This news story linking amylin build up to diabetes, based on new research conducted jointly in Auckland, New Zealand and Manchester England, makes the case reasonably clearly:
What does it mean?
Diabetes is defined as the loss of beta-cells, so that insulin production ceases - the insulin dependent stage. Interestingly, amylin allegedly plays the same role in type 1 and 2 diabetes, and the aggregates of amylin are amyloid formations similar to those seen in alzheimers. Before you start thinking of type 3 diabetes, though, the amyloids in Alzheimers aren't made of amylin. Amyloid just means "starch-like". What they have in common is beta-sheet protein structures (nothing to do with beta-cells) misfolding in a contagious, prion-like process.
Amylin, AKA Islet Amyloid Polypeptide (IAPP) is a protein produced by beta-cells in tandem with insulin. Insulin promotes glucose uptake and metabolism in cells, amylin slows glucose - and other food - uptake from the gut, by delaying gastric emptying, and decreases appetite; it also seems to be responsible from the switch from muscle glycogenogenesis to adipose lipogenesis, so probably has a role in obesity. According to wiki the ratio is 100:1 in favour of amylin (unless I've read it wrong and it's the other way round). Is the ratio always constant? Does amylin have any independence from insulin? In any case, amylin plus glucose represents a two-pronged approach to preventing systemic over-exposure to glucose; insulin pulls glucose out of, amylin slows access into, the blood.
Amylin overproduction results in incompleted (proamylin, or proIAPP) molecules being retained in cells; these serve to promote crystalization of further amylin beta-sheets in the cell, the amyloid clumps then initiate apoptosis, killing of the beta-cells with eventual loss of insulin - and amylin - production.
This looks like the end-stage of hyperinsulinaemia; this over-production of amylin (and insulin) is being driven by excess glucose intake, the amylin incompletion and insulin resistance may also indicate an overall micronutrient deficiency, and the out-of-phase insulin response to begin with indicates a) salivary amylase polymorphism, b) presence of excess omega 6 -> PGE2, c) absence of factors inhibiting PGE2, e.g. omega 3, CLA (the most likely reason for the diabetes-protective effect of dairy fat, or, if you don't eat dairy, ruminant fats).
Note that PGE2, an omega 6 series prostaglandin, inhibits glucose uptake, and to a somewhat lesser extent fructose, in sheep, but increases it in rats - probably the better model for human response. If this is the case, PGE2 is increasing glucose uptake as it decreases first phase insulin response, which is already diminished in individuals with fewer salivary amylase gene copies. The compensatory rise in second phase insulin response - exacerbated by 12-HETE, an omega 6 series eicosanoid - results in a larger area under the curve for both glucose and insulin; i.e., in hyperglycaemia and hyperinsulinaemia.
It is interesting that this hyperinsulinaemia-related problem with amylin is also causative in type 1 diabetes. It means that lower carb diets can be recommended to those at risk of this disease.
The T1D connection is really interesting. .I have a friend who is T1 diabetic; he said when he was 12 he got a craving to eat dry Milo (Nestles chocolate flavoured drink sweetened with maltose, i.e. glucose), ate a big tin of it in one day, crashed into a coma and woke up in hospital on insulin. That seems to show pronounced hyperinsulinaemia immediately preceding beta-cell burn out. Perhaps a combination of sudden hyperinsulinaemia from an inflammatory, infective, or autoimmune cause together with a low tolerance for amylin production.
Both insulin and amylin contain disulfide bridges (Cys-S-S-Cys) and this is interesting as the bridges are only meant to form outside the cell (the peptide is expressed as a string from the reducing environment of the cell, where the cysteine residues exist as 2x Cys-S-H, into the more oxidising serum enviroment, where the sulfur bonds snap together as the cysteine residues are oxidised to Cys-S-S-Cys plus H2O). If insulin output is very high, this puts a heavy demand on the reducing systems of the cell; glutathione, glutathione reductase, thioredoxin reductase and so on. Hydrogen sulfide - H2S - is also protective in the beta-cell for some reason. These are mostly selenocysteine enzymes, and selenocysteine is also required for a protein folding enzyme. (note though, supplying 200mcg Se as selenomethionine in America, not overall a selenium deficient country, has been associated with double the rate of diabetes in one study that was not directed at glycemic endpoints).
Here is the Wiki page on amylin. It's interesting, has the background to the latest research, and I wonder why we never hear about this hormone in diet-health discussions? I guess that from now on we will be hearing more of it.
The take home - keep insulin production under control by counting carbs and avoiding vegetable seed oils, and amylin should tag right along, ensuring beta-cell function lasts a lifetime.