Open Access
Add to BibSonomy
Abstract
The detection of circulating tumor cells (CTCs) from peripheral blood is considered as great significance for the diagnosis and prognosis of cancer patients. Raman spectroscopy is a highly sensitive optical detection technique that can provide fingerprint molecular identification information. In this paper, the silver film substrate surface-enhanced Raman scattering (SERS) was used to research several tumor cells, immortalized cells, clinical cancer cells isolated from cancer patient’s tissue and blood cells. The results display that there is great difference for the nucleic acid characteristic peaks of those cells. The red blood cells have almost none nucleic acid characteristic peak and the SERS signals of white blood cells are only a slight increase. Except for immortalized cells and few tumor cells, the nucleic acid characteristic peaks of some tumor cells have huge enhancement. Nucleic acid characteristic peaks of clinical cancer cells also have greater enhancement. The discriminant model established by the intensity ratio of the nucleic acid characteristic peak 730 cm−1 to the substrate background peak 900 cm−1 shows that some tumor cells and clinical sample cells can be separated from white blood cells, but tumor cells with relatively low-DNA index cannot be differentiated from white blood cells. This study demonstrates that thin-film SERS technology can distinguish between blood cells and some types of tumor cells. This study opens up a new possible method for the detection of CTCs with label-free SERS spectra.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
It is reported by the World Health Organization (WHO)'s annual World Health Statistics that approximately 8 million people die from cancer worldwide every year, and 90% of these deaths are caused by systemic diseases. In most countries, cancer is one of the leading cause of death [1]. Early cancer screening can significantly improve the survival rate of cancer patients. In recent years, the circulating tumor cells (CTCs) in peripheral blood have been widely concerned. CTCs, the general name of all kinds of tumor cells in the peripheral blood, are tumor cells falling from solid tumors into the bloodstream by spontaneous cause or treatment operations. It can be used as a tumor marker for early cancer detection, prognosis evaluation, monitoring cancer drug efficacy and so on [2–4]. At present, a variety of CTCs detection techniques have been developed, such as reverse transcription polymerase chain reaction (RT-PCR), fluorescence in situ hybridization, flow cytometry (FCM), and so on [5–9]. However, PCR technology is complex to operate and time-consuming. Specificity and sensitivity of fluorescence in situ hybridization techniques can be affected by the expression of surface proteins and false positive. Therefore, it is necessary to establish non-invasive, rapid and sensitive methods for early cancer screening.
Recently, the rapid development of Raman spectroscopy brings light for the non-invasive and rapid identification of CTCs. Raman spectroscopy is a unique vibrational spectroscopic technique that measures inelastic light scattering processes. It can provide specific spectroscopic fingerprints of bimolecular structures and compositions of tissues [10–16]. Raman spectroscopy has been applied to the cancer diagnosis by discovering the difference of characteristic spectra existing in normal and cancerous biological samples. However, due to the limitation of Raman scattering cross sections, the conventional Raman spectroscopy remains quite low scattering efficiency, which greatly restricts the practical application of Raman spectroscopy. Development of Surface-enhanced Raman scattering (SERS) technique makes up the deficiency of conventional Raman spectroscopy. When the sample molecules are adsorbed on the surface of some costly metal, such as silver, gold and copper, their Raman scattering signal will be greatly enhanced, this phenomenon is named SERS [11,17–23].
In order to develop a highly sensitive and non-invasive method to rapidly detect CTCs, here a silver film SERS substrate was employed to research SERS spectra of blood cells, several tumor cells, immortalized cells and clinical sample cells. This thin film SERS method can non-destructively detect cell samples and rapidly identify blood cells and tumor cells. We believe this study can lay a favorable foundation for the label-free detection of CTCs, and we hope this study can open up a new approach for the detection of CTCs.
2. Materials and methods
2.1 Treatment of cell
The human breast cancer (MCF-7) cells, human neuroblastoma (SH-SY5Y) cells, cervical cancer (Hela) cells, hepatocellular carcinoma (SMMC7721) cells, (HO-8910) cells, lung cancer (NCI-H446) cells, immortalized cells (16HEB) and laryngeal cancer (Hep2) cells were used in this experiment. Before the measurement with SERS, tumor cells were processed as follow: the tumor cells were digested with trypsin in the culture medium and then centrifuged at 1500 rpm for 5 minutes. After the pancreatin was removed, the tumor cells were washed and centrifuged repeatedly 3 times by phosphate buffered solution (PBS).
A total of 19 blood samples were collected from Sun Yat-sen University Cancer Center. The method of white blood cells isolation was as follows: an appropriate amount of EDTA anticoagulant and haemolysin were added into centrifuge tube with blood sample. After standing for 15 minutes at room temperature for the dissolution of red blood cytoplasm, these mixtures were centrifuged at 1500 rpm for 5 minutes, and then the upper liquid was removed. Some PBS was put into the remained mixtures, after mixing, centrifuging, the upper liquid was removed again. Finally, the white blood cells suspension was successfully obtained with the addition of another PBS.
2.2 Separation of red blood cells
An appropriate amount of separation solution was added into centrifuge tube, and then blood sample was carefully put to the above of separation solution. After standing for 30 to 40 minutes, the sample was centrifuged and the mixture was subsequently divided into four layers. The bottom layer is just the red blood cells.
2.3 Clinical cell samples
Three cancer tissues were also collected from Sun Yat-sen University Cancer Center. After fresh cancer tissues resected from patients were immediately washed three times with PBS, they were cut, ground and sterilized with penicillin and streptomycin. Then 0.25% trypsin was added to digest tissues in cell incubator at 37°C for 100 minutes, and the culture medium was added to terminate digestion process. Next the mixture was filtered and centrifuged, the isolated cancer cells were obtained at once. After they were cultivated three generations, the higher purity of cancer cells was finally obtained for experiments. Prior to research, all volunteers signed an informed consent to permit.
A brief treatment scheme of all samples preparation is shown in Table 1.
Table 1. A simple table of samples preparation
2.4 SERS substrate
The SERS substrate was purchased from Yun Yang Photoelectric Technology Co., Ltd. It was fabricated by coating a thin layer of poly (vinylidene fluoride) (PVDF)on the tip of the poly (methyl methacrylate) PMMA fiber with a length of 1.5 cm, then the silver mirror reaction was implemented to grow AgNPs until a SERS active area was formed on the tip surface [24]. Figure 1 is the Atomic Force Microscope (AFM) topography of the substrate surface. The scale bar of Figs. 1(a) and 1(b) is 20.0μm and 400.0 nm, respectively. It is displayed that the substrate surface is very rough, which is advantageous to enhance Raman scattering signals.
Fig. 1 Atomic Force Microscope (AFM) topography of the silver film substrate, Fig. 1(a) Scale bar of 20.0μm, Fig. 1(b) Scale bar of 400.0 nm.
2.5 Instrumentation
The Raman spectroscopy was recorded with a confocal Raman microscopy (Renishaw, in Via, United Kingdom) in the range of 610 to 1700 cm−1 with a spectral resolution about 1 cm−1 under a 785 nm diode laser excitation. The spectra were collected in back-scattered geometry using a Leica DM2500 microscope equipped with objective 20 × . The power of laser exposed on sample was about 1 mw with a spot diameter about 5 μm. The software package WIRE 3.2 (Renishaw) was employed for spectral acquisition and analysis. For each experiment, 2μl sample solution was dropped onto the base column surface with pipette gun, and 5 spectra were measured at different positions of each substrate. Each Raman spectrum was acquired twice with an integration time of 10 s. All data was collected under the same conditions.
2.6 Data preprocessing and analysis
The acquired Raman spectra contain many auto fluorescence and background noises [25]. A fifth-order polynomial was employed to fit and then this polynomial was subtracted from original spectra. In order to compare the changes of spectral shapes and relative peak intensities among different cells samples, an area normalized under the curve was employed. Vancouver Raman algorithm was employed for spectra smoothing and baseline correction. It is an automated auto fluorescence background subtraction algorithm based on modified multi-polynomial fitting by presetting of the relevant parameters such as size of Boxcar smooth, order of polynomial fit and stop criteria [26,27].
3. Results and discussion
To compare the enhanced performance of silver film substrate SERS and conventional Raman scattering, the Raman spectra of tumor cells (SH-SY5Y) are measured on the surface of silver film and aluminum sheet, respectively, under the same conditions. The measured Raman spectra are showed in Fig. 2. It is obvious that the Raman spectral peaks detected on the aluminum sheets are almost invisible. However, the spectral peaks of 730cm−1(C–H bending Adenine, Coenzyme)and 1330cm−1(CH3CH2 wagging, Tryptophan, Adenine, Guanine)measured on silver film substrate are greatly enhanced. In order to observe the influence of background peaks come from silver film and PBS, another Raman spectrum displayed in Fig. 2 is obtained by scanning the substrate surface where pure PBS is dropped on. It clearly indicates that there are two main peaks located at 900 and 1400cm−1, and other two small peaks 800~850cm−1 and 1020~1040cm−1.
Fig. 2 The comparison of traditional Raman spectrum (Aluminum sheet substrate) and sliver film substrate SERS spectrum. Red lines represent average SERS spectra. The shaded lines represent one standard deviation.
For the sake of evaluating the spectral reproducibility of SERS substrate, 30 spectra of human neuroblastoma cells (SH-SY5Y) are scanned continuously with the same parameters and substrate. Figure 3 is three-dimensional display (3D) SERS spectra of human neuroblastoma cells (SH-SY5Y). As is shown in Fig. 3 that the intensity of most characteristic peaks such as 960cm−1, 1330cm−1, 1410cm−1, and so on, has few change in the thirty spectra, and only characteristic peak 730cm−1 has a little difference in intensity. The results indicate that the spectral reproducibility of silver film substrate is excellent.
Figure 4 is SERS spectra of red blood cells, white blood cells and human neuroblastoma cells (SH-SY5Y). It can be seen from the Fig. 4 that there are strongest peaks for tumor cells and weaker peaks for white blood cells at 730cm−1 and 1330cm−1. However, there is no peak for red blood cells. Both the 730cm−1 and 1330cm−1 are characteristic peaks of nucleic acid. Tumor cells are usually metabolically active with rapidly proliferating, and contain high nucleic acid contents, thus, their SERS signals for nucleic acid are extremely enhanced. White blood cells consist of lymphocytes, neutrophils, eosinophils, basophils, and so on. These cells have a little nucleic acid, so their nucleic acid peaks are weak. For mature red blood cells, there hardly have any nucleic acid, because they haven’t nucleus, which is the reason why red blood cells have no characteristic peaks of nucleic acid. Our results are consistent with the outcomes of other groups [28]. For example, B.R. Wood group use multi-wavelength (488nm, 514nm, 568nm and 633nm) Confocal Raman to research the Raman spectra of red blood cells, also, there is no Raman characteristic peak 730cm−1 in their results [29]. G.J. Puppels group research the Raman spectra of granulocytes. In their results, nucleic acid characteristic peak 731cm−1 is very weak [30]. These results indicate that the Raman spectra of these cells are consistent with their biological characteristics.
Fig. 4 The comparison of SERS spectra among tumor cells(SH-SY5Y), white blood cells and red blood cells. Red lines represent average SERS spectra. The shaded lines represent one standard deviation.
In this study, the SERS spectra of seven kinds of tumor cells are measured. The tumor cell lines are as follows: Ovarian cancer cells (HO-8910), lung cancer cells (NCI-H446) breast cancer cells (MCF-7), hepatocellular carcinoma cells (SMMC7721), cervical cancer cells (Hela), human neuroblastoma cells (SH-SY5Y) and laryngeal cancer cells (Hep2). A total of 185 spectra are collected from seven kinds of tumor cells and each kinds of tumor cells are measured about 25~30 spectra. The average SERS spectra of seven kinds of tumor cells are shown in Fig. 5, which discloses that nucleic acid characteristic peaks 730cm−1 and 1330cm−1 for ovarian cancer cells (HO-8910) are quite low, and nucleic acid characteristic peaks of other six tumor cells are very strong.
Fig. 5 Average SERS spectra of seven kinds of tumor cells, SH-SY5Y, Hep2, Hela, SMMC7721, MCF-7, NCL-H446 and HO-8910. Red lines represent average SERS spectra. The shaded lines represent one standard deviation.
In order to further observe the spectral differences among tumor cells on the whole, we divided 7 kinds of measured tumor cells SERS spectra into two groups: group (a) and group (b). Group (a), including breast cancer cells (MCF-7), human neuroblastoma cells (SH-SY5Y), cervical cancer cells (Hela), hepatocellular carcinoma cells (SMMC7721), laryngeal cancer cells (Hep2) and lung cancer cells (NCL-H446), have 155 spectra in total. Group (b) is ovarian cancer cells (HO-8910), which has 30 spectra. Figure 6 shows the average SERS spectra of white blood cells and tumor cells, as well as the corresponding mean square error represented with shadow parts. 121 spectra of white blood cells are collected from 20 volunteers. It is displayed from Fig. 6 that there are another three characteristic peaks of 960cm−1(C–C α-helix, Proline, Valine), 630cm−1(C–S, L- tyrosine, Lactose)and 660cm−1 besides the main peak of 1330 cm–1 and 730 cm–1 for the SERS spectra of tumor cells group (a). However, for the spectra of tumor cells group (b), nucleic acid characteristic peaks 730 cm−1 and 1330 cm−1 are quite weak, and other peaks have slight differences compared with tumor cells group (a). For the white blood cells there are another two characteristic peaks of 1224cm−1 (C–C6H5 phenylalanine, Tryptophan) and 1585cm−1(C = C bending, Phenylalanine, Acetoacetate) in addition to main peaks of nucleic acid and SERS substrate peaks [13,15–17,24,27–32]. The characteristic peaks 730cm−1 and 1330cm−1in tumor cells group (a) are far higher than in white blood cells and tumor cells group (b). The characteristic peak 960cm−1 appears only in tumor cells group (a), but characteristic peaks 1224cm−1 and 1585cm−1 in white blood cells and tumor cells group (b) are higher than in tumor cells group (a). The significant differences of SERS spectra between tumor cells group (a) and white blood cells indicate the feasibilities of distinguishing some tumor cells and white blood cells with label-free SERS spectra. But the great differences between tumor cells group (a) and tumor cells group (b) suggest there are considerable differences in tumor cells structure and molecular composition.
Fig. 6 The comparison of SERS spectra between tumor c ells and white blood cells. Tumor cells group (a) include breast cancer cells (MCF-7), human neuroblastoma cells (SY5Y), cervical cancer cells (Hela), hepatocellular carcinoma cells (SMMC7721), laryngeal cancer cells (Hep2) and lung cancer cells (NCL-H449); Tumor cells group (b) is ovarian cancer cells (HO-8910). Red lines represent average SERS spectra of all tumor cells or white blood cells. The shaded lines represent one standard deviation.
So as to classify tumor cells group (a) and white blood cells the intensity ratio of nucleic acid characteristic peak 730 cm−1 to substrate background peak 900 cm−1 is calculated in Fig. 7. The intensity ratio of white blood cells is below 0.44. The intensity ratio ranges of ovarian cancer cells (HO-8910) lie in 0.03~0.11, which overlaps with that of white blood cells, thus they are cannot be distinguished with white blood cells. The intensity ratio scope of breast cancer cells (MCF-7), lung cancer cells (NCI-H446), hepatocellular carcinoma cells (SMMC7721)cervical cancer cells (Hela), laryngeal cancer cells (Hep2) and human neuroblastoma cells(SH-SY5Y)are 0.49~2.02, 0.54~1.62, 1.67~8.58, 1.76 ~5.28, 3.15~7.33 and 5.37~26.87, respectively. According to these range of ratios tumor cells group (a) and white blood cells can be distinguished. Finally, we implement statistical analysis to the above intensity ratio with SPSS 23.0. Kruskal-Wallis test showed significant differences among overall groups. Then Nemenyi post-hoc test was used for pairwise comparison. The results showed that tumor cells were significantly different from the white blood cells except HO-8910 (P<0.01).
The above researches find that the SERS spectra of some types of tumor cells and white blood cells have enormous differences, which can be differentiated with intensity ratio method. Aim to validate the effectiveness of this method, the SERS spectra of clinical cancer cells isolated from three lung cancer patients are measured in Fig. 8. Their intensity ratio locates in 0.90~4.40.According to the intensity ratio range of white blood cells and some tumor cells, they can be classified as tumor cells. Histopathological diagnosis results of three clinic tissue samples are all small cell lung cancer, which is consistent with the type of measured lung cancer lines (NCI-H446). The results demonstrate that the SERS spectra discriminant model established by laboratory cultured cancer cell lines can be effectively applied to the identification of clinical cancer cell samples. In order to observe the differences between other normal cells and tumor cells, the SERS spectra of lung immortalized cells (16HBE) are measured in Fig. 8. Their intensity ratio is 0.07~0.36, which cannot be distinguished with white blood cells, so they cannot be classified as tumor cells. The results disclose that this tumor cells identification method of SERS spectra can distinguish immortalized cells and tumor cells.
Fig. 8 The average SERS spectra of lung cancer cells (NCI-H446), immortalized cells (16HBE), clinical sample cells and white blood cells. Red lines represent average SERS spectra. The shaded lines represent one standard deviation.
The identification of tumor cells is an important part in CTCs detections. Currently most methods of identifying CTCs are based on immune recognition, which has deficiencies of high cost and low efficiency in multiple cancer screening because of the one by one specific identification for each tumor [1,33]. It is important to develop a label-free CTCs detection method. In this study, the SERS spectra with silver film substrate is employed to detect tumor cells and blood cells. The research discloses that there is great difference for the nuclide aside feature spectra between some tumor cells and blood cells [34]. The SERS spectra of some types of tumors cells and blood cells can be distinguished by feature peaks intensity ratio method. The validation experiments of clinical cancer cells SERS spectra indicate that intensity ratio method not only can identify SERS spectra of some tumor cell lines, but also can be used in clinical cell samples identification. This study lays a favorable foundation for the label-free identification of tumor cells. We hope this study can open up a new approach for the label-free detection of CTCs.
Substrate is closely associated with SERS signal enhancement. Generally, the metal nanoparticles are used as SERS substrate in the study of cells SERS spectra. There are two means to import nanoparticles into cells. On the one hand, the SERS nanoparticles are mixed with cells for a period of time waiting for the cells’ endocytosis. On the other hand, the SERS particles are directly injected into cells through the holes drilled with electroporation [30,35–37]. The two methods can greatly enhance Raman spectra of cells, but it takes long time to wait for cells endocytosis, or it must damage cells with electroporation, both shortcomings are unfavorable for the CTCs detection.
In this study the silver film substrate is employed to enhance cells SERS spectra. The results reveal that the substrate not only enhances Raman scattering signals of biomolecules on cell surfaces, but also enhances Raman scattering signals of intracellular biomolecules, especially the nucleic acid feature peaks 730cm−1 and 1330cm−1. Metal film has no Raman peak, but the silver film substrate in this study has Raman background peaks 900cm−1 and 1400cm−1. Obviously, the two peaks originate from Raman scattering signals of non-metallic part of substrate. The intensity of cells characteristic peaks just can be scaled by the two peaks as intensity reference. In fact, the silver film substrate has been synthesized firstly by Jyisy Y group in 2012 [38]. It has been successfully used to detect melamine in milk and bacterial pathogens in peritonitis patients [38,39]. The silver film substrate has many advantages such as high sensitivity, good reproducibility and excellent hydrophobicity. These performances have been confirmed in this study as shown in Fig. 2 and Fig. 3.
In this work, the SERS spectra of several kinds of cells are researched, including red blood cells, white blood cells, several kinds of tumor cells, clinical cancer cells and normal cells (lung immortalized cells). The results discover that red blood cells have no nucleic acid characteristic peak and white blood cells have weak nucleic acid characteristic peaks. Some types of tumor cells have extremely strong nucleic acid characteristic peaks. These results can be reasonably explained by the biophysical properties of cells. Mature red blood cells have no cell nucleus, and they account for about 98% of whole blood in normal bodies. Therefore, red blood cells almost have no nucleic acid characteristic peak. In the following study, we only research SERS spectra of white blood cells and tumor cells, because red blood cells are easy to be distinguished from other cells due to no nucleic acid characteristic peak. Most of white blood cells consist of single nucleus cells, and their nucleic acid contents are not abundant, so their nucleic acid characteristic peaks are weak. Tumor cells have strong nucleic acid characteristic peaks owing to the fact that they are metabolically active and their nuclear cytoplasmic ratio is much higher than that of normal cells. However, it is strange that the nucleic acid characteristic peaks of tumor cell line HO-8910 are very weak. We have not found an exact explanation until now. We speculate the possible reason is that the nucleic acid content of tumor cell line HO-8910 is not high. Because its DNA index, the ratio of the DNA average content of a group of cells to that of the normal cells, is 1.42, which indicates that its DNA content is only 42% higher than that of normal cells. In general, the possibility of cancer is considered when the cell's DNA index exceeds 2.5.
To classify the SERS spectra of tumor cells and blood cells, the intensity ratio of nucleic acid characteristic peaks 730cm−1 to substrate background 900cm−1 is employed as evaluation reference in Fig. 7. The intensity ratio of white blood cells is below 0.44, and the intensity ratio of ovarian cancer cells (HO-8910) in the scope of 0.03~0.11. It is difficult of to differentiate them from white blood cells. The intensity ratio of other six tumor cell lines lie in 0.49~26.87, so they can be absolutely distinguished from white blood cells. If the principal component analysis (PCA) method is used, the SERS spectra of seven tumor cell lines and white blood cells can be completely separated. But the intensity ratio method is more simple and intuitive with clear physical meaning.
In the course of establishing tumor cells identification method with SERS, all data models come from tumor cell lines. In order to verify practicality of this method, clinical tissue samples from three cases of lung cancer patients are collected. The clinical lung cancer cells with higher purity are obtained through isolating tissue and culturing three generations. The measured results display that the intensity ratio of SERS spectra from clinical lung cancer cells is in the range of 0.90~4.40, which can be classified as tumor cells. We want to directly measure the SERS spectra of CTCs, but it is difficult to capture CTCs. Thus we have to detect the tumor cells which are isolated from clinical tissue samples. The verification results of clinical rumor cells are consistent with the tumors cells identification model, which demonstrates that the tumor cells identification method with SERS is effective in practical applications. In the course of building the tumor cells discrimination model, only tumor cells are researched, however, the properties of SERS spectra of normal cells are not clear. Thus the SERS spectra of lung immortalized cells (16HBE) are measured. The results show that its intensity ratio is 0.07~0.36, so it can’t be classified as tumor cells. The results just conform to biophysical properties of immortalized cells. Because immortalized cells don’t belong to tumor cells with relatively lower nucleic acid content, the nucleic acid characteristic peaks are weaker. The SERS spectra properties of immortalizes cells further verify the effectiveness of identifying tumor cells with silver film substrate SERS method.
In this study, the silver film substrate is used to successfully identify blood cells and some kinds of tumor cells. The study displays important reference value to the label-free detecting CTCs. If this tumor cells identification method can be developed into a practical CTCs detection technology, it will be of great significance to the doctors and patients. In health screening, different kinds of tumor cells in blood can be conveniently and rapidly identified by one-time label-free CTCs detection, which can save all kinds of consumptions when tagging antibody screaming method is used. Although this study has made some encouraging progress, there are many problems to be solved. For example, all tumor cell lines are cultivated with the uniform standard operation, however, why there are such obvious differences in their nucleic acid characteristic peaks. What differentiation and connection are there for the biophysical properties in these tumor cells? These deep-seated problems deserve further exploration.
4. Conclusions
This study has measured SRES spectra of blood cells, different kinds of tumor cells, clinical cancer cells and immortalized cells by rough silver film substrate. The experimental results display that the characteristic peaks 730cm−1 and 1330cm−1 associated with nucleic acids exhibit significant differences in these cells. The SERS signals detected from some kinds of tumor cells are significantly enhanced, only a slightly increased in white blood cells, immortalized cells and some tumor cells, but no peak is detected from red blood cells. The spectral characteristics exhibited from these cells are approximately consistent with their nucleic acid content. The discriminant model established by the intensity ratio of the nucleic acid characteristic peak 730 cm−1 to the substrate background peak 900 cm−1 can effectively classify some kinds of tumor cells. The results demonstrate that the label-free SERS technology can effectively identify blood cells and some kinds of tumor cells. This study opens up a new possible method for the detection of CTCs with label-free SERS spectra.
Funding
National Natural Science Foundation of China (81571724); Natural Science Foundation of Guangdong Province (2017A030313387)
Acknowledgments
The authors would like to acknowledge the financial support of the National Natural Science Foundation of China (81571724), the Natural Science Foundation of Guangdong Province (2017A030313387).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. A. Kowalik, M. Kowalewska, and S. Góźdź, “Current approaches for avoiding the limitations of circulating tumor cell detection methods-implications for diagnosis and treatment of patients with solid tumors,” Transl. Res. 185, 58–84 (2017). [CrossRef] [PubMed]
2. M. Ilie, V. Hofman, E. Long, O. Bordone, E. Selva, K. Washetine, C. H. Marquette, and P. Hofman, “Current challenges for detection of circulating tumor cells and cell-free circulating nucleic acids, and their characterization in non-small cell lung carcinoma patients. What is the best blood substrate for personalized medicine?” Ann. Transl. Med. 2(11), 107 (2014). [PubMed]
3. X. Wang, X. Qian, J. J. Beitler, Z. G. Chen, F. R. Khuri, M. M. Lewis, H. J. Shin, S. Nie, and D. M. Shin, “Detection of circulating tumor cells in human peripheral blood using surface-enhanced Raman scattering nanoparticles,” Cancer Res. 71(5), 1526–1532 (2011). [CrossRef] [PubMed]
4. M. Lin, J.-F. Chen, Y. T. Lu, Y. Zhang, J. Song, S. Hou, Z. Ke, and H. R. Tseng, “Nanostructure embedded microchips for detection, isolation, and characterization of circulating tumor cells,” Acc. Chem. Res. 47(10), 2941–2950 (2014). [CrossRef] [PubMed]
5. F. Zhan, X. Liao, F. Gao, W. Qiu, and Q. Wang, “Electroactive crown ester-Cu2+ complex with in-situ modification at molecular beacon probe serving as a facile electrochemical DNA biosensor for the detection of CaMV 35s,” Biosens. Bioelectron. 92, 589–595 (2017). [CrossRef] [PubMed]
6. C. L. O’Connell, R. Nooney, and C. McDonagh, “Cyanine5-doped silica nanoparticles as ultra-bright immunospecific labels for model circulating tumour cells in flow cytometry and microscopy,” Biosens. Bioelectron. 91, 190–198 (2017). [CrossRef] [PubMed]
7. X. Chen, F. Zhou, X. Li, G. Yang, L. Zhang, S. Ren, C. Zhao, Q. Deng, W. Li, G. Gao, A. Li, and C. Zhou, “Folate Receptor-Positive Circulating Tumor Cell Detected by LT-PCR-Based Method as a Diagnostic Biomarker for Non-Small-Cell Lung Cancer,” J. Thorac. Oncol. 10(8), 1163–1171 (2015). [CrossRef] [PubMed]
8. G. Premasekharan, E. Gilbert, R. A. Okimoto, A. Hamirani, K. J. Lindquist, V. T. Ngo, R. Roy, J. Hough, M. Edwards, R. Paz, A. Foye, R. Sood, K. A. Copren, M. Gubens, E. J. Small, T. G. Bivona, E. A. Collisson, T. W. Friedlander, and P. L. Paris, “An improved CTC isolation scheme for pairing with downstream genomics: Demonstrating clinical utility in metastatic prostate, lung and pancreatic cancer,” Cancer Lett. 380(1), 144–152 (2016). [CrossRef] [PubMed]
9. X. Wu, Y. Xia, Y. Huang, J. Li, H. Ruan, T. Chen, L. Luo, Z. Shen, and A. Wu, “Improved SERS-Active Nanoparticles with Various Shapes for CTC Detection without Enrichment Process with Supersensitivity and High Specificity,” ACS Appl. Mater. Interfaces 8(31), 19928–19938 (2016). [CrossRef] [PubMed]
10. Z. Huang, H. Lui, D. I. McLean, M. Korbelik, and H. Zeng, “cRaman spectroscopy in combination with background near-infrared autofluorescence enhances the in vivo assessment of malignant tissues,” J. Photochem. Photobiol. B 81(5), 1219–1226 (2005). [CrossRef]
11. K. W. Short, S. Carpenter, J. P. Freyer, and J. R. Mourant, “Raman spectroscopy detects biochemical changes due to proliferation in mammalian cell cultures,” Biophys. J. 88(6), 4274–4288 (2005). [CrossRef] [PubMed]
12. Z. Huang, A. McWilliams, H. Lui, D. I. McLean, S. Lam, and H. Zeng, “Near-infrared Raman spectroscopy for optical diagnosis of lung cancer,” Int. J. Cancer 107(6), 1047–1052 (2003). [CrossRef] [PubMed]
13. M. S. Bergholt, W. Zheng, K. Lin, K. Y. Ho, M. Teh, K. G. Yeoh, J. B. Yan So, and Z. Huang, “In vivo diagnosis of gastric cancer using Raman endoscopy and ant colony optimization techniques,” Int. J. Cancer 128(11), 2673–2680 (2011). [CrossRef] [PubMed]
14. S. Li, G. Chen, Y. Zhang, Z. Guo, Z. Liu, J. Xu, X. Li, and L. Lin, “Identification and characterization of colorectal cancer using Raman spectroscopy and feature selection techniques,” Opt. Express 22(21), 25895–25908 (2014). [CrossRef] [PubMed]
15. S. X. Li, Q. Y. Chen, Y. J. Zhang, Z. M. Liu, H. L. Xiong, Z. Y. Guo, H. Q. Mai, and S. H. Liu, “Detection of nasopharyngeal cancer using confocal Raman spectroscopy and genetic algorithm technique,” J. Biomed. Opt. 17(12), 125003 (2012). [CrossRef] [PubMed]
16. S. K. Teh, W. Zheng, K. Y. Ho, M. Teh, K. G. Yeoh, and Z. Huang, “Near-infrared Raman spectroscopy for optical diagnosis in the stomach: identification of Helicobacter-pylori infection and intestinal metaplasia,” Int. J. Cancer 126(8), 1920–1927 (2010). [CrossRef] [PubMed]
17. H. J. Butler, L. Ashton, B. Bird, G. Cinque, K. Curtis, J. Dorney, K. Esmonde-White, N. J. Fullwood, B. Gardner, P. L. Martin-Hirsch, M. J. Walsh, M. R. McAinsh, N. Stone, and F. L. Martin, “Using Raman spectroscopy to characterize biological materials,” Nat. Protoc. 11(4), 664–687 (2016). [CrossRef] [PubMed]
18. D. Cialla-May, X. S. Zheng, K. Weber, and J. Popp, “Recent progress in surface-enhanced Raman spectroscopy for biological and biomedical applications: from cells to clinics,” Chem. Soc. Rev. 46(13), 3945–3961 (2017). [CrossRef] [PubMed]
19. J. E. Kim, J. H. Choi, M. Colas, D. H. Kim, and H. Lee, “Gold-based hybrid nanomaterials for biosensing and molecular diagnostic applications,” Biosens. Bioelectron. 80, 543–559 (2016). [CrossRef] [PubMed]
20. J. F. Li, Y. F. Huang, Y. Ding, Z. L. Yang, S. B. Li, X. S. Zhou, F. R. Fan, W. Zhang, Z. Y. Zhou, D. Y. Wu, B. Ren, Z. L. Wang, and Z. Q. Tian, “Shell-isolated nanoparticle-enhanced Raman spectroscopy,” Nature 464(7287), 392–395 (2010). [CrossRef] [PubMed]
21. S. Li, L. Li, Q. Zeng, Y. Zhang, Z. Guo, Z. Liu, M. Jin, C. Su, L. Lin, J. Xu, and S. Liu, “Characterization and noninvasive diagnosis of bladder cancer with serum surface enhanced Raman spectroscopy and genetic algorithms,” Sci. Rep. 5(1), 9582 (2015). [CrossRef] [PubMed]
22. Z. Liu, Z. Guo, H. Zhong, X. Qin, M. Wan, and B. Yang, “Graphene oxide based surface-enhanced Raman scattering probes for cancer cell imaging,” Phys. Chem. Chem. Phys. 15(8), 2961–2966 (2013). [CrossRef] [PubMed]
23. S. Wachsmann-Hogiu, T. Weeks, and T. Huser, “Chemical analysis in vivo and in vitro by Raman spectroscopy--from single cells to humans,” Curr. Opin. Biotechnol. 20(1), 63–73 (2009). [CrossRef] [PubMed]
24. P. Rajapandiyan and J. Yang, “Sensitive cylindrical SERS substrate array for rapid microanalysis of nucleobases,” Anal. Chem. 84(23), 10277–10282 (2012). [CrossRef] [PubMed]
25. Z. G. Zhao, W. Zhao, G. Claire, and S. L. Huang, “A spectrum signals detection method for surface enhanced raman scattering under high fluorescence and background noise,” Guangpuxue Yu Guangpu Fenxi 30(8), 2146–2150 (2010). [PubMed]
26. J. Zhao, H. Lui, D. I. McLean, and H. Zeng, “Automated autofluorescence background subtraction algorithm for biomedical Raman spectroscopy,” Appl. Spectrosc. 61(11), 1225–1232 (2007). [CrossRef] [PubMed]
27. S. Feng, R. Chen, J. Lin, J. Pan, G. Chen, Y. Li, M. Cheng, Z. Huang, J. Chen, and H. Zeng, “Nasopharyngeal cancer detection based on blood plasma surface-enhanced Raman spectroscopy and multivariate analysis,” Biosens. Bioelectron. 25(11), 2414–2419 (2010). [CrossRef] [PubMed]
28. C. G. Atkins, K. Buckley, M. W. Blades, and R. F. B. Turner, “Raman Spectroscopy of Blood and Blood Components,” Appl. Spectrosc. 71(5), 767–793 (2017). [CrossRef] [PubMed]
29. B. R. Wood and D. McNaughton, “Raman Excitation Wavelength Investigation of Single Red Blood Cells In Vivo,” J. Raman Spectrosc. 33(7), 517–523 (2002). [CrossRef]
30. G. J. Puppels, F. F. de Mul, C. Otto, J. Greve, M. Robert-Nicoud, D. J. Arndt-Jovin, and T. M. Jovin, “Studying Single Living Cells and Chromosomes by Confocal Raman Microspectroscopy,” Nature 347(6290), 301–303 (1990). [CrossRef] [PubMed]
31. A. Giuliani, “The application of principal component analysis to drug discovery and biomedical data,” Drug Discov. Today 22(7), 1069–1076 (2017). [CrossRef] [PubMed]
32. M. S. Bergholt, W. Zheng, K. Lin, K. Y. Ho, M. Teh, K. G. Yeoh, J. B. So, and Z. Huang, “Combining near-infrared-excited autofluorescence and Raman spectroscopy improves in vivo diagnosis of gastric cancer,” Biosens. Bioelectron. 26(10), 4104–4110 (2011). [CrossRef] [PubMed]
33. A. Huefner, W. L. Kuan, K. H. Müller, J. N. Skepper, R. A. Barker, and S. Mahajan, “Characterization and Visualization of Vesicles in the Endo-Lysosomal Pathway with Surface-Enhanced Raman Spectroscopy and Chemometrics,” ACS Nano 10(1), 307–316 (2016). [CrossRef] [PubMed]
34. S. A. Hosseini, M. Abdolahad, S. Zanganeh, M. Dahmardeh, M. Gharooni, H. Abiri, A. Alikhani, S. Mohajerzadeh, and O. Mashinchian, “Nanoelectromechanical chip (NELMEC) combination of nanoelectronics and microfluidics to diagnose epithelial and mesenchymal circulating tumor cells from leukocytes,” Small 12(7), 883–891 (2016). [CrossRef] [PubMed]
35. J. Lin, R. Chen, S. Feng, Y. Li, Z. Huang, S. Xie, Y. Yu, M. Cheng, and H. Zeng, “Rapid delivery of silver nanoparticles into living cells by electroporation for surface-enhanced Raman spectroscopy,” Biosens. Bioelectron. 25(2), 388–394 (2009). [CrossRef] [PubMed]
36. N. Beije, A. Jager, and S. Sleijfer, “Circulating tumor cell enumeration by the CellSearch system: the clinician’s guide to breast cancer treatment?” Cancer Treat. Rev. 41(2), 144–150 (2015). [CrossRef] [PubMed]
37. A. Huefner, W. L. Kuan, R. A. Barker, and S. Mahajan, “Intracellular SERS nanoprobes for distinction of different neuronal cell types,” Nano Lett. 13(6), 2463–2470 (2013). [CrossRef] [PubMed]
38. P. Rajapandiyan, W.-L. Tang, and J. Yang, “Rapid detection of melamine in milk liquid and powder by surface-enhanced Raman scattering substrate array,” Food Control 56, 155–160 (2015). [CrossRef]
39. N. Tien, H. C. Chen, S. L. Gau, T. H. Lin, H. S. Lin, B. J. You, P. C. Tsai, I. R. Chen, M. F. Tsai, I. K. Wang, C. J. Chen, and C. T. Chang, “Diagnosis of bacterial pathogens in the dialysate of peritoneal dialysis patients with peritonitis using surface-enhanced Raman spectroscopy,” Clin. Chim. Acta 461, 69–75 (2016). [CrossRef] [PubMed]
