Insulin secretion and sensitivity

Type 1 diabetes results from insulin deficiency, although some residual insulin secretion is present at diagnosis and a minority of people retain limited function even after many years of diabetes. Preservation or regeneration of beta cells is a prime goal of diabetes research. Healthy individuals vary in their insulin sensitivity, i.e. the amount of insulin they need to maintain normal metabolic function; people with type 1 diabetes do not differ from the non-diabetic population in this respect. People with type 2 diabetes are less sensitive to insulin, but not when correction is made for BMI. In type 1 diabetes, insulin production by the beta cells diminishes long before clinical onset of hyperglycaemia, and this process is marked by loss of pulsatile secretion and first phase insulin response (FPIR) to intravenous glucose; declining efficiency of insulin action results in reduced insulin sensitivity and progressive glucose intolerance. Insulin insensitive individuals need more insulin, and therefore present earlier than the insulin sensitive. Changes in glucagon secretion from the alpha cells also contribute to the metabolic disorder. Although insulin secretion and sensitivity are key measures of metabolic dysfunction, routine tests are relatively crude, and we still lack a reliable measure of beta cell reserve capacity.


Harold Himsworth
Harold Himsworth
In the 1930s Sir Harold Himsworth devised a primitive test of glucose disposal in response to insulin injection, and made the key observation that lean young people with or without diabetes respond similarly, whereas older overweight people with diabetes require much more insulin to achieve the same effect. From this he inferred that there were two types of diabetes: an insulin-sensitive form due to simple insulin deficiency, and an insulin-insensitive form in which the tissues were resistant to the actions of insulin.[1]

In recent decades, observations in high-risk relatives have shown that clinical onset of type 1 diabetes is preceded by progressive glucose intolerance, loss of the FPIR, and loss of pulsatile insulin secretion. Progression to diabetes is more rapid in those who are less sensitive to insulin.

Insulin secretion

There are sub-populations of beta cells within healthy islets, and these have varying levels of responsiveness to glucose. Those with a low threshold for response are more active at normal glucose levels; others cut in at higher glucose levels.[2] Fully functional beta cells are metabolically very active, shedding and replacing 30–50% of their surface membrane daily in the course of insulin secretion.

A lean healthy individual might secrete about 35 units of insulin per day, yet will have about 10 times this amount stored within his pancreas. By contrast, an obese insulin-resistant person might need to produce >100 units daily to maintain normal blood glucose levels.

Type 1 diabetes results from progressive beta cell loss by apoptosis, thus increasing the work-load of the residue. A further consequence is loss of beta to beta cell communication and an altered cell-to-cell (paracrine) interaction between beta cells and glucagon-producing alpha cells.

Coordinated pulsatile release of insulin deteriorates during the type 1 diabetes prodrome, and results in loss of efficacy in regulating glucose output by liver cells (hepatic insulin resistance). Blood glucose rises, increasing the workload of the remaining beta cells, and may further impair their function by the effect known as glucose toxicity.

There are indications that functional defects are present at diagnosis of type 1 diabetes, and may recover to some extent in the period following diagnosis.

Studying islet function

Beta cells function within a highly structured environment, communicate directly with one another, and indirectly with other islet cell types in the vicinity. They receive a rich blood supply and the activities of the islets are linked by sympathetic nerve fibres.

The pancreas is an inaccessible organ, and islets cannot easily be studied in intact humans or animals. The alternatives are isolated perfused pancreases, islet preparations or dispersed cells. Beta cell function becomes progressively easier to measure with each step, but also becomes less representative of normal function in an intact animal.

Studying insulin secretion

Insulin secretion is hard to evaluate on the basis of blood samples taken from a peripheral vein. This is because a high proportion of the insulin secreted by the pancreas is cleared on first passage through the liver. The liver is thus exposed to higher concentrations of insulin, delivered in pulses. The systemic circulation sees lower insulin levels with small pulses, since these are largely eliminated in passage through the liver. Furthermore, insulin has a half-life of about 6 minutes in the circulation, so very frequent sampling is needed to monitor its fluctuations.

Last but not least, people on insulin typically have low levels of insulin antibodies in the circulation, which complicates accurate measurement in the insulin assay. These limitations can largely be overcome by measuring C-peptide in the plasma or urine. C, or connecting, peptide, is a short peptide sequence released from the proinsulin molecule on a one-to-one ratio as insulin is secreted. C-peptide is not cleared by liver (it is excreted via the kidney) and has a plasma half-life of about 30 minutes.

Stressed beta cells accelerate insulin production and secretion, and one consequence is that the proportion of proinsulin to cleaved insulin entering the circulation is increased. A raised proinsulin:insulin ratio is thus a marker of beta cell distress and, since proinsulin has less metabolic effect than cleaved insulin, insulin action is also reduced.

Studying insulin sensitivity

Insulin sensitivity can be measured directly in intact humans or animals, in isolated organs, or in isolated or cultured cells. Whole-body insulin sensitivity can be measured by glucose clamp techniques. The principle of the most commonly used variant is to infuse insulin at a constant rate and to infuse glucose variably to 'clamp' glucose at a predetermined level. This procedure, although highly unphysiological, allows direct comparison of insulin sensitivity between individuals, or within the same individual over time. Other direct measures include the insulin suppression test, discussed elsewhere

Indirect measures of insulin sensitivity are derived from modelling techniques based upon measure of fasting or stimulated glucose and insulin. The best known of these is the homeostasis model assessment (HOMA). HOMA models interactions between glucose and insulin to predict fasting steady-state glucose and insulin concentrations for a wide range of possible combinations of insulin resistance and beta cell function, and predicts fasting steady-state levels of plasma glucose and insulin for any given combination of pancreatic beta cell function and insulin sensitivity. Its great advantage is that it can be derived from simple measures of fasting glucose and insulin; its disadvantage is that it infers a dynamic interaction from steady-state measures.

Paracrine influences

Seventy percent of beta cells are in intimate contact with alpha cells. In health, insulin and glucagon secretion are reciprocal and balance one another: thus insulin secretion inhibits glucagon production and vice versa.

In type 1 diabetes, reduced insulin production results in unopposed glucagon secretion, resulting in increased hepatic glucose output. It has been argued that type 1 diabetes should therefore be viewed as a two-hormone disorder.


Insulin secretion is exquisitely regulated to meet the requirements of the body. Failing insulin secretion is aggravated by defects in both the timing and amount of the insulin produced. Loss of pulsatility results in hepatic insulin resistance; loss of beta cell mass leads to unopposed glucagon action; and hyperglycaemia may in itself inhibit insulin secretion. These forward-feeding mechanisms result in an acceleration towards hyperglycaemia.

Current techniques for measuring insulin secretion and sensitivity are nonetheless relatively crude, and have limited clinical application. The main exception is measurement of C-peptide, although this too should be interpreted with care. Tests of insulin secretion identify functional beta cell capacity, but there are indications that not all beta cells are fully active in type 1 diabetes. This is an important issue for future research, but there is currently no reliable means of measuring beta cell reserve, i.e. the full potential capacity of a diabetic pancreas to secrete insulin.


  1. ^ Himsworth HP. Diabetes mellitus: its differentiation into insulin sensitive and insulin-insensitive types. Lancet 1936;227:127–30

  2. ^ Pipeleers D. Heterogeneity in pancreatic beta-cell function. Diabetes 1992;41:777–81


Nobody has commented on this article

Commenting is only available for registered Diapedia users. Please log in or register first.