nanogaps dominate the overall spectrum.26 Therefore, the exosomal surface proteins located near the nanogaps in a SERS substrate can

ACS Paragon Plus Environment

ACS Sensors 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

produce enhanced Raman scattering profiles. These profiles provide information regarding the protein composition on an exosome. If the surface composition of a cancerous exosome differs from that of a normal cell-derived exosome, the unique Raman scattering profiles of cancerous exosomes may show noticeable differences.

Page 2 of 9

exosome surface is heterogeneous. Therefore, we applied principal component analysis (PCA) to identify the major patterns in the Raman spectra of the exosomes. PCA is a statistical analysis method utilized to investigate major patterns in datasets. We identified the unique spectral peaks of cancerous exosomes utilizing PCA and ratiometric analysis on mixtures containing normal and cancerous exosomes. To investigate the origin of the peaks, we analyzed the Raman scattering profiles of several exosomal surface protein markers. CD9 and CD81 were chosen as common protein markers expressed on the surfaces of all exosomes.11 Additionally, as representative diagnostic and prognostic markers for NSCLC exosomes, the Raman profiles of epithelial cell adhesion molecules (EpCAM) and epidermal growth factor receptors (EGFR) were measured. EpCAM is a protein on the cell membrane that is likely present on exosome surfaces and associated with lung cancer stem cells and carcinogenesis.27,28 Therefore, EpCAM has been known as an attractive potential prognostic factor29. EGFR has been reported as one of the few membrane-bound proteins among NSCLC exosome markers.11 The EGFR mutation status in NSCLC has been reported to be correlated with survival rates.29 We identified an association between the Raman spectra of these protein markers and NSCLC exosomes. Additionally, we verified our results through immunoblotting to confirm protein marker expression.

MATERIALS AND METHODS

Figure 1. Detection of unique Raman scattering profiles of lung cancer cell-derived exosomes and comparison to the profiles of their potential surface protein markers. Cancerous exosomespecific protein markers are associated in terms of signal similarity.

In this study, we identified the unique SERS profiles of non-small-cell lung cancer (NSCLC) cell-derived exosomes and potential protein markers that likely contribute to these unique profiles (Figure 1). For SERS characterization, an aggregated gold nanoparticles (GNPs) substrate was fabricated to form nanogaps. A strong EM field forms in the nanogaps between aggregated nanoparticles.22 We obtained exosomal SERS spectra in a liquid state to prevent deformation of the signal through damage to the exosome or salt formation. Therefore, the SERS substrate was coated with cysteamine to induce the electrostatic adsorption of exosomes. Based on the anionic surfaces of exosomes, the cationic amino groups of the cysteamine can capture exosomes in solution. However, each spectrum can exhibit heterogeneous peak compositions, even in the case of exosomes derived from the same cell, because the molecular composition on the

Materials. An 80-nm gold nanoparticle colloidal solution was purchased from BBI Solutions (UK). 1-Ethyl-3-(3(dimethylamino)propyl)carbodiimide (EDC) was purchased from Daejung chemicals. Cysteamine, APTES, bovine serum albumin, and N-hydroxysuccinimide (NHS) were purchased from Sigma-Aldrich (St. Louis, MO). AntiCD9 (sc-13118) and EGFR (sc-373746) antibodies were purchased from Santa Cruz Biotechnology (Dallas, TX). Anti-CD81 (NB100-65805SS) and EpCAM (NBP2-45448) were purchased from Novus Biologicals (Littleton, CO). All antibody solutions were prepared at 200 mg/ml in PBS. Isolation and validation of exosomes. Exosomes in cell culture media were isolated through size-exclusion column chromatography, as described in a prior study.30 The column was filled with 2B sepharose beads. The cell culture media was centrifuged at 500 × g, 5,000 × g, and 10,000 × g to remove the detached cells and debris. The resulting supernatant was concentrated through a 100-kDa filter, followed by loading into the column. Each eluted fraction contained different sizes of particles. We measured particle sizes via DLS and NTA. The zeta potentials of the exosomes were measured by a zeta potential analyzer (ELSZ-1000, Otsuka Electronics). Fractions containing 100 nm particles were utilized for SERS characterization and stored at -80 °C prior to analysis.

2 ACS Paragon Plus Environment

Page 3 of 9 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

ACS Sensors Next, the substrate was incubated in a 1 mM APTES ethanoic solution for 1 h, then rinsed with ethanol and dried with N2 gas. A 10 OD GNP colloidal solution was dropped onto the substrate and thoroughly dried. The substrate was washed with deionized water to remove physically adsorbed GNP and then dried. The resulting GNP substrate was functionalized with cysteamine in a 20 mM ethanolic solution for 1 hr. SERS characterization. For SERS characterization of the exosomes, we prepared an exosome solution at a pH of 6.5 because the amine groups of cysteamine are expected to be positively charged by protonation under acidic conditions.32 Next, the diluted exosome solution containing 1 x 108 particles/ml was dropped onto the substrate, followed by spectrum measurements after 2 h. For protein markers, the SERS substrate was functionalized with 100 μl of 4-mM EDC and 10-mM NHS solutions, then incubated for 10 min. Next, 10 μl of the antibody solution was added and incubated at 4 °C for 24 h. The substrate was thoroughly washed four times. 1% (w/v) BSA for blocking was applied for 1 h. The substrate was then washed four more times. We obtained the Raman spectra of the antibodies on the SERS substrate prior to conjugating target proteins. Next, 100 μl of the lysate solution of exosomes was added onto the substrate and incubated overnight at 4 °C. We washed off the residual lysate and measured the Raman spectra at the same location. The SERS spectra were measured by an inverted-type microscope (Axio Observer D1, Zeiss) equipped with a spectrometer (Acton SP2300, Princeton Instruments). A 5 mW 785-nm laser was utilized to irradiate the aggregated GNPs on the substrate and a 1,340 × 400 pixels cooled spectrograph detector (PIXIS400, Princeton Instruments) collected the scattering light that passed through the 785 nm filter. The acquisition time was 10 s. Baseline correction and denoising operations were applied to all spectra utilizing the chromatogram baseline estimation and denoising using sparsity (BEADS) toolbox and built-in functions of MATLAB (Figure S1).33

Results and discussion

Figure 2. SERS characterization and PCA for investigating the specific Raman scattering signals of exosomes derived from NSCLC cells. (A) SERS spectra of exosomes derived from HPAEC (normal) and H1299, PC9 (lung cancer) cell lines, with PBS as a control. Grey box represents characteristic Raman bands estimated to be derived from proteins. (B) PCA score plot of the SERS data. The 90% confidence ellipses are illustrated. (C) PC1 loading data of (B). The black triangles indicate 26 peaks as candidates for the unique Raman bands of NSCLC exosomes.

Fabrication of SERS substrate. A cover glass was immersed in a piranha solution (H2SO4: H2O2 = 3:1) for 30 min to remove organic matter from the surface. The resulting substrate was rinsed with deionized water and ethanol, followed by thorough drying with pure N2 gas.

Isolation and validation of exosomes. We isolated exosomes from the human pulmonary alveolar epithelial cell (HPAEC) and NSCLC cell (PC9 and H1299) culture supernatants (Figure S2). Utilizing dynamic light scattering (DLS), particles with sizes of approximately 100 nm were eluted in fractions 8–10. In our nanoparticle tracking analysis (NTA) results, the particle concentrations were 1– 10 × 109 particles/ml. In our western blotting results, exosome markers (CD9) were observed in fractions 8–10. We utilized mixed fractions 8 and 9 for experimental samples. The zeta potential of the exosomes was negatively charged at a pH of 6.5 (Table S1). We verified that the exosomes were not damaged at this condition (Figure S3). Based on the surface charges of the exosomes, we determined that the extent to which the exosomes attached to cysteamine-coated gold nanoparticles was largely unaffected by the secreted cells.

ACS Paragon Plus Environment

3

ACS Sensors 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 9

Figure 3. Ratiometric analysis of mixtures. The mixtures contained different ratios of exosomes from NSCLC cell lines (PC9) and normal alveolar cell lines (HPAEC). (A) SERS spectra of exosomes in the mixtures. Each graph represents the average of 15 SERS spectra. Blend percentages are based on the particle numbers of exosomes in the mixtures, as measured by NTA. (B) The intensities of the Raman peaks gradually increased as the cancerous exosome ratio increased by 25%, 50%, and 75%. R2 represents the coefficient of determination.

Verification of SERS substrate. Next, we verified the SERS substrate. The dried GNPs on the 3aminopropyltriethoxysilane (APTES)-coated substrate formed aggregated clusters through an edge based on the coffee-stain effect (Figure S4). Following the

functionalization of cysteamine for the electrostatic adsorption of exosomes, strong characteristic Raman bands of the cysteamine appeared at 640–650 cm-1 and 725–740 cm-1 (Figure S5).34

4 ACS Paragon Plus Environment

Page 5 of 9 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

ACS Sensors

Figure 4. Identification of specific Raman bands of exosomal protein markers and their similarity to NSCLC exosomes. (A) PC1 scores of corresponding protein markers. (B) PC1 loading data of the protein markers. The red lines represent the ranges of the unique Raman bands of the NSCLC exosomes. (C) Similarity to NSCLC exosomes. Similarity was calculated as the reciprocal of Euclidean distance.

However, the deformation of analytes on the SERS substrate caused by photothermal effects is a critical issue. For verifying the photothermal effects on our SERS substrate, we monitored SERS signal variation at the same location under continuous irradiation by a laser (Figure S6). When continuous signal detection was performed on bovine serum albumin (BSA)-coated GNP substrate, signal variations in the major peaks were moderate over 60 s of irradiation. SERS characterization of exosomes. Figure 2A presents representative SERS spectra of HPAECs, NSCLC cells, and H1299- and PC9-derived exosomes, with phosphatebuffered saline (PBS) as a control. The spectra show many sharp peaks ranging from 470–1,800 cm-1. In SERS, a nonuniform substrate can be a critical obstacle to signal analysis because of the variability it causes in EM field enhancement. To overcome this issue, we normalized the data based on the intensity of a spectral peak shared by all data. Our SERS substrate also has its own spectral peaks, which are caused by functionalized cysteamine and stabilizers such as citrate. Therefore, we normalized the intensity of each spectrum such that the peak intensity of approximately 725 cm-1, which a major Raman band of cysteamine, was set to 1. Typical spectra of the exosomes had notable peaks near 640, 1,440, and 1,580 cm-1. According to previous studies, these peaks are thought to be associated with the characteristic Raman bands of proteins: 640 cm-1 (tyrosine), 1,440 cm-1 (CH2, CH3 deformation), and 1,580 cm-1 (amide II bond).35,36 Each type of exosome had several distinct peaks, but their individual spectra exhibited heterogeneous peak compositions. We then utilized PCA to identify major patterns in the spectra. Figure 2B presents a PCA scores plot graph. We utilized 25 data for each of the HPAEC, H1299, and PC9 cellderived exosomes, as well as the control. Each dot in the plot

represents one spectrum with dimension reduction. The first principal component (PC1) and second principal component (PC2) reveal the trends in each dataset. Each dataset was clustered while maximizing covariance. The variance levels in PC1 and PC2 were 61.7% and 11.7%, respectively. The clusters of exosome data were distinguishable from the PBS control. The cancerous exosomes derived from the H1299 and PC9 cells overlapped. However, the 90% confidence ellipses of these cancerous exosomes were separated from the normal exosomes. Based on PC1, this difference is statistically significant (P < 0.001). Additionally, the cancerous exosomes tended to be located on the positive side of PC1. Therefore, we could evaluate the dominant patterns based on the PC1 loading data. The loading data graph for PC1 shows various sharp and narrow peaks with positive values (Figure 2C). These 26 peaks maximize the difference between normal and cancerous exosomes. Therefore, we selected these peaks as candidates for unique Raman band patterns of the cancerous exosomes. Additionally, we performed ratiometric analysis of mixtures containing normal and cancerous exosomes to verify the Raman patterns. The total amount of exosomes was fixed at 108 particles/ml. The intensities of several Raman bands increased with the ratio of cancerous exosomes in the mixtures (Figure 3A). We examined the intensities of the unique Raman bands derived by PCA. The intensities of 21 of the unique bands gradually increased with the particle concentration of cancerous exosomes (Figure 3B), whereas the other five bands showed irregular or opposite changes (Figure S7). We selected the 13 bands that correlated well with cancerous exosome ratios (R2 > 90%) as the unique Raman bands for NSCLC exosomes (553, 584, 812, 827, 854, 1,010, 1,223, 1298, 1,387, 1,436, 1,588, 1,701, and 1,771 cm-1).

5 ACS Paragon Plus Environment

ACS Sensors 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

Correlation between cancerous exosomes and protein markers. We next examined the Raman bands of exosomal protein markers (CD9, CD81, EGFR and EpCAM) to investigate their correlation with the unique Raman bands of NSCLC exosomes. In our analysis of exosomal protein markers, we utilized exosome lysate. However, there are limitations to collecting abundant amounts of exosomes to be utilized for the isolation and purification of proteins. Therefore, we attempted to compare the spectra of an antibody-coated SERS substrate before and after the conjugation of an antigen. As both cases shared the other molecules including BSA and linking reagent, their signals were exhibited in common. Thus, the main difference between the cases may originate from the conjugated proteins. We utilized PCA to classify these cases. To verify this detection approach, we performed a comparative analysis between cases with and without target proteins (Figure S8). For the non-target protein sample, there was no difference between before and the after treatment of the sample. However, in the sample containing the target protein, there were significant differences. Utilizing this approach, we identified specific Raman bands for the exosomal protein markers. We obtained thirty spectra of before and after the conjugation of the protein markers. In the PCA of exosomal protein markers, the spectra conjugated with each protein marker were distinguished from the antibody-only SERS spectra (Figure S9). In particular, the PC1 scores could be well classified dependent on protein conjugation (Figure 4a). Therefore, the PC1 loading data may be indicative of the specific Raman bands of the proteins (Figure 4b). All spectra have common Raman spectral patterns for proteins35, 36: 653-657 cm-1 (tyrosine); 831–865 cm-1 (tyrosine); 962–965 cm-1 (tryptophan, valine); 1,034–1,047 cm-1 (Phenylalanine); 1,447–1,460 cm-1 (C-H); 1,580 cm-1 (amide II bond). The overall peak compositions of the markers were similar, but there was a specific distinction in terms of intensity. We identified which protein markers have high similarity with the unique Raman bands of NSCLC exosomes. In Figure 4b, we marked the ranges of the unique Raman bands of NSCLC exosomes. To investigate these similarities, we calculated the Euclidean distance between the loading data of the protein markers and that of the NSCLC exosomes in the above ranges (Figure S10).37 The similarity was calculated as the reciprocal of the Euclidean distance: 1 ∑ 𝑛 (𝑥 ― 𝑦 )2 𝑆𝑖𝑚𝑖𝑙𝑎𝑟𝑖𝑡𝑦 = 𝑖 𝑖=1 𝑖 where 𝑥𝑖 and 𝑦𝑖 are the value of loading data of the protein marker and the NSCLC exosomes in the ranges, respectively. As a result, EGFR showed the highest similarity, while the other protein markers had poor similarity. Based on this result, we can conclude that EGFR is highly correlated with the unique Raman bands of the NSCLC exosomes. If EGFR is a primary variable of NSCLC exosomes, it must exist in the NSCLC exosomes dominantly.

Page 6 of 9

Figure 5. Immunoblotting of exosomes derived from HPAEC, PC9, and H1299. Each sample contained the same amount of total protein (15 μg).

These associations are consistent with our immunoblotting results (Figure 5). CD9 and CD81, which were less correlated with the unique patterns of the NSCLC exosomes, were commonly expressed in all exosomes. EGFR showed dominant expression in both the PC9 and H1299 cell-derived exosomes. EpCAM, which is also less correlated as a protein marker, was only expressed in the PC9 exosomes. This indicates that EpCAM may not contribute significantly to the signal differences between NSCLC exosomes and normal exosomes. Finally, we determined that a protein marker considered to be highly relevant in Raman signal analysis showed higher protein expression in the NSCLC exosomes. However, the unique Raman bands of the NSCLC cellderived exosomes may not have originated from these protein markers exclusively. Further studies utilizing other proteins may be required to determine which markers show significant differences between normal and cancerous exosomes. This would provide a deeper understanding of the biochemical compositions of cancerous exosomes.

CONCLUSIONS We correlated the unique Raman bands of NSCLC cellderived exosomes and their protein markers. SERS yielded enhanced Raman signals for the exosomes. We analyzed several unique Raman scattering peaks utilizing PCA and ratiometric experiments. To understand the origin of the unique peaks, we compared the peaks to the Raman signals of their associated protein markers. Several markers had a strong association with the unique Raman bands of NSCLC exosomes. Immunoblotting results also supported this correlation. These results will aid researchers in studies on exosomal markers for the diagnosis and prognostic prediction of cancers.

6 ACS Paragon Plus Environment

Page 7 of 9 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

ACS Sensors

ASSOCIATED CONTENT Cell culture condition, western blotting of the isolated exosome, data pre-processing of SERS spectrum, exosome isolation, exosomal surface charges, validation of the SERS substrate (SEM images, functionalization, and photothermal effect), irregular changes at the unique Raman band of NSCLC exosome, comparison between the case of with and without target protein, and PCA analysis of the exosomal protein markers. This material is available free of charge via the Internet at http://pubs.acs.org.

AUTHOR INFORMATION Corresponding Author *E-mail: [email protected]

Notes

There are no conflicts to declare.

ACKNOWLEDGMENT This research was supported by a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health & Welfare, Republic of Korea (Grant No. HR14C-0007060018) and Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (Grant No. NRF2017R1E1A1A01075147).

REFERENCES (1) Hofman, P. Liquid biopsy for early detection of lung cancer. Curr. Opin. Oncol. 2017, 29, 73-78. DOI: 10.1097/CCO.0000000000000343 (2) Lin, E. C. Radiation risk from medical imaging. Mayo. Clin. Proc. 2010, 85, 1142-1146. DOI: 10.4065/mcp.2010.0260 (3) Midthun, D. E. Early diagnosis of lung cancer. F1000Prime Reports 2013, 5, 12. DOI: 10.12703/P5-12 (4) Cheng, Y.-L.; Lee, S.-C.; Harn, H.-J.; Chen, C.-J.; Chang, Y.C.; Chen, J. C.; Yu, C. P. Prognostic prediction of the immunohistochemical expression of p53 and p16 in resected non-small cell lung cancer. Eur. J. Cardiothorac. Surg. 2003, 23, 221-228. DOI: 10.1016/S10107940(02)00749-2 (5) Thery, C. Exosomes: secreted vesicles and intercellular communications. F1000 Biol. Rep., 2011, 3, 15. DOI: 10.3410/B3-15 (6) Lin, J.; Li, J.; Huang, B.; Liu, J.; Chen, X.; Chen, X.-M.; Xu, Y.M.; Huang, L.-F; Wang, X.-Z. Exosomes: novel biomarkers for clinical diagnosis. Scientific World Journal 2015, 2015, 8. DOI: 10.1155/2015/657086 (7) Sheridan, C. Exosome cancer diagnostic reaches market. Nat. Biotechnol. 2016, 34, 359-360. DOI: 10.1038/nbt0416359 (8) Zhang, X.; Yuan, X.; Shi, H.; Wu, L.; Qian, H.; Xu, W. Exosomes in cancer: small particle, big player. J. Hematol. Oncol. 2015, 8, 83. DOI: 10.1186/s13045-015-0181-x (9) Madhavan, B.; Yue, S.; Galli, U.; Rana, S.; Gross, W.; Muller, M.; Giese, N.-A.; Kalthoff, H.; Becker, T.; Buchler, M.-

W.; Zoller, M. Combined evaluation of a panel of protein and miRNA serum-exosome biomarkers for pancreatic cancer diagnosis increases sensitivity and specificity. Int. J. Cancer 2015, 136, 2616-2627. DOI: 10.1002/ijc.29324 (10) Taylor, D.-D.; Gercel-Taylor, C. MicroRNA signatures of tumor-derived exosomes as diagnostic biomarkers of ovarian cancer. Gynecol. Oncol. 2008, 110, 13-21. DOI: 10.1016/j.ygyno.2008.04.033. (11) Sandfeld-Paulsen, B.; Aggerholm-Pedersen, N.; Boek, R.; Jakobsen, K.-R.; Melolgaard, P.; Folkersen, B.-H.; Rasmussen, T.-R.; Varming, K.; Jorgensen, M.-M.; Sorensen, B.-S. Exosomal proteins as prognostic biomarkers in nonsmall cell lung cancer. Mol. Oncol. 2016, 10, 1595-1602. DOI: 10.1016/j.molonc.2016.10.003. (12) 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. DOI: 10.1039/c5lc01117e. (13) Kibria, G.; Ramos, E.-K.; Lee, K.-E.; Bedoyan, S.; Huang, S.-M.; Samaeekia, R.; Athman, J.-J.; Harding, C.-V.; Lotvall, J.; Harris, L.; Thompson, C.-L.; Liu, H.-P. A rapid, automated surface protein profiling of single circulating exosomes in human blood. Sci. Rep. 2016, 6, 36502. DOI: 10.1038/srep36502 (2016). (14) Raimondo, F.; Corbetta, S.; Morosi, L.; Chinello, C.; Gianazza, E.; Castoldi, G.; Di Gioia, C.; Bombardi, C.; Stella, A.; Battaglia, C.; Bianchi, C.; Magni, F.; Pitto, M. Urinary exosomes and diabetic nephropathy: a proteomic approach. Mol. Biosyst. 2013, 9, 1139-1146. DOI: 10.1039/c2mb25396h. (15) Coi, D.-S.; Kim, D.-K.; Kim, Y.-K.; Gho, Y.-S. Proteomics of extracellular vesicles: Exosomes and ectosomes. Mass. Spectrom. Rev. 2015, 34, 474-490. DOI: 10.1002/mas.21420 (16) Zarovni, N.; Corrado, A.; Guazzi, P.; Zocco, D.; Lari, E.; Radano, G.; Muhhina, J.; Fondelli, C.; Gavrilova, J.; Chiesi. A. Integrated isolation and quantitative analysis of exosome shuttled proteins and nucleic acids using immunocapture approaches. Methods 2015, 87, 46-58. DOI: 10.1016/j.ymeth.2015.05.028 (17) Akagi, T.; Kato, K.; Hanamura, N.; Kobayashi, M.; Ichiki, T. Evaluation of desialylation effect on zeta potential of extracellular vesicles secreted from human prostate cancer cells by on-chip microcapillary electrophoresis. Jpn. J. Appl. Phys. 2014, 53, 06JL01-1. DOI: 10.7567/JJAP.53.06JL01 (18) He, M.; Crow, J.; Roth, M.; Zeng, Y.; Godwin, A.-K. Integrated immunoisolation and protein analysis of circulating exosomes using microfluidic technology. Lab Chip 2014, 14, 3773-3780. DOI: 10.1039/C4LC00662C (19) Higginbotham, J.-N.; Zhang, Q.; Jeppesen, D.-K.; Scott, A.-M.; Manning, H.-C.; Ochieng, J.; Franklin, J.-L.; Coffey, R.-J. Identification and characterization of EGF receptor in individual exosomes by fluorescence-activated vesicle sorting. J. Extracell. Vesicles 2016, 5, 29254. DOI: 10.3402/jev.v5.29254 (20) Im, H.; Shao, H.-L.; 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. DOI: 10.1038/nbt.2886

7 ACS Paragon Plus Environment

ACS Sensors 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

(21) Hong, S.; Shim, O.; Kwon, H.; Choi, Y. Autoenhanced Raman spectroscopy via plasmonic trapping for molecular sensing. Anal. Chem. 2016, 88, 7633-7638. DOI: 10.1021/acs.analchem.6b01451 (22) Lim, J.-Y.; Nam, J.-S.; Yang, S.-E.; Shin, H.; Jang, Y.-H.; Bae, G.-U.; Kang, T.; Lim, K.-I.; Choi, Y. Identification of newly emerging influenza viruses by surface-enhanced Raman spectroscopy. Anal. Chem. 2015, 87, 11652-11659. DOI: 10.1021/acs.analchem.5b02661 (23) Kwizera, E. A.; O'Connor, R.; Vinduska, V.; Williams, M.; Butch, E. R.; Snyder, S. E.; Chen, X.; Huang, X. Molecular Detection and Analysis of Exosomes Using SurfaceEnhanced Raman Scattering Gold Nanorods and a Miniaturized Device. Theranostics. 2018, 8, 2722-2738. DOI: 10.7150/thno.21358 (24) Huang, X.; O'Connor, R.; Kwizera, E. A. Gold Nanoparticle Based Platforms for Circulating Cancer Marker Detection. Nanotheranostics. 2017, 1, 80-102. DOI: 10.7150/ntno.18216 (25) Chaffin, E.; O'Connor, R. T.; Barr, J.; Huang, X.; Wang, Y. Dependence of SERS enhancement on the chemical composition and structure of Ag/Au hybrid nanoparticles. J. Chem. Phys. 2016, 145, 054706. DOI: 10.1063/1.4960052 (26) Cialla-May, D.; Zheng, X.-S.; Weber, K.; Popp, Recent progress in surface-enhanced Raman spectroscopy for biological and biomedical applications: from cells to clinics. J. Chem. Soc. Rev. 2017, 46, 3945-3961. DOI: 10.1039/C7CS00172J (27) Zakaria, N.; Yusoff, N.-M.; Zakaria, Z.; Lim, M.-N.; Baharuddin, P.-J.-N.; Fakiruddin, K.-S.; Yahaya, B. Human non-small cell lung cancer expresses putative cancer stem cell markers and exhibits the transcriptomic profile of multipotent cells. Bmc Cancer 2015, 15. DOI: 10.1186/s12885-015-1086-3 (28) Rabinowits, G.; Gercel-Taylor, C.; Day, J.-M.; Taylor, D. D.; Kloecker, G.-H. Exosomal microRNA: a diagnostic marker for lung cancer. Clin. Lung Cancer 2009, 10, 42-46. DOI: 10.3816/CLC.2009.n.006 (29) Pak, M.-G.,; Shin, D.-H.; Lee, C.-H.; Lee, M.-K. Significance of EpCAM and TROP2 expression in non-small cell lung cancer. World J. Surg. Oncol. 2012, 10. DOI: 10.1186/14777819-10-53 (30) Bethune, G.; Bethune, D.; Ridgway, N.; Xu, Z. Epidermal growth factor receptor (EGFR) in lung cancer: an overview and update. J. Thorac. Dis. 2010, 2, 48-51. (31) Park, J.; Hwang, M.; Choi, B.; Jeong, H.; Jung, J.-H., Kim, H.-K.; Hong, S.; Park, J.-H.; Choi, Y. Exosome classification by pattern analysis of surface-enhanced Raman spectroscopy data for lung cancer diagnosis. Anal. Chem. 2017, 89, 66956701. DOI: 10.1021/acs.analchem.7b00911 (32) Ma, Y.; Jiang, L.; Mei, Y.; Song, R.; Tian, D.; Huang, H. Colorimetric sensing strategy for mercury(II) and melamine utilizing cysteamine-modified gold nanoparticles. Analyst 2013, 138, 5338–5343. DOI: 10.1039/C3AN00690E (33) Ning, X.; Selesnick I.; Duval, L. Chromatogram baseline estimation and denoising using sparsity (BEADS). Chemom. Intell. Lab. Syst. 2014, 139, 156-167. DOI: 10.1016/j.chemolab.2014.09.014

Page 8 of 9

(34) Kudelski A.; Hill W. Raman Study on the Structure of Cysteamine Monolayers on Silver. Langmuir 1999, 15, 3162–3168. DOI: 10.1021/la9811463 (35) Maiti, N.-C.; Apetri, M.-M.; Zagorski, M.-G.; Carey, P.-R.; Anderson, V.-E. Raman Spectroscopic Characterization of Secondary Structure in Natively Unfolded Proteins:  αSynuclein. J. Am. Chem. Soc. 2004, 126, 2399-2408. DOI: 10.1021/ja0356176 (36) Rygula, A.; Majzner, K.; Marzec, K.-M.; Kaczor, A.; Pilarczyk, M.; Baranska, M. Raman spectroscopy of proteins: a review. J. Raman Spectrosc. 2013, 44, 1061-1076. DOI: 10.1002/jrs.4335 (37) Xu, Z.; Xia, M. Distance and similarity measures for hesitant fuzzy sets. Inf. Sci. 2011, 181, 2128-2138. DOI: 10.1016/j.ins.2011.01.028

8 ACS Paragon Plus Environment

Page 9 of 9 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

ACS Sensors

For TOC image

9 ACS Paragon Plus Environment