Diabetes mellitus has been recognized for centuries as a debilitating disease, characterized by the excretion of “sweet urine” (mellituria), polydipsia, wasting of tissue, and the development of ketoacidosis, hyperosmolar coma, and death. The role of the pancreas in the development of diabetes was first recognized in 1886 when Minkowski and von Mering produced diabetes by total pancreatectomy in the dog. In 1921, Banting and Best extracted the active compound from pancreas and showed that it was able to control hyperglycemia in diabetic dogs and humans. Insulin was prepared for the first time in crystalline form in 1926 by Abel, and it took 34 more years until its amino acid sequence was demonstrated by Sanger. Finally, in 1963 insulin was synthesized by Meyenhofer and coworkers, and in 1964 by Katsoyannis and coworkers. In 1967, Steiner discovered that insulin was synthesized as a larger precursor, proinsulin. An even larger molecule, named preproinsulin, was identified as a precursor for proinsulin in 1976 by Chan and coworkers (for a detailed review see Bliss, 1983). These research efforts have culminated in the biosynthesis of human insulin by recombinant DNA techniques, allowing large-scale production of the hormone in 1983 (Frank and Chance, 1983; Johnson, 1983).

Spontaneous diabetes has been reported in many species. Diabetes is a major health problem and one of the leading causes of death in humans. In animals, it occurs most frequently in the dog and cat, with an incidence of approximately 0.5–1%. Overt diabetes is rare in horses, cattle, and sheep, although obese horses may develop metabolic syndrome, which might include abnormal glucose concentrations. Isolated cases of diabetes have been reported in mules, ferrets, pigs, newts, buffalos, monkeys, and fish (Stogdale, 1986). Diabetes also has been reported in several breeds of birds (Stogdale, 1986). Diabetes in humans has been classified into two major categories, which are of interest to veterinary medicine: type 1 diabetes, previously called insulin-dependent diabetes mellitus or juvenile-onset diabetes, and type 2 diabetes, previously called noninsulin-dependent diabetes mellitus or maturity-onset diabetes. Based on glucose-tolerance testing (Kaneko et al., 1977; O’Brien et al., 1985), it has been determined that most diabetic dogs and cats are insulin deficient.

Insulin is produced in the beta cells of the islets of Langerhans in the endocrine pancreas. The islet also contains glucagon-secreting alpha cells, somatostatin-secreting delta cells and PP or F cells, which secrete pancreatic polypeptide. The beta cells compose about 60–80% of the islet. The endocrine cells are arranged in a nonrandom distribution, with beta cells forming a central core surrounded by a mantle of the other three cell types in some animals and humans (Bonner-Weir and Orci, 1982). In the cat, the beta cells are located in the periphery (O’Brien et al., 1986), while in the horse, the alpha cells form a central core (Helmstaedter et al., 1976) surrounded by beta cells.

The secretion of insulin from the beta cell occurs through the process of exocytosis in which there is fusion of the secretory granule with the plasma membrane. Insulin secretion is highly regulated primarily by glucose, but also by other fuels, hormones, and neurotransmitters (Porte and Halter, 1981). On a cellular level, insulin release stimulated by glucose and other fuels is thought to be initiated by closure of ATP (adenosine triphosphate)-dependent K channels, which leads to depolarization of the beta cell and influx of calcium into the cytosol through voltage-gated calcium channels (Arkhammar et al., 1987; Cook et al., 1988). The increase in cytosolic free calcium triggers pulsatile insulin secretion. In addition, augmentation of this response occurs by both a K⁺-ATP channel-independent Ca²⁺-dependent pathway and K⁺-ATP channel-independent Ca²⁺-independent pathways of glucose action (Henquin, 2011). The pathway that is Ca²⁺ dependent but independent of ATP-K channels is called the amplification pathway. The exact mechanism is not understood (for review see Straub and Sharp, 2002). Insulin is released from the beta cell in a pulsatile fashion. It is thought that these rapid oscillations are generated through intraislet mechanisms, because they can be seen even in isolated islets and beta cells (Berman et al., 1993). Pulsatile secretion becomes abnormal in diabetes; however, pulsatility can be restored by beta cell rest (Laedtke et al., 2000).

In both type 1 and type 2 diabetes, islet function and insulin secretion are abnormal. It is now clear that in both types a loss of beta-cell mass has occurred and that the fundamental difference between the two major types of diabetes lies in the etiology of the beta-cell dysfunction and its severity but not its presence or absence (Porte, 1991). It is controversial whether a reduction in beta cell mass or function play the major role in the pathogenesis of type 2 diabetes (for review see Meier and Bonadonna, 2013). It appears that both occur together and are tightly related. Insulin resistance, frequently seen with obesity in humans and cats (Kahn et al., 2006; Hoenig et al., 2006b) is an important concomitant factor in type 2 diabetes and puts an additional stress on the beta cell. However, the insulin resistance of obesity alone is not sufficient to cause diabetes. This is shown by the fact that over 80% of obese people are not diabetic; a similar assessment has not been made in cats. It is not known how many obese cats actually become diabetic. Obesity increases the risk of diabetes; however, most obese cats have normal glucose control because they counteract peripheral insulin resistance with a decrease in endogenous glucose output by the liver (Kley et al., 2009; Hoenig et al., 2011, 2012, 2013). It is interesting that, while the prevalence of obesity in people and cats is about equal, about 8% of humans develop type 2 diabetes, whereas the prevalence of diabetes in cats is <1%. The reason for this discrepancy is not known.

Evidence suggests that disordered insulin secretion may contribute to development of insulin resistance and hence represent an initiating factor in the progression to type 2 diabetes (Schofield and Sutherland, 2012). An early marker of beta cell dysfunction is a loss of pulsatility of insulin secretion. This leads to hepatic insulin resistance and higher uncleaved proinsulin concentrations (for review see Wahren et al., 2012). An early marker is also a change in the proinsulin/insulin secretion ratio (Haffner et al., 1997). In cats and humans, among a few others, the increase in insulin release from a reduced beta-cell mass in type 2 diabetes is associated with release and deposition of islet amyloid polypeptide, leading to further loss of beta cells (Porte and Kahn, 1989; Hoenig et al., 2000a). It is thought that the majority of dogs have a form of diabetes similar to type 1 diabetes in people, that is involving a genetic component, which is influenced by environmental input. The possibility of a genetic basis for canine diabetes has been suggested by the occurrence of the disease in dog families, and the presence of DLA haplotypes that seem to confer risk of diabetes. In type 1 diabetes in people, the loss of beta cells is caused primarily by immune-mediated destruction (MacLaren et al., 1989). Beta-cell antibodies are present in approximately 50% of naïve diabetic dogs (Hoenig and Dawe, 1992). In later studies, 4/30 dogs were found to be positive for antibodies against the 65-kDa isoform of canine glutamic acid decarboxylase and 3/30 also had antibodies against the insulinoma-associated antigen-2 (IA-2), both frequently found in human type 1 diabetics (Davison et al., 2008a). In another study, 5/40 newly diagnosed dogs had antibodies against insulin (Davison et al., 2008b). Beta-cell antibodies have not been demonstrated in diabetic cats (Hoenig et al., 2000b).

Insulin secretion can also become autonomous, resulting in hyperinsulinemia associated with hypoglycemia. Insulin-secreting tumors (adenomas or carcinomas; also called insulinomas) occur infrequently and have been described in the dog, cat, ferret, cattle, and a pony (Capen, 1990). The diagnosis of an insulinoma can be difficult, but frequently it can be made on the basis of an inappropriately high serum insulin concentration. In humans, the diagnosis is often made based on high proinsulin and/or C-peptide concentrations (Chammas et al., 2003).

Insulin is the most potent physiological anabolic agent. In insulin-responsive tissues, it facilitates cellular uptake and metabolism of glucose. Insulin promotes the synthesis of glycogen, protein, and fat and is involved in the uptake of ions such as K into the cell. Insulin exerts its effect by binding to specific receptors on the plasma membrane of cells. The insulin receptor is a tetrameric structure consisting of two subunits: an alpha and a beta subunit. Two of each of these subunits are joined together via disulfide bonds. Both subunits are glycosylated and exposed to the extracellular milieu; only the beta subunit is exposed to the intracellular environment. The insulin receptor has tyrosine kinase activity. Binding of insulin leads to autophosphorylation of the receptor and these phosphorylated Tyr residues serve as docking sites for downstream effectors (Le Roith and Zick, 2001; Schinner et al., 2005). Once the insulin receptor kinase is activated, the presence of insulin is not necessary for its continued activity. The receptor is inactivated by dephosphorylation.

Despite our knowledge of insulin–receptor interaction, our understanding of molecular events that link the insulin–receptor to the regulation of cellular metabolism is poor. Some, but not all, of the actions are due to the phosphorylation or dephosphorylation of target proteins on serine and threonine residues. The mediation of the mitogenic responses involves the mitogen-activated protein (MAP) kinase pathway, whereas the metabolic responses of insulin, such as glucose transport, involve phosphatidyl-inositol-3 (PI3)-kinase. Diabetes has also been linked to alterations in the function of both endoplasmic reticulum (Kaneto et al., 2005) as well as mitochondria (Lowell and Shulman, 2005).

The existence of a relatively specific insulin-degrading enzyme (IDE, insulinase, insulin-specific protease) has been reported by several investigators (for a review see Duckworth, 1990). Insulin-degrading activity has been found in essentially all tissues examined, and high activity has been seen in liver, kidney, muscle, brain, fibroblasts, and red blood cells. The enzyme seems to be located predominantly in the cytoplasm but can be found in other subcellular compartments. Plasma membrane preparations also contain this enzyme. Some receptor-bound insulin is degraded on the membrane, while some insulin is released intact from the receptor, and some is internalized with the receptor, and insulin degradation is then initiated in endosomes. The degradation of insulin via this process is rapid. Other enzymes may also be involved in the degradation of insulin, and their role and factors controlling insulin-degrading activity in various tissues and disease states needs to be examined.