The Carbohydrate Hypothesis II: Diabetes Mellitus
( Part I)
When you eat carbohydrates, your digestive system converts them to glucose which is then transported into your bloodstream. Glucose is the most readily combustible fuel your body uses, and like oxygen it's as dangerous as it is precious. As a first step to understanding the importance of diabetes mellitus, it helps to understand why your body handles glucose the way it does, because the role of insulin has been largely defined by glucose's dual nature as a valuable resource and a toxin.
Glucose, like other reducing sugars, is hazardous to proteins because the electron-poor aldehyde (C1) of the sugar likes to grab hold of the electron-rich amino groups on proteins, forming an awkward conglomeration referred to as a Schiff base. Schiff bases are somewhat unstable and eventually rearrange into ketosamines, which are very readily oxidized. This whole process is collectively called the Maillard reaction, and it's the same process involved in browning things when you cook them; in a sense, blood sugars cook your proteins by sticking to them, which is why I think of it as biomolecular napalm.
So you don't want any more of this stuff floating around in your blood than is necessary: at low glucose concentrations the first step in the reaction is relatively uncommon, because while glucose is constantly shape-shifting between a (reactive) chain configuration and a (non-reactive) ring coniguration, it spends more of its time as a ring. The higher the glucose concentration the higher the absolute number of reducing (chain) species, the more Maillard reactions occur. The lower the concentration, the less often the reaction happens. So when carbohydrates are present in the diet, your body works to maintain a stable optimum concentration of glucose in the blood, and it does this using hormones.
Organs involved in gross metabolic regulation express a special high-volume glucose transporter that makes them particularly sensitive to glucose levels. As glucose pours through these transporters in the insulin-secreting cells of your pancreas, it gets consumed to make ATP. As the amount of ATP in the cells rises, increasing numbers of potassium channels open up, permitting positively charged potassium ions to flow out of the cell and resulting in a depolarization of the voltage across the cell membrane. The beta cells function like neurons and muscle cells in this sense -- when they depolarize past a certain threshold, voltage-sensitive calcium channels open up, enabling an influx of calcium ions into the cell, which in secreting tissues leads to packets of signaling peptides being shoved and fused into the cell wall, emptying their contents into the extracellular space. In this case, the most important peptide released is insulin.
Insulin is essentially the only hormone in your body that has the effect of shifting glucose out of the bloodstream, and this is the aspect of it that's been focused on in medical research and practice. But this is something of an accident of history that might have been very different if, say, the acetone smell of diabetic urine had been noticed before the sweetness was. Insulin isn't "for" keeping your blood from turning to caramel syrup because no ancestor of ours ever had to deal with enough carbohydrates in their diet for that to even be a live possibility; everything it does is intimately involved with regulating the balance of energy storage vs. energy use, and to understand the full import of diabetes as more than just a disease of blood glucose dysregulation we have to look at it from this angle.
Insulin's most immediate direct effect on the liver, muscles and fat tissue is to enhance their uptake of glucose and its conversion to glycogen -- which can be converted back to glucose in a snap as needed -- by a rapid up-regulation of another special glucose receptor that normally is inactive. But its next effect is just as important: in adipose tissue insulin signaling up-regulates an enzyme that's responsible for cleaving triglycerides into free fatty acids and shifting them from the bloodstream into cells; in muscle tissue, insulin has the precise reverse effect of down-regulating this enzyme.
Fatty acids are what your cells usually burn when carbs aren't around, and flow in and out of your cells relatively freely through special receptors. But it's also not good to have much lipids floating around in your blood for reasons I'll return to in subsequent posts, so normally your body packages them up to be carted around specially -- first by binding them into bundles of three bound to a glycerol molecule (i.e. as triglycerides), then stuffing these into special protein packages that cart them around through the blood and dump them off directly at cells with the help of the aforementioned enzyme. Once fatty acids are transferred into adipose tissue they get bundled back up again as triglycerides, and another of insulin's effects is to increase the conversion rate. It also increases the rate of de novo synthesis of fatty acids (palmitic acid, specifically) from glucose in fat and liver tissue.
That's insulin's role as an acute response to carb intake, but when we're not eating it's just as important: as your liver degrades the circulating insulin (which has a half-life of about 1-2 hours), all these effects kick into reverse and your body shifts back toward burning all this newly stored glycogen and fatty acids for fuel. But it doesn't stop completely: there's a basal pulse rate of minimal insulin release that keeps you from shifting too far in the opposite direction, which would result in inefficient glucose uptake and excessive production of free fatty acids, leading to the spikes of hypo- and hyperglycemia and ketoacidosis that are the bane of people with Type I diabetes. In Type I the problem is a lack of insulin production, but Type II gets comparable results by the less direct route of insulin resistance.
As I said in Part I, hormones are chemical signals that coordinate processes across tissues in response to environmental conditions, and if the signal becomes miscalibrated you get maladaptive physiological responses. It's a fundamental fact of cell signaling that negative feedback loops are ubiquitous, which is to be expected based on Claude Bernard's insight that the organism attempts to maintain a stable milieu interieur: too much of anything can break the modular, synchronized functioning of subsystems in the body, and a simple way to combat this in a receptor-mediated signaling system is to directly tie the amount of receptors present on tissues to the amount of signaling integrated over time. When insulin binds to a receptor, the whole hormone-receptor complex gets pulled into the cell and processed by enzymes that break it down; new receptors get manufactured and placed back on the cell membrane at a slower rate, which automatically limits the amount of signal that can get through to any given cell. Different tissues have different rates at which this adaptation occurs, and this fact is very important as we'll see, but the process is fundamentally the same across all multi-cellular life.
You now have everything you need to see how Type II diabetes happens: anything that causes blood glucose to rise will cause corresponding rises in insulin release, which will cause corresponding downregulation of insulin receptors, which will tend to attenuate the body's response to insulin. If the integral of blood insulin concentration over time is small, the power of the insulin signal will stay approximately constant, oscillating mildly around a settling point. If on the other hand the integral is large, then the power of the signal will attenuate over time; since one of the effects of this will be less effective clearance of glucose from the bloodstream, the pancreas will respond by continually pumping out ever more insulin in order to get the job done. Eventually this hits a point where glycemic control is so bad as to merit an official diagnosis.
This attenuation of insulin sensitivity is ultimately what underlies all the knock-on problems that statistically accompany Type II diabetes: obesity is one we've already seen, and is particularly prominent due to the fact that fat tissue is typically the most sensitive tissue to insulin and thus tends to keep performing dutifully for longer than other tissues. But in subsequent posts I'll demonstrate how atherosclerosis, hypertension, cancer, dementia and a whole host of other diseases can be attributed in most cases to much the same mechanism. This, in a nutshell, is the carbohydrate hypothesis: any health problem that Type II diabetics suffer more than the general population is likely due, directly or indirectly, to insulin dysregulation caused by excessive glycemic load.
In the next installment, we'll apply this to a constellation of metabolic disorders that correlate so strongly as to be considered a unified disease.
* * * * *
Ashcroft & Randle (1968) "Control of insulin release by glucose"
Bunn & Higgins (1981), "Reaction of monosaccharides with proteins: possible evolutionary significance"
MacDonald et al (2005), "Glucose-sensing mechanisms in pancreatic beta-cells"
McGarry (1992), "What if Minkowski had been ageusic? An alternative angle on diabetes"
Randle (1963), "Endocrine control of Metabolism"
Randle (1964), "The Interrelationships of Hormones, Fatty Acid and Glucose in the Provision of Energy"
Randle (1995), "Metabolic fuel selection: general integration at the whole-body level"
When you eat carbohydrates, your digestive system converts them to glucose which is then transported into your bloodstream. Glucose is the most readily combustible fuel your body uses, and like oxygen it's as dangerous as it is precious. As a first step to understanding the importance of diabetes mellitus, it helps to understand why your body handles glucose the way it does, because the role of insulin has been largely defined by glucose's dual nature as a valuable resource and a toxin.
Glucose, like other reducing sugars, is hazardous to proteins because the electron-poor aldehyde (C1) of the sugar likes to grab hold of the electron-rich amino groups on proteins, forming an awkward conglomeration referred to as a Schiff base. Schiff bases are somewhat unstable and eventually rearrange into ketosamines, which are very readily oxidized. This whole process is collectively called the Maillard reaction, and it's the same process involved in browning things when you cook them; in a sense, blood sugars cook your proteins by sticking to them, which is why I think of it as biomolecular napalm.
So you don't want any more of this stuff floating around in your blood than is necessary: at low glucose concentrations the first step in the reaction is relatively uncommon, because while glucose is constantly shape-shifting between a (reactive) chain configuration and a (non-reactive) ring coniguration, it spends more of its time as a ring. The higher the glucose concentration the higher the absolute number of reducing (chain) species, the more Maillard reactions occur. The lower the concentration, the less often the reaction happens. So when carbohydrates are present in the diet, your body works to maintain a stable optimum concentration of glucose in the blood, and it does this using hormones.
Organs involved in gross metabolic regulation express a special high-volume glucose transporter that makes them particularly sensitive to glucose levels. As glucose pours through these transporters in the insulin-secreting cells of your pancreas, it gets consumed to make ATP. As the amount of ATP in the cells rises, increasing numbers of potassium channels open up, permitting positively charged potassium ions to flow out of the cell and resulting in a depolarization of the voltage across the cell membrane. The beta cells function like neurons and muscle cells in this sense -- when they depolarize past a certain threshold, voltage-sensitive calcium channels open up, enabling an influx of calcium ions into the cell, which in secreting tissues leads to packets of signaling peptides being shoved and fused into the cell wall, emptying their contents into the extracellular space. In this case, the most important peptide released is insulin.
Insulin is essentially the only hormone in your body that has the effect of shifting glucose out of the bloodstream, and this is the aspect of it that's been focused on in medical research and practice. But this is something of an accident of history that might have been very different if, say, the acetone smell of diabetic urine had been noticed before the sweetness was. Insulin isn't "for" keeping your blood from turning to caramel syrup because no ancestor of ours ever had to deal with enough carbohydrates in their diet for that to even be a live possibility; everything it does is intimately involved with regulating the balance of energy storage vs. energy use, and to understand the full import of diabetes as more than just a disease of blood glucose dysregulation we have to look at it from this angle.
Insulin's most immediate direct effect on the liver, muscles and fat tissue is to enhance their uptake of glucose and its conversion to glycogen -- which can be converted back to glucose in a snap as needed -- by a rapid up-regulation of another special glucose receptor that normally is inactive. But its next effect is just as important: in adipose tissue insulin signaling up-regulates an enzyme that's responsible for cleaving triglycerides into free fatty acids and shifting them from the bloodstream into cells; in muscle tissue, insulin has the precise reverse effect of down-regulating this enzyme.
Fatty acids are what your cells usually burn when carbs aren't around, and flow in and out of your cells relatively freely through special receptors. But it's also not good to have much lipids floating around in your blood for reasons I'll return to in subsequent posts, so normally your body packages them up to be carted around specially -- first by binding them into bundles of three bound to a glycerol molecule (i.e. as triglycerides), then stuffing these into special protein packages that cart them around through the blood and dump them off directly at cells with the help of the aforementioned enzyme. Once fatty acids are transferred into adipose tissue they get bundled back up again as triglycerides, and another of insulin's effects is to increase the conversion rate. It also increases the rate of de novo synthesis of fatty acids (palmitic acid, specifically) from glucose in fat and liver tissue.
That's insulin's role as an acute response to carb intake, but when we're not eating it's just as important: as your liver degrades the circulating insulin (which has a half-life of about 1-2 hours), all these effects kick into reverse and your body shifts back toward burning all this newly stored glycogen and fatty acids for fuel. But it doesn't stop completely: there's a basal pulse rate of minimal insulin release that keeps you from shifting too far in the opposite direction, which would result in inefficient glucose uptake and excessive production of free fatty acids, leading to the spikes of hypo- and hyperglycemia and ketoacidosis that are the bane of people with Type I diabetes. In Type I the problem is a lack of insulin production, but Type II gets comparable results by the less direct route of insulin resistance.
As I said in Part I, hormones are chemical signals that coordinate processes across tissues in response to environmental conditions, and if the signal becomes miscalibrated you get maladaptive physiological responses. It's a fundamental fact of cell signaling that negative feedback loops are ubiquitous, which is to be expected based on Claude Bernard's insight that the organism attempts to maintain a stable milieu interieur: too much of anything can break the modular, synchronized functioning of subsystems in the body, and a simple way to combat this in a receptor-mediated signaling system is to directly tie the amount of receptors present on tissues to the amount of signaling integrated over time. When insulin binds to a receptor, the whole hormone-receptor complex gets pulled into the cell and processed by enzymes that break it down; new receptors get manufactured and placed back on the cell membrane at a slower rate, which automatically limits the amount of signal that can get through to any given cell. Different tissues have different rates at which this adaptation occurs, and this fact is very important as we'll see, but the process is fundamentally the same across all multi-cellular life.
You now have everything you need to see how Type II diabetes happens: anything that causes blood glucose to rise will cause corresponding rises in insulin release, which will cause corresponding downregulation of insulin receptors, which will tend to attenuate the body's response to insulin. If the integral of blood insulin concentration over time is small, the power of the insulin signal will stay approximately constant, oscillating mildly around a settling point. If on the other hand the integral is large, then the power of the signal will attenuate over time; since one of the effects of this will be less effective clearance of glucose from the bloodstream, the pancreas will respond by continually pumping out ever more insulin in order to get the job done. Eventually this hits a point where glycemic control is so bad as to merit an official diagnosis.
This attenuation of insulin sensitivity is ultimately what underlies all the knock-on problems that statistically accompany Type II diabetes: obesity is one we've already seen, and is particularly prominent due to the fact that fat tissue is typically the most sensitive tissue to insulin and thus tends to keep performing dutifully for longer than other tissues. But in subsequent posts I'll demonstrate how atherosclerosis, hypertension, cancer, dementia and a whole host of other diseases can be attributed in most cases to much the same mechanism. This, in a nutshell, is the carbohydrate hypothesis: any health problem that Type II diabetics suffer more than the general population is likely due, directly or indirectly, to insulin dysregulation caused by excessive glycemic load.
In the next installment, we'll apply this to a constellation of metabolic disorders that correlate so strongly as to be considered a unified disease.
* * * * *
Ashcroft & Randle (1968) "Control of insulin release by glucose"
Bunn & Higgins (1981), "Reaction of monosaccharides with proteins: possible evolutionary significance"
MacDonald et al (2005), "Glucose-sensing mechanisms in pancreatic beta-cells"
McGarry (1992), "What if Minkowski had been ageusic? An alternative angle on diabetes"
Randle (1963), "Endocrine control of Metabolism"
Randle (1964), "The Interrelationships of Hormones, Fatty Acid and Glucose in the Provision of Energy"
Randle (1995), "Metabolic fuel selection: general integration at the whole-body level"