An evaluation of enzymic and autolytic assays for cyanide in cassava (Manihot esculenta, crantz)

Abstract The total cyanide contents of cassava parenchymal tissue (peeled roots), cassava cortex (peel) and cassava leaves were evaluated by autolytic and enzymic assays. Autolysis of parenchymal tissue was studied under different conditions of pH, time and temperature and addition of exogenous enzyme and antibiotic. Optimal conditions were determined to be 24 h at 37°C in acetate buffer (0.1M; pH 5.5) with 0.1 mg ml−1 chloramphenicol. Total cyanide contents similar (about 90%) to those obtained by enzymic assay could be achieved only by the use of small sample sizes: < 1 g of parenchymal tissue, < 0.3 g of cortex and < 0.1 g of leaf tissue. This caused sampling problems because of the presence of cyanide gradients in cassava tissues, which could only be resolved by tissue homogenisation prior to analysis. A study using the enzymic assay of cyanide stability in such homogenates, at different pH values and temperatures, has indicated that subsampling must be done within 15 min so as to prevent appreciable losses in the measured cyanide contents. The rate of loss of total cyanide was found to depend not only on the proportion of total cyanide which is non‐glucosidic (free), but also on the proportion of the non‐glucosidic cyanide present as cyanohydrins. The implications with regard to residual cyanide contents on cassava processing and the advantages of the enzymic assay over autolytic methods are discussed.