The time course of the appearance of pancreatic amyloid mirrors the appearance of clinical diabetes (Ohsawa et al., 1992). A relatively restricted number of mammalian species exhibit a propensity to form amyloid in pancreatic islets; these are the same species that are susceptible to type 2 diabetes. In addition to humans (Westermark, 1972) and macaque monkeys (Clark et al., 1991; de Koning et al., 1993; Howard, 1988), islet amyloid is found in domestic cats (Betsholtz et al., 1990; Westermark et al., 1987b) as well as in tigers, lions, lynx, raccoons (Jakob, 1970), and cougars (Johnson et al., 1991b). It is not found in islets of dogs or other members of the Canidae (wolf, jackal, fox) (Jakob, 1970). Except for Octodon degu, which is a special case (Hellman et al., 1990), amyloid is not found in the islets of rodents. However, human islets transplanted into mice form amyloid (Westermark etal., 1995), suggesting that it is a species-specific characteristic of the peptide itself that leads to amyloid formation (Ashburn and Lansbury, 1993). This idea is supported by the observation that transgenic mice over-expressing human amylin form amyloid (Soeller et al., 1996b; Verchere et al., 1996), but mice overexpressing mouse amylin do not (Soeller et al., 1996b).
Based upon analysis of sequence divergence (Betsholtz et al., 1989) and propensity of subpeptides to form fibrils (Westermark et al., 1990), residues 20-29, especially 24-29, which are common to humans, cats, and raccoons, were predicted by some authors to be amyloidogenic (Johnson et al., 1992; Jordan et al., 1994; Westermark et al., 1990). This prediction does not, however, account for the absence of islet amyloid in dogs (except in insulinoma; Jordan et al., 1990), since dog and cat amylin are identical from residues 20-37. It appears that more than just an amyloidogenic molecule is required, and that some stimulus, perhaps associated with ¿-cell hypersecretion, may be necessary. For example, mice containing the human amylin transgene usually do not spontaneously develop amyloid (Verchere et al., 1997) but can be induced to do so by such maneuvers as feeding a high-fat diet (Verchere et al., 1996), crossing with an insulin-resistant strain (Hull et al., 2003; Soeller et al., 1996a), induction of insulin resistance with dexamethasone and growth hormone (Couce et al., 1996), or oophorectomy (Kahn et al., 2000). These observations tend to support the idea that ¿-cell hypersecretion (of an amyloidogenic molecular species) promotes amyloid formation. The resistance of human amylin transgenic mice to amyloid formation when they simultaneously carry ¿-cell glucokinase deficiency (which limits ¿-cell secretion) is also consistent with this idea. Some workers in the field conclude, however, that amyloidogenicity involves more than simple ¿-cell hypersecretion (Marzban et al., 2003) and may include ¿-cell ''strain,'' in which secretory rate exceeds prohormone convertase capacity, resulting in increased prevalence of prohormone forms of amylin (de Koning et al., 1999) as well as insulin (MacNamara et al., 2000).
The role of amyloid formation, mechanical disruption, and possible cytotoxic effects of amyloid in the pathogenesis of islet secretory failure and diabetes has been covered in numerous reviews (Artozqui et al., 1993; Betsholtz et al., 1993; Butler, 1996; Clark, 1992; Clark et al., 1991, 1995, 1996a,b; Hansen, 1996; Johnson et al., 1988, 1989, 1991a,c; Kamaeva, 1993; O'Brien et al., 1993a,b; Porte et al., 1991; Weir and Bonner-Weir, 1996; Westermark, 1994; Westermark and Johnson, 1988; Westermark et al., 1988, 1991, 1992; Wolffen-Buttel and Van Haeften, 1993) and is not covered in further detail here. These reviews, which constitute ^30% of the ongoing literature, do not generally address a functional (receptor mediated, hormonal) role of amylin.
Was this article helpful?