Accounting for co-extractable compounds (blank correction) in spectrophotometric measurement of extractable and total-bound proanthocyanidin in Leucaena spp

Methods to account for the spectral interference of co-extractable compounds (blank correction) in the spectrophotometric analysis of both extractable and bound proanthocyanidin (PA) using the proanthocyanidin (butanol/HCl) assay were evaluated. Crude extractable and bound PA sample matrices of PA-free Leucaena magnifica were used. Extractable PA blanks generated in heated 95% butanol/5% H2O reagent underestimated the optical density (absorbance) of co-extractable compounds by 24% (P<0.01), whereas unheated 95% butanol/5% HCl blanks, incubated at room temperature, accurately measured the absorbance of the background matrix (P<0.01). Current procedures that estimate bound PA concentrations using the proanthocyanidin assay produce intensely coloured background matrices. Recovery measurements from total-bound PA extracts spiked with 1071 and 2142 mug anthocyanidin per tube indicated that existing analytical procedures that do not account for the spectral interference of co-extractable compounds overestimated (P<0.01) bound PA concentrations by 69 and 38% respectively. An innovative technique that generated an internal correction factor for each sample, using wavelength-scanning spectrophotometry and nonlinear curve-fitting computer software, was developed. This procedure recovered 100% of added anthocyanidins from bound PA matrices. (C) 2002 Society of Chemical Industry.