Characterization of structural changes occurring in insulin at different time intervals at room temperature by surface-enhanced Raman spectroscopy

Insulin storage above the temperature recommended by food and drug administration (FDA) causes decrease in its functional efficacy due to degradation and aggregation of its protein based active pharmaceutical ingredient (API) that results poor glycemic control in diabetic patients. The aggregation of protein causes serious neurodegenerative diseases such as type-2 diabetes, Huntington disease, Parkinson's disease, and Alzheimer's disease. Surface-enhanced Raman spectroscopy (SERS) has been employed for the denaturation study of many proteins at the temperature above the recommendations of food and drug administration (FDA) (above 30 °C) which indicates potential of technique for such studies. SERS along with multivariate discriminating analysis techniques-based analysis of degradation of liquid pharmaceutical insulin protein after regular intervals of time at room temperature to analyze the structural changes in this protein during the storage of insulin pharmaceutical at room temperature. Silver nanoparticles (Ag-NPs) prepared by chemical reduction method are used as SERS active substrate for the surface enhancement of the insulin spectral signal. SERS spectral measurements of insulin were collected from eight different samples of insulin in the time range of 7 pm to 7 am first at fridge temperature (5 °C), second after half hour and next six with the time difference of 2 h each time at room temperature. The acquired SERS spectral data was preprocessed and analyzed. SERS structural transformations detection and discrimination potential in insulin was further confirmed by applying multivariate discriminating analysis techniques including principal component analysis (PCA) and Partial least square regression analysis (PLSR). SERS significantly detects the structural changes produced in insulin even after 2 h of insulin placement at room temperature. PCA successfully differentiates the insulin spectral data obtained after regular intervals of time according to PC-1 (77 %) explained variance. Application of PLSR model provides quantitative confirmation of SERS efficiency, by providing insulin data regression coefficients plot, efficient prediction of time with calibration data set having 0.77 mean square absolute error of calibration (RMSAEC), validation data set with 0.80 mean square absolute error of prediction (RMSAEP) and 0.98 coefficient of determination ( R 2 ) for both calibration and validation data set. SERS is proved as a highly sensitive and discriminating technique to detect and discriminate insulin structural changes after regular intervals of time at room temperature.