Drugs 44 (Suppl . 3): 1-12, 1992 ()() 12-6667/92/0300-0001/$6.00/0 © Adis International Limited. All rights reserved . DRSUP3199

Pathophysiology of Diabetes

A Review of Selected Recent Developments and Their Impact on Treatment Harry Shamoon Division of Endocrinology and Diabetes Research Center, Albert Einstein College of Medicine of Yeshiva University, Bronx, New York, USA

Summary

Recent developments in epidemiology, physiology,anatomy and molecular biology have greatly increased our knowledge of the aetiology and immunological mechanisms involved in diabetes mellitus . This understanding has, in turn, facilitated progress in the diagnosis and treatment of the disease. It is generally accepted that both genetic and environmental factors have a role in the pathogenesis of insulin- and non-insulin-dependent diabetes mellitus. The contribution of insulin resistance or decreased insulin secretion to the pathogenesis of non-insulin-dependent diabetes remains controversial but it is likely that both have a role to play. Counterregulatory hormones, principally adrenaline (epinephrine) and glucagon, prevent blood glucose levels falling to extreme levels by antagonising the effect of insulin during hypoglycaemia, and inducing hepatic glucose production. Patients with insulin-dependent diabetes frequently exhibit impaired glucose counterregulation and , although its aetiology is uncertain in some patients , intensification of insulin therapy per se has been implicated. Secondary failure of oral hypoglycaemic agents in patients with non-insulin-dependent diabetes is a major and often inevitable problem, necessitating combined use of sulphonylurea and insulin in most patients. Recently, new treatments for patients with diabetes have been developed , including insulin analogues administered by a variety of novel methods, pancreatic grafts and transplantation of islet cells. Although promising, the clinical viability of these techniques remains to be demonstrated.

The past 15 years have witnessed an enormous increase in our knowledge of diabetes. New techniques in epidemiology , whole body physiology, cellular anatomy, and molecular biology have emerged during this period, allowing insights into the mechanisms of disease, the glucose transport system, the insulin receptor, and glucoregulatory physiology. This review highlights selected areas of recent progress. In some instances, these advances have led to modalities of treatment or diagnosis not previously possible. Regardless of their applic-

ability at the present time, however, the future of clinical diabetes management will surely be dramatically altered.

1. Immunological Basis o/IDDM Our view of the natural history of insulindependent diabetes mellitus (lOOM) is evolving as it becomes clearer that this disease has a long preclinical period (Gorsuch et al. 1981 ; Srikanta et al. 1985). During the early phase, one or more im-

2

mune markers for the presence of 100M emerge. Because at least 80 to 90% of islet {3-cell function must be destroyed before hyperglycaemia ensues, even the most sensitive tests of insulin secretion remain normal despite ongoing immune destruction (Ward et al. 1988). It is now virtually certain that the susceptibility to 100M is linked to genes within the human leucocyte antigen (HLA) complex on the short arm of chromosome 6. The risk of developing 100M in relatives of the proband is estimated to be about 3% for parents, 7% for nonHLA-identical siblings, and 5% for children (Palmer & Lernmark 1990). In HLA-identical siblings, the concordance rate for 100M is 30 to 40%, suggesting that environmental factors probably play a significant role in its pathogenesis. The HLA complex consists of the class I region (HLA-E, E, C and B), the class II region (HLA-OR, OQ and OP), and class III region (complement C4 and C2 genes, properdin Bf gene). Numerous studies have shown that serological specificities for HLA-OR 3 and/or 4 are extremely frequent in 100M (90 vs 60% in the general population). Class II immune response gene products may control or modify the autoimmune reaction directed toward {3-cells in susceptible persons. As a result of linkage disequilibrium, other HLA class II molecules may be commonly associated with HLA-OR4. Recent advances in molecular genetics have made it possible to study the structure of the HLA class II molecule genes as well as determine their location. Using restriction length fragment polymorphism (RLFP) analysis, a number of fragments differing in frequency between HLA-ORidentical and control individuals have been detected with HLA-DQ-{3 chain gene probes. (The DQ molecule is a heterodimer made up of polymorphic ex and {3 chains which are present on the surfaces of (3 cells, macrophages and activated T cells. Its function appears to be the presentation of small peptides to T cells bearing the C04 marker. The C04 T cell recognises the peptide-class II molecule complex, and this then triggers an immunological cascade.) The net effect of these analyses was to demonstrate that HLA-OQ appears closer to the 100M

Drugs 44 (Suppl. 3) 1992

susceptibility gene than HLA-OR. However, further studies at the nucleotide sequence level demonstrated more heterogeneity, with OR4 being associated with several different OQ-{3 genes. These studies implicate an allelic variation of the {3-chain, OQw3.2 (Cohen-Haguenaur et al. 1985; Michelsen & Lernmark 1987; Owerbach et al. 1988; Todd et al. 1987). Recent studies have suggested that susceptibility to diabetes might be related to the amino acid in position 57 of DQ-{3, since aspartic acid in that position was rarely found in patients with diabetes. The NOO mouse model ofIOOM also lacks the protective aspartic acid residue at position 57 of its corresponding {3-chain (Acha-Orbea et al. 1987). Since all OQ-{3 sequences found to date in subjects with diabetes have also been found in nondiabetic controls, the major histocompatibility complex (MHC)-linked polymorphism is not a 'mutant' variant of the normal allele. In addition, the data suggest that OQw8 specificity confers the highest risk of 100M only among HLA-OR4 positive persons. On the other hand, 5% of patients with 100M have been found positive for the HLADQ-{3 3.1 allele, and in Japanese patients with 100M, the OQw4 and OQw9-{3 chains associated with diabetes are known to contain aspartic acid at position 57 (Aparicio et al. 1988). This implies that the search for a diabetes-linked susceptibility gene will not stop at the Asp 57 of the OQ-{3 molecule. As with other autoimmune diseases, it is presumed that type 1 diabetes occurs when T lymphocytes become activated by 'self' recognition antigens. Like all antigens recognised by T cells, these antigens are held in the cleft of an MHC molecule on the surface of T cells. These are Class I MHC molecules in the case of C08 T cells and Class II in the case of C04 T cells. The mechanism behind {3-cell specific autoimmunity is uncertain . One attractive hypothesis is that Class II molecules (not usually expressed by nonlymphoid cells) may be expressed by pancreatic {3-cells of susceptible individuals. It is possible that such expression may be mediated by cytokines such as IL-l, IL-2, interferon, TNF and others (Pujol-Borrell et al. 1987). Such cytokines have also been implicated in the

Pathophysiology of Diabetes

islet destruction or 'insulitis' phase of the disease (Palmer & Lernmark 1990). Clearly, cell-mediated immune mechanisms are likely to play the final role in ~-cell destruction. In animal models, genes both within the MHC complex as well as genes on other chromosomes are necessary for the development of diabetes. Although such independent, non-MHC-linked diabetogenic genes have not been identified in human subjects, recent reports suggest an association of human 100M with T cell receptor variants as well as with the insulin gene (Bell et al. 1983; Millward et al. 1988). It is of interest that C04-positive T cells, the most highly activated T cells present in the blood of patients with diabetes, have been cloned from a patient with 100M, and were found to have the capacity to destroy histocompatible pancreatic cells provided that the pancreatic cells had been cytokine-induced to express HLA Class II (De Berardinis et al. 1988). One powerful new tool in the study of the role of antigen expression on islet ~-cells is the transgenic mouse model. By engineering transgenic lines of mice that express gamma interferon or the simian virus T antigen, both diabetes and 'insulitis' seem to develop (Adams et al. 1987). Although these experiments suggest alternative hypotheses, it is still most likely that the 'diabetogenic' gene(s) in human subjects induce the disease by direct effects on the immune system. 'Insulitis' occurs promptly upon haploidentical pancreatic transplant into the diabetic recipient, and passive transfer of diabetes with lymphocytes into both diabetes-prone and resistant strains of BB rats have been documented. Prediction of the development of 100M is a primary goal, because of the need for the earliest possible prevention of autoimmune destruction if alternative therapies are to be used. For example, both cyclosporin (Canadian-European Randomized Control Trial Group 1988) and intensive insulin therapy (Shah et al. 1989) may preserve residual ~-cell function in recent onset 100M. There are now several such 'early' markers for 100M in human subjects. Cytoplasmic islet cell autoantibodies (lCAs) are IgGs that can be detected by specific binding to sections of human or animal pan-

3

creata (Colman et al. 1988). Recent international workshops have standardised these assays, a step that will resolve many conflicting results. Although the target of the ICAs is not known, some laboratories have demonstrated a remarkably high predictive value for a positive ICA test at almost 10 years of follow-up (Dib et al. 1986). Antibodies against insulin (lAA) can also be detected at diagnosis and have also been reported to have a high specificity and positive predictive power (Vardi et al. 1987). However, since the IAAs are likely to be a reflection of ongoing islet ~-cell destruction, the utility of this test may be questionable. More closely linked in time to the onset of overt diabetes are tests of islet ~-cell insulin reserve. Variants of intravenous glucose administration (either intravenous glucose tolerance tests or a modified minimal model analysis) suggest that diminished firstphase insulin secretion will also predict diabetes with a high degree of confidence (Wilson & Eisenbarth 1990). The use of all of the above tests in prediction of diabetes is a research subject of great importance, but clinical application remains a future prospect.

2. Contributions of {j-Cell, Muscle and Hepatic Defects in Type II Diabetes Despite being 10 to 15 times more common than 100M, the aetiology and pathogenesis of type II or non-insulin-dependent diabetes mellitus (NIOOM) remains obscure. The influence of a genetic predisposition is underscored by the virtually 100% concordance of the disease in identical twins. Nevertheless, there is evidence for considerable genetic heterogeneity and for the influence of environmental factors such as obesity. In more genetically isolated populations (such as the Pima Indians of the American Southwest), the prevalence of the disease may reach 60 to 80% in obese adults. In the absence of a clear delineation of the primary lesion in NIOOM, controversy continues as to whether decreased insulin secretion or insulin resistance is the principal factor in the pathogenesis of the disease. Recent studies in a subgroup of nondiabetic

Drugs 44 (Suppl . 3) 1992

4

Pima Indians suggest that insulin resistance may antedate the onset of diabetes (Lillioja et al. 1988). The change from normal to impaired glucose tolerance was associated with a decline in insulinmediated glucose uptake. Whether these findings can be extrapolated to other patients with NIDDM is uncertain because of the highly inbred nature of this population. However, studies of small groups of patients and first-degree relatives in other populations tend to confirm this suspicion . A recent report indicates that nondiabetic relatives of NIDDM probands were insulin resistant, largely as a result of impaired glycogen storage in muscle (see below) [Eriksson et al. 1989]. The magnitude of the defect in insulin action in patients with NIDDM is impressive. Most estimates obtained during euglycaemic insulin clamp experiments suggest that glucose disposal in

NIDDM is about half that in nondiabetic individuals (Golay et al. 1988). Both glucose oxidation and storage are impaired, although the quantitative contribution of the latter is likely to have a greater impact on fasting plasma glucose. In molecular terms, defective glucose disposal may reflect defects intrinsic to the glucose transport system (Kono et al. 1982). Reduced numbers and functional activity of glucose-transporter proteins have been demonstrated in patients with NIDDM (table I; Garvey et al. I987b). There may be multiple components to the transport defect, with both reversible and irreversible aspects (Garvey et al. 1989). Hyperglycaemia may provoke or worsen defects in glucose transport, as has been shown in vitro (Garvey et al. 1987a). Most studies, however, reveal that impaired glucose transport activity is not reversed completely by any of the known thera-

Table I. Glucose transporters in mammalian tissues (adapted from Kahn & Flier 1990) Nomenclatures Type

Function

Regulatory factors

Tissue with transporter present

in vitro

in vivo

Oncogenes . phorbol esters, growth factors , insulin, cAMP, protein kinase C, vanadate, sulphonylurea, glucose, dexamethasone, cellular differentiation

Fasting, refeeding, high-fat diet, genetic obesity (Zucker rat), insulin , exercise , developmental stage

GLUT 1

Erythrocyte

Basal uptake

Human RBCs, bloodbrain barrier, placenta, transformed cells in culture

GLUT 2

Liver

Liver, j3-cells, kidney, intestine (basolateral membrane)

Fasting, refeeding, streptozotocin diabetes

GLUT 3

Brain

Hepatic transport, j3-cell, basolateral epithelium of kidney and intestine Basal uptake

Developmental stage in muscle

GLUT 4

Muscle/fat

Insulin-stimulated uptake

Brain , placenta, kidney (low levels in adult skeletal muscle) White adipose tissue, brown fat, red and white muscle, heart, smooth muscle (?)

GLUT 5

Small intestine

Dietary absorption Small intestine, kidney, skeletal muscle , adipose tissue

a Glucose transporters in order in which they were cloned . cyclic adenosine monophosphate; RBCs Abbreviations: cAMP

=

= red blood

Cellular differentiation

cells .

Insulin, streptozotocin diabetes, fasting, refeeding, high-fat diet, genetic obesity (Zucker rat), exercise

Pathophysiology of Diabetes

peutic modalities such as weight reduction, oral drug therapy or insulin. It is of note that a defect in an insulin-responsive glucose transporter may be involved in NIDDM; a recent study in rats with streptozotocin-induced diabetes demonstrated depletion of- both the glucose transporter and its mRNA in adipose cells, and both effects were reversed by insulin (Kahn & Flier 1990). One of the issues to be resolved is whether these glucose transport defects in NIDDM occur in all or only some of the insulin-sensitive target tissues, i.e. muscle and fat, and, particularly, whether they occur in muscle since it plays a greater quantitative role in glucose disposal than fat. Over the long term , we will need to understand whether defects in the glucose transport system are causes of or consequences of diabetes. Instead of being due to hyperglycaemia, hyperinsulinaemia may reflect subtle impairment of insulin sensitivity . Defective insulin-stimulated glucose disposal (primarily due to nonoxidative glucose metabolism in muscle) is present early in the course ofNIDDM and, indeed, may be present in unaffected family members (Eriksson et al. 1989). Insulin resistance could result from defects at the level of the insulin receptor, activation of the receptor tyrosine kinase, or intracellular processes distal to these steps and as yet poorly understood. A decrease in tyrosine kinase activity in adipocyte insulin receptors from NIDDM patients has been demonstrated in some studies (Freidenberg et al. 1987). Other groups have found this defect in skeletal muscle insulin receptor preparations from obese, nondiabetic individuals; thus, reduced tyrosine kinase activity may not be unique to NIDDM (Caro et al. 1987). The discrepancies in some of the data may derive from the presence of subpopulations of insulin receptors capable of autophosphorylation and possibly affected in diabetes. With regard to the defect in the effector system (i.e. those steps in insulin action distal to the interaction of the hormone with the receptor), part of the action of insulin in stimulating glucose transport resides in its ability to recruit intracellular glucose transporters to the plasma membrane. In

5

patients with NIDDM, basal numbers of transporters and their recruitment by insulin appear to be subnormal. There is also a reduction in the functional activity of these transporters, as measured by cytochalasin-B binding (Ciaraldi et al. 1982). The mechanisms responsible for the depletion of glucose transporters and their structural defects remain unclear. Furthermore, there may be additional post-transport intracellular defects in glucose metabolism that contribute to abnormal glucose homeostasis (fig. I). Recent studies confirm the notion that defective insulin secretion must also playa critical role in the development of overt diabetes. Although fasting plasma insulin concentrations are frequently elevated in patients with NIDDM, particularly early in the disease course, insulin secretory defects can also be demonstrated (Porte 1991). In most patients with fasting hyperglycaemia, there is loss of firstphase insulin release in response to intravenous glucose (reminiscent of the earliest demonstrable secretory defect in 100M). The cellular mechanisms for this defect are not known, but abnormal function of the islet {3-cell glucokinase in transducing the signal or some aberration of the glucose transport system may playa role (Porte 1991). More recent studies in a cohort of 16 French families with maturity-onset diabetes of the young (MODY) have shown a strong linkage between glucokinase gene locus on chromosome 7p and diabetes (Froguel et al. 1992). New insights into the molecular mechanisms which regulate insulin secretion have also been recently generated. For example, the glucose 'sensing' of the {3-cell may be mediated by the GLUT-2 transporter, and this transporter is reduced in activity in models of both type I and type 2 diabetes (Unger 1991). Studies by Weir and colleagues (1986) have shown that the simple reduction of {3-cell mass in normal animals, which is not immediately diabetogenic, may increase plasma glucose by a minute extent and induce abnormal insulin secretory dynamics reminiscent ofNIDDM (Weir et al. 1986). Thus, it is likely that both conditions - insulin resistance and defective insulin secretion - must exist for the development of diabetes.

6

Drugs 44 (Suppl . 3) 1992

FFA

Gluconeogen ic subs trates (alanine, lactate)

LIVER

c:=:J

Glucose Uptake

1 \.

Glucose Oul pul

High Blood Glucose

MUSCLE

Glucose Transpor l

r---1

,

.----.--,- :'j IJ·CELL

Fig. 1. Interactions between impaired insulin secret ion, hepatic glucose overproduction, and impaired muscle glucose uptake in the pathogenesis of non-insulin-dependent diabetes mellitus. Gluconeogenic substrate flux is partly responsible for glucose overproduction, and hyperglycaemia in turn may further worsen impaired l3-cell insulin secretion .

3. Counterregulatory Hormones and Hypoglycaemia in IDDM During the past decade, the clinical management of diabetes has attempted to achieve near normalisation of glycaemia with intensified insulin therapy and lower glycaemic goals in IDDM and NIDDM patients. The consequence of such therapy has been an increase in the frequency and severity of hypoglycaemia, quantified at least for IDDM patients in the preliminary results from the Diabetes Control and Complication Trial (DCCT) [DCCT Research Group, 1987]. The data from the DCCT regarding severe hypoglycaemia (i.e. that associated with seizures, coma or loss of consciousness) indicate an incidence of about 17 episodes per 100 patient-years in patients treated conventionally and a 3-fold higher incidence in the group receiving intensive insulin therapy. These may indeed be minimum estimates, given the selection bias toward healthier patients with a lesser duration of diabetes. Despite advances in the monitoring of blood glucose and insulin delivery systems, hypoglycaemia remains the major obstacle in safely achieving close metabolic control. The plasma glucose threshold for release of

counterregulatory hormones in healthy persons ranges between 2.8 and 3.9 mmol/L, although in some patients with diabetes or chronic hypoglycaemic states these thresholds may be reduced (Amiel et al. 1988; Davis & Shamoon 1991).Hypoglycaemia and counterregulatory hormone responses appear to occur at plasma glucose levels below 2.8 mmol/L in intensively treated patients with type I diabetes (Amiel et al. 1988). Conversely, poorly controlled subjects with type I diabetes appear to exhibit both symptoms and counterregulatory hormone release at higher than normal plasma glucose concentrations (Amiel et al. 1988; Boyle et al. 1988), suggesting that antecedent changes in plasma glucose concentration may influence subsequent reduction of the plasma glucose level. Thus, brief hypoglycaemia seems to reduce the magnitude of the counterregulatory response to subsequent episodes of hypoglycaemia (Davis & Shamoon 1989; Heller & Cryer 1991; fig. 2). From the foregoing, it can be appreciated that glucose counterregulatory hormone secretion seems to be triggered at plasma glucose concentrations higher than the level producing symptoms of hypoglycaemia (Mitrakou et at. 1991). This observation establishes the relative importance of glucose

7

Pathophysiology of Diabetes

Iatrogenic mild hypoglycaemia

Severe hypoglycaemia

L~.r g,.t.ShOldS/I ~ fucose for counterregulatory h t' ormone secre Ion

Decreased awareness

p ~ . ceunterregulatlen by liver

~I/ Decreased plasma glucagon and adrenaline levels during hypoglycaemia

Fig. 2. Proposed mechanisms underlying severe hypoglycaemia in insulin-dependent diabetes mellitus include the existence of lowered glycaemic thresholds for symptoms of hypoglycaernia and the secretion of counterregulatory hormones, both of which could be traced to antecedent hypoglycaemia. Impaired hepatic glucose release ultimately accounts for the failure of blood glucose to be restored in time to prevent neuroglycopenia.

counterregulatory mechanisms in preventing a dangerous decline in blood glucose to levels that threaten the brain. Impaired glucose counterregulation may be defined as the failure to restore plasma glucose concentrations to normal or to prevent any further decline. Counterregulatory hormone secretion during hypoglycaemia in humans has been extensively reviewed (Cryer 1981 ; Gerich & Campbell 1988). The suppression of hepatic glucose release and increased peripheral utilisation of glucose induced by insulin - the inciting factors leading to hypoglycaemia - are partly reversed by declining plasma insulin concentrations. This mechanism is critical to the restoration of the plasma glucose concentration following hypoglycaemia in insulin-treated diabetes (Cryer & Gerich 1985). Counterregulatory hormones can overcome the effects of insulin even when the plasma insulin concentration does not decline (Bolli et al. 1983). Among the counterregulatory hormones, adrenaline (epinephrine) [released from the adrenal medullae] and glucagon (secreted by the alpha cells of the Islets of Langer-

hans) appear to be the most potent insulin antagonists. These hormones are promptly secreted after plasma glucose levels fall, and both induce a rapid increase in hepatic glucose production (Gerich et al. 1980; Shamoon et al. 1981). The contributions of glucagon and adrenaline to glycaemic recovery are, in part, redundant. Acutely, both hormones trigger glycogenolysis and gluconeogenesis. When both hormones are deficient, defects in counterregulation are dramatic. This is best demonstrated in patients with type 1 diabetes, most of whom have a marked reduction in glucagon secretion during hypoglycaemia. Conversely, glucose counterregulation appears to be normal in patients after adrenalectomy with adequate glucocorticoid replacement and in subjects receiving combined (¥- and {3adrenergic antagonists. Finally, in many patients with IDDM of greater than a few years' duration, adrenaline secretion in response to hypoglycaemia is predominant; the combined deficiencies of both glucagon and adrenaline are most likely to be responsible for the greater propensity for severe and/ or prolonged hypoglycaemia in such patients (Cryer

8

1981 ; Gerich & Campbell 1988; Kleinbaum & Shamoon 1983). The contributions of the other major counterregulatory hormones seem to be less critical for the initial response « 30 to 60 minutes) but are nonetheless important in the long term . Cortisol and growth hormone both have significant late effects on the ability of the liver to sustain glucose output in the face of hyperinsulinaemia (De Feo et al. 1989a,b). In addition, these hormones reduce peripheral glucose utilisation during recovery from hypoglycaemia, an action shared by adrenaline. These effects are direct and indirect (e.g. by stimulation of free fatty acid release). In patients with 100M and impaired hepatic glucose release, the reduction in peripheral glucose uptake is vital to glycaemic recovery from hypoglycaemia. Impaired glucose counterregulation has emerged as a clinical syndrome of considerable importance in 100M patients. As noted above, combined defects in the secretion of glucagon and adrenaline lead to inadequate hepatic glucose production during recovery from insulin-induced hypoglycaemia. The aetiology of these hormonal defects is uncertain. Although some patients clearly suffer from autonomic neuropathy (implicated in defective adrenergic responses to hypoglycaemia), most patients with impaired glucose counterregulation do not exhibit other clinical symptoms of autonomic neuropathy. There is evidence that both the defect in glucagon secretion and adrenaline release may be specific to the hypoglycaemic stimulus per se and that both hormones can be secreted by other stimuli (Gerich et al. 1973; Hirsch & Shamoon 1987). The recent observations that intensive insulin therapy in patients with diabetes further impairs adrenaline secretion (as well as the secretory responses of cortisol and growth hormone) suggests that 'functional' defects in the hormonal responses to hypoglycaemia may be common (Simonson et ~I. 1985). A clinical hallmark of impaired glucose counterregulation is 'hypoglycaemic unawareness'. In view of the deficient adrenaline response in such patients, the lack of adrenergic warning symptoms exacerbates the risk of prolonged and/or more severe hypoglycaemia. It is of interest that not all of

Drugs 44 (Suppl. 3) /992

the adrenergic symptoms of hypoglycaemia can be attributed to adrenaline secretion, suggesting that there may be defects in the sympathetic response to hypoglycaemia other than reduced adrenomedullary adrenaline secretion.

4. Secondary Failure of Oral Antidiabetic Therapy A major problem in the long term management of NIOOM is so-called secondary failure of oral hypoglycaemic agents. Although the number of patients with secondary failures seems to increase with a longer duration of diabetes, various studies have yielded conflicting estimates of the frequency of this phenomenon (Groop et al. 1989). This may stem from inaccuracies in the diagnosis of the type of diabetes or the lack of a common definition of secondary failure. Patient factors such as diet, stress, and illness may contribute to the decreased responsiveness to drugs. Alternatively, worsening of {J-cell function over time could account for metabolic deterioration. Since most patients with NIOOM continue to secrete at least some insulin even late in the disease (Lomasky et al. 1990), it is possible that further defects in insulin action might also play a role. Recent studies in small groups of patients with NIOOM who no longer responded to drug therapy suggest that impaired nonoxidative glucose metabolism together with a further deficiency in insulin secretion and augmented hepatic glucose production accounted for the loss of responsiveness (Groop et aI. 1989). It is interesting to note that dietary intervention late in the course ofNIOOM can also ameliorate metabolic control, as has also been shown for intensive insulin treatment and combination therapy with insulin and sulphonylureas. A number of studies have documented that improvement of hyperglycaemia can improve residual islet {J-cell function (Foley et al. 1983). These observations suggest that, while secondary drug failure may be inevitable in most patients with NIDOM, this may reflect the inherent difficulty in achieving truly 'normal' glycaemic patterns even in the early phases of

Pathophysiology of Diabetes

NIOOM, rather than a preprogrammed deterioration of islet (j-cell function. One approach to the treatment of 'sulphonylurea failure' has been the use of combined sulphonylurea and insulin therapy. The rationale for this combination therapy is the heterogeneity of patients with NIOOM, all of whom suffer from some combination of insulin deficiency and insulin secretion (Banerji & Lebovitz 1989). Since dietary management alone fails to normalise insulin secretion and activity in most patients, the addition of sulphonylurea agents is required in the majority of patients with NIDOM in the US. The precise proportion in whom insulin therapy is ultimately required is not known, but is probably > 50% (Lebovitz 1990). Conversely, insulin treatment is not without complications in NIOOM and does not guarantee adequate regulation of carbohydrate and fat metabolism in those NIOOM patients with obesity and significant insulin resistance. Since some evidence supports an extrapancreatic, insulin-sensitising effect of sulphonylurea drugs, a variety of studies have examined the use of combined treatment (Groop et al. 1990; Lebovitz & Pasmantier 1990). In general, results in small studies have shown a modest improvement in glycaemia compared with the effects of one agent alone. However, when counterbalanced against the need for lower insulin doses in these patients, addition of sulphonylurea to insulin may be helpful. However, only long term randomised clinical trials of such an approach will yield important clues for the treatment of selected patients. More studies will also be needed to establish the potential long term deleterious effects of insulin. One recent study confirms that atherosclerosis-like lesions can develop in normal rats made hyperinsulinaemic for I year (Sato et al. 1989).

5. New Treatment Approaches in [DDM A variety of promising new approaches to the treatment of 100M have appeared in the past few years. Based on the knowledge of the molecular structure of insulin, recombinant DNA technology has been applied to develop insulin analogues with

9

substituted amino acids. Since the insulin in commercially produced soluble insulin preparations is polymeric, it has been suggested that monomeric insulin would be more rapidly absorbed from subcutaneous injection sites. One such analogue, in which the B9 and B27 residues were substituted, resulted in more rapid absorption and greater hypoglycaemic potency in vivo (Vora et al. 1988).The absorbability of nasally administered insulin has also been enhanced. These approaches will probably come into clinical use in the near future, especially as developments in alternative methods of insulin delivery continue to be slow. Although reliable long term implantable glucose sensors have yet to be developed , programmable insulin delivery systems have been showing promising results. One recent study examined the effect of an implanted, externally programmable insulin infusion system for I year in a multicentre trial (Point Study Group 1988). The major technical hurdles of catheter obstruction by precipitated insulin or clot and the long range reliability of such systems appear to be largely resolved. With respect to transplantation, there have been recent promising developments in both whole pancreas grafts as well as islet cell transplantation. With the advances in surgical techniques for whole pancreas grafts in the previous decade, it has now become possible to see the benefits of combined kidney and pancreas transplants . In recent studies from one group, the survival rate of both grafts was 83% after 2 years (Sollinger et al. 1988). At the very least, the near normal metabolic control possible with functioning islet tissue in 100M with kidney transplants suggests that hyperglycaemia-induced failure of the transplanted kidney may be preventable (Mauer et al. 1989). In the long run, however, whole pancreatic graft transplantation may be replaced by islet transplantation with its (theoretically) lower morbidity. Islet transplantation continues to be a major goal of several laboratories, and while there have been no entirely successful islet transplants in human subjects, progress has been made (Weir et al. 1990). Adequate long term function of islet autografts in dogs and monkeys (for up to 18 months after he-

Drugs 44 (Suppl. 3) 1992

10

patic implantation) suggeststhat transplanted islets could normal ise glucose metabolism in human subjects with diabetes (Gray et al. 1987; Kneteman et al. 1987; Warnock et al. 1989, 1991). In human subjects, the clinical application of these techniques has lagged. In part, this problem relates to how much ~-cell mass is needed for adequate glycaemic control and the amounts of purified islets that can be isolated from one cadaver donor. On the basis of studies from segmental pancreatic transplants, it would seem that 50%of normal islet mass may be needed to normalise glucose dynamics. A variety of immune, non-immune (e.g. hyperglycaemia, anoxia) and mechanical considerations will determine the ultimate viability of transplanted islets and whether enough will survive the transplantation process. A number of laboratories have reported isolation and purification techniques that can yield functioning human islets (Fujioka et al. 1990; Ricordi et al. 1988). Some of the immunological barriers may indeed be overcome by the use of pooled multiple human donors (Gotoh et al. 1988), which would also help deal with the problem of sufficient quantities of islets for transplantation. It has been recently reported that frozen islets can survive in adequate numbers (Kneteman et al. 1989), suggesting that cryopreservation may also help to overcome the barrier oflimited supply. Recent reports of patients in whom insulin could be withdrawn after islet cell transplantation, though only briefly, are very encouraging (Scharp et al. 1990; Tzakis et al. 1990). Whether islet survival in subsequent patients can sustain glucose homeostasis long enough to become a clinically viable treatment remains to be determined.

References Acha-Orbea H, McDevitt HO. The first external domain of the nonobese diabetic mouse class II I-A {J chain is unique. Proceedings of the National Academy of Sciences of the USA 84: 2435, 1987 Adams E, Alpert S, Hanahan D. Non-tolerance and autoantibodies to a transgenic self-antigen expressed in pancreatic {J cells. Nature 325: 233, 1987 Amiel SA, Sherwin RS, Simonson DC. et al. Effect of intensive insulin therapy on glycemic thresholds for counterregulatory hormone release. Diabetes 37: 901, 1988 Aparicio JMR . Wakisaka A. Takada A. HLA-DQ system and insulin dependent diabetes mellitus in Japanese : does it contrib-

ute to the development of IDDM as it does in caucasians? Immunogenetics 28: 240-246, 1988 Banerji MA, Lebovitz HE. Insulin-sensitive and insulin resistant variants in NIODM . Diabetes 38: 784-792, 1989 Bell GI , Horita S, Karam JH . A polymorphic locus near the human insulin gene is associated with insulin dependent diabetes mellitus. Diabetes 33: 176-183, 1983 Bolli G, De Feo P, Compagnucci P, et al. Abnormal glucose counterregulation in insulin-dependent diabetes mellitus: interaction of anti-insulin antibodies and impaired glucagon and epinephrine secretion. Diabetes 32: 134, 1983 Boyle P, Schwartz NS, Shah et al. Plasma glucose concentrations at the onset of hypoglycemicsymptoms in patients with poorly controlled diabetes and in non-diabetics. New England Journal of Medicine 318: 1487, 1988 Canadian-European Randomized Control Trial Group . Cyclesporin-induced remission of IDDM after early intervention. Diabetes 37: 1574-1582, 1988 Caro JF, Sinha MK, Raju SM, et al. Insulin receptor kinase in human skeletal muscle from obese subjects with and without NIODM. Journal of Clinical Investigation 79: 1330-1337, 1987 Ciaraldi TP, Kolterman OG, Scarlett JA, et al. Role of glucose transport system in the post receptor defect of NIDDM. Diabetes 31: 1016-1022, 1982 Cohen-Haguenaur 0 , Robbins E, Massari C, et al. A systemic study of HCA class II {J DNA restriction fragments in insulin dependent diabetes mellitus. Proceedings of the National Academy of Sciences of the United States of America 82: 3335, 1985 Colman PG, Tautkus M, Rabizadeh A, Cahill C, Einsenbarth GS. Assay for islet cell antibodies with rat pancreas and peroxidase conjugated protein A. Diabetes Care II : 367-368, 1988 Cryer PE. Glucose counterregulation in man. Diabetes 30: 261, 1981 Cryer PE, Gerich JE. Glucose counterregulation , hypoglycemia and intensive insulin therap y in diabetes mellitus. New England Journal of Medicine 313: 232, 1985 Davis MR, Shamoon H. Adaptive counterregulatory responses to hypoglycemia in nondiabetic humans . Diabetes 38: 5A, 1989 Davis MR, Shamoon H. Deficient counterregulatory hormone responses during hypoglycemia in a patient with insulinoma. Journal of Clinical Endocrinology and Metabolism 72: 788792, 1991 De Berardinis P, Londei M, James RFL, et al. Do CD4-positiv e cytotoxic T cells damage islet {J cells in type I diabetes? Lancet 2: 823-824, 1988 DCCT Research Group . The Diabetes Control and Complicat ion Trial: result of a feasibility study. Diabetes Care 10: 1-19, 1987 De Feo P, Perriello G, Torlone E, et al. Contribution of cortisol to glucose counterregulation in humans . American Journal of Physiology 20: E35, 1989b De Feo P, Perriello G, Torlone E, et al. Demonstration of a role for growth hormone in glucose counterregulation . American Journal of Physiology 19: E835, 1989a Dib S, Vardi P, Connelly J, Eisenbarth GS, Soeldner JS. Absence of transient immunologic or metabolic abnormalities in short term discordant monozygotic twins for IODM. British Medical Journal 292: 1670, 1986 Erikson J, Franssila-Kallunki A, Ekstrand A, Saloranta C, Widen E, et al. Early metabolic defects in persons at increased risk for non-insulin dependent diabetes mellitus. New England Journal of Medicine 321: 337-343, 1989 Foley JE, Kashiwagi A, Verso MA, et al. Improvement in in vitro insulin action after I month of insulin therapy in obese NIODM . Journal of Clinical Investigation 72: 1901-1909, 1983 Freidenberg GR , Henry RR, Klein HH, et al. Decreased kinase activity of insulin receptors from adipocytes of non-insulin dependent diabetic subjects. Journal of Clinical Investigation 79: 240-250, 1987

Pathophysiology of Diabetes

Fujioka T, Terasaki PI, Heintz R, et al. Rapid purification of islets using magnetic microspheres coated with anti-acinar cell monoclonal antibodies. Transplantation 49: 404-407, 1990 Garvey WT, Hueckstead t TP, Birnbaum MJ. Pretranslational suppression of an insulin-responsive glucose transporter in rats with diabetes mellitus. Science 245: 60-63, 1989 Garve y WT, Huecksteadt TP, Matthaei S, Olefsky JM. The role of glucose transporters in the cellular insulin resistance of type II noninsulin-dependent diabetes mellitus. Journal of Clinical Investigation 81: 1528-1536, 1987b Garvey L, Olefsky JM, Matthaei S, Marshall S. Glucose and insulin coregulate the glucose transport system in primary cultured adipocytes: a new mechanism of insulin resistance. Journal of Biological Chemistry 262: 189-197, 1987a Ger ich JE, Campbell PJ. Overview of counterregulation and its abnormalities in diabetes mellitus and other conditions. Diabetes/Metabolism Reviews 4: 93, 1988 Gerich JE, Langlois M, Noacco C, et al. Lack of glucagon response to hypoglycemia in diabetes: evidence for an intrinsi c pancreatic alpha-cell defect. Science 182: 171, 1973 Gerich JE, Rizza R, Haymond M, et al. Hormon e mechanisms in acute glucose counterregulation: the relative roles of glucagon, epinephrine, norepinephrine growth hormones and cortisol. Metabolism 29 (Suppl. I): 1164, 1980 Golay A, Felber JP , Jequier E, DeFronzo RA, Ferrannini E. Metabolic basis of obesity and noninsulin-dependent diabetes mellitus. Diabetes/Metabolism Reviews 4: 727-747, 1988 Gorsuch AN, Spencer KM, Lister J, et al. Evidence for a long prediabetic period in type I (insulin-dependent) diabetes mellitus. Lancet I: 1363-1365, 1981 Gotoh M, Porter J, Kanai T, et al. Multiple donor allotransplantation : a new approach to pancreatic islet transplantation. Transplantation 45: 1008-1012, 1988 Gray DW, Warnock GL, Sulton R, et al. Successful autotransplantation of isolated islets of Langerhands in the cynomolgus monkey. Transplantation Proceedings 19: 965-966, 1987 Groop L, Schalin C, Franssila-Kallunki A, et al. Characteristics of non-insulin-dependent diabetic patients with secondary failure to oral ant i-diabetic therapy . American Journal of Medicine 87: 183-190, 1989 Groop LC, Gray PH, Stenman S. Combined insulin-sulfonylurea therapy in treatment ofNIDDM. Diabetes Care 13 (Suppl. 3): 47-52, 1990 Heller S, Cryer PE. Reduced neuroendocrine and symptomatic responses to subsequent hypoglycemia after I episode of hypoglycemia in nondiabetic humans . Diabetes 40: 223-226, 1991 Hirsch BR, Shamoon H. Defective epinephrine and growth hormone responses in IDDM are stimulus-specific. Diabetes 36: 20, 1987 Kahn BB, Flier JS. Regulation of glucose transporter gene expression in vitro and in vivo. Diabetes Care 13: 548-564, 1990 K1einbaum J, Shamoon H. Impaired counterregulation ofhypogIycemia in insulin-dependent diabetes mellitus. Diabetes 32: 493, 1983 Kneternan NM, Alderson D, Scharp DW. Cyclosporine A immunosuppression of allotransplanted canine pancreati c islets. Transplantation Proceedings 19: 950-951, 1987 Kneteman NM, Alderson D, Scharp DW. Prolonged cryopreservation of purified human pancreatic islets. Diabetes 38: 176178,1989 Kono T, Robinson FW, Blerins FW, et al. Evidence that translocation of the glucose transport activity is the major mechanism of insulin action on glucose transporter in fat cells. Journal of Biological Chemistry 257: 19042-19047, 1982 ' .ebovitz HE, Diabetes Mellitus: Theory and Practice, Rifkin H, Porte D (Eds), 4th ed., pp. 554-574, Elsevier, New York, 1990 i...ebovitz HE, Pasmantier R. Combination insulin sulfonylurea therapy . Diabetes Care 13: 667-675, 1990 Lillioja S, Molt DM, Howard BV, Bennett PH, et al. Impaired

11

glucose tolerance of insulin action: longitudinal and cross-sectional studies in Pima Indians . New England Journal of Medicine 318: 1217-1225, 1988 Lomasky S, D'Eramo G, Fleischer N, Shamoon H. Insulin response to oral glucose predicts glycemic outcome of diet in type II diabetes. Archives of Internal Medicine 150: 169-172, 1990 Mauer SM, Goetz FC, McHugh LE, Sutherland DER, Barbosa J, et al. Long-term study of normal kidneys transplanted into patients with type I diabetes. Diabetes 38: 516-523, 1989 Michelsen B, Lernmark A. Molecular cloning of a polymorphic DNA endonuclease fragment associated IDDM with HLA-DQ. Journal of Clinical Investigation 79: I 144-I 152, 1987 Millward BA, Welsh KI, Leslie RDG, et al. Susceptibility to insulin dependent diabetes is associated with a T cell receptor B chain polymorphism and HLA-DR. Diabetes 37: 26A, 1988 Mitrakou A, Ryan C, Veneman T, Mokan M, Jenssen T, et al. Hierarchy of glycemic thresholds for activation of counterregulatory hormone secretion, symptoms and cerebral dysfunction. American Journal of Physiology 260: E67-E74, 1991 Owerbach D, Lernmark A, Platz P, et al. HCA-D region B chain DNA endonuclease fragments differ between HLA-DR identical healthy and insulin-dependent diabetic individuals . Nature 303: 815-8I9, 1988 Palmer JP , Lernmark A. Pathophysiology of type I diabetes. In Rifkin Hand, Pore D (Eds). Diabetes mellitus: theory and practice, 4th ed., Elsevier, New York, 1990 Point Study Group . One year trial of a remote-controlled implantable insulin infusion system in type I diabetic patients. Lancet 2: 866-869, 1988 Porte D. B-cells in type II diabetes mellitus. Diabetes 40: 166180,1991 Pujol-Borrell R, Todd I, Doshi M, et al. HLA class II induction in human islet cells by interferongamma plus tumour necrosis factor or Iymphotoxin . Nature 326: 304-306, 1987 Ricordi C, Lacey PE, Finke EH, et al. Automated method for isolation of human pancreatic islets. Diabetes 37: 413-420, 1988 Sato Y, Shiraishi S, Oshida Y, et al. Experimental atherosclerosislike lesions induced by hyperinsulinism in Wistar rats. Diabetes 38: 9 I-96, 1989 Scharp DW, Lacy PE, Santiago IV, McCullough CS, et al. Insulin independence after islet transplantation into type I diabetic patient. Diabetes 39: 515-5I8, 1990 Shah SC, Malone JI, Simpson NE. A randomized trial of intensive insulin therapy in newly diagnosed insulin-dependent diabetes mellitus. New England Journal of Medicine 320: 550554, 1989 Shamoon H, Hendler R, Sherwin R. Synergisticinteractions among anti-insulin hormones in the pathogenesis of stress hyperglycemia in hormones. Journal of Clinical Endocrinology and Metabolism 52: 1235, 1981 Simonson DC, Tamborlane WV, DeFronzo RA, et al. Intensive insulin therapy reduces counterregulatory hormone responses to hypoglycemia in patients with type I diabetes. Annals of Internal Medicine 103: 184, 1985 Sollinger HW, Stratta RJ, D'Alessandro AM, et al. Experience with simultaneous pancreas-kidney transplantation. Annals of Surgery 208: 475-482, 1988 Srikanta S, Ganda 0, Rabizadeh A, Soeldner JS, Eisenbarth GS. First-degree relatives of patients with type I diabetes mellitus. Islet-cell antibodies and abnormal insulin secretion. New England Journal of Medicine 313: 461-464, 1985 Todd JA, Bell JI, McDevitt HO. HLA-DQ gene contributes to susceptibility and resistance to insulin dependent diabetes mellitus. Nature 329: 599-604, 1987 Tzakis A, Ricordi C, Alejandro R, Zeng Y, Fung JJ , et al. Pancreatic islet transplantation after upper abdominal exenteration and liver replacement. Lancet 336: 402-405, 1990

12

Unger RH. Diabetic hyperglycaemia : link to impaired glucose transport in pancreatic tl-cells. Science 251: 1200-1205, 1991 Vardi P, Dibsa SA, Tuttleman M, Connelly JE, Grinbergs M, et al. Competitive insulin autoantibody assay: prospective evaluation of subjects at high risk for development of type I diabetes mellitus. Diabetes 36: 1286-1291, 1987 Vora JP , Owens DR, Dolben J, et al. Recombinant DNA derived monomenic insulin analogue: comparison with soluble human insulin in normal subjects. British Medical Journal 297: 12361239, 1988 Ward WK, Wallum BJ, Beard JC, et al. Reduction of glycemic potentiation. Sensitive indicator of beta-cell loss in partially pancreatectomized dogs. Diabetes 37: 723-729, 1988 Warnock GL, Kneteman NM, Ryan EA, et al. Continued function of pancreatic islets after transplantation in type I diabetes. Lancet 2: 570-572, 1989 Warnock GL, Kneteman NM, Ryan E, Sellis RGA, Robinovitch

Drugs 44 (Supp/. 3) /992

A, et al. Normoglycemia after transplantation of freshly isolated and cryopreserved pancreatic islets in type I (insulindependent) diabetes mellitus. Diabetologia 34: 55-58, 1991 Weir GC, Bonner-Weir S, Leahy JL. Islet mass and functions in diabetes and transplantation. Diabetes 39: 401-405, 1990 Weir GC, Leaky JL, Bonner-Weir S. Experimental reduction of B-cell mass: implications for the pathogenesis of diabetes. Diabetes/Metabolism Reviews 2: 125-161, 1986 Wilson K, Eisenbarth GS. Immunopathogenesis and immunotherapy of type I diabetes . Annual Review of Medicine 41: 497-508, 1990

Correspondence and reprints : Prof. Harry Sham oon, Diabetes Research Center, Albert Einstein College of Medicine, 1300 Morris Park Ave, Bronx, NY 10461, USA.