br Fig Characterization of IR
Fig. 1. Characterization of IR spectra of saliva (A – determination of intensity and peak area, B – construction of a calibration curve for the determination of lipids; C
– an example of the IR spectrum of saliva before extraction, D – an example of the IR spectrum of saliva after extraction).
for lipid determination was preliminarily constructed, and a glycerin solution with a concentration of 2.29 mmol/l (Vector-Best, Novosibirsk, Russia) was used as a standard. Extraction was carried out according to the scheme described above, using 25, 50, 100, and 150 μl of standard (solutions 1–4, respectively, Fig. 1B). IR spectra before and after lipid extraction are shown in Fig.1C and D.
2.6. Biochemical analysis of saliva samples
All saliva samples were evaluated for the level of acyl hydroper-oxides . To do that, 0.200 mL of saliva were placed in a separate tube, where 4 mL of heptane-isopropanol mixture (1:1) were added, and the contents were shaken for 10–15 min. Then, 1 mL of HCl solu-tion (рH 2) and 2 mL of heptane were added, the mixture was then intensively shaken and allowed to stand for 20–25 min. Out of this mixture, split into phases, the upper heptane layer was taken to mea-sure the concentration of acyl hydroperoxides by degree of light ab-sorption at the wavelengths of 220, 232, 278, and 400 nm. It should be mentioned that light Minocycline HCl at the wavelength of 220 nm (E220) corresponds to absorption of isolated double bonds. Light absorption at the wavelength of 232 nm (E232) corresponds to absorption of diene conjugates; light absorption at the wavelength of 278 nm (E278) cor-responds to absorption of triene conjugates; absorption at the wave-length of 400 nm (E400) corresponds to absorption of Schiﬀ’s bases. To increase precision of measurements and eliminate indicated errors, ratios of E232/E220, E278/E220, and E400/E220 were calculated; le-vels of lipid peroxidation products were expressed with relative units, namely, diene conjugates as E232/E220, triene conjugates as Е278/ Е220, and Schiﬀ’s bases as Е400/Е220.
The method to measure malondialdehyde (MDA) is based on the fact that at high temperatures in an acidic medium, MDA reacts with thiobarbituric acid, making the coloured pink trimethine complex with the highest absorption at 535 nm. 2 mL of the distilled water and 1 mL
of 0.6% TBA in glacial acetic acid were added to 0.300 mL of saliva. The solution was boiled for 30 min, cooled, then 1 mL of 5% potassium hydroxide and 2 mL of isopropanol were added. The solution was centrifuged at 6000 rpm for 20 min, and measured at 535 and 580 nm versus the control sample, containing the water instead of the saliva. All tests were carried out in duplicate or in triplicate if the results could not be reproduced.
2.7. Statistical methods
The statistical analysis of the obtained data was performed by means of Statistica 10.0 (StatSoft) program and R package (version 3.2.3) using the non-parametric method and Wilcoxon criterion in de-pendent groups and Mann-Whitney U-criterion in independent groups. Questionable results (outliers) were rejected using the Grubbs test. The sample was described by calculating the median (Me) and interquartile range in the form of the 25th and 75th percentile [LQ; UQ]. The dif-ferences were considered to be statistically significant at p < 0.05. The statistical interrelationships were studied by means of the nonpara-metric correlation analysis using Spearman correlation coeﬃcients (R).
Intensity (h) and area (s) of absorption bands in the IR spectra of saliva of patients with endometrial and ovarian cancer and the control group.
Endometrial Cancer, Ovarian Cancer, Control,
Note. p - statistically significant diﬀerences compared with the control group.
Best, Novosibirsk) showed that the error of determination using IR spectroscopy does not exceed 10%. Statistically significant diﬀerences were revealed only for the group of patients with non-malignant ovarian tumors and the control group (p = 0.0253).
Nevertheless, the intensity of the absorption bands of lipids after the extraction is significantly increased, which in turn provides greater accuracy in determining the characteristics of the spectra (peak height, peak area). This can provide valuable information about changes in the structure of lipids that appear on the background of various pathologies of the human body. Thus, it is shown that the intensity and area of all absorption bands in the spectra of patients with cancer are lower than for the control group (Table 1). At the same time, the characteristics of the absorption bands of 1396 and 1458 cm−1 are more reduced with endometrial cancer, while the characteristics of the absorption bands of 2853 and 2923 cm−1 decrease with ovarian cancer.