Some starches, notably the high-amylose types, contain IM. This material has gained interest because it appears to contribute to the fraction of resistant starch (Li et al., 2008), which is considered beneficial for human health. Besides mutant maize varieties (Gerard et al., 2001, 2002; Shi et al., 1998; Wang et al., 1993a; Boyer and Liu, 1985; Baba and Arai, 1984; Vamadevan et al., 2014; Li et al., 2008; Ikawa et al., 1978; Inouchi et al., 1983; Perera et al., 2001; Yun and Matheson, 1992) and wrinkled peas (Banks et al., 1974; Matheson and Welsh, 1988; Lloyd et al., 1996; Colonna and Mercier, 1984; Bertoft et al., 1993a), IM was also described in high-amylose varieties of barley (Banks et al., 1974) and potato (Schwall et al., 2000). Moreover, IM was found in nonmutant (normal) starches from potatoes (Yoon and Lim, 2003), maize (Klucinec and Thompson, 1998), oats (Banks and Greenwood, 1967; Yoon and Lim, 2003; Paton, 1979; Wang and White, 1994b), wheat (Banks and Greenwood, 1967; Dais and Perlin, 1982), rye, and barley (Banks and Greenwood, 1967). The nature of this material is somewhat confusing and might be unique for each type of starch (Banks et al., 1974). The method of fractionation of the starch could also influence the result, especially when IM is isolated together with either the amylose or the amylopectin fraction. In such cases, IM might even remain unknown to the investigator and influence the structural analysis of the other components. In several cases, the measured amylose content of high-amylose starches by iodine binding is largely overestimated due to IM with a high-IA or amylopectin of longer CL than normal (Ge´rard et al., 2001; Boyer et al., 1980; Perera et al., 2001; Wang et al., 1993b), or both (Shi et al., 1998; Wang et al., 1993a,b; Baba and Arai, 1984; Wang and White, 1994b). Because IM is structurally related to either amylose or amylopectin, it is analyzed by the same methods.
Banks and Greenwood (1967) described a method adopted to cereal starches, in which an amylose-thymol complex was separated from the amylopectin component. The complex was reprecipitated in butanol into amylose and a soluble fraction containing anomalous amylose and/or amylopectin, the latter with unusually long CL. Adkins and Greenwood (1969), working with maize starches, precipitated amylose in 1-butanol and from the supernatant (the amylopectin fraction) they precipitated a glucan in complex with iodine. The amount of the complex increased with an increasing amylose content of the maize and was characterized as short-chain amylose material. Wang et al. (1993b) also found IM included in the supernatant fraction together with amylopectin that they isolated from a range of mutant maize samples. IM was fractionated by GPC on Sepharose CL 2B and found to be a branched component smaller than amylopectin (Li et al., 2008). It possessed similar types of chains as in amylopectin, but in different proportions depending on the type of mutation (Wang et al., 1993b). Klucinec and Thompson (1998) precipitated amylose together with an intermediate fraction from the amylopectin in an aqueous mixture of 6% 1-butanol and 6% isoamyl alcohol. The amylose was then reprecipitated in 1-butanol, whereas IM remained in the supernatant. They found that IM in normal maize had rather similar types of chains as amylopectin, but in high-amylose starches the composition changed, and IM possessed a large group of long chains. The apparently true amylopectin component also contained increased amounts of the long chains (Klucinec and Thompson, 1998). More, and/or longer, long chains is a typical feature associated with amylopectin in maize starches containing the amylose-extender mutation (Boyer and Liu, 1985; Baba and Arai, 1984; Vamadevan et al., 2014; Ikawa et al., 1978; Inouchi et al., 1983; Wang et al., 1993b; Liu et al., 2012). Similar results have been found for high amylose-containing rice starches (Asaoka et al., 1986; Nishi et al., 2001; Kubo et al., 2010). Klucinec and Thompson (1998) concluded that IM, like amylopectin, is branched, though its structural features results in altered physical behavior, as shown by its precipitation in the 1-butanol-isoamyl alcohol mixture. In a double (dull:waxy) and triple mutant (amylose-extender:dull:waxy) maize starch, an IM component was found in high concentration (40% and 80%, respectively; Bertoft et al., 2000). The material, which contained only slightly altered chain distribution, was apparently more resistant to α-amylase attack than the normal amylopectin and the behavior of the α-dextrins in methanol was different. Bertoft et al. (2000) suggested that this IM possessed a more regularly branched structure that prevented the action of the α-amylase and rendered the molecule the altered behavior.
Colonna and Mercier (1984) isolated a very low-molecular weight component from wrinkled pea starch (also a high-amylose type starch). This IM was branched and possessed similar types of chains as the amylopectin component, but the IA was high and the S:L chain ratio low. Bertoft et al. (1993a) size-fractionated the IM and found that the proportion of the long chains slightly increased with decreasing molecular weight. In fact, the molecular weight of the small IM resembled that of the clusters of the amylopectin component and was suggested to be composed of small, clusterlike structures interconnected by the long chains, thus increasing the proportion of these chains (Bertoft et al., 1993a). The IM of wrinkled pea has also been described as a mixture of very short, linear amylose chains and branched, either normal (Banks et al., 1974) or long-chained amylopectin (Boyer et al., 1980). Biliaderis et al. (1981) also reported long-chained materials and Matheson (1990) found that both ECL and ICL were larger than in the normal amylopectin of smooth pea starch. The different opinions regarding the nature of IM might reflect differences among varieties of wrinkled peas.
In the developing endosperm of wheat kernel starch, material with sizes corresponding to amylose was considered as IM due to its intermediate iodinestaining properties (Waduge et al., 2014). At maturity the material was not present anymore, which suggested that it was a precursor molecule to either amylose or amylopectin. In fact, it was suggested that the linear structure of amylose and the backbone structure of amylopectin are related (Bertoft, 2013). The linear amylose molecule is an extreme example of a backbone with comparatively low DP and high-IA. Branched amylose has shorter chains that combine into a backbone and few side branches to the backbone, together having higher DP and lower IA than linear amylose. Amylopectin is the other extreme example with a backbone consisting of a large number of comparatively short chains and extensive side branches, which render the molecule a high-DP and low-IA. In this scenario, IM possess a structure in between the extremes with an intermediate backbone, DP, and iodine-staining properties.