METABOLIC CHANGES IN DIABETES

R. D. G. LESLIE
London

SUMMARY glycogenolysis. The main substrates for gluconeogenesis

Diabetes is not a single disease but a group of diseases are alanine and glutamine, glycerol, lactate and pytuvate.
characterised by hyperglycaemia. The most important Glycogenolysis is controlled by the synthesis or degra­
regulator of glucose uptake from the blood is the hor­ dation of glycogen; the hormone glucagon stimulates gly­
mone insulin, which is produced by islet beta cells and cogenolysis while the hormone insulin inhibits it. Glucose
acts on insulin receptors to promote nutrient uptake and is taken up from the circulation by the splanchnic bed, adi­
processing. A decrease in either insulin secretion or sensi­ pose tissue, skeletal muscle and brain. Glucose utilisation
tivity can cause diabetes. Exposure to prolonged hyper­ in skeletal muscle is by oxidative and non-oxidative pro­
glycaemia causes reversible and then irreversible cesses.1 The non-oxidative pathway determines, in the
changes to tissue metabolism and structure. These main, the rate of glucose uptake and glucose incorporation
changes may be responsible for the potentially devas­
into glycogen. The most important regulator of glucose
tating complications of diabetes.
uptake is the hormone insulin.2
Industrialisation in the last century has altered the balance Insulin is coded by a gene on chromosome II, initially
of major diseases. One hundred years ago, epidemics of processed as a prohormone (preproinsulin) and secreted
infectious diseases were the main cause of mortality. by beta cells in the islets of Langerhans of the pancreas.3
Today, non-communicable diseases including cancer, Conversion of proinsulin to insulin involves the removal
arterial disease and diabetes are the major causes of death of a connecting peptide, C-peptide, in the Golgi apparatus
in the Western world. There is a very real prospect that by enzymes with trypsin- and carboxypeptidase-like
industrialisation of Asia and Africa will herald an epi­ activity. Insulin then co-precipitates with zinc ions as
demic of these non-communicable diseases. microcrystals in the secretory granules. Insulin secretion
Diabetes is characterised by a persistently increased follows stimulation of the islet beta cell by glucose or
blood glucose level. There are many causes of diabetes, other secretagogues. Glucose metabolism is required for
just as there are many causes of anaemia, including endo­ insulin release to be stimulated and this triggers a
crine diseases, liver diseases or drugs. The commonest sequence of events including generation of ATP, closure
causes of diabetes are insulin-dependent or type I diabetes of K+ channels in the beta cell membrane and increased
(IDDM) and non-insulin-dependent or type 2 diabetes Ca2+ entry into the cell.
(NIDDM). Insulin acts on target cells via insulin receptors. These
Diabetes is a major international health problem affect­ receptors, coded by genes on chromosome 19, mediate the
ing possibly some 60 million people world-wide. In the intracellular action of insulin to stimulate nutrient uptake
industrialised Western world, diabetes is the commonest and processing.4 Insulin receptors lie on the target cell sur­
cause of blindness during working life, the second com­ face and straddle it. The receptor is composed of two alpha
monest cause of renal failure, and a major risk factor for subunits linked covalently to two beta subunits. Insulin
leg amputations. On average diabetes reduces life expec­ binds to the alpha subunits initiating a series of steps
tancy by a decade, an effect largely attributable to macro­ involving phosphorylation and autophosphorylation of
vascular disease. Thus, diabetes causes both tyrosine residues. The receptor-hormone complex is
microvascular and macrovascular disease.
internalised, insulin is degraded and the receptor is recy­
NORMAL GLUCOSE METABOLISM cled to the cell surface.
Glucose is produced by the liver by gluconeogenesis and A decrease in either the secretion of insulin or the sensi­
tivity to insulin causes a rise in blood glucose through
Correspondence to: Dr. R. D. O. Leslie, Department of Diabetes and
Metabolism, 3rd Floor, Dominion House, 59 Bartholomew Close, West reduced glucose utilisation. The diagnosis of diabetes
Smithfield, London ECIA 7BE, UK mellitus is esiablished by a raised blood glucose level: the

Eye (1993) 7, 205-208

206 R. D. G. LESLIE

venous whole blood glucose must be over 10 mmol/l in ance not beta cell dysfunction. However, it is widely
samples taken at random or 2 hours after a 75 g oral glu­ believed that neither abnormal insulin secretion nor
cose load. In IDDM hyperglycaemia results from abnormal sensitivity to insulin alone can explain the glu­
decreased insulin secretion due to beta cell destruction. In cose intolerance of NIDDM.
NIDDM a decrease in both insulin secretion and insulin
sensitivity may operate to cause hyperglycaemia and the Aetiology
relative contribution of each defect can vary from individ­ The most powerful evidence that NIDDM is predom­
ual to individual. inantly inherited comes from the study of identical twins:
more identical than non-identical twins of diabetic
NON-INSULIN-DEPENDENT DIABETES patients are concordant for NIDDM (see Table 1).9,/0
Non-insulin-dependent diabetes affects at least 2% of the Recent evidence suggests that some individuals with
population and no race is immune from the disease; maturity onset diabetes in the young (MODY) have a
among the Pima Indians of Arizona and Nauruans from defect (both missense and nonsense mutations have been
Polynesia, half the adult population is diabetic. The preva­ reported) in their glucokinase gene promoter region which
lence of NIDDM increases with increasing age: in the might account for up to 40% of these cases. Rare genetic
elderly the prevalence can be striking, reaching 45% in defects in the insulin gene and in the insulin receptor have
I ,
men aged 75-79 years in east Finland. In part this increase also been described which may cause diabetes. 1 1 2
may be due to an age-related deterioration in glucose The estimated heritability of NIDDM is about 82%
tolerance. Thus, while fasting blood glucose levels remain (Table I). In assessing the role of the environment in caus­
fairly constant with age the glucose level 2 hours after oral ing NIDDM it is important to understand that heritability
glucose rises steadily. In the absence of prospective stud­ is not an invariant index of a genetic influence. It describes
ies this age-related effect is ignored for the purposes of the genetic effect under particular environmental condi­
defining the disease. tions. Different estimates of heritability might be obtained
if twins were studied in different environments. A number
Pathogenesis of non-genetic factors have been implicated in the aetiol­
The islets of Langerhans in patients with NIDDM appear ogy of this disease, including nutrition, obesity, ageing
normal except for amyloid deposits and a reduction in the and reduced exercise.
beta cell mass to about 60% of normal. In general, the Nutrition, The incidence of NIDDM decreases during
secreted insulin has a normal structure though there is a food shortage. However, high carbohydrate diets improve
tendency to secrete a relative excess of proinsulin. The insulin sensitivity in both normal and NIDDM subjects
disease is characterised by a decreased beta cell secretory through physiological adaptation to an altered fuel supply
capacity, insulin resistance and hepatic glucose overpro­ and not reversal of a pathological process. The change
duction.4-6 The onset of the diabetes is insidious, probably from hunter-gatherer to a modern diet may be responsible,
occurring several years before the clinical diagnosis. in part, for the virtual epidemic of diabetes in migrant
Since hyperglycaemia itself can induce beta cell dys­ populations and previously isolated communities. At
function and insulin insensitivity it has proved impossible present, however, there is no direct evidence that dietary
to distinguish primary changes leading to diabetes from factors cause NIDDM although they may influence rate of
those which are secondary to the disease. progression to clinical symptoms. Recent studies do indi­
To address this problem, studies have been performed cate a relationship between low birth weight and impaired
on non-diabetic children of diabetic patients. These chil­ glucose tolerance in later life; 1 3 if confirmed, this associ­
dren show fasting hyperglycaemia, impaired glucose ation could be due to poor nutrition in utero limiting pan­
tolerance, decreased glucose clearance, fasting hyperin­ creatic development.
sulinaemia, either decreased or increased insulin Obesity, Obesity is not a major factor but can potentiate
responses to glucose, and impaired insulin-mediated glu­ NIDDM in genetically susceptible individuals. Offspring
cose disposal due to reduced non-oxidative glucose of NIDDM patients who subsequently develop NIDDM
metabolism. It remains to be determined whether these themselves are more likely to be obese when young. In one
changes presage diabetes. The one study which followed study, those subjects with low birth weight who became
patients prospectively found that offspring who developed obese were particularly prone to diabetes.1 3 Individuals
diabetes, as compared with those who did not, were with upper body obesity are also at particularly high risk
initially more obese and had decreased glucose tolerance
Table I. Proband concordance of non-insulin-dependent diabetes
and glucose clearance, fasting hyperinsulinaemia and mellitus after first and second oral glucose tolerance tests
increased second phase insulin responses.7 The develop­
ment of NIDDM was both preceded and predicted by First test' Second testb
(%) (%)
defects in insulin-dependent and insulin-independent glu­
cose uptake, changes which could precede the onset of Identical twins 29 58
hyperglycaemia and diabetes by more than a decade.8 Non-identical twins 14 17

These observations are consistent with the finding that the " First test: subjects aged 42-55 years,
initial lesion in NIDDM is due to peripheral insulin resist- h Second test: subjects aged 52-65 years,
METABOLIC CHANGES IN DIABETES 207

of NIOOM. Nevertheless, there are no differences in the been called the New World Syndrome, the Matabolic Syn­
relative contributions of decreased insulin secretion and drome or Syndrome X.
decreased insulin sensitivity between obese and lean The cause of microvascular disease is not clearly
NIDOM subjects. defined but the belief is that hyperglycaemia is a major
Ageing and Exercise. Ageing is not associated with factor and, probably the major factor. Microvascular com­
decreased insulin secretion nor with decreased insulin plications are not simply genetically determined since the
sensitivity in physically active subjects. However, lack of non-diabetic identical twins of diabetic patients do not get
exercise in the elderly may hasten the appearance of them. For complications to develop hyperglycaemia must
hyperglycaemia. Recent prospective studies of popula­ be present. The cause of the hyperglycaemia, i.e. the cause
tions at risk of diabetes suggest that physical activity pro­ of the diabetes, is irrelevant since microvascular compli­
tects against the development of NIOOM. cations are a feature of all types of diabetes. As diabetes is
defined by hyperglycaemia it reasonable to anticipate that
INSULIN-DEPENDENT DIABETES these microvascular complications should result from this
There is worldwide variability in the average annual inci­ hyperglycaemia. It is clear that the risk of diabetic compli­
dence oflOOM under the age of 15 years ranging from 1.7 cations is related to the duration of the disease.14 A number
per 100 000 person-years in Japan to 29.5 per 100 000 of studies have demonstrated a relationship between the
person-years in Finland. Incidence rates in Western indus­ level of blood glucose and the risk of developing compli­
trialised countries establish 100M as the second com­ cations. Thus, in one study an average blood glucose more
monest chronic childhood illness after asthma. than 50% above the normal range was associated with a
40% risk of developing severe retinopathy at 14 years: in
Pathogenesis contrast, this risk was only 5% in those patients with blood
Insulin-dependent diabetes is due to destruction of the glucose levels close to the normal range.14
beta cells in the islets of Langerhans. The disease is caused
by environmental factors operating in a genetically sus­ Pregnancy
ceptible host in early childhood to initiate the destruction It is known that hyperglycaemia can influence fetal
of the insulin secreting cells, probably by an immune pro­ development. IS The incidence of major and minor congen­
cess.9 In some genetically susceptible individuals this ital anomalies in children of patients with diabetes is
immune process can persist in association with chronic between 6% and 9%, that is up to 3 times greater than in
progressive beta cell destruction over many months, even the general population. The most prevalent congenital
years, and lead to 100M, but in others it may remit sponta­ anomalies in children of diabetic patients include caudal
neously without diabetes developing.9 regression syndrome, neural tube defects and cardiac
At diagnosis about 80% of islets contain no beta cells anomalies. The excess in malformations is confined to
and the islets may be heavily infiltrated with lymphocytes. patients whose diabetes antedates their pregnancy. In
There is no evidence that the exocrine pancreatic cells or addition, the malformations arise from developmental
the other islet cells are involved in this destructive process. changes likely to have occurred before the seventh week
The limited secretion of insulin by patients with 100M of gestation. It was proposed that the excess congenital
results in them being prone to increased ketoneogenesis. anomalies in children of patients with diabetes was due to
In the absence of insulin treatment such patients will die in hyperglycaemia in early fetal life. This hypothesis was
diabetic ketoacidosis. tested by measuring glycated haemoglobin, an index of
blood glucose levels, over the previous 2 months. Children
METABOLIC CONSEQUENCIES OF of patients with high glycated haemoglobin levels had a
HYPERGLYCAEMIA striking excess of congenital anomalies which reached
A common health aim is to reduce the morbidity and mor­ 22% if the glycated haemoglobin was greater than 10%.
tality associated with diabetes. Risk factors for macrovas­ The risk of major malformations can be reduced to non­
cular disease are well established in the non-diabetic diabetic levels if the diabetic mother is treated to obtain
population and include hypercholesterolaemia, increased normal glycated haemoglobin levels before conception. IS
plasma fibrinogen, smoking, obesity and hypertension. The mechanism of this embryopathy is not clear. Pre­
The evidence is that in the diabetic population the same pregnancy counselling is now routine in diabetic clinics
risk factors operate but, if anything, diabetes has an addi­ and patients are advised to obtain near-normal blood glu­
tive effect with them. In addition, in NIOOM there is a cose levels before conception.
tendency for some of these risk factors to aggregate; these
patients are particularly at risk of obesity, dyslipidaemia Mechanisms
and hypertension. A common feature of these changes is Exposure to hyperglycaemia can cause acute reversible
their association with insensitivity to insulin, an obser­ metabolic changes and, if prolonged, cumulative irrevers­
vation which has led to the proposal that insulin resistance ible changes. Three broad mechanisms have been
is the single unifying factor causing an excess risk of described for glucose-induced damage.16 First, glucose
macrovascular disease in Western society.3 The associ­ and other sugars can bond with any exposed lysine resi­
ation of major risk factors for macrovascular disease has dues or (in the case of haemoglobin) valine, or any protein.
208 R. D. G. LESLIE

This process of glycation can alter the structure and func­ 3. Howell SL, Bird G St J. Biosynthesis and secretion of
insulin. Br Med Bull 1989;45:19-36.
tion of the protein. Further changes can lead to glycation
4. Houslay MD, Siddle K. Molecular basis of insulin receptor
products with extensive cross-linkage called advanced function. Br Med Bull 1989;45:264-84.
glycation end products - an irreversible change. These 5. DeFronzo RA. Pathogenesis of type 2 (non-insulin depend­
molecules may lead to the production of free oxygen rad­ ent) diabetes mellitus: a balanced overview. Diabetologia
icals which could themselves cause tissue damage. The 1992;35:389-97.
6. Temple RC, Carrington CA, Luzio SD et al. Insulin defi­
second mechanism is the production of excess sorbitol
ciency in non-insulin dependent diabetes. Lancet 1989;I:
though a normally redundant pathway involving the
293-5.
enzyme aldose reductase. Sorbitol cannot readily leave a 7. Warram JH, Martin BC, Krolewski AS, Soeldner JS, Kahn
cell and accumulation of the alcohol sugar could lead to CR. Slow glucose removal rate and hyperinsulinemia pre­
osmotically driven overhydration of the tissue and dam­ cede the development of type II diabetes in the offspring of
age. The third mechanism involves the direct competition diabetic parents. Ann Intern Med 1990;113:909-15.
8. Martin BC, Warram lH, Krolewski AS, Bergman RN,
between glucose and myoinositol. Myoinositol is an
Soeldner JS, Kahn CR. Role of glucose and insulin resist­
important substrate in cellular energy production, and its ance in development of type 2 diabetes mellitus: results of a
structure is very similar to that of glucose. Excess glucose 25-year follow up study. Lancet 1992;340:925-9.
can therefore compete for myoinositol uptake by a cell, 9. Lo SSS, Tun RYM, Hawa M, Leslie RDG. Studies of dia­
leading to myoinositol depletion. betic twins, Diabetes Metab Rev 1991;7:223-38.
10. Newman B, Selby JV, King M-C, Slemenda C, Fabsitz R,
Friedman GD. Concordance for type 2 (non-insulin depend­
CONCLUSION
ent) diabetes mellitus in male twins. Diabetologia 1978;30:
These observations suggest that hyperglycaemia, but also 763-8.
the level of other major risk factors, plays an important II. Haneda M, Polansky KS, Berganstal RM, et al. Familial
role in producing the complications of diabetes. The days hyperinsulinaemia due to a structurally abnormal insulin:
definition of an emerging new clinical syndrome. N Engl J
of regarding diabetes as simply a sugar problem are gone.
Med 1984;310:1288-94.
In the absence of a successful primary prevention policy 12. Taylor SI, Cama A, Accili D, et al. Molecular genetics of
for diabetes we seek to reverse the level of these risk fac­ insulin resistant diabetes mellitus. J Clin Endocrinol Metab
tors towards normal. 1991;73:1158-63.
13. Hales CN, Barker DJP. Type 2 (non-insulin dependent)
Key words: Diabetes, Glucose, Glycation, Insulin.
diabetes mellitus: the thrifty phenotype hypothesis . Diabe­
tologia 1992;35:595-601.
REFERENCES 14. Chase HP, Jackson WE, Hoops SL, Cockerham RS, Archer
1. Shulman GI, Rothman DL, Jue T, et al. Quantitation of PG, O'Brien D. Glucose control and the renal and retinal
muscle glycogen synthesis in normal subjects and subjects complications of insulin-dependent diabetes. JAMA 1989;
with non-insulin dependent diabetes by DC nuclear mag­ 261:1155-60.
netic reasonance spectroscopy. N Engl J Med 1989;322: 15. Fuhrmann K, Reiher H, Semmler K, Fischer F, Fischer M,
223-8. Glockner E. Prevention of congenital malformations in
2. Reaven GM. Insulin resistance, hyperinsulinemia, hypertri­ infants of insulin-dependent diabetic mothers. Diabetes
glyceridemia and hypertension: parallels between human Care 1983;6:219-23.
disease and rodent models. Diabetes Care 1991;14: 16. Larkins RG, Dunlop ME. The link between hyperglycaemia
195-202. and diabetic nephropathy. Diabetologia 1992;6:499-504.