High-throughput Analysis Of Physical And Chemical Properties Of Reduced Glutathione Tablets With Near Infrared Spectroscopy

This paper establishes a high-throughput and high selective method todetermine the impurity named oxidized glutathione (GSSG) and radialtensile strength (RTS) of reduced glutathione (GSH) tablets based on nearinfrared (NIR) spectroscopy and partial least squares (PLS). In order tobuild and evaluate the calibration models, the NIR diffuse reflectancespectra (DRS) and transmittance spectra (TS) for330GSH tablets wereaccurately measured by using the optimized parameter values. For analyzingGSSG or RTS of GSH tablets, the NIR-DRS or NIR-TS were selected,subdivided reasonably into calibration and prediction sets, and processedappropriately with chemometric techniques. After selecting spectralsub-ranges and neglecting spectrum outliers, the PLS calibration modelswere built and the factor numbers were optimized. Then, the PLS modelswere evaluated by the root mean square errors of calibration (RMSEC),cross-validation (RMSECV) and prediction (RMSEP), and by thecorrelation coefficients of calibration (Rc) and prediction (Rp). The results indicate that the proposed models have good performances. It is thus clearthat the NIR-PLS can simultaneously, selectively, nondestructively andrapidly analyze the GSSG and RTS of GSH tablets, although the contents ofGSSG impurity were quite low while those of GSH active pharmaceuticalingredient (API) quite high. This strategy can be an important complementto the common NIR methods used in the on-line analysis of API inpharmaceutical preparations. And this work expands the NIR applications inthe high-throughput and extraordinarily selective analysis.OBJECTIVES:1. Establish the rapid determination of GSSG, the main impurity ofGSH tablets, by NIR-DRS, to quality control.2. Establish the rapid determination of RTS, an important physicalparameter of GSH tablets, by NIR-DRS, to quality control.METHODS:1. Measurement of spectraOne side of each cylindrical flat-faced tablet was placed on thecircular detection window of integrating sphere module of FT-NIR analyzer.The position of the tablet was adjusted to be concentric with the detectionwindow by the sample holder. The NIR-DRS and NIR-TS were acquiredsimultaneously. And then, the other side of this tablet was detected in thesame way. All spectra were recorded under optimal conditions: a resolutionof8cm-1and64scans. The scan range for NIR-DRS was10,000–4000 cm-1, one for NIR-TS10,000–6000cm-1. The air temperature was about25, and the humidity was (55±5)%RH. The measurement background,the gold-plated inner wall of the integrating sphere, was detected toautomatically deduct the interferences of water vapor and CO2in air.2. Determination of GSSGThe reference values of GSSG contents were determined by HPLC andcalculated using the external standard method. The separation wasperformed on a Phenomenex Gemini C18column (250mm4.6mm,5μm),whose temperature was maintained at30, by using methanol-buffer(water containing10mmol/L sodium1-heptanesulfonate and50mmol/Lsodium dihydrogen phosphate, pH3.0)(4:96, v/v) as the mobile phase at aflow rate of1.0mL/min. The GSSG was detected at210nm. The injectionvolume was20μL. And the HPLC method was validated according to thevalidation of analytical procedures in ICH-Q2(R1).3. Determination of RTSThe reference values of RTS were calculated by formula (1) after thehardness was measured with a tablet hardness tester, both diameter andthickness with a caliper.4. Build and evaluate the PLS prediction model of GSSG based onNIR-DRSThe NIR-DRS or NIR-TS were selected, subdivided reasonably intocalibration and prediction sets, and processed appropriately with chemometric techniques. After selecting spectral sub-ranges and neglectingspectrum outliers, the PLS calibration models were built and the factornumbers were optimized. Then, the PLS models were evaluated by the rootmean square errors of calibration (RMSEC), cross-validation (RMSECV)and prediction (RMSEP), and by the correlation coefficients of calibration(Rc) and prediction (Rp).5. Build and evaluate the PLS prediction model of RTS based onNIR-DRSThe NIR-DRS or NIR-TS were selected, subdivided reasonably intocalibration and prediction sets, and processed appropriately withchemometric techniques. After selecting spectral sub-ranges and neglectingspectrum outliers, the PLS calibration models were built and the factornumbers were optimized. Then, the PLS models were evaluated by the rootmean square errors of calibration (RMSEC), cross-validation (RMSECV)and prediction (RMSEP), and by the correlation coefficients of calibration(Rc) and prediction (Rp).RESULTS:1. Measurement of spectra660NIR-DRS and660NIR-TS for330samples of GSH tablets wereacquired.2. Determination of GSSGThe regression equation y=16.8520x-0.8236(R2=1.0000) reveals a good linear relationship between the peak area and GSSG concentration inthe test range of0.44–16.34μg/mL. The average recovery is100.3%with1.3%RSD (n=9).The GSSG reference values of186samples were distributed over therange of15.81–22.20mg/g.3. Determination of RTSThe RTS reference values of318samples were distributed overtherange of270.76–1145.14kPa.4. Model evaluation of GSSGFor the optimal NIR-DRS model, of RMSEC, RMSECV and RMSEPare separately0.575,0.729and0.607mg/g, and the values of Rcand Rp0.9173and0.9182. The regression equation is y=0.8614x+2.6982,represents a good linear relationship between the prediction and referencevalues of the prediction samples.5. Model evaluation of RTSFor the optimal NIR-DRS model, of RMSEC, RMSECV and RMSEPare separately58.2,61.3and69.0kPa, and the values of Rcand Rp0.9393and0.9151. The regression equation is y=0.8211x+113.3224, represents agood linear relationship between the prediction and reference values of theprediction samples.CONCLUSIONS:This paper establishes a high selective NIR-PLS method to determine the GSSG impurity and RTS of GSH tablets, simultaneously,nondestructively and rapidly. The built models have low values of RMSEC,RMSECV and RMSEP, and high ones of Rcand Rp. It was demonstrated thatthe NIR-PLS could be used to simultaneously and selectively measure thelow content of the impurity as well as the pharmaceutical property of thetablets. The proposed strategy might be used for monitoring the tabletmanufacturing process.