Abstract
Starch provides approximately half of humans' food energy, and its structural features influence human health. The most important structural feature is the chain length distribution (CLD), which affects properties such as the digestibility of starch-containing foods. The rate of digestion of such foods has a strong correlation with the prevalence and treatment of diseases such as diabetes, cardiovascular disease and obesity. Starch CLDs can be divided into multiple regions of degrees of polymerization, wherein the CLD in a given region is predominantly, but not exclusively, formed by a particular set of starch biosynthesis enzymes: starch synthases, starch branching enzymes and debranching enzymes. Biosynthesis-based models have been developed relating the ratios of the various enzyme activities in each set to the CLD component produced by that set. Fitting the observed CLDs to these models yields a small number of biosynthesis-related parameters, which, taken together, describe the entire CLD. This review highlights how CLDs can be measured and how the model-based parameters obtained from fitting these distributions are related to the properties of starch-based foods significant for health, and it considers how this knowledge could be used to develop plant varieties to provide foods with improved properties.