Insulin does not aggregate its own receptor

Diabetologia, Apr 1983

K. R. Lyen

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

Insulin does not aggregate its own receptor

Exercise induced proteinuria in diabetic chil- dren. Diabetologia L e t t e r s to the E d i t o r Insulin Does Not Aggregate Its Own Receptor 0 1 0 Dr. G. C. Viberti Unit for Metabolic Medicine Guy's Hospital Medical School London SE1 9RT, U K 1 Yours sincerely , D. Mackintosh, G. C. Viberti and H. Keen Dear Sir, We would like to make a few comments of a technical nature concerning the statements in the letter by Dr. Simpson in the August 1982 issue [ 1 ]. When assessing proteinuria as a marker of glomerular function, it is essential to quantitate specifically and accurately one or more proteins, the renal handling characteristics of which are known and understood. Albumin is the obvious choice as it is the major filtered plasma protein in normal urine, and in mildly elevated 'Albustix'-negative proteinuria (total protein excretion < 0.5 g/24 h) comprises 7%-15% of the total. Moreover, in 'Albustix'-positive clinical proteinuria (total protein excretion > 0.5 g/24 h) albumin is the major protein component of any origin, representing 40%-50% of the total [23]. The filtration of albumin is greatly restricted by the glomernlus. In the rat, the plasma concentration is about 40 g/l, that found in the Bowman space is only 15 mg/1. O f this approximately 90% is reabsorbed by the proximal tubules [ 4 ]. However, the reabsorption process is easily saturated by relatively small increases in filtration [ 5 ]. Even under conditions of non-saturation the excretion of albumin in the urine should change in proportion to the amount filtered if, as seems likely, the reabsorption o f proteins is proportional to the filtered load [ 6 ]. Therefore, as a general rule, urinary excretion reflects glomernlar filtration. Using sodium dodecyl sulphate polyacrylamide gel electrophoresis to establish 'patterns' of proteins (unidentified save by molecular weight) in normal urine, and expecting t o make definitive statements concerning the properties of the glomerular filter, is unrealistic. Though this technique has been used in 'clinical' proteinuria to provide information supplementary to albumin measurements, one must be extremely cautious when applying this technique to normal or mildly elevated proteinuria. Under these circumstances, by far the greatest proportion of the protein reaches the urine by post-glomerular secretion and has absolutely no relevance to glomerular filtration or the nature of the glomerular barrier. These secreted proteins include the dimeric secretory form of IgA (molecular weight 385,000) believed to be secreted by the tubules; urokinase, a fibrinolytic enzyme (molecular weight 55,000) found only in urine and not in plasma; B-galactosidase (molecular weight 230,000), found in normal urine in similar quantities to albumin; and, most important of all, the TammHorsfall mucoprotein, which alone comprises 30%-50% of the total protein in normal urine [ 7 ]. This is a very large molecule (molecular weight 23,000,000) excreted by the distal convoluted tubules and the loops of Henle [ 8 ]. This protein has been shown to be disaggregated into individual subunits with a molecular weight of 80,000-100,000 by a number of agents including sodium dodecyl sulphate [ 9 ]. Normal urine also contains a large proportion of secreted heterogenous mucoproteins which are yet to be adequately characterised. We must stress, therefore, that without further analysis of these high molecular weight proteins (170,000-215,000), noted by Dr. Simpson in the urine of different groups of subjects with otherwise normal kidney function, he cannot make inferences about changes at the glomerular filtration level. Diabetic kidney disease is primarily a glomernlopathy, and proteinuria has recognised clinical importance as an early indicator of this pathological process. Thus in a study of diabetic proteinuria the main line of approach should involve methods, the results of which can be interpreted unambiguously to provide insights into the nature o f the glomernlar lesion(s), leaving other techniques as a possible supplementary approach. Dear Sir, The concept that insulin can aggregate its own receptor is used by Jeffrey [ 1 ] to formulate a hypothesis of the mechanism of insulin action. However, experiments [ 2 ] using insulin labelled with ferritin as an ultrastructural marker for the insulin receptor, do not support the idea of hormone-induced receptor aggregation. The validity of monomeric ferritin-insulin (Fm-I) as a receptor marker is based on studies [ 3 ] demonstrating that it binds specifically to the insulin receptor, is equivalent in immunological reactivity to insulin, and stimulates glucose oxidation to the same extent as native insulin. Fm-I occupied receptor sites on adipocyte plasma membranes occur in the natural state as groups of two to six molecules [ 4 ]. Pre-fixation of the membranes before the addition of Fm-I does not change the pattern of receptor distribution [ 4 ], and therefore argues against a ligand-induced rec'eptor aggregation. In contrast to adipocyte membranes, Fm-I occupied receptors on liver membranes exist predominantly as single sites [ 2 ]. This distribution pattern is not altered over a fourfold range of Fm-I concentration [ 5 ], thus further upholding the counter-proposal that insulin does not aggregate its own receptor. The fluorescence microscopy observations [ 6 ] that are often quoted in support of the notion of receptor aggregation require re-appraisal. Unlike Fm-I which has nearly 100% of insulin's activity, rhodamine conjugated insulin (rhodamine-lactalbumin-insulin) used for these studies retains < 2% of the activity of native insulin [ 7 ]. It is therefore difficult to see how one can arrive at physiologically significant conclusions from these experiments. Furthermore, the fluorescence is visualized using image intensification microscopy, and the clustering observed could be an order of magnitude greater than the micro-aggregation seen by electron microscopy. Hence the data obtained with rhodamine-insulin might not be comparable to that of Fro-I, and molecular models of receptor aggregation [ 1 ] may not be applicably to these fluorescence studies. Jeffrey [ 1 ] also cites the work of Kahn et al. [ 8 ] to support the tenet of insulin-induced receptor aggregation. Monovalent Fab' fragments of anti-receptor antibody were shown to bind to the insulin receptor but were unable to stimulate glucose oxidation. Addition of anti-Fab' fragments restored the ability of the Fab' fragments to stimulate glucose oxidation. Kahn et al. concluded [ 8 ] that anti-receptor antibody acts by aggregating the insulin receptor. However, their data may have alternative interpretations. For instance, once can postulate that there are separate binding and biological response sites for the insulin receptor, with corresponding complementary sites for the insulin molecule and the anti-receptor antibody. When the anti-receptor antibody is split into its Fab' fragments, a minor conformational change might occur, resulting in the retention of binding activity, but loss of the ability to stimulate biochemical activity. The addition of anti-Fab' antibody can perhaps expose the sites for biological activity on the Fab' fragments by restoring the original conformation. There is then no need to invoke receptor aggregation to explain the biochemical activity of the anti-receptor antibody or insulin. Any hypothesis concering the role of groups of insulin receptors must account for all the pertinent morphological and biochemical observations including the following. First the natural state of distribution of insulin receptors appears to be tissue specific [ 2, 4, 9 ]. Secondly dithiothreitol, a reducing agent, disperses the groups of Fm-] occupied receptors on adipocyte membranes [ 10 ], yet increases 125I-insulin binding to these membranes about threefold [ 11 ]. Thirdly anti-insulin antibody [ 5 ] induces aggregation of the largely single Fm-I occupied receptors on liver membranes, and causes up to a 15-fold increase in a25I-insulin binding to liver membranes [ 12 ]. In conclusion, the validity of the interpretations from experimental data [ 6, 8 ] used to support the concept of insulin-induced receptor aggregation, can be questioned. There are experimental findings to endorse the opposite view that insulin does not aggregate its own receptor [ 2, 4, 5 ]. Receptor aggregation, nevertheless, does occur in the natural state of certain tissues [ 2, 9 ] and may not be necessary as a primary mechanism for insulin action. However, there might be a secondary role for receptor aggregation in modulating the biochemical responses to insulin [ 5, 11,12 ]. Studies are needed to determine whether the pattern of aggregation of the insulin receptor seen in the natural state is an intrinsic property of the receptor structure or the membrane micro-environment for each individual tissue. Yours sincerely, K.R.Lyen Dear Sir, We wish to comment on the Letter to the Editor by Janka and Mehnert [ 1 ] on the role of antiplatelet agents in diabetic ketoacidosis. There have recently been several reports of haemostatic changes in diabetic coma [2, 31 and, while a coagulopathy may exist, its association with fatal thromboembolic events has not been established. In the hope of decreasing mortality from fatal thromboembolism and, in turn, lowering overall mortality from uncontrolled diabetes (up to 50% in the elderly [ 4 ]) some investigators have proposed prophylactic anticoagulation in these patients [ 5-8 ]. This has gained increased support in the literature and while antiplatelet drugs are not recommended, it has been suggested that heparin may be warranted in high risk uncontrolled diabetic patients. In a recent review of 275 cases of uncontrolled diabetes (unpublished observations), we noted one cerebrovascular accident in a hyperosmolar patient as well as one mesenteric arterial occlusion and one cerebrovascular accident in ketoacidotic patients. In reviewing several large series and case reports including 251 cases of hyperosmolar coma and 1,764 cases of ketoacidosis, we found that major fatal or near fatal thromboembolic (pulmonary embolus, cerebravascular occlusion, mesenteric arterial occlusion or limb loss due to arterial occlusion) represented less than 1% of events in ketoacidosis and 6% in hyperosmolar coma (references available on request). Many of these events occurred with severe hypotension and might just as easily be explained by increased serum viscosity with low flow in an already compromised circulation rather Diabetologia ( 1983 ) 24 : 305 - Letters to the Editor 1. Jeffrey PD ( 1982 ) The interaction of insulin with its receptor: cross-linking via insulin association as the source of receptor clustering . Diabetologia 23 : 381 - 385 2. Jarett L , Schweitzer JB , Smith RM ( 1980 ) Insulin receptors: differences in structural organization on adipocyte and liver plasma membranes . Science 210 : 112 % 1128 3. Smith RM , Jarett L ( 1982 ) A simplified method producing biologically active monomeric ferritin-insulin for use as a high resolution ultrastructural marker for occupied insulin receptors . J Histochem Cytochem 30 : 650 - 656 4. Jarett L , Smith RM ( 1977 ) The natural occurrence of insulin receptors in groups on adipocyte plasma membranes as demonstrated with monomeric ferritin-insulin . J Supramol Struct 6 : 45 - 59 5. Lyen KR , Smith RM Jarett L ( 1982 ) Aggregation of monomeric ferritin-insulin occupied receptor sites on liver plasma membranes caused by anti-insulin antibody . J Receptor Res 2 : 523 530 6. Schlessinger J , Shechter Y , Willingham MC , Pastan I ( 1978 ) Direct visualization of binding, aggregation, and internalization of insulin and epidermal growth factor on living fibroblastic cells . Proc Natl Acad Sci USA 75 : 2659 - 2663 7. Shechter Y , Schlessinger J , Jacobs S , Chang KJ , Cuatrecasas P ( 1978 ) Fluorescent labeling of hormone receptors in viable cells: preparation and properties of highly fluorescent derivatives of epidermal growth factor and insulin . Proc Natl Acad Sci USA 75 : 2135 - 2139 8. Kahn CR , Baird KL , Jarrett DB , Flier JS ( 1978 ) Direct demonstration that receptor crosslinking or aggregation is important in insulin action . Proc Natl Acad Sci USA 75 : 2659 - 2663 9. Nelson DM , Smith RM , Jarett L ( 1978 ) Non-uniform distribution and grouping of insulin receptors on the surface of human placental syncitial trophoblast . Diabetes 27 : 530 - 538 10. Jarett L , Smith RM , Schweitzer JB ( 1981 ) Disulfide bonds help maintain naturally occuring groups of insulin receptors on adipocyte plasma membranes: an ultrastructural study . Diabetes 30 : 53A 11. Schweitzer JB , Smith RM , Jarett L ( 1980 ) Differences in organizational structure of insulin receptor on rat adipocyte and liver membranes: role of disulfide bonds . Proc Natl Acad Sci USA 77 : 4692 - 4696 12. Schechter Y , Chang K , Jacobs S , Cuatrecasas P ( 1979 ) Modulation of binding and bioactivity of insulin by anti-insulin antibody: relation to possible role of receptor self-aggregation in hormone action . Proc Natl Acad Sci USA 76 : 2720 - 2724

This is a preview of a remote PDF:

K. R. Lyen. Insulin does not aggregate its own receptor, Diabetologia, 1983, 304-305, DOI: 10.1007/BF00282720