Diabetes, Development and Causes

Acrobat PDF file can be downloaded by clicking here.

Diabetes mellitus is usually thought of as an illness involving sugar metabolism, as implied by its name.  "Diabetes" refers to excessive urination and "mellitus" to "honey sweet", together indicating loss of  large volumes of urine containing lots of sugar.  The basic cause of most cases of diabetes is either a loss of production of the hormone insulin (diabetes mellitus type 1 or DM1) or a reduced cellular response to the hormone (diabetes mellitus type 2 or DM2).  In fact, diabetes mellitus involves almost all angles of metabolism.  Lipid and protein metabolism as well as carbohydrate metabolism are dependent upon "physiologically correct" levels of and responses to insulin. 

We are accustomed to think of diabetes as a disease involving high blood sugar levels as a result of restricted tissue uptake of sugar from food.  However, today we know that the loss of regulation of hepatic gluconeogenesis is an equally important element in deficient control of blood sugar levels. 

I will take up the different forms of diabetes further on, but first let us look at blood glucose and the dangers which follow loss of control over this physiologically important element.


Blood sugar values, normal and diabetic.

The current ADA norms (2003)* for blood glucose levels are shown in the following table.  The upper limit for normal levels of blood glucose is currently set at  less than 5.6 mmol/l while the reference level 2 hours after a glucose load is less than 7.8 mmol/l.  Note that interpretation of the data is "temporary".  Glucose levels swing somewhat from day to day and from individual to individual.  Therefore, if high glucose levels are found, it is advised that measurement should be repeated over the following three days.   Only then should hyperglycemia be noted as an enduring symptom and treatment initiated.

Fasting glucose levels between 5.6 mmol/l and 7.0 mmol/l indicate "impaired glucose tolerance" (IGT) while higher levels signify diabetes.  IGT can be viewed as "pre-diabetes" as most individuals in this group develop diabetes type 2 with time. 


* In 2003 an expert group suggested reducing the cut point for impaired glucose tolerance  from 6.1 mmoles/l to  5.6 mmoles/l. This was expected to increase the number of individuals diagnosed as prediabetics by about 20%.  Many of these people have a high risk for developing diabetes type 2 (DM2) later in life.  They should, therefore, be advised to revise their lifestyle and work for reduction of their hyperglycemia. 

Fasting blood glucose levels have been suggested to be the leading indication of the metabolic state.  This assay is a rapid, inexpensive and accurate measure of the individual's metabolic situation.  The ADA committee suggested that fasting blood sugar be determined every 3rd year after reaching 45 years of age.  A glucose tolerance test was deemed unnecessary as long as fasting glucose does not exceed 5.6 mmoles/liter.  However, it has recently been pointed out that many DM2 patients achieve normal fasting blood sugar levels but have an increased level after eating and that this can last throughout the day.  That is, glucose levels may well reach "correct" levels while sleeping, but that insulin response may remain inadequate after meals.  Since high blood glucose levels are associated with long-term diabetic damage, it may be wise to occasionally control two-hour blood sugar levels in all IGT and DM2 patients.


What's wrong with a little extra blood sugar?

Glucose intolerance and the development of disease.

The suggested blood glucose standards are based upon statistical analysis of the correlation between blood glucose levels and the development of pathological states.  Among these is retinopathy, which can lead to blindness.  One might have expected a linear relationship between the level of blood sugar and the development of glucose-related diseases.  However, this is not the case, as shown in the next figure.  Data from three independent studies are shown in which elderly Americans, Pima Indians and Egyptians were observed.  These groups are known to be especially prone to development of diabetes type 2.  In all cases, an abrupt increase in the frequency of retinopathy was noted when blood sugar levels exceeded 6.0-6.1 mmol/l.  This toxic effect of glucose may result from:

1.  Sorbitol formation.

2.  Glycation of proteins, a non-enzymatic coupling of glucose to proteins. 

3.  Activation of the hexosamine signaling pathway.  Increased blood sugar levels drive this pathway forward with formation of glucosamine-6-phosphate from fructose-6-phosphate.  This then reacts with UTP or UDPG to form the high-energy compound uridine diphosphate N-acetylglucosamine (UDP-N-acetylglucosamine). This activated form of glucosamine then reacts with many proteins and modifies their functions.  It does so by O-linking glucosamine to serine and threonine residues.  Note that these are the same amino acid residues that are phosphorylated by protein kinases.  Thus, O-glucosylation may be as important as phosphorylation in the control of enzyme activity and metabolism. 


Glucose is toxic.


Three pathological processes seem to follow hyperglycemia.  Each of these may be able to initiate impaired glucose tolerance or diabetes type 2.  However, they "overlap" one another and probably occur simultaneously.  It may be impossible to specifically identify the  prcis "cause" of DM2 in each isolated case.  The factors listed here are only a few of those involved in this complex disease.



1.  Conversion of glucose to sorbitol with ensuing osmotic disturbance.  

Sorbitol is formed in many tissues when blood glucose levels are increased.  This occurs through the "polyol pathway", a  reaction sequence found in most tissues. Glucose is first reduced to sorbitol in the aldol reductase reaction.  It is then oxidized to fructose by sorbitol dehydrogenase.  The whole pathway is only found in testes.  Unlike most tissues with the exception of the liver, sperm cells possess fructokinase.  They can, therefore, utilize the fructose in seminal fluid for energy while traveling to their target; the egg.  The women's tissues cannot compete for this energy substrate as they lack fructokinase. 








Sorbitol is not usually formed in other tissues owing to the aldose reductase's high Km for glucose.  However, the high blood glucose concentrations found in diabetic patients activate the reductase, leading to sorbitol formation in several organs, notably the kidneys, nerves and the lens of the eye.  Sorbitol accumulates in these tissues because they lack a membrane transport system for this sugar.  They can reduce glucose, but cannot utilize or be quit the resulting sorbitol. 

Activation  of aldose reductase, therefore, leads to an accumulation of its product and this increases the osmolarity of cell's cytoplasm. Influx of water and swelling follows the increasing sugar content.  There is often a parallel denaturation of cellular proteins. 

Cataract formation in the lens is believed to follow sorbitol accumulation. Renal  damage and neuropathy seen in diabetic patients is thought to be at least partially triggered by sorbitol accumulation in these tissues.







2.  Spontaneous non-enzymatic glucosylation of proteins.

Glucose is a reactive compound because of its carbonyl group at carbon 1.  The acrylic (straight chain) form of the sugar reacts with itself following solution in water, forming alpha and beta pyranose rings.  The straight chain also splits to some extent, forming glyoxal and methylglyoxal.  These 2 and 3-carbon compounds both have two carbonyl groups that are quite reactive. 

Glucose, glyoxal and methylglyoxal react with proteins to form stable addition products. This glucosylation (or glycation) is a normal process which follows reaction of the carbonyl compounds with amino groups in proteins.  The process is non-enzymatic and the rate of reaction is proportional to the concentration of glucose in the blood.   There seems to be no specificity here; most proteins have been found to become glucosylated to a limited degree.  The glycoproteins formed undergo cross-linking, forming "AGE" or advanced glycation end-products.  This is, in effect, protein degradation as coupling of sugar molecules to proteins markedly affects their metabolic and structural functions.  The label, AGE, passes well with the fact that levels of cross-linked glucosylated proteins do seem to increase with age!






Hemoglobin A1C.

Hemoglobin is one of many proteins which become increasingly glucosylated as blood sugar levels rise.  Glucosylated hemoglobin is known as HbA1cThis glucosylated derivative  normally makes up about 5-6 % of the circulating hemoglobin.  Hyperglycemia can result in a two to three-fold increase in HbA1c .  Since the turnover rate for hemoglobin is approximately 100 days, HbA1c values give a good picture of the patient's mean blood glucose levels over this period.

The figure to the left shows formation of  HbA1c .  The carbonyl group in glucose reacts with a valine in hemoglobin forming an unstable intermediate.  This then rearranges (Amadori rearrangement) and forms the stable HbA1c.





There is a near correlation between blood sugar  and HbA1c level as shown in the following table.  HbA1c levels respond markedly to changes in glucose levels.  Normal blood glucose lie between 4.5 and 5.5 millimoles/liter.  This gives HbA1c of approximately 5.5 % of total hemoglobin concentration.  Glucose levels that exceed "normal" concentrations yield easily measured changes in HbA1c .

Establishing stable "correct" glucose levels over time in diabetic patients is not trouble-free.  HbA1c is often used to follow treatment progress.  A well-accepted norm for treatment of diabetes is to reduce HbA1c levels to less than 7.0 %, corresponding to average blood glucose levels 8.3 mmol/l.







Marked improvements are seen in patients with small changes in HbA1c.  In the following table we can see that a 1% fall in HbA1c resulted in improvement in retinopathy and kidney, nerve and cardiac function in the large studies quoted here.  












3.  Activation of the "Hexosamine Signaling Pathway".

The third pathologic reaction called forth by high blood glucose levels is initiated by conversion of the glycolytic intermediate fructose-6-phosphate (F-6-P) to glucosamine as shown in the next figure.    A small amount of the F-6-P (between 2 and 5 %) binds an amino group which is released from glutamine.  The product is glucosamine-6-P.  The details of hexosamine metabolism can be called up by clicking here.  I shall hold my discussion here to the simplified diagram below.  Here you can follow synthesis of the active substance, UDP-N-acetylglucosamine.  This O-glucosylates  many serine and threonine groups in a number of cytosolic and nuclear proteins.  These are the same as those phosphorylated by cyclic nucleotide-activated protein kinases.  O-glucosylation is catalyzed by UDP-N-acetylglucosamine transferase (OGT) while the reverse reaction is catalyzed by -N-acetylglucosaminidase (O-GlcNAcase). 

Once again, O-glucosylation occurs at the same amino groups that are subject to phosphorylation by protein kinases.  Thus, many hormone-regulated enzymes and processes are controlled both by cyclic nucleotides and protein phosphorylation and by o-glucosylation.

It would appear the O-glucosylation, like phosphorylation, is a major determinant of transcription and metabolic activity. 

The hexosamine signaling pathway has become a field of major interest.  The past few year's many publications  attest to this fact.  I suggest the following references for those who wish go further: 

1.  The hexosamine Signaling Pathway:  Deciphering the "O-GlcNAc Code", D. C. Love and J. A. Hanover, http://stke.sciencemag.org/cgi/content/full/sigtrans;2005/312/re13

2.  Role of Insulin, Adipocyte Hormones, and Nutrient-Sensing pathways in Regulating Fuel Metabolism and Energy Homeostasis: A Nutritional Perspective of Diabetes, Obesity and Cancer, Stephan Marshall, http://www.stke.org/cgi/content/full/sigtrans;2006/346/re7.  Just click on the Internet addresses to call up these references.




Aberrant Signaling at High Blood Glucose Levels.

Increased blood glucose concentrations lead to a rise in glucose uptake and cytoplasmic glucose levels in a number of tissues.  Conversion of F-6-P to N-acetylglucosamine and UDP-N-acetylglucosamine or (UDP-GlcNAc) follows this augmented cellular glucose uptake.  O-glucosylation of many proteins ensues with many modifications in metabolism.  At a first glance this would seem to be a protective reaction to counter the high glucose.

1.  Insulin sensitivity is reduced and less glucose will be taken up in many tissues.  This is what we experience in IGT.

The kidneys, nervous tissue and the retina take up excessive glucose in spite of the generalized reduction in sensitivity to insulin. This may explain why these are the main sites of glucose-induced tissue damage.

2.  Lipid synthesis increases, leading to elevated blood lipid levels and, in time, weight gain.  Increased plasma lipids are associated with cardiovascular disease which is frequent in diabetic patients.

3.  Regulation of the synthesis of a number of hormones, especially those of the adipocyte. Increased hexosamine signaling markedly increases leptin and adiponectin, both key factors in appetite and weight regulation.

Again, while this increase in O-glucosylation does, at first, seem to protect against high glucose, the results seem to kick off the development of DM2.  Please go to the suggested references for more information. 

Three very recent (2008) publications indicate that O-glucosylation may be involved in development of diabetes type 2 (DM2).

1 and 2. 

The 7. March issue of Science presented an article "Hepatic Glucose Sensing via the CREB Coactivator CRTC2"  R. Dentin et al, Science 319 1402-1405:2008 with a news article covering this publication "Sweet Conundrum", M. J. Birnbaum Science 319 1348-1349:2008.  Click on the addresses to call up these publications.  In essence, this work indicates that increased O-glucosylation coupled to high blood glucose levels increases hepatic glucose output through activation of synthesis of the two rate-limiting enzymes in gluconeogenesis, namely G-6-Pase and PEPCK.


"Phosphoinositide Signalling links O-GlcNAc Transferase to Insulin Resistance", Yang et al, Nature451 964-970:2008.  This article presents findings which clearly show linking between  insulin signaling through phosphatidylinositides and O-GlcNAc transferase.  These studies demonstrate that high blood glucose levels bring about hepatic insulin resistance and dyslipidemia through O-glucosylation of the insulin signalling system in the plasma membrane. 

Insulin, Fat Metabolism and Ketogenesis.

Triacyglycerides are split into free fatty acids and glycerol in adipocytes.  However, this glycerol can not be recycled back to triacylglycerides as these cells lack glycerol kinase.  They must take up glucose from the circulation via glucose-transport-protein 4 (GLUT4) and convert this to α-glycerol phosphate via glycolysis.   Insulin is required to activate GLUT4.

Lipolysis or release of fatty acids from triacylglycerol is also under control of insulin.  More correctly stated, catecholamines and glucagon activate lipolysis while insulin inhibits this process.  The degree of fatty acid accumulation and release is coupled to metabolic requirements through the balance between these three hormones.  This is shown in the figure to the left.  You can pick up the details about insulin's actions in adipocytes by clicking here.

Note that a reduced response to insulin due to reduced secretion or resistance will greatly disturb the balance between these hormones.   Faulty insulin signaling leads to disproportionate release of fatty acids.  These are then converted to "ketones" (acetoacetate and -hydroxybutyrate) by the liver.  In diabetic patients, this can proceed to ketoacidosis and serious CNS effects.








Classification of diabetes


Most diabetic patients fall into one of two classifications, diabetes type 1 and diabetes type 2.  In the USA and Europe about 80% of all diabetic patients have the type 2 variant. 

DM1.  Patients with diabetes type 1 are afflicted by an autoimmune illness which results in complete destruction of pancreatic insulin-producing cell, the so-called beta cells of the islets of Langerhans.  These chronically ill persons must receive insulin to survive.  Given proper guidance they can experience a relative normal life.  However, this requires that they take over the responsibilities of the beta-cell "manually".  That is, they must monitor their blood sugar level, plan their physical activities and meals and calculate the amount and type of insulin that they must administer.  No easy situation!  Some succeed well in regulation of metabolism, other suffer from instability.  Both hyperglycemic and hypoglycemic episodes occur in some patients. 

DM2.  Type 2 diabetes and its precursor "impaired glucose tolerance" (DM2 and IGT) is often initially characterized by a faulty response to insulin (or so-called insulin resistance).  The body automatically compensates for this by increasing the amount of the secreted hormone.  Individuals with glucose intolerance or type 2 diabetes are often initially treated by modifying lifestyle and diet.  However, DM2 and IGT are progressive illnesses.  That is, resistance to insulin action develops over a period of years.  While changes in diet and motion can initially hinder development of hyperglycemia, many of these patients become dependent upon medications with time.  Beta cell insulin production often fails after some years with naturally augmented or drug-stimulated synthesis.  Insulin treatment is often required with time.

Gestational diabetes is a passing hyperglycemia in pregnancy.  This is a genetically determined condition and is usually handled without insulin treatment.  Women who acquire gestational diabetes often develop diabetes type 2 later in life.


Diabetes type 1.

Diabetes type 1 was a fatal disease prior to the work of  Banting and Best.  They first identified the hormone insulin in pancreatic extracts in 1922.  The first patient who received insulin (also 1922) was 14 year old Leonard Thompson.  We are accustomed to assuming that diabetes patients can and will be treated and can live a good life.  We are often ignorant of the fact that many patients struggle to normalize their blood sugar levels and often suffer from serious "late diabetic symptoms".  It is also shocking for us to find that treatment of diabetes with insulin is not economically possible for many people in underdeveloped countries. 

Diabetes type 1 results from an autoimmune destruction of the -cells of the pancreas.  It has earlier been commonly accepted that this is a genetically controlled process associated with HLA alleles.  However, current knowledge has shown that development of DM1 is not strongly associated with heredity, while there is a strong genetic component in development of DM2.

There is a age-dependent spreading of the disease and this has changed with time.  Increasing evidence points to a combination of factors leading to beta cell destruction.  The disease most commonly develops before 20 years of age as is shown in the figure to the left.  However, incidence (cases/100,000) of the disease has changed in the latter half of the 20th century.  Patients develop diabetes at an earlier age and the frequency has increased.  This implies that there are other factors than the genetic which lead to development of DM1.  The figure and data are from Norway and can be found in "The Rise of  Childhood Type 1 Diabetes in the 20th Century", Medscape  or in the original publication (Diabetes 51, 3353, 2002). 



A further indication that factors other than the genetic are involved in development of diabetes type 1 is shown in the next figure.  The data are also from Norway and show a clear decrease in the frequency of diabetes type 1 in adults during World War 2 (cases/100,000).  This strongly suggests that social and economic factors are involved in development of the disease.  The scanty diet and need for physical activity experienced during the war led to weight reduction in many adults.  Overweight has been clearly shown to be involved in development of DM2.  Is this also a factor in development of DM1?  Note that most children who develop DM1 are thin when symptoms develop due to loss of food energy and body mass as urinary sugar. 

The abrupt initiation of DM1 is usually noted following a viral infection.  It has been assumed that beta-cell destruction is an ongoing process but that it is accelerated by such infections.  The mechanism involved remains unknown.




Autoimmune attack on the Langerhans Islets.

It has been commonly assumed that antibodies and T cells specifically attacked beta cells, leading to insulin deficiency and destruction.  However, in a recent publication, Shawn Winer et al have shown that Schwann cells which surround the Langerhans islet are autoimmune targeted and destroyed before the autoimmune attack on beta cells and loss of beta cell activity.  It appears that antigen-presenting cells move from islets to the pancreatic lymph mode where cell-specific antigens are produced.  These attack Schwann cells surrounding the islets, destroying the Schwann cell capsule.  Beta cell destruction follows. (Nature Medicine 9, 198-205, 2003). 

Most DM1 patients are today completely dependent upon injections of insulin to sustain life.  Current research is aimed at developing oral agents capable of activating the insulin-signal pathway downstream of the insulin receptor.  Recently, inhaled insulin preparations have been developed and shown to be quite effective.  Click here for information about these.  The need for injected insulin may perhaps be eliminated in the future.





 Why don't we replace lost -cells?

One might think that we should be able to utilize stem cells to replace destroyed -cells and thereby restore insulin secretion.   After all, blood cells, skin and intestines do not disappear after loss of tissue.  This was the question a group of researchers at Harvard put forward in a recent study.  They investigated generation of -cells in mice and looked for formation of new pancreatic islets in adult animals.  Surprisingly, they found that no new islets could be identified.  In adult mice, -cells are formed only by self-duplication and appeared not to arise from stem cells.  The permanent loss of -cells seen in diabetes type 1 may, therefore, occur because only pre-existing -cells give rise to new insulin-producing cells in the adult.  The loss of insulin-producing -cells seen with time in diabetes type 2 may also develop through this mechanism.  Go to the original article for more information.  Click here to call up "Adult pancreatic -cells are formed by self-duplication rather than stem-cell differentiation", Dor et al, Nature 429, 41 (2004).

Is replacement therapy possible?

A fascinating approach to diabetes type 1 therapy would be to replace pancreatic beta-cells with new cells produced in the laboratory.  Is this possible?  Can we produce human beta cells in vitro?  In a recent publication in Science Express 25, November 2004 1-10, Gershengorn et al (click here) present data showing conversion of human pancreatic fibroblast-like cells to insulin-producing cells.   Cell replacement therapy just might be possible, but it remains to be seem whether or not fibroblasts in vivo can give rise to viable insulin-producing cells. 


Diabetes type 2.

Diabetes mellitus type 2 arises from a reduced response of target tissues to insulin (so-called insulin resistance).  Insulin levels are often quite high in early type 2 diabetes and resistance to the hormone is counteracted by increased stimulation of the hormone receptor.  The etiology of insulin resistance remains unclear, although many hypotheses have been set forth.  I will go through some of the most interesting findings below. You can click here for information about insulin's mechanism of action and possible sites development of resistance to this crucial hormone.  

We have some insight into of the physiological changes that give rise to DM2.  There is a very clear association between overweight (BMI 25-30) and obesity (BMI>30), the metabolic syndrome, and development of diabetes type 2.  A genetic factor is also clearly involved as many ethnic groups are predisposed for the disease.  More detailed information about the causes of and classification of diabetes mellitus can be found here. 

Diabetes type 2 (DM2), was earlier called "maturity -onset diabetes".  Typically, it developed over a period of many years.  It usually begins parallel with weight-gain, goes through a period of reduced response to ingested sugar (glucose intolerance or IGT), and ends with the full-blown diabetic state.  In contrast to diabetes type 1, the disease develops slowly and goes through several stages as shown in the following figure.  The patient typically goes through a period of increasing weight parallel with a loss of sensitivity to insulin, so-called insulin resistance.  The body normalizes the metabolic state by producing higher levels of insulin after meals.  Usually, the patient is unaware of his or her situation.  Following a period of many years, the pancreatic -cells no longer manage to continue their enhanced insulin secretion. Insulin levels fall in the face of continuing insulin resistance, resulting in increased fasting blood glucose levels and markedly increased postprandial levels of serum glucose. 

The patient has become clearly diabetic!

I pointed out that DM2 used to be called "maturity onset diabetes".  That is, people became diabetic from around 30-40 years of age.  This is no longer the case.  It should be noted that the global wave of obesity we now experience applies also to young children (10-2 years old).  This is known as MODY or "Maturity Onset Diabetes of the Young".  Social factors (fear of kidnapping, parents who drive kids to school and after-school activities, passive TV and PC use, etc.) contribute to reduced physical activity and increased overweight and obesity in children.  Simultaneously, energy intake is often increased through an ever-growing consumption of fast foods, soda pop, sugar coated breakfast food, snacks and confection.  Decreased physical activity coupled to increased food intake must and does lead to pronounced weight gain.   You can download a good review article concerning  DM2 in children here.


Overweight and diabetes type 2.

As I have previously stated, there is a striking correlation between central obesity (or body weight) and the incidence of DM2.  We can look at some examples of this in the following two figures.  In the first we see the incidence of obesity among Americans state by state.  In the course of a nine-year period, the incidence of obesity (BMI>30) rose from 10-14% to greater that 20% for many Americans.









If we then look at development of diabetes in the same period, we find that the frequency of diabetes (mostly DM2) had risen from <4% to more than 6% for most of the USA.  These figures are a minimum, many individuals have impaired glucose tolerance or "pre-diabetes" for many years before they are diagnosed as having DM2. 

Now, correlations are one thing, proof of cause is another.  However, the evidence that overweight and obesity are coupled to development of diabetes type 2 is very strong.  This is very well presented and discussed in a CME program that you can find at Medscape.com by clicking here.







Global perspectives of the incidence of diabetes.

Predictions of the global frequency of metabolic syndrome, insulin resistance and diabetes are frightening.  The consequences of urbanization and overweight are much more threatening that the AIDS/HIV epidemic.  One of many estimates of the spreading of this global epidemic are shown in the following figure.  Most of the increase in diabetes will be that of type 2.  Overweight with associated insulin resistance and metabolic syndrome will lead to an almost 50% increase in the number of individuals suffering from diabetes within the current decade.  Over 200 million persons will be involved.  Most of this expansion will occur in developing countries where health services will be unable to cope with the coming situation.  Without indispensable guidance and treatment, these people will develop coronary and vascular disease, chronic kidney disorders, hypertension and the many other complications of diabetes.  Predictions for the middle of the century indicate that at least 300 million persons will develop diabetes type 2. 






The situation is well-described in the following quotations from Paul Zimmet's article entitled:

"The global scope of diabetes and obesity -- an epidemic in progress: paradise lost"

" Type 2 diabetes is poised to become one of the major challenges to public health in the 21st Century and will result in a huge economic burden, particularly in developing nations, through premature morbidity and mortality.

 there will be more than 230 million people with diabetes by 2010. The majority of the new cases will be those with type 2 diabetes.

 ... Type 2 diabetes is the tip of the iceberg of a cluster of cardiovascular disease (CVD) risk factors, including  obesity, hypertension, and dyslipidemia, otherwise known as the "Metabolic Syndrome," "New World Syndrome," or "Deadly Quartet." The natural consequence will be an epidemic of cardiovascular complications, such as coronary heart disease and stroke as well as microvascular complications.

  Over 60% of the adult population of the United States and Australia are either overweight (body mass index [BMI] of 25-29.9 kg/m2) or obese   In some developing countries -- as well as in disadvantaged groups in developed countries (Mexican Americans, African Americans, and Australian Aborigines) -- an even more extreme situation exists. In Samoa and Nauru, more than 70% of adults fit the obesity criteria.

An important and alarming feature of the diabetes epidemic is that type 2 diabetes is increasing in these younger age groups. In China, Japan, and in the Pacific Islands, more than 70% of children presenting with diabetes have the type 2 form.


Obesity and lack of exercise have been implicated in this trend, that has been labeled as "Nintendonization". Children are often driven to and from school and then come home and race to the computer or computer-game station instead of playing games or sports outdoors".

Paul Zimmet MD, PhD

60th Scientific Sessions of the American Diabetes Association
June 10, 2000




A very short summary of insulin's actions and the effects of reduced secretion and resistance.

Insulin's major effects actions which are disturbed in both DM1 and DM2 are listed in the table to the left. 

1.  Decreased uptake of sugar from the blood after meals lead to hyperglycemia, increased urine formation and loss of sugar and ions.  The effects of high blood glucose levels have been discussed above.  Loss of K+ can lead to cardiac disturbances.

2.  Disturbance of the balance between hepatic gluconeogenesis and glycolysis may be a major cause of hyperglycemia in type 2 diabetes (DM2).

3.  Deficient insulin response in adipocytes leads to loss of control of lipolysis.  This is the basis for the overproduction and release of fatty acids to the circulation.  The excess fatty acids are converted to acetoacetate and -hydroxybutyrate ("ketones") in the liver.  This acid production can exceed the body's pH-buffer capacity and lead to ketoacidosis.




What Causes Diabetes type 2?


1.  The beta cell and insulin secretion.

There has been a debate in the medical literature about this question for many years.  Some have suggested that pancreatic beta cells in DM2 patients are damaged early in the course of the illness.  Let's look at the pattern of release of insulin from these cells.

  In the upper left you can see a trace of the concentration of the hormone after a glucose challenge.  Note the very quick rise which occurs during the first minutes of release.  This is known as "phase 1".  Thereafter, a slow but sustained hormone release continues, reaching a top (not shown) after approximately 1 hour.  This is "phase 2".  Details about insulin secretion can be found here.

Phase 1: The very rapid hormone release, uses isolated insulin granules lying close to the beta cell's outer membrane.  Inflowing calcium ions increase "docking" and exocytosis (a "kind of spitting out" of the content of those insulin granules). 

Phase 2:  The slow process.  This recruits insulin granules from storage sites, brings them to the membrane (using energy) and delivers them to the circulation via exocytosis.   It is the delivery from storage sites that is slow and which characterizes phase 2. 



Insulin Secretion in IGT and DM2.

The pattern of insulin secretion in "impaired glucose tolerance" and DM2 or type 2 diabetes is shown in the two following figures.   The first shows blood insulin levels after oral glucose in normal, IGT and type 2 diabetes patients.  In the first figure you can clearly see that the amount of insulin released is quite similar in all of the cases shown.  However, the release pattern differs markedly!  The "top" in insulin secretion and blood levels of the hormone came later in persons with "impaired glucose tolerance" and  diabetes type 2. The implication was that something is amiss with the phase 1 release of the hormone.    










This was more clearly demonstrated in the figure below, taken from Medscape.  Again, the first and second phases of insulin release following a glucose challenge are clearly shown in the non-diabetic.  The diabetic patient did not exhibit the phase 1 release.

We can conclude:

1.  In cases of "Pre-diabetes" or IGT and diabetes type 2 there exists an error in beta cell  insulin secretion.  Release of insulin, measured as amount of the hormone, was adequate, but it came "too late".  That is, the secretion of insulin and the actions of the hormone come after the peak of circulating glucose (not shown).  Phase 1 insulin release triggers tissue uptake of glucose and rapidly reduces hepatic glucose production.  The lack of phase 1 release may partially explain why diabetics and "pre-diabetics" experience high levels of blood sugar after a meal. 

2.  The patient's pancreatic beta cells do seem to be altered quite early on during development of diabetes type 2.  The mechanism for this is not known.  However, treatment of patients with diabetes type 2 often relies on medications that directly trigger insulin release (sulfonylurea compounds) or which sensitize those cells to glucose (exenatide).  These often work only during a limited time course as the pancreatic beta cells seem to become "exhausted" with time.  Many diabetes type 2 patients must go over to insulin treatment after some years.  Is this because of early beta cell damage as is suggested in the two reports shown above?


2.  Reduced Sensitivity to Insulin or "Insulin Resistance".

Why does insulin miss its effect?  What happens at target organs which limits their response to the hormone?  Honestly, we cannot give an adequate answer to these questions today.  I shall try to list up some possible answers in the following chapter.  You will in any case get an idea of known control elements andthe complexity of the system.

The Insulin Receptor.

We can begin with something from the past.  As a student I learned that all hormones act at specific receptors and that these were subject to damage and feedback control.  OK, lack of a response to insulin must have followed receptor injury!  That has been ruled out!  Less than 1 % of those with DM2 have antibodies to the insulin receptor.  Furthermore, receptor-initiated autophosphorylation appears to be quite normal in most diabetics.  The advances in the insulin signaling field up to 1979 are clearly depicted in the drawing to the left (taken from TIBS).









Integrated Control Systems.

Those "somethings" have virtually exploded into a huge amount of information.  I will try to present some of this below.  The major conclusion of most studies today is that a series of closely integrated systems are involved in insulin's many actions. 

1.  Regulation of phosphorylation of system proteins.

This was presented in a understandable diagram by Biddinger and Kahn just a few years ago.  They divided up insulin effects into three groups.  Actions of all of these are initiated by autophosphorylation of the insulin receptor.  This follows hormone-receptor coupling. 

1.  Actions on carbohydrate and protein metabolism.  These follow phosphorylation of insulin-receptor-substrates (IRS 1-4) and activation of Pi-3-kinase.  This phosphorylates phosphatidyl inositol-3-phosphate and forms phosphatidyl inositol 3,4,5,triphosphate.  Activation of Akt (or protein kinase B) leads to further signal generation and metabolic control.

2.  Control of lipid synthesis appears to proceed through PKCλ.

3.  Cell growth and differentiation is also an area regulated by insulin.  Biddinger and Kahn indicate that this is regulated through MEK and MAP kinase. 




These authors go further and point out possible sites for development of insulin resistance, given the systems they have described.  This is summarized the next figureThey indicate several sites which can be modified by protein phosphatases, where tyrosine phosphorylation can be inhibited through serine phosphorylation, and suggest that the the insulin receptor level can, in fact, be regulated through control of  synthesis and degradation.










I have attempted to simplify this in the following figure.  Biddinger and Kahn have pointed out 3 main control elements, all of which are subject to extensive specific and and integrated control elements.  These are the hormone receptor and its protein substrates, the kinase system for phosphatidyl inositol-3-phosphate and protein kinase B (Akt).  

Changes in any of these elements could conceivably give rise to insulin resistance. 









A little more precise (and detailed) mechanism, please!

1.  Insulin, PI3-K and PtdIns(3,4,5) "one more time".

D. Alessi and colleagues have presented the data about intermediates in insulin signaling in a manner similar to Biddinger and Kahn.  This is shown in the next figure.  I want you to note one special point here.  The product of the phosphatidylinositol-3-kinase (PI3-K) is triphosphate phosphatidylinositol 3,4,5 triphosphate or PtdIns(3,4,5)P3.  This is a lipid-soluble material and becomes "anchored" in the insulin target cell's plasma membrane.   Now, I am not trying to make life difficult, but this may be the key to insulin resistance.  That is why I take it up now.  We will soon see that there may be a coupling between chronically elevated blood sugar levels, activation of the "hexosamine signaling pathway" and, precisely,  membrane-bound PtdIns(3,4,5)P3.








2.  Integration of  Hormone and Nutrient Signaling in Regulation of Energy Metabolism.

Professor Stephen Marshall has presented an exciting review article which I have referred to previously.  Here, he discusses control of the physiological and metabolic state through integration of key metabolic "watchdogs" into a "regulatory network".  A whole battery of protein kinases and small "G-proteins" (GTP-binding proteins) is involved in regulation of energy metabolism, appetite and growth. 

The following concepts, figure and table are taken from "Role of Insulin, Adipocyte Hormones, and Nutrient-Sensing pathways in Regulating Fuel Metabolism and Energy Homeostasis: A Nutritional Perspective of Diabetes, Obesity and Cancer, Stephan Marshall, http://www.stke.org/cgi/content/full/sigtrans;2006/346/re7 and reprinted with permission from AAAS.

Professor Marshall suggests that four tightly integrated systems participate in this control network:

1.  The insulin stimulated signaling pathway which responds to changes in blood-carried nutrients.

2.  The mTOR or "mammalian target of rapamycin" which measures availability of amino acids to be used in protein synthesis.

3.  The AMPK (adenosine monophosphate kinase) system which monitors the cellular energy level.

4.  The hexosamine system which monitors nutrients and regulates hormone secretion and sensitivity in accordance with this. 

The very tight integration of these in a single metabolically active cell is depicted in the following figure.


The metabolic pathways monitored by these systems and the resulting control of physiological processes is well-summarized in the following table, taken by permission from Marshall's review article.

Again, please go to the original article to come further with these concepts.

Note the following restrictions from AAAS:  "Readers may view, browse and/or download for temporary coping purposes only, provided these uses are for noncommercial personal purposes.  Except as provided by law, this material may not be further reproduced, distributed, transmitted, modified, adapted, performed, displayed, published, or sold in whole or in part, without prior written permission from the publisher".

3.  Does Hexosamine Formation Control Insulin Sensitivity?

 Phosphoinositide signaling links O-GlcNAc transferase to insulin resistance is the title of a paper by Yang and colleagues, published in Nature 451,964-970 2008.  You can download the paper here if you have library services.  They report that binding of the end product of the hexosamine signaling pathway (O-GlcNAc) to proteins involved in generation of PtdIns(3,4,5)P3.  Protein phosphorylation of IRS-1 (an insulin receptor substrate) and the protein kinase Akt are influenced by binding O-GlcNAc.  This is a dynamic reaction where the forward reaction is catalyzed by OGT and the removal of the glucosamine group is under control of O-GlcNAcase.   That will say that the hexosamine signaling pathway is closely involved in regulation of signal transduction, transcription and proteasomal degradation.  The implication is that disorders of the hexosamine signaling pathway may be involved in many serious chronic illnesses such as diabetes, neurodegeneration and cancer.

It will be exciting to see if the O-GlcNAc binding will be found to be a major factor in development of insulin resistance. 

The point to remember is that insulin signaling and its control is most complex.  I would suggest that this "network" offers many opportunities for failure and that the "cause" of insulin resistance in individuals will be difficult to identify. 




Do fatty acids themselves control sensitivity to insulin?

It is generally accepted that there is a correlation between overweight, the level of fatty acids in the circulation and development of type 2 diabetes.  Most studies of this phenomenon have emphasized the roles of adipokines (small peptide hormones produced in fat cells) in this process.  However, recent reports have indicated that fatty acids released from fat cells can themselves regulate sensitivity to insulin.  Jerrold M. Olefsky (in a introductory paper in Cell 134(6) 914-916, 2008) points out that different fatty acids have divergent effects on insulin sensitivity:  while saturated fatty acids cause insulin resistance, polyunsaturated  fatty acids have little or no effect on insulin signaling and omega 3 fatty acids may improve insulin sensitivity.  The figure at the left, taken from Olefsky's paper, illustrates fatty acid synthesis in adipocytes and summarizes the observation that saturated fatty acids inhibit insulin action while palmitoleate stimulates insulin's actions. 






Olefsky's introduction sets the stage for another report in the same issue of Cell:  Cao et al, Identification of a Lipokine, a Lipid Hormone Linking Adipose Tissue to Systemic Metabolism, Cell 134(6) 933-944 2008.  In essence, these authors demonstrate that adipose tissue's fatty acid synthesis differs from that of the liver.  While hepatic fatty acid synthesis produces mostly saturated fatty acids (palmitate), adipose tissue synthesizes both saturated and unsaturated fatty acids.  One of the latter is palmitoleate (C16:1n7-palmitoleate).  This unsaturated adipose tissue specific fatty acid is characterized as a lipid hormone, a lipokine, by these authors.  While palmitic acid acid leads to insulin resistance, palmitoleate increases the actions of insulin.  Thus, palmitoleate inhibits hepatic lipogenesis and stimulates insulin's various actions on muscle cells.  This may be the represent the normal control of insulin sensitivity, adjusting hormone response to energy intake.  Excessive hepatic palmitate release following "overeating" may then override this action of adipocytes.  Palmitoleate might just be the link between obesity and  diabetes in individuals with a genetically defined susceptibility to type 2 diabetes.  

Note that this publication reports data from studies in mice and may not give a true picture of human metabolism.  It will be exciting to follow future developments in this story.







Genetic Aspects of Type 2 Diabetes.

Genetic background is a strong component in the development of type 2 diabetes.    People of western European origin are less sensitive to the effects of overweight than non-Europeans.  In the USA, blacks and Latinos have a higher frequency of diabetes type 2 than whites.  Asians, especially Chinese and Japanese, develop metabolic syndrome and diabetes type 2 at lower body weights than whites.  Thus, these groups develop diabetes 2 in the BMI range from 25-30. 

This "puzzle" of sensitivity to diet and development of type 2 diabetes was taken up in a recent article by Jared Diamond in Nature.  A figure from this article is presented below.  Prevalence of diabetes is relatively low in Europeans and native people living in their accustomed environment.  However, native people who convert to a western diet or urban life present a markedly increased incidence type 2 diabetes.  Genetic variation can explain differing frequencies of diabetes development in diverse racial groups.  However, development of type 2 diabetes among urbanized groups must be a response to the environment.  Genes do not change so quickly! 

One explanation for this has been the development of a set of "thrifty genes", a genetic adaptation to a poor milieu with a meager diet.   A switch to an urban diet and activity level is suggested to rapidly lead to obesity and associated illnesses in such people.   Please go to the original article for discussion of this data.


An excellent review article concerning the development and possible causes of the global diabetes epidemic can be found here