Starch is the primary mechanism of energy storage in plants, is the single most important source of human food energy and is commonly used as an animal feed. It also has many significant industrial uses in adhesives, packaging materials, pharmaceutical excipients and in the production of textiles and papers. Due to its ubiquity, both as a food and as a food additive, starch structure has been examined in relation to its physical and nutritional properties extensively. There has also been significant success in understanding the biochemical/biosynthetic pathways of aspects of its production and how these affect individual starch structures. The various structural features of starch can be divided into several structural levels, the first four of which are most relevant to this research. Level 1 starch structure is the chain length distribution of starch. Level 2 structure relates to the whole individual molecule, such as whole molecule size and branching pattern. Level 3 structure is the aggregated structures of the molecules such as double helices, crystallites and crystalline-amorphous lamellae. Level 4 starch structures relate to the compartmentalisation of the granule into the hilum and semi-crystalline and amorphous growth rings. This project aims to increase understanding of the influence of the molecular structures (levels 1 and 2) of starch upon the larger scale aggregated structures (level 3) that are present in native starch. The crystalline structure as well as the crystalline-amorphous (C-A) lamellae are known to be produced by the amylopectin molecule while the amylose molecules modifies the size and stability of these structures. Thus, waxy starches (those without amylose) can be used to explore the influence of molecular properties of amylopectin on native starch structure. These results can be used to give an indication about which native starch structures are influenced by the underlying molecular structures and which are mediated by biological changes during synthesis that do not affect the molecular structure. The chain length distribution (CLD) of the amylopectin is observed via debranching followed by size characterization, and parameterised by two different techniques to describe the changes in the branch pattern of the different waxy starches. One technique is the empirically based division iii of chains, the other is based upon biosynthetic modelling. Enzymatic techniques are used to probe the assembly of the branches that produce the whole amylopectin molecule. The enzymes degrade the exterior starch chains allowing the observation of the core of the amylopectin molecule which contains all of the branching. These molecular parameters are then to be compared to the aggregated native structures of starch which are most likely to be related to the molecular structure parameters. The first of these aggregated structures are the crystalline properties of the starches, with an observation of the proportion of crystallinity, the proportion of double helices which make up these crystallites as well as the thermal stability of both the double helices and crystallites. The majority of the eleven waxy starches used, which are from a variety of botanical sources, are A-type starches with a single B-type starch which is used to observe whether B-type starches follow the same trends as A-type starches.
The results showed that there was a strong correlation between the molecular parameters of amylopectin, the aggregated starch parameters and the crystalline-amorphous lamellae of native starch. The amylopectin molecule chains that were branched were more likely to be the shoulder length chains in the CLD; increases in these branched shoulder length chains increased the size of the C-A lamellae. No other branch parameters of amylopectin correlated with the CLD of the amylopectin. This indicates that the commonly held belief, that A-chains are more likely to have a lower degree of polymerisation (DP) than B-chains, is questionable. The increase in lamellar size was linked to an increase in the amorphous region of the lamellae which contains the majority of the branch points. The crystalline region of the lamellae increased in size with an increase in the number of longer chains in the first, approximately linear, region of the CLD. The number of longer chains in the linear region increases the likelihood that the starch may produce longer helices, which leads to an increase the crystalline region of the lamellae. Increasing proportions of DP 13-24 and DP 25-36 chains were observed to correlate with an increase in the size of the lamellae while the DP 6-12 chains correlated with a decrease. The DP 13-24 and DP 25-36 chains are likely related to the approximately linear crystalline regions, as iv above, while the DP 6-12 chains decrease the lamellar size by making the production of longer helices unlikely. An increase in the degree of branching of starch was linked to an increase in the number of crystalline defects within starch that is removed with annealing. This was related to the smaller size of the internal chains of the amylopectin molecule which decrease the mobility of the helices during synthesis preventing their correct alignment, which is ameliorated by the annealing process.
Speculation about the biosynthesis of starch in light of these discoveries is given with a focus upon starch branching enzyme 1 and the role of the rate of starch synthesis. Starch branching enzyme 1 seems to be the enzyme that specifically produces the shoulder length CLD chains. This results in an increase in the proportion of the B-chains and the placement of these chains in the amorphous lamellar region. There is an apparent lack of difference in the amylopectin molecules in different areas of the granule. The role of the changing rate of starch synthesis in the production of the amorphous and semi-crystalline growth rings is suggested as a reason for the similarity in molecular structure despite the large differences in granular structure.