2.4. High Performance Liquid Chromatography Instrument, Calibration and Validation

HPLC system was equipped with a Refractive Index Detector (RID) and the analysis of carbohydrates was performed on a Rezex™ RSO-Oligosaccharide Ag 4%, LC Column 200 × 10 mm, preceded by a Rezex™ RSO-Oligosaccharide Ag 4%, LC Guard Column 60 x 10 mm (Phenomenex^®, Torrance, California, CA, USA). Milli-Q^® water was used as mobile phase, at a flow rate of 0.3 mL/min. The column was maintained in a thermostatic oven at a constant temperature of 65 °C. Injections of standards and samples (5 μL) were made after filtering through 0.22 μm filters for aqueous solutions (Nylon syringe filters, 0.22 µm diameter Branchia, Labbox, Barcelona, Spain). Ultrapure water (Resistivity 18.2 MΩ cm at 25 °C; Milli-Q Integral 5 Water Purification System from Millipore, Merck KGaA, Darmstadt, Germany) was used for preparing dilutions, as well as, mobile phases for chromatography analysis.

Thirteen carbohydrate standards with different average molecular weight (MW) were used for retention time (RT) and calibration curve calculation: Pullulan 100 (100 kDa), Pullulan 50 (50 kDa), Pullullan 20 (20 kDa) and Pullulan 10 (10 kDa) from a Shodex pullulan standard P-82 kit were obtained from Waters, Madrid, Spain. Inulin (5.94 kDa), Verbascose (0.83 kDa), Stachyose (0.67 kDa), Cellotriose (0.50 kDa), Raffinose (0.50 kDa), Glucose (0.18 kDa) and Fructose (0.18 kDa) were obtained from Sigma, (Alcobendas, Madrid, Spain). Cellobiose (0.34 kDa) and Sucrose (0.34 kDa) were obtained from Merck (Darmstadt, Germany). Each standard was injected in triplicate at diverse concentrations (0.125–1 mg/mL) into the HPLC-RID. Linear regression standard curves were calculated using, the standard concentration versus the area obtained from the RID (in Nriu *s) and, the linearity (R²), were assessed. Furthermore, equations for high molecular weight carbohydrates (HMWC) and for oligosaccharides molecular weight estimations were calculated. These equations were used estimating the average molecular weight of standards for validation.

Precision of the HPLC-RID method was assessed by intra- and inter-day repeatability assay. Results were expressed as relative standard deviation (RSD) percent of the retention time. Moreover, the limits of detection (LOD) and quantitation (LOQ) were calculated according to the signal-to-noise ratio from standard solutions with known low concentrations (10 µg/mL) and blank solutions. For LOD calculation, signal-to-noise ratio equals 3. For LOQ, signal-to-noise ratio equals 10 [30].