Research On Blood Protein And Erythrocyte Membrane From Type Ⅱ Diabetes Patients Based On Raman Spectroscopy

Raman spectroscopy (i.e. fingerprint spectroscopy) is capable of probing structural information of various biomolecules and is widely applied in biomedical research field. In this study, Raman spectroscopy was employed to detect plasma albumin, hemoglobin and erythrocyte membrane variation in type II diabetic development without using exogenous reagents. With the aid of multivariate statistical method, we tried to explore a simple, high-efficiency and label-free detection method for improving the early and accurate screening of type Ⅱ diabetes. The main research results are as follows:1. Surface-enhanced Raman scattering (SERS) spectroscopy combined with membrane electrophoresis (ME) was employed to analyze type II diabetic plasma albumin. Albumins were first purified from diabetic (n=35) and healthy (n=45) plasma by ME and then mixed with silver nanoparticles to perform SERS spectral analysis. Based on principal component analysis and linear discriminant analysis (PCA-LDA), the diagnostic accuracy of albumin SERS spectra for diabetes detection is 98.8%, which is much higher than that of untreated plasma SERS spectra (88.8%).2. SERS spectroscopy was firstly employed to detect hemoglobin (Hb) variation in type II diabetic development without using exogenous reagents. High-quality SERS spectra were obtained from blood Hb samples of 49 diabetic patients and 40 healthy volunteers. Then PCA-LDA diagnostic algorithm was employed to analyze and classify the SERS spectra acquired from diabetic and healthy Hb, yielding the diagnostic accuracy of 95.5%. Furthermore, Hb SERS spectra were used to distinguish diabetes without complication (n=22) from diabetes with complication (n=27), and the diagnostic accuracy is 91.8%. Ultimately, partial least-squares (PLS) regression algorithm was employed to predict diabetes, and the prediction accuracy is 90%.3. High-wavenumber Raman spectroscopy (HWRS) was employed to detect erythrocyte membrane (EM) variation in type II diabetic development. Raman spectra in high-wavenumber and low-wavenumber regions were obtained from EM samples of 45 diabetic patients and 38 healthy volunteers, respectively. Based on PCA-LDA statistical method, the diagnostic accuracy of HW region for diabetes detection is 98.8%, which is much higher than that of low-wavenumber (LW) region (82.9%).This exploratory work suggests that albumin SERS spectra, Hb SERS spectra and EM HWRS in combination with multivariate statistical analysis have great potential for the simple, high-efficiency, early and accurate mass screening of type II diabetes.