诊断

Research Article

Analysis on intensities and profile of Raman spectroscopy for CNE1 and CNE2

Jianghua Li1,2*,  Yong Du3*,  Musheng Zeng3

 

1School of Michanical and Electrical Engineering, Shenzhen Institute of Information Technology, Shenzhen, 518172, China

2School for Information and Optoelectronic Science and Engineering, South China Normal  University, Guangzhou 510006, China

3Department of Radiation Oncology, Cancer Center, Sun Yat-sen University of Medical Sciences,Guangzhou 510060, China

Corresponding author: Jianghua Li, E-mail: lijianghua25@163.com; Mushen Zeng, E-mail: zengmsh@mail.sysu.edu.cn

OCIS codes: 170.5660, 170.1530

*Contributed equally to this study

 

Citation: Li JH, Du Y, Zeng MS. Analysis on intensities and profile of Raman spectroscopy for CNE1 and CNE2. J Nasopharyng Carcinoma, 2014, 1(17): e17. doi:10.15383/jnpc.17.

Funding: This work was funded by the National Natural Science Foundation of China (No. 31200629) and China National Funds for Distinguished Young Scientists (No. 81025014).

Competing interests:The authors have declared that no competing interests exist.

Conflict of interest: None.

Copyright:image001.gif2014 By the Editorial Department of Journal of Nasopharyngeal Carcinoma. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

 

Abstract: Raman spectroscopy (RS) has proved to be very effective in tracing the distribution of biological molecules within nasopharyngeal carcinoma (NPC) cells. In this paper, the representative radiotherapy model of NPC cell lines CNE1 and CNE2 were measured by Raman spectroscopy in the range of 750-3000cm-1. The scatter plots of intensity ratios of protein, lipid, and nucleic acids showed some overlap using some Raman markers. The spectral data were further evaluated using 19 intensities data set (selected intensity points) and the whole spectra data set (whole intensity points) by principal component analysis and linear discriminate analysis, yielding a diagnostic accuracy of 78% (56/72) and of 94% (68/72) to differentiate the two NPC cell lines, respectively. Our findings suggested that the whole Raman spectra are more optimistic accuracies than selected intensities for diagnosing the NPC cell lines and the information in the shape of RS could be further revealed by intensities.

Keywords: Raman spectroscopy; Nasopharyngeal carcinoma; CNE1; CNE2; PCA LDA

  

 

1. Introduction

        Nasopharyngeal carcinoma (NPC) is a disease with a remarkable geographical and racial distribution. It is vastly more common in certain regions of East Asia and Africa than elsewhere [1-2]. The 5-year survival rate depends greatly on the detection staging of the disease, so early diagnosis is crucial to improve survival rate. However, owing to the deep location of nasopharynx and vague symptoms of the disease, most NPC patients are diagnosed at the late stages. Nowadays, standard treatments for NPC patients include radiation therapy, chemotherapy and surgery. Among these, radiation therapy is the most popular. However, NPC cells have the categories of radiosensitive (e.g.CNE1) or radio-resistant (e.g.CNE2) according to both the characterization of the NPC cell and their radiotherapy reaction [3]. The target treatment is a better way to improve 5-year survival rate except for early diagnosis. Therefore, accurate diagnosis of cell types will release the pain of NPC patients without unnecessary treatment and improve the therapy effect.

         Raman scattering is an inelastic scattering process that occurs when cell is illuminated with laser light, photons interact with the intramolecular bonds present; the photon then donates energy to or receives energy from the bond, producing a change in the bond’s vibrational state. An electron jumps to an excited state then falls back to a higher or lower vibrational energy level with accompanied release of a new photon. The energy transfer (‘Raman shift’) is proportional to a specific vibrational mode of the molecular, so Raman spectrum is independent of excitation wavelength; each peak can be associated with specific vibration in molecular bond [4]. Thus, Raman spectroscopy (RS) can provide detail information on chemical composition, molecular structure and molecular interactions such as proteins, nucleic acids, and lipids in sample [5]. The characteristic of RS makes it ideal for probing cell because numerous biological molecules undergo some Raman scattering. It is particularly suited for diagnosing cancer because of its sensitivity in detecting small molecular changes that are associated with physiology such as an increased nucleus-to-cytoplasm ratio, disordered chromatin, higher metabolic activity [6]. Both in vivo and in situ measurements by RS have demonstrated the great abilities for differential diagnosis of cancer and cancer cell lines with minimum or no sample preparation [7-8].

           Recently, the discrimination of NPC including cell lines and tissue [9-12] by RS was complemented. Spectral differences were observed between the cancer and normal subjects; Nine Raman band ratios were used for classification [10] using principal component analysis (PCA) and linear discriminant analysis (LDA), the sensitivity and specificity were reported to be 92% and 82%, respectively. Our group differentiated the four NPC cell lines with PCA-LDA, the overlap rate of the PC scores was serious and we could sort them at two steps [13]. However, all of them focused on the “fingerprint” spectral region with wavenumber from 800-1800 cm-1. Lipid was noted in the C-H stretching band (2800-3030 cm-1), which was reported to be presented in increased numbers in cancer as compared to normal tissue [14]. Thus, we investigated on cell lines CNE1 and CNE2 extending to the region of high wavenumber (800-3000 cm-1). The cell line CNE1, which came from well-differentiated squamous cell carcinoma, and CNE2 from poor differentiated squamous cell carcinoma, have an obvious discrepancy in morphological aspect and biological behaviour. So this pair of cell lines has already become a common model in the study of radiobiology [15]. The aim is to test which parameters of cell lines could be more effectively discriminated CNE1 and CNE2 using PCA-LDA, compared the main selected intensities data set or the whole spectra data set.

 

2. Materials and methods

2.1. Sample Preparation

CNE1 and CNE2 were supplied by the Cancer Center of Sun Yat-sen University. They were cultured in DMEM medium supplemented with 10% fetal calf serum (HyClone); cell suspensions were prepared by trispsination. All the cells were grown in an incubator humidified 5% CO2 atmosphere at 37◦C. All the cells were suspended in phosphate-buffered saline (PBS) to maintain the cells’ activities prior to Raman measurement in a logarithmic growth phase.  Left and right views of Fig. 1 are 200× representative histological sections stained with H&E for CNE1 and CNE2, respectively.

2.2.  RS measurement of Single Cell

       A 10-μl drop of suspended live cells in PBS was deposited on pure aluminum flake, where the cells can be allowed to settle and become immobilized after dry, ~15μm above the aluminum surface. A cell is positioned near the focus of the laser beam by moving the manual stage. Raman spectra of individual cells were probed at random three locations using a confocal Raman micro-spectrometer (Renishaw, Great Britain) in the range of 750-3000cm−1  under a 785nm diode laser excitation (about 10mW of power); Such power and excitation wavelength would not result in any notable cell degeneration[16]. The experiment was repeated for three times, at each time the random five cells were detected. The spectra were collected in backscattering geometry using a microscope equipped with a Leica 50× objective connected to a Renishaw 2000 spectrometer with a spectral resolution of 2cm−1(N.A.=0.7), resulting in a diffraction limited spot of 1 µm diameter at the laser focus. The detection of Raman signal was carried out with a Peltier cooled charge-coupled device (CCD) camera. Each cell spectrum was processed to remove cosmic rays, reduce noise via spectral smoothing, estimate and subtract a baseline arising from the aluminum substrate and biological fluorescence and normalize to the total amount of biological material within the sampling volume. The software package WIRE 2.0 was employed for spectral acquisition and analysis. The Raman spectra were acquired with 10s integration time. Peak frequencies and rapid checking of instrumental performance were calibrated with the silicon phonon line at 520 cm−1.

2.3. Data Analysis

      Firstly, Raman spectra of CNE1 and CNE2 were collected; then the background correction was operated by a background-correction algorithm named adaptive iteratively reweighted Penalized Least Squares (airPLS), which was implemented in R language [17-18]. Finally, the steady band at 1004 cm−1 assigned as Phenylalanine was used for normalizing the RS curves [19], which enable a better comparison of the spectral shapes and relative Raman band intensities among different organs and anatomical regions. The mean spectra and   different cell-types’ spectra with the standard deviation were calculated to determine cell-to-cell variability within a particular cell type and were defined using student’s t-test statistics. Subsequently, a principal component analysis (PCA) and linear discriminate analysis (LDA) were performed on all spectra to extract persistent features of spectra from individual cell classes and to compare different cell classes against each other. Leave-one-out and cross-validation methods (LOOCV) are then employed to validate the classification performance. The data analysis were performed with  the  Matlab  software  (The  Mathworks,  Inc.,  Natick,  MA, USA), version 7.5.

 

3. Results and discussion

3.1. Analysis of the cells’ RS

      The average normalized Raman spectra of CNE1 (n=36) and CNE2 (n=36) in the range of 800-1750 cm−1 and 2800-3000 cm−1 are shown in Fig.2(A-B), respectively. The obvious bands of CNE1 and CNE2 have been reported:[9-10]  it included the bands  at 1001cm−1 (representing the protein), 1123cm−1 (protein, lipids), 1336cm−1(DNA), 1446cm−1 (protein and phospholipids), 1660cm−1 (protein), 2857cm−1 (lipid),  2879cm−1 ( lipid), 2935cm−1 ( protein). After the removal of fluorescence background from the original data, the rich spectra acquired in the nucleus and cytoplasts could be observed. The obtained nineteen peaks and their assignments of CNE1 and CNE2 are listed in table 1[9, 20-28]. Compared to the previous reports, there were some band shifts which might be caused by the different culture methods. However, the two cell lines exhibited almost a similar spectral profile with subtle differences, which indicated that they were composed of the same components with content variation of certain bands. Comparing with CNE1, it could be seen that a red-shift was observed in CNE2 cell at 1208, 1578 cm−1 assigned to proteins and nucleic acids/protein, respectively.  However, the very weak bands which were not labelled were located at 879, 939, 1068, 1078, 1173 cm−1, they were related to L-proline, α-helix, DNA, (lipids),  and protein(Tyrosine, Phenylalanine), respectively. But the bands at 979, 1103, 1258, 1578, 2857cm−1 could be observed which were missed in other papers.

      Overall, we had identified Raman bands in organs of CNE1 and CNE2, which were highly associated with proteins, DNA, and lipids. The spectra with the subtle interanatomical variability indicated that the overall biomolecular and biochemical constituents of the two cell lines were very similar. CNE1 and CNE2 are differentiated squamous cell carcinoma of the nasopharynx, highly and poorly, respectively. However, there were no studies on differentiating them by RS in 750-3000cm-1. Viewing the differences between the normal tissue and NPC tissue had been reported. The significance of cancerous tissue transformation differs in the Raman bands at 1094, 1260-1340, 1530-1580 cm-1 etc, where the malignant tissue spectrum had relatively strong intensities. In malignant spectra the intensity ratio was greater with respect to that in the normal samples, indicating increased collagen levels and/or cellular nuclear content in squamous cell carcinoma compared with that of normal tissue. In contrast to the tissue, these normalized intensity differences between CNE1 and CNE2 could be viewed more clearly in the difference spectra with ±1 standard deviations (SD) ,which were shown to be different by Student’s t-test at >90.0% confidence.(Fig.2.(C-D)). It could be seen that the intensities of the majority bands of CNE1(820-920 cm-1, 1004 cm-1, 1250-1280 cm-1, 1440-1460 cm-1) were greater with respect to the CNE2, including lipid in CNE1 was greater than in CNE2 (2800-3030 cm-1). This was consistent with the theory that CNE1 had more abundant cytoplasm in nucleus [29]; which might be the reason that CNE1 was more radioresistant than CNE2 [30].

Table 1. Band assignment for Raman spectra of CNE1 and CNE2 cell lines

                          Band(cm-1)                                 Assignments

 

 CNE1        CNE2                                         

829               828                            DNA (PO2 asymmetric stretching [23]

855               853                              δ(CCH) ring breathing of collagen [17]  

979               978              (v(C–C) of  a-helix conformation for proteins 21]                       

1004                                        (n(C–C) ring   breathing of phenylalanine[26]

1034             1033                        Phenylalanine, Protein C-N stretching[25]

1103             1104                          C–C stretching of phospholipids, [21]

1128             1128                                            protein:C-N stretching

1158             1158                                         DNA:Ribose-phosphate

1208             1211                   Tyrosine, phenylalanine, protein: Amide

1258             1257                amide III of protein   [24]

1320             1318                  Guanine, protein: C-H deformation vibration

1341             1339         Adenine, guanine, protein: C-H deformation vibration

                                           CH3CH2 twisting of nucleic acids  [26]

1451             1451         Lipids/protein:CH3 deformation, CH2 bending modes

1578             1582               C=C bending of pyrimidine ring of nucleic acids

1608             1609                  Phenylalanine, tyrosine, protein: C=C

1661             1659                amide I v(C=O) of proteins, a-helix conformation

2857             2857                           (lipid)[27],

2879             2877                             CH2 asymmetric stretching of lipid

2935             2933                                              CH3 stretching of protein

 

        The Raman active biochemical profiles of two cell lines were further assessed by their intensity ratios. The relative intensity ratio of bands I1257/I1608 vs I855/I829, I1208/I1158 vs I855/I829, I2935/I1578 vs I1451/I1578, and I2879/I979 vs I2935/I979 were used as the diagnostic parameters. 2-D scatter plots of the distribution of their intensity ratios were shown in Fig.3. T-test analysis further confirmed this difference (p<0.05). The Raman peaks at 829, 1158 cm−1 were highly associated with DNA contents. Whereas the peaks at 855, 1608, 1208 and 1257 cm−1 representing the protein contents. And the peaks at 979, 1578, 2879, 2935 cm−1 were assigned to carbohydrate, nucleus acid, lipid and protein.  This indicated that CNE1 and CNE2 had the different component ratios of protein, DNA, nucleus acid, lipid and carbohydrate. These bimolecular changes suggested the increased nuclear activity, which was of considerable pathological value for tissue diagnosis and characterization. Proteins and lipid could also be related to increasing mitotic activities [31].  The overlap plots showed that

intensities were not sensitive enough to differentiate the two cell lines of CNE1 and CNE2.

3.2.  Multivariate analysis PCA-LDA

       To test the capability of obtained Raman spectra for differentiating CNE1 and CNE2, PCA-LDA algorithm was performed on the measured Raman spectra [32]. All selected intensities data set (20 intensities listed in table1) and the whole spectra data set (all points make up of the profile of RS) were fed into the Matlab 7.5 for PCA-LDA analysis: after a reduction of the data to a small subset, linear discriminant scores for CNE1 versus CNE2 group were obtained, where p<0.05 under T-test analysis. These diagnostic models together with leave-one out, cross validation technique yielded diagnostic accuracy of 94% (68/72) for distinction of CNE1 and CNE2 using the whole spectra shown in Fig.4A; while Fig.4B  showed an accuracy of 78% (56/72). As shown in Fig.4A and Fig.4B, the zero line separated CNE1 from CNE2 more clearly in former. The main information obtained from the PCA is described by the first twenty principal components for case A (80% of variance) and by the first ten principal components for case B (99.5 % of variance).

Fig. 5 displayed receiver operating characteristic (ROC) curves illustrating the true-positive rate against the false-positive rate to further evaluate the performance of the recursive partitioning diagnostic model. The ROC gave an area under curve of 0.99 and 0.90 by solid line and dashed line, representing the curve drawn by the whole spectra data set and all intensities data set, respectively. The classification accuracy of the whole data set was always higher with respect to intensity selected data set. PCA-LDA diagnostic algorithm gave rise to the sensitivity of 97% (70%) and specificity of 92% (83%) for case A(B), respectively.

 3.3. Discussion

        The NPC cell lines, CNE1 and CNE2, which were both derived from squamous cell carcinoma, exhibiting prominent difference in morphology and biological behaviour. It has been demonstrated that CNE2 is more radiosensitive than CNE1. Wand et al found that ATM(ataxia-telangiectasia mutant), a protein of DNA repair, was higher in CNE1 than that in CNE2 [33]. So this pair of cell lines has already become a common model in the study of radiobiology in NPC.

      Representative bands for DNA consist of the DNA backbone vibrations at 830 and 1093 cm−1 as well as adenine (1490 cm−1) and guanine (1579 cm−1). The Raman spectrum of a protein is characterized by Amide I (1680-1682 cm−1), Amide II (~1607 cm−1) and Amide III (1248-1249 cm−1) vibrations of the peptide backbone [34-36]. Amide I and III bands are classical Raman markers for protein conformation as the Amide II band is usually weak in a Raman spectrum and can in practice usually not be used as a marker. Spectroscopic markers for lipids can be divided into bands originating from the acyl chain and the lipid head group. Acyl chain markers include a prominent band at 1440 cm−1 corresponding to CH2 bending vibrations, while the 1460 cm−1 band is representative of CH3 bending mode. All transskeletal vibrations are represented around 1133 cm−1and 1064 cm−1 by C-C stretching bands. Head group markers include the band at ~ 860 cm−1 which is assigned to a phosphate group characteristic to most phospholipids [36]. The above analysis showed that Raman biomarkers are not separately and unique. It was therefore the intensity ratios of protein, lipid and acid had serious overlap in the scatter plot of the distribution of CNE1 and CNE2. Furthermore, the study compared to differentiate the two cell lines CNE1 and CNE2 by RS in the range of 750-3000cm-1 using the whole data set and the intensity selected data set. The findings showed that both of them can be used to identify the two cell lines but the former had the higher accuracy. These might be the same reason that some valuable biochemical information was not constructed, thus the information in the shape of RS could be further revealed in the future.   

 

4. Conclusions

Raman spectroscopy is a rapid and non-destructive technique for cell and neither labels nor fixation of cells is required while measurements can be taken in physiological conditions. Proteins, nucleic acids and lipids can be studied within cells as Raman spectroscopy is sensitive to chemical and physical changes of biological molecules.  In this study, we tested how the intensities data set and the whole spectra data set made the contributions on the accuracy of the NPC cell lines CNE1 and CNE2 using PCA-LDA algorithm. The two cell lines could be identified with the a diagnostic accuracy of 94% (68/72) for distinction of CNE1 and CNE2 using the whole spectra; while an accuracy of 78% (56/72) clustering of the two kinds of cell lines using the selected intensities data, respectively, demonstrating that combination of RS with the classification procedure on the whole spectra could help the pathologist to improve diagnostic accuracy.

 

Acknowledgements

This work was funded by the National Natural Science Foundation of China (No. 31200629) and China National Funds for Distinguished Young Scientists (No. 81025014)

 

References:

[1] V.P. Lutzky, D.J. Moss, D. Chin, W.B. Coman, P.G. Parsons, and G.M. Boyle, Biomarkers for Cancers of the Head and Neck. Clinical Medicine Insights: Ear, Nose and Throat 1, 5-15 (2008).

[2] K.W.Lo, K.F.To, D.P.Huang, Focus on Nasopharyngeal Carcinoma. Cancer Cell, 5 (5), 423-428 (2004).

[3] Z.Q.Li, Y.F.Xia, Q.Liu, W.Yi, X.F.Liu, F.Han, W.Luo, T.X Lu, Radiotherapy-related typing in 842 patients in canton with nasopharyngeal carcinoma. Int. J Radiat Oncol Biol Phys. 66, 1011-1116 (2006).

[4]   M.D.      Keller,   E.M.      Kanter,   A.     Mahadevan-Jansen, Mahadevan-Jansen, Raman Spectroscopy for Cancer  Diagnosis. A. Spectrosc.  21(11), 33-41  (2006).

[5] G.J.Puppels, F.F.M. de Mul, C.Otto, J.Greve, M.Robert-Nicoud, D.J.Arndt-Jovin, T.M. Jovin, Studying single living cells and chromosomes by confocal Raman microspectroscopy. Nature   347, 301-303  (1990).

[6] R.Liu, J.F. Liu, X.X.Zhou, G.B.Jiang, Applications of Raman-based techniques to on-site and in-vivo analysis. Trends Anal. Chem. 30( 9), 1462  (2011).

[7]  W. S. Iwan and  H. Thomas , Methods and Applications of Raman Microspectroscopy to Single-Cell Analysis. Appl. Spectrosc. 67( 8), 813-828  (2013).

[8] K. Charlotte, A. L. Max, B. Hugh, W. James, H. Joanne, K. Catherine, S. Nick, Advances in the clinical application of Raman spectroscopy for cancer diagnostics. Photodiagnosis and Photodynamic Therapy  10, 207-219  (2013).  

[9] Y.H.Ye, Y.Chen, Y.Z Li, Y.Su, C.Y.Zou, Y.P.Chen, L.Ou, R.Chen, and H.S.Zeng,  Characterization and discrimination of nasopharyngeal carcinoma and nasopharyngeal normal cell lines using confocal Raman microspectroscopy. Spectroscopy  25, 217-224  (2011).

[10] Y.Z. Li , J. J. Pan , G. N. Chen , C. Li , S. J. Lin , Y. H. Shao , S.Y. Feng , Z.F. Huang , S. S.Xie , H.S. Zeng , R. Chen,  Micro-Raman spectroscopy study of cancerous and normal nasopharyngeal tissues. J Biomed Opt. 18(2), 270-273  (2013).

[11] H. Huang, H. Shi, S.Y. Feng, W.W. Chen, Y. Yu,  D. Lin and R. Chen, Confocal Raman spectroscopic analysis of the cytotoxic response to cisplatin in nasopharyngeal carcinoma cells. Anal. Methods, 5, 260-266  (2013).

[12] J.Lin, R.Chen, S.Feng, Y.Li, Z.Huang, S.Xie, Y.Yu, M.Cheng, H.Zeng,  Rapid delivery of silver nanoparticles into living cells by electroporation for surface-enhanced Raman spectroscopy. Biosens. Bioelectron. 25, 388-394  (2009). 

[13] P. Liu, J. H. Li, Z. W. Tu, et al. Discrimination of NPC Cell Lines Associated with Malignant Types Using Raman Spectroscopy. Laser Physics. 22(11), 1682-1688  (2012). 

[14] S.H. Claudia, U. Martin, H. Carmen, R. Angelika, Investigation of lipid bodies in a colon carcinoma cell line by confocal Raman microscopy. Medical Laser Application.  26, 152-157  (2011).

[15] H.M. Wang, X.R. Wu, Y.F. Xia, et al.  Expression of ATM protein in nasopharyngeal carcinoma cell lines with different radiosensitivity. Chin J Cancer. 22, 579-581  (2003).

[16] C.X.Yu, E.Gestl, et al. Characterization of human breast epithelial cells by confocal Raman microspectroscopy. Cancer Detection and Prevention. 30, 515-522  (2006).

[17] Z.M.Zhang, S.Chen, Y.Z.Liang, Z.X Liu, Q.M.Zhang, L.X.Ding, F.Ye, and H. J.Zhou, An intelligent background-correction algorithm for highly fluorescent samples in Raman spectroscopy.  J Raman Spectrosc. 41, 659-669  (2009).

[18] Z.M.Zhang, S.Chen, Y.Z.Liang, Baseline correction using adaptive iteratively reweighted penalized least Squares.  Analyst. 135, 1138-1146  (2010).

[19] X.L.Yan, R.X.Dong, L.Zhang, X.J.Zhang, Z.W Zhang, Raman spectra of single cell from gastrointestinal cancer patients. World J Gastroenterol. 11(21), 3290-3292  (2005).

[20] J.W. Zhang, A.G. Shen, Y. Wei, X.H. Wang, et al.  Study of normal mucosa and gastric carcinoma by confocal Raman microspectroscopy. J Biomed Eng. 21(6), 910-912  (2004).

[21] J.D.Gelder, K.D.Gussem, P.Vandenabeele, and L.Moens, Reference database of Raman spectra of biological molecules. J. Raman Spectrosc. 38, 1133-1147 (2007).

[22] S.K.The, W.Zheng, K.Y.Ho, M.The, K.G.Yeoh, Z.W.Huang, Diagnosis of gastric cancer using near-infrared Raman spectroscopy and classification and regression tree techniques.  J. Biomed. Opt. 13(3), 13-34  (2008).

[23] Z.W.Huang, S.K The, W.Zheng, K.Lin, K.Y.Ho, M.The, and K.G.Yeoh, In vivo detection of epithelial neoplasia in the stomach using image-guided Raman endoscopy. Biosens. Bioelectron. 26, 383-389  (2010).

[24] P.Crow, B.Barrass, C.Kendall, M.Hart-Prieto, M.Wright, R.Persad, and N Stone,  The use of Raman spectroscopy to differentiate between different prostatic adenocarcinoma cell lines. Br J Cancer. 92(12), 2166-2170   (2005).

[25] M.S.Bergholt, W.Zheng, K.Lin, K.Y.Ho, M.The, K.G.Yeoh, J.B.Y. So, and Z.Huang, In vivo diagnosis of gastric cancer using Raman endoscopy and ant colony optimization techniques. Int. J. Cancer. 128, 2673-2680  (2010). 

[26] M.S.Bergholt, W.Zheng, K.Lin, K.Y.Ho, M.The, K.G.Yeoh, J.B.Y. So, and Z.Huang, Characterizing variability in in vivo Raman spectra of different anatomical locations in the upper gastrointestinal tract toward cancer detection. J. Biomed. Opt. 16(3), 037003   (2011).

[27] K.W.Short, S.S.Carpenter, J.P.Freyer, and J.R.Mourant, Raman Spectroscopy Detects Biochemical Changes Due to Proliferation in Mammalian Cell Cultures. Biophysical Journal. 88, 4274-4288  (2005).

[28] K.Hamada,   K.Fujita,   N.I.Smith,   M.Kobayashi,  Y.Inouye,   S.Kawata, Raman microscopy for dynamic molecular imaging of living cells. J Biomed Opt. 13(4), 044027  (2008).

 [29] Hunan college of medicine cancer research centre, Nasopharyngeal carcinoma cytological map, People’s health press, Beijing (1981).

[30] P.P.Zhong, Y.X.HE, Y.F.Xia, S.S.Yan,  Expression and subcellular localization of DNA-PK in nasopharyngeal carcinoma cell lines CNE1 and CNE2 with different radiosensitivity.    Chinese   Journal   of   Cancer    Research. 18(2),77-82  (2006).

[31] Z.Huang, M.S.Bergholt, W.Zheng, K.Lin, K.Y.Ho, M.The, and K.G.Yeoh, In vivo early diagnosis of gastric dysplasia using narrow-band image-guided Raman endoscopy. J. Biomed. Opt., 15, 037017  (2010).

[32] I.Notingher, G.Jell, P.L.Notingher, I.Bisson, O.Tsigkou, J.M.Polak, M.M.Stevens, L.L.Hench, Multivariate analysis of Raman spectra for in vitro non-invasive studies of living cells. Journal of Molecular Structure. 744, 179-185  (2005).

[33] Y.X. He, P.P. Zhong, S.S. Yan, L. Liu,  H.L. Shi,  M.S. Zeng,  Y.F. Xia, DNA-dependent protein kinase activity and radiosensitivity of nasopharyngeal carcinoma cell lines CNE1/CNE2. Sheng Li Xue Bao. 59(4), 524-533  (2007).

[34] C. Krafft, T. Knetschke, A. Siegner, R.H.W. Funk, R. Salzer, Mapping of single cells by near infrared Raman microspectroscopy. Vib. Spectrosc. 32, 75-83  (2003).

[35]  P.V. Zinin, A. Misra, L. Kamemoto, Q. Yu, N. Hu, S.K. Sharma,  Visible, near-infrared, and ultraviolet laser-excited Raman spectroscopy of themonocytes/macrophages (U937) cells.  J. Raman Spectrosc. 41,  268-274 (2010)

 

 

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.

eISSN: 2312-0398

Asia Press is a professional Science, Technology and Medicine publisher, who owns rapid publication, Peer-Reviewed, Open Access Journals. Asia Press aims to promote “knowledge sharing”. As you know, the main barrier for free “knowledge sharing” is the cost of publishing and transfer. In order to encourage scholars and scientists to the max, and devote whole power to realize the aim of “knowledge sharing” and the benefit of “all” mankind, Asia Press performs a permanent policy of no charge for publication and access, and always open its door for authors worldwide.

© 2013-2017 by the Asia Press. All rights reserved.