Professor, Department of Veterinary Clinical Science, The Royal Veterinary College
North Mymms, Hatfield, Hertfordshire, UK
The majority of cats with diabetes mellitus have a form of the disease characterised by both subnormal insulin secreting capacity resulting in either a relative or absolute insulin deficiency and variable degrees of insulin resistance. Although non-diabetic cats can have a wide range of insulin sensitivities, most diabetic cats are roughly six times less sensitive to insulin than healthy cats.
Decreased insulin sensitivity can be a result of many factors including obesity, various primary endocrinopathies that inhibit insulin's peripheral action, virtually any systemic disease, pregnancy, drugs such as glucocorticoids and various progestins as well as, most frustratingly, hyperglycaemia itself.
Additionally other hormones may have a role in the pathogenesis of insulin resistance and the development of diabetes. Leptin has recently been shown to be important for fatty acid homeostasis and leptin resistance may lead to fat deposits in non-adipose tissue causing dysfunction. Plasma leptin concentrations increase in obese cats that are insulin resistant correlating highly with the degree of obesity. It is assumed that these obese cats have leptin resistance since the increase in leptin is not associated with a decrease in food intake.
Hyperglucagonaemia is also a well known feature of obesity and type-2 diabetes in other species and is thought to be secondary to the reduction of insulin action on alpha cell activity. Glucagon concentrations are significantly higher in obese than in lean cats and may be important in the progression from obesity and insulin resistance to diabetes, as glucagon increases insulin resistance and may hasten beta cell exhaustion.
Insulin sensitivity varies dramatically in healthy cats and those animals with lower insulin sensitivities have been shown to be at increased risk of developing impaired glucose intolerance if other factors (such as obesity) exacerbate their insulin insensitivity.
Insulin Secretory Capacity--'Functional' and 'Structural' Disorders
While the cause of the reduced insulin secreting capacity found in diabetic cats remains unclear, it is almost certainly multifactorial and includes impairment of insulin secretion because of functional abnormalities as well as possible but poorly defined alterations in the beta cells themselves. In other words there is both a 'functional' defect in beta cell activity and a 'structural' one.
Functional islet disorders. This so-called functional defect is clearly demonstrated by the alteration in insulin secreting capacity seen in some diabetic cats where a subnormal insulin secretory profile to various insulinogenic stimuli can normalise as a result of periods where insulin demand is substantially reduced through the administration of exogenous insulin and/or a reduction in the level of insulin resistance. Diabetic cats with no insulin secreting capacity in response to a glucagon stimulation responded normally once they were no longer diabetic. It is worth noting that this apparently 'reversible' lack of insulin secretory capacity means 'insulin stimulation tests' are of questionable value in assessing the presence or absence of the underlying insulin secreting capacity of feline diabetics in the clinical setting.
At least part of this reversible or 'functional' impaired islet capacity is due to so called glucose toxicity--the name given to the impaired capacity of beta cells exposed to chronic (greater than 3-10 days) hyperglycaemia to produce insulin.
There is also evidence that fatty acids and triglycerides may play a role in the deterioration of beta cell function and may be involved in the progression from insulin resistant to diabetic states. Glucose inhibits fatty acid oxidation and fatty acids inhibit glucose oxidation. Glucose must be metabolised in order to cause insulin secretion, therefore decreased glucose metabolism due to increased fatty acid concentrations might explain the lowering of the early phase insulin secretion in insulin resistant patients.
Although most diabetic cats have minimal insulin secreting capacity at the time of diagnosis, many will recover substantial levels of insulin output with appropriate treatment aimed at reducing insulin demand. This is generally best achieved through a combination of exogenous insulin and reduction in the dietary glucose load. Remission can occur in anywhere from 30 to 80% of diabetic cats depending upon the sample of diabetics cats examined and the method or methods used to 'unload the endocrine pancreas'.
Structural islet disorders. Over the last five years various researchers have suggested islet amyloid may have a role in the development of impaired insulin secreting capacity in diabetic cats. While islet amyloid can be found in the pancreatic tissue of many diabetic cats, over 50% of non-diabetic aged cats have islet amyloid as well. Not all glucose-intolerant cats have islet amyloid and various studies have failed to demonstrate any relationship between the severity of islet dysfunction and amyloid deposition.
Additionally, the demonstration of a lack of any difference between endocrine cell volume in diabetic and non-diabetic cats suggests the initial defect in insulin secretion, essential for the clinical expression of diabetes, occurs independently of alterations in beta cell mass.
It is of course possible that the most appropriate response to insulin resistance would be beta cell hyperplasia and could be argued that the unremarkable beta cell mass in diabetic cats is actually 'inappropriately normal'.
Some studies have found evidence of structural damage in islets during the progression from glucose-intolerance to overt diabetes, suggesting the hypersecretion that occurs in response to insulin resistance leads to exhaustion and apoptosis of beta cells over time. However, histological evidence for this is difficult to interpret as at least one study demonstrated one or more combinations of amyloidosis, vacuolar degeneration, reduced islet number or reduced islet cell mass in cats with transient diabetes. Despite all these beta cell abnormalities, the cats still returned to euglycaemia with no need of exogenous insulin.
Regardless of the microscopic appearance of 'stressed islets' and the relative importance of functional verses structural changes on islet endocrine activity, chronic hypersecretion of insulin from beta cells occurs in response to insulin resistance and can be substantially amplified if this inherent insulin insensitivity is exacerbated by diets high in readily available carbohydrates.
The Carnivore Connection and Predisposition to Diabetes Mellitus
While the level of insulin resistance is certainly greater in cats with glucose intolerance or diabetes mellitus than it is in normal cats, it has been suggested that as a strict carnivore, the cat is inherently more insensitive to insulin and less able to cope with carbohydrate loads than other more omnivorous species.
It has been proposed that during its evolutionary development the cat's natural diet of food of animal origin only has resulted in it becoming markedly adapted to a diet high in protein (approximately 54% of dry matter) and low in carbohydrates (approximately 8% of dry matter). This adaptation is reflected by the cat's unique metabolism of various nutrients, making it a true and strict carnivore. When comparing carbohydrate metabolism of the cat with those of other, more omnivorous species, there are a number of specific adaptations evident. These include altered levels of enzymes responsible for digestion and uptake of both starches and sugars in the intestine, an altered capacity to handle glucose loads including both a slower incorporation rate of glucose to glycogen and elongation of glucose elimination times with standard glucose tolerance tests, the effective absence of hepatic fructokinase and, perhaps most tellingly, the minimal hepatic glucokinase activity present in the cat. This low level of glucokinase activity limits the cat's ability to metabolise large glucose loads, as glucokinase has a far lower Km than hepatic hexokinase and hence is more readily able to respond to changes in blood glucose.
According to the carnivore connection theory propagated by Brand Miller and Colagiuri, chronic ingestion of a low carbohydrate-high protein diet results in selection pressure favouring animals with a tendency for increased hepatic glucose production and decreased peripheral glucose utilisation, i.e., insulin resistance. Both the ability of insulin to inhibit hepatic glucose production and to augment tissue glucose disposal are therefore impaired.
The increased hepatic glucose production is the result of the high protein intake and is mediated through an increased carbon flux through the gluconeogenic pathways. This increased carbon flux may be mediated by a number of different mechanisms including a mass action affect of increased concentrations of gluconeogenic substrates, an increase in glucagon levels that stimulate gluconeogenesis and/or the activation of a number of key enzymes in the gluconeogenic pathway.
The decreased insulin stimulated glucose disposal by peripheral tissues is largely due to the decrease in carbohydrate intake and the consequent hypoinsulinaemia and/or reduced insulin efficacy peripherally, i.e., peripheral insulin resistance.
In other words a predominantly carnivorous diet (or expressed another way a high protein-low carbohydrate diet) may produce metabolic adaptation which is effectively expressed as insulin resistance, both in the liver and peripheral tissues.
As previously mentioned, insulin resistance in man is now recognised as the earliest metabolic defect in those destined to develop non-insulin dependent diabetes mellitus and enhanced insulin resistance is a feature of many diabetic cats. It has been proposed by the devotees of the carnivore connection theory that insulin resistance was the normal phenotype for an obligate or strict carnivore and this very insulin resistance increases the likelihood of the development of diabetes in strict carnivores fed a diet high in carbohydrate for any protracted period of time. Such diets, through evoking higher post prandial insulin responses, might lead to over stimulation of the pancreatic beta cells and ultimately result in their 'exhaustion' as well as of course reducing their functional capacity through such processes as glucose toxicity.
When allowed to graze ad libitum, cats do not exhibit a post-prandial rise in blood glucose and hepatic glucokinase activity does not increase in response to increased carbohydrate feeding. Additional support for the cat's adaptation to a carnivorous diet is found with the levels of gluconeogenic enzymes present in feline hepatocytes.
When a diet contains low amounts of glucose, hepatic gluconeogenesis is predicted to be the major pathway for maintaining blood glucose. Consistent with this latter expectation, the activities of key gluconeogenic enzymes, (glucose-6- phosphatase, fructose-1,6 bisphosphatase and pyruvate carboxylase) are increased in the liver of normal cats. Additionally, unlike the situation in rodents and man, the gluconeogenic capacity of the feline liver is not inhibited by glucose. The recently reported finding that in cats, stress hyperglycaemia is caused by enhanced hepatic glucose output rather than, as previously postulated, insulin resistance underscores the gluconeogenic potential of the feline liver and suggests its possible role in the genesis of pathological hyperglycaemia such as is observed in diabetes mellitus.
Interestingly the low carbohydrate of the carnivore's diet may not be the only important factor in the development of impaired insulin secreting capacity. A recent study evaluating the effect of a high fat diet on glucose tolerance in intact male cats demonstrated a reduction in the acute insulin response to a glucose tolerance test suggesting diminished pancreatic insulin secretion and/or beta cell responsiveness to glucose as a result of high fat diets.
Consequently the very adaptive processes that have favoured selection for the obligate carnivore also favour the development of hyperinsulinaemia and a chronic state of increased demand for insulin production being placed upon the beta cells of the pancreatic islets. While in its most overt form this may manifest itself as progressive islet destruction, in the cat, beta cell dysfunction appears to precede any obvious evidence for structural islet changes that can be correlated with this impaired function.