[Baseline correction of Raman spectrum based on piecewise linear fitting].

The baseline drifts of Raman spectra occur in many types of instrumental measurements. It is an important part and a routine step to correct the baseline drift for the data preprocessing. In the present work, the limitations of the baseline correction method based on polynomial fitting were highlighted and a modified polynomial fitting method, i. e. piecewise linear fitting method, was proposed. Combined with the computer, this method could eliminate the baseline automatically. A series of Raman spectra of single polystyrene bead, red blood cell or yeast cell acquired by laser tweezers Raman spectroscopy were preprocessed by this method and its efficiency was verified. The results demonstrated that piecewise linear fitting can correct the baseline shifts effectively and provides more accurate information for further data analysis. It is a feasible method for correction of Raman spectrum.

[Identification of Cortex Phellodendri by Fourier-transform infrared spectroscopy and principal component analysis].

Fourier transform infrared spectrometer was used to collect infrared spectra of Cortex Phellodendri from six different regions. Original spectra were preprocessed by carrying out appropriate baseline correction and five-points smoothing, and the averaged spectra of Cortex Phellodendri from the six origins were analyzed. As a result, the averaged spectra looked quite similar. The normalized spectra were selected to construct principal component analysis model in the range of fingerprint region 1800 - 500 cm(-1), and according to the model, the first three principal components accounted for 98% of the variance information in the fingerprint region, and each sample was able to form distinct cluster in the principal component space, then the identification of Cortex Phellodendri from the six regions was basically achieved; besides, to some extent, the sparse density of the samples distribution reflected the genetic relationship. The loading factors of the model were analyzed, and the results indicated that the differences between Cortex Phellodendri samples mostly depended on the contents of protein, carbohydrates, lipids, alkaloids, sterols, obaculactone, oba-cunone, and obacunonlc acid. On the whole, combined with principal component analysis, FTIR provides an effective way to evaluate the herbal Cortex Phellodendri rapidly and nondestructively, which also reflects the content difference of material composition.

[Raman tweezers-based analysis of carotenoid synthesis in Rhodotorula glutinis].

Carotenoid synthesis in Rhodotorula glutinis was investigated with Raman tweezers in order to find the effect of nitrogen and carbon resource on carotenoid yield. The cells in fermentation terminus were harvested, and then divided into two parts, one for UV analysis, the other for Raman tweezers detection. Original spectra were preprocessed by carrying out background elimination and baseline correction, and the averaged spectra of cells cultivated in different fermentation medium were analyzed qualitatively. The results showed that the Raman intensity of carotenoid were obviously different. There was a high correlation between UV results and Raman peak height data, the correlation coefficients of fitted parameters were 0.907 8 and 0.912 1, respectively. Quantitative analysis of 1 508 cm(-1) peak height indicated that the appropriate nitrogen and carbon resources for the growth of Rhodotorula glutinis cells and synthesis of carotenoid were yeast extract + tryptone, and glucose, respectively. The above results suggest that Raman tweezers can provide information about carotenoids in Rhodotorula glutinis cells and serve as an effective tool for real time measurement of carotenoid synthesis and optimization of fermentation medium.

[Sorting oleaginous yeast by using optical manipulation and Raman spectroscopy].

Extensive research has been carried out in an effort to screen the oleaginous microorganisms. Here, Raman spectroscopy and laser tweezers were used to sort oleaginous yeast from mixed yeast cells. The preprocessing of subtracted background, 17 points S-G smoothing filter, polynomial fitting baseline correction and vector normalization were performed and the main features information of intracellular substances from the Raman spectroscopy of yeast cells was extracted by combining principal component analysis. Based on the distinguished composition of oleaginous yeast and non-oleaginous different yeast, a sorting model was established. The test yeast cell in optical trapping was distinguished real-time by the model referring to its Raman spectra. The cells distinguished as oleaginous yeast were collected by means of optical manipulation. The sorted oleaginous yeast cells were verified by microbial culture and Sudan black B test. The result illustrates that Raman spectroscopy combined with optical manipulation is an effective technique for sorting oleaginous yeast and other economic microorganisms.

[Analysis of components difference of yeast strains based on laser tweezers Raman spectroscopy combined with multistatistical analysis].

The combination of laser tweezers and con-focusing Raman spectroscopy (LTRS) has made it possible to investigate single cells in aqueous media. In our experiments, a modified method of Percoll density gradient centrifugation was developed for isolating synchronous cells from six yeast strains and Raman spectra of single yeast cells were collected by LTRS as well, in which multiple statistical analysis, principal component analyses (PCA) and discriminant function analysis (DFA) were applied to distinguish synchronous cells between yeast strains statistically. The result showed that Raman spectra scattered from the trapping yeast cells could provide intrinsic molecular information, and there were remarkable difference among those of six yeast strains in Raman spectrogram which correspond to various biomacromolecule, the difference of protein as well as lipid were significantly higher than nucleinic acid between two of yeast strains randomly, and among the six strains, synchronized yeast cells can be discriminated using PCA and DFA based on 14 most contribution bands, 706, 862, 918, 997, 1 073, 1 127, 1 269, 1 291, 1 305, 1 429, 1 465, 1 591, 1 602 and 1 652 cm(-1), 10 bands of which were from protein, 3 bands were from lipoid, and 1 band was from nucleinic acid. To validate their significance, these variables were used to reperform DFA analysis and the replotted PC-DFA was the same as in the previous PC-DFA analysis, and the data would be considered validated. The authors show that the approach of laser tweezers Raman spectroscopy combined with multistatistical analysis has the potential to study difference among yeast strains.

[Using Raman spectroscopy to analyze apoptosis of gastric cancer cells induced by cisplatin].

Apoptosis of gastric cancer cells induced by cisplatin was investigated using laser Raman spectroscopy. Gastric cancer cells (SGC7901) were treated with 10 microg x mL(-1) cisplatin for 24, 48 and 72 hours, then were divided into two parts, one for fluorescence staining, the other for collection of Raman spectra by means of scanning. The acquired spectra were then preprocessed by background elimination, smoothing, normalization, baseline correction, and peak fitting. Fluorescence staining result showed that the nucleuses from untreated group were uniformly stained, while those from the group treated for 72 hours were densely stained and broken. The spectra results revealed that the intensity of peaks associated with nucleic acid and protein decreased after the cells were incubated with cisplatin for 24, 48 and 72 hours. The intensity of peaks at 783, 1002 and 1343 cm(-1) respectively fell to 52, 64 and 76 percent of the original value after 72 h of treatment, which indicated that cisplatin could induce apoptosis of gastric cancer cells and reduce the amount of nucleic acid and protein in the cells. The above results suggest that Raman spectra can provide abundant information about the changes in materials in cells and serve as an effective method for real time measurement of apoptosis.

[Laser tweezers Raman spectroscopy analysis of liver cancer tissue].

Single cell laser tweezers Raman spectroscopy (LTRS) has been applied to biology field. In the present article, the authors measured the spectra of liver cancer cells, para-cancer cells and normal hepatocytes using single cell laser tweezer Raman spectroscopy (LTRS) system and compared their average spectra changes. The results showed that the laser tweezers Raman spectroscopy could differentiate specimens of different pathological changes from liver tissue studied. The 1 070 and 1 266 cm(-1) peaks obtained from normal hepatocytes were more visible than the same two peaks obtained from liver cancer and para-cancer specimen. The 1 445 cm(-1) peak of normal hepatocytes was higher than that of liver cancer cells and para-cancer cells. It is known that the 1 070 cm(-1) peak represents lipids and nucleic acids, while 1 266 and 1 445 cm(-1) peaks represent lipids and proteins. So, these peak changes may directly reflect the changed biomaterials related to liver carcinogenesis. Thus, single cell laser tweezer Raman spectroscopy may be a nondestructive, rapid and good method to measure and analyze different pathological specimens from liver cancer.

[The effect of abnormal cell shape on the spectral distinguishing of erythrocytes using laser tweezers Raman spectroscopy].

Erythrocyte is a mature blood cell that contains hemoglobin to carry oxygen to the bodily tissues. Erythrocyte, which takes on a biconcave disc that has no nucleus, is flexible and changeful. Erythrocyte is so sensitive to the environment that the shape of cell goes crimpy, even acanthoid. A laser tweezers Raman spectroscopy (LTRS) setup was used to trap single erythrocyte from healthy donors and patients with thalassemia and to collect the Raman scattering of trapping cell. Normal shape, crimpy erythrocytes and acanthoid erythrocytes were tested, and the averaged spectra, and principal component analysis (PCA) which detailed the spectral difference and the change of hemoglobin, were used to evaluate the effects of different cell shape on the spectral distinguishing of erythrocyte. The results reveal that in normal physiological environment the change in cell shape does not effect the spectral distinguishing of abnormal erythrocyte.

[FTIR-HATR to identify beta-thalassemia and its mechanism study].

Fourier transform infrared spectroscopy (FTIR) associated with horizontal attenuated total reflectance (HATR) was firstly used to diagnose beta-thalassemia patients. With excellent linearity (r = 0.997) and reproducibility (RSD < 4%), FTIR-HATR shows an order-of-magnitude increase in IR absorption bands over the single-path transmission FTIR Based on above, spectra from 37 patients' and 68 health samples indicated several observable differences in IR vibrational spectra of the Hb lysates between the beta-thalassemia major patients and control: (1)Because of decreasing hemoglobin, the peak intensities are obviously lower in beta-thalassemia group that is consistent with index from routine hemoglobin diagnosis. (2) In 1750-1500 cm(-1) region, slight decrease at 1652 cm(-1) (alpha-helix), 1638 and 1 628 cm(-1) bands but mild increase at 1682 cm(-1) all demonstrate structure changes by both Fourier self-deconvolution and second derivative spectra. (3) More importantly, difference spectra substantially demonstrate decreased intensities at 1440, 1453, 1479 cm(-1) bands arising from CH2/CH3 deformation vibration of phospholipids but increased intensities at 1150 cm(-1) band originating from C--O stretching vibration of carbohydrate and at 1081 and 1053 cm(-1) bands that are attributed to 2,3-diphosphoglycerate (DPG) in beta-thalassemia major group. Statistical analysis demonstrates significant difference of DPG/phospholipids ratio between two groups. All the samples can be 100% correctly classified into groups on the basis of this ratio. These finding could help understand possible mechanism for diagnosing thalassemia. It makes large-scale screening of thalassemia by FTIR a possibility.

[Study of Raman spectra of single carcinoma of nasopharynx cell].

The Raman spectra from carcinoma of nasopharynx cell lines (CNE2) and normal airway epithelial cell lines (HBE) were investigated using a laser tweezers Raman spectroscopy (LTRS). The Raman scattering measurements were obtained from three different places in every single cell. Visual inspection of the spectra shows that the differences observed in spectra of the cancer cells and normal cells are obvious. The peak ratio I1 304/I1 336 is 1.05 for the normal cell and 1.22 for the cancer cell. Using a combination of principal component analysis (PCA) and discriminant function analysis (DFA), the authors are able to predict cancer cells, and normal cells and the DFA is better for single Raman spectrum. The sampling locations did not seriously affect the result of PCA and DFA. PCA and DFA also show that the uniformity of normal cells is better than that of cancer cells. The results indicate that the Raman spectra may offer the experimental basis for colorectal cancer diagnosis.

[Raman micro-spectroscopy of single blood platelets].

The authors collected the Raman spectra of single blood platelets of human, pig, rat and rabbit suspended in saline so-lution by using a laser tweezers Raman spectroscopy (LTRS) setup. A single platelet cell was trapped in the focus of a near-infrared laser beam at 785 nm and the excited Raman spectrum was acquired. For each species, the Raman spectra of up to 20 platelet cells were acquired and were used to perform a principal components analysis (PCA) or a discriminate function analysis (DFA). The average Raman spectra indicate that the vibration bands at 1 524 and 1 157 cm(-1) of human platelets are obviously different from those of the platelets from pig, rat and rabbit. The Raman intensities at 1 157 and 1 524 cm(-1) bands are significantly high for human platelets. The ratio I1 157 / I1 003 of human platelets was 0.795, but those of pig, rat and rabbit were 0.532, 0.502 and 0.485, respectively. In addition, the platelets from four different species can be discriminated with multivariate analysis. These findings demonstrate that the LTRS, combined with multivariate analysis, could be used to rapidly discriminate platelets from various species and may find valuable application in rapid sensing of biochemical changes in a single cell.