In response to my post on the Kitavan diet, which includes plenty of fruit and likely 36g of fructose daily, "gn" asked me if I think it plausible to suppose that human ethnic groups differ in ability to deal with fructose loads, so that daily fruit won't harm dark-skinned Kitavans living in tropical climate (constant summer) for hundreds of generations, but will harm people of European descent whose ancestors had fructose supply only in late summer and fall seasons.
From this question, which I will address below, I get the impression that the Lustig lecture embedded above has led many to believe that daily ingestion of fructose in any quantity has toxic effects, such as non-alcoholic-fatty-liver-syndrome (NAFLS), particularly to people of European descent.
Unfortunately, Lustig's lecture contains misleading hyperbole or exaggeration (e.g. within the first five minutes he says "The Japanese diet is all carbs and no fat, and the Atkins diet is all fat and no carbs" and later he claims that Japanese eat no sugar, a falsehood) and doesn't give important details about research on fructose metabolism. The main problem lies in a failure to define the dose of fructose required to produce the adverse effects. Let's take a look at some of the research to see what people have missed.
Le et al investigated the effect of fructose overconsumption on blood lipids and ectopic lipid deposition in healthy subjects with and without a family history of type 2 diabetes. In this study they found that fructose overconsumption "increased ectopic lipid deposition in liver and muscle and fasting VLDL-triacylglycerols and decreased hepatic insulin sensitivity" and the effect appeared greater in offspring of parents who had type 2 diabetes.
To get this effect, they used a hypercaloric high-fructose diet supplying 3.5g fructose per kg of fat free mass, amounting to more than 35% of energy intake. For a lean individual like myself, that would require putting down more than 222g/888kcal of fructose daily while also consuming more calories than I expend. This is an unrealistic intake of fructose for a person eating a whole foods diet. It would require consuming more than 444g of table sugar, 12 non-diet sodas, or about 18 medium size apples or bananas.
For another example, Ackerman et al studied fructose-induced fatty liver disease in rats. To induce FLD in the rats, they used a diet consisting of 60% fructose by weight, which supplied all the carbohydrate in the diet. Humans do not typically consume diets in which fructose is the sole carbohydrate and the major macronutrient by weight.
Brown et al looked at the effect of fructose on blood pressure in young healthy people. They had 15 volunteers drink a 500ml beverage containing a single dose of 60g fructose or glucose. Both sugars produced an increase in heart rate, and the fructose dose produced an elevation in blood pressure that persisted for at least 2 hours. To get a single dose of 60g fructose from sucrose (table sugar) would required ingesting about 120g sugar in one sitting--or, 157g of high-fructose corn syrup, or more than 4 apples or 4 bananas. This says nothing about the effects of consuming the more typical 15g fructose per serving found in a serving of soda, let alone eating whole fruits, which contain other components (e.g. potassium) that tend to reduce blood pressure. For my n=1 experiment, I have routinely eaten 3-4 pieces or servings (cups) of fruit in a day for more than 10 years in a row and my blood pressure remains at 110/60.
Swarbrick et al claimed to show that consumption of fructose-sweetened beverages increases postprandial triglycerides and fasting apoB concentrations, and suggest that long-term consumption of diets high in fructose could lead to an increased risk of CVD. To produce their results, they fed study subjects diets providing 25% of total energy as fructose. For a 2000 kcalorie diet, that is 500 kcalories or 125g of fructose. Again, this would require consuming a total of 250g of sugar from sucrose or high-fructose corn syrup, i.e. more than 6 non-diet sodas, or at least 10 pieces or cups of fruit in a day.
As John White noted in the American Journal of Clinical Nutrition, in the typical U.S. diet, fructose contributes about 200–250 kcal/d, which amounts to about 7–8% of the current 2700-kcal/d per capita total calorie intake, a much smaller intake than used in these experiments, so they provide no evidence that typical intakes of fructose induce fatty liver disease. He added:
"Although examples of pure fructose causing metabolic upset at high concentrations abound, especially when fed as the sole carbohydrate source, there is no evidence that the common fructose-glucose sweeteners do the same. Thus, studies using extreme carbohydrate diets may be useful for probing biochemical pathways, but they have no relevance to the human diet or to current consumption."I could go on but instead I will refer you to an excellent critical review of Lustig's lecture by Alan Aragon: The bitter truth about fructose alarmism. After reviewing 19 papers on the effects of dietary fructose disputing some of the anti-fructose claims made by Lustig, and finding Lustig's presentation lacking, Alan comments:
"So, what’s the upper safe limit of fructose per day (all sources considered)? Again, this depends on a number of variables, not the least of which are an individual’s physical activity level and lean body mass.Currently in the literature is a liberal camp reporting that fructose intakes up to 90 grams per day have a beneficial effect on HbA(1c), and no significant effects are seen for fasting triacylglycerol or body weight with intakes up to 100 grams per day in adults . The conservative camp suggests that the safe range is much less than this; roughly 25-40 grams per day . Figuring that both sides are biased, the middle figure between the two camps is roughly 50 grams for active adults."Back To The Original Question
As I noted above, "gn" asked me if I find it plausible that Europeans have less tolerance for fructose than Kitavans. Short answer: No. Long answer: All of the above studies were done on people of European descent, illustrating that Europeans have a quite high tolerance for fructose. Further, fructose tolerance developed millions of years ago in the African primate lineage from which we all hale, due to primate consumption of fruit as a dietary staple. When humans moved out of Africa about 50 thousand years ago, they would have lost fructose tolerance only if maintaining it proved a disadvantage in northern climates. In other words, they would have lost fructose tolerance only if maintaining it resulted in death before reproduction.
I can't imagine a scenario in which the environment would select against fructose tolerance, i.e. in which maintenance of fructose tolerance despite a fructose-poor environment, would cause a person to lose fertility or die before having a chance to reproduce. I also can't imagine a scenario in which the environment (seasonal variations in supply of fructose) would select for fructose-intolerance, i.e. favor the reproduction of fructose-intolerant individuals. On the contrary, seasonal supply of fructose would continue to select for those who could eat naturally large amounts of fructose (i.e. fruits) when seasonally available, because those people maintaining the deeply ingrained primate ability to metabolize fructose would even in the north have a greater total available food supply than fructose-intolerant individuals.
I routinely eat three to five servings of fruits daily, i.e. 30 to 50g of fructose, and have done so most days for the past 10 years that I have eaten a practically paleo diet. My ethnic background consists of Hungarian, French, and German. My last blood profile showed my total lipoproteins at 231 mg/dL, my HDL at 85, and my triglycerides at 47. Using the Friedewald equation they calculated the LDL at 138, but since I have very low triglycerides, using the Iranian formula calculator I calculate my LDL equals 104. Since I have nearly twice as much HDL as triglycerides, and low fasting glucose, I have extremely low heart disease risk. My liver enzyme levels and bilirubin all fell in low normal values, indicating no liver dysfunction.