With the growing research surrounding fructose consumption, I thought it'd be a interesting to discuss the role of fructose consumption throughout human evolution, and address of some of the theories surrounding it.
Chimpanzees, our closest living primate relatives, are considered ripe-fruit specialists, even more so than other primate species (1). Studies of the feeding ecology of wild chimpanzees report varied percentages of the diet coming from fruit, varying from 40-90% (2). The amount and type of fruit consumed is widely variable among different populations of chimps of observed, and particularly effected by seasonality of fruits. Dietary Diversity varies inversely with fruit availability (3). Chimpanzees, getting a large percentage of their calories from fruits, were getting a large percentage of their calories from fructose. It's difficult, as is always the case with plant foods, to understand fruit consumption throughout the rest of the hominid timeline - it's highly likely that fruit/fructose was consumed, but how much and by which geographically distinct hominid species can't be determined. However, we do know that modern hunter-gatherers, like the Hadza, consume not only fruit but honey (16), providing relatively significant amounts of calories depending, again, on the season.
So what's the buzz around fructose, if primates have been consuming it for millennia?
One theory that has arisen in the biological anthropology field, and transferred to the Paleolithic diet community, is that the loss of uricase activity interacted with fructose consumption to be selectively advantageous, promoting a fat storage state (8, 21) - this loss has also become conjectured to have had a combined effect with the loss of endogenous vitamin C synthesis.
Many organisms express uricase, the enzyme that converts uric acid to allantoin to eventually be excreted as urea (5). However, throughout hominid evolution, 2 nonsense mutations arose in the promoter regions of the uricase gene (6). Humans and higher primates do not appear to show any uricase activity, and in other Old/New World Monkeys, uricase activity is moderate, anywhere from 2-4 times lower than mice and rabbits. It's estimated that the loss of activity occurred during a time 14-15MYA in the middle Miocene (5). Given that two distinct mutations have been found in divergent lineages of the primate tree of life (6) - the one in gibbons being different than orangs and chimps - it's not strange to think that there was some selective advantage to losing uricase activity.
The theory posited by some anthropologists/Paleo'ers for this loss of uricase activity goes like this: fructose is special because it enters glycolysis past the rate limiting step of phosphofructokinase, and is processed primarily in the liver ,because most cells don't express Glut-5 transporters (4). Once inside the liver, fructose is driven through glycolysis and can (spoiler) become a source of glycerol-3-p for use in triglyceride formation, or acetyl-CoA, for de novo fatty acid synthesis. Since fructose passes this rate limiting step and must be phosphorylated, the ATP that donates this phosphorous group, ultimately leads to a depletion in Liver ATP and an increase in ADP/AMP. Excess adenosine can be converted to inosine, and drive through the uric acid pathway. Without uricase, increased consumption of fructose would drive increased uric acid production.
From this, it's been suggested that the loss of uricase, leading to an increase in uric acid, combined with high fruit (fructose) consumption would've led to a metabolic state that promoted fat storage, and would've conveyed a survival advantage, for times of food shortage. In rats, oxonic acid, which inhibits uricase activity, combined with sweetened beverage consumption (11% fructose-glucose), induces hypertension, insulin resistance, heightened plasma glucose, and hepatic triglcyeride accumulation (7), whereas just Oxonic Acid induces glomerular hypertension and just sweetened beverage consumption induced insulin resistance. It's also been shown that knockout mice models (xanthine oxidoreductase) that cannot produce uric acid have a 50% reduction in adipose mass relative to wild types (22).
As a recap, the theory goes: high fructose consumption, combined with the loss of uricase, leads to a metabolic state that promotes fat storage, and would've allowed for greater chance of survival during the Miocene climate shift (12), when food availability may have been limited. Some (21) have added the loss of endogenous vitamin C synthesis into this equation, citing that vitamin C supplementation prevents fructose induced hypertension in rats (23) and fruits, at their ripest when fructose is ripest, also have the lowest vitamin C content (24).
These theories, while interesting, have some major flaws:
1. Animal models - Humans have significantly lower rates of de novo lipogenesis (DNL) than rats (17, 18). Much of this theory is based off of models showing this impaired metabolic effect in animal models where DNL is high. Most fructose in human consumption is not shuttled towards DNL (20).
2. Fruit vs Isolated Fructose - studies of fructose overfeeding occur in the context of isolated fructose overfeeding. Unless one is arguing that most of the fructose consumed comes from honey (which may have protective factors of its own), fruit would've been the major provider of fructose. Overfeeding fruit might be quite difficult, due to its higher fiber content. Fruit also contains polyphenols/flavanoids, which have been shown to attenuate the detrimental metabolic effects of fructose (9,10,11).
3. Chimpanzees - Chimpanzees also lack uricase activity - however, there is little evidence to suggest that chimpanzees are regularly insulin resistant; C-peptide levels are more determined by dominance hierarchy (25). Those that do become diabetic are captive and usually obese (26). More data from wild chimpanzees would be needed to back up this theory, though I've never seen data that suggests chimpanzees specifically overeat fruit during times of fruit abundance (28).
4. Human data - Stephen Guyenet cited a study (27) recently that showed Humans consuming 82% of their kcals from fruit improved their metabolic profiles and lost or maintained their weights. I'm extremely skeptical that hominid ancestors would be able to consume enough fruit to induce insulin resistance and store a significant amount of fat to convey a survival advantage.
An alternative theory has been suggested for the loss of uricase - Blood Pressure Maintenance. In the Miocene environment (12), hominids would've had very low salt and high potassium intake d representing a challenge for maintaining blood pressure (13). High uric acid levels are associated with hypertension and CVD risk (14), and acute/chronic hyperuricemia has been shown to maintain blood pressure, by stimulating the renin-angiotensin system/inducing salt sensitivity. This is a more interesting theory, though I don't know enough about comparative herbivore physiology to comment on why this would've occurred in humans and not any other herbivore species.
How is all of this relevant to modern nutritional sciences? Animal models studying fructose may not be representative of the effects of fructose overfeeding in humans. Not only are mice/rat not appropriate models due to uricase, they also have more efficient rates of de novo lipogenesis compared to humans (17,18). Apart from that, we can tell very little - as always, inferring anything about health from evolutionary theorizing is quite difficult and unreliable.
Modern fructose hysteria, leading some Paleo dieters to limit fruit intake, seems to take the context of fructose and dosage out of the conversation. Most studies of fructose are done with isolated fructose and glucose feedings (19), whereas we almost never consume isolated fructose in the food supply. Fructose does not promote weight gain, anymore than other calories, in humans (15). Most fructose does not primarily end up as fat in humans (20, 29). For more up-to-date discussions on fructose, see here and here.
-Side note: These gene mutations also predisposed humans to higher incidence of gout.
1. http://www.indiana.edu/~semliki/PDFs/WranghamEtAlFruit.pdf
2. Feeding Ecology In Apes and Other Primates. 2012. Hohmann et al.
3. http://onlinelibrary.wiley.com/doi/10.1002/ajp.21016/abstract
4. http://www.ncbi.nlm.nih.gov/pubmed/15971409
5. http://www.ncbi.nlm.nih.gov/pubmed/11961098
6. http://link.springer.com/article/10.1007%2FBF00163854
7. http://www.ncbi.nlm.nih.gov/pubmed/23303409
8. http://onlinelibrary.wiley.com/doi/10.1002/evan.20266/abstract
9.http://www.ncbi.nlm.nih.gov/pubmed/23275372
10.http://www.ncbi.nlm.nih.gov/pubmed/21062538
11. http://jn.nutrition.org/content/early/2013/12/11/jn.113.184358.full.pdf
12. http://www.nature.com/nature/journal/v360/n6405/abs/360641a0.html
13. http://hyper.ahajournals.org/content/40/3/355.long
14. http://www.nejm.org/doi/full/10.1056/NEJMra0800885
15. http://annals.org/article.aspx?articleid=1132642
16.http://www.ncbi.nlm.nih.gov/pubmed/19350623
16. http://advances.nutrition.org/content/4/5/527.full
17. http://ajcn.nutrition.org/content/74/6/707.full
18. http://www.ncbi.nlm.nih.gov/pubmed/10831168
19.http://www.ncbi.nlm.nih.gov/pubmed/19381015
20.http://www.ncbi.nlm.nih.gov/pubmed/20086073
21. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2917125/
22.https://www.ncbi.nlm.nih.gov/pubmed/17276354
23. https://www.ncbi.nlm.nih.gov/pubmed/12482032
24. https://www.ncbi.nlm.nih.gov/pubmed/7358939
25. http://dash.harvard.edu/bitstream/handle/1/3693702/C-peptide%20Thompson%202008.pdf?sequence=1
26. http://ilarjournal.oxfordjournals.org/content/47/3/259.full
27.http://www.ncbi.nlm.nih.gov/pubmed/4928686
28.http://www.ncbi.nlm.nih.gov/pubmed/3578495
29.http://www.nutritionandmetabolism.com/content/9/1/89
Chimpanzees, our closest living primate relatives, are considered ripe-fruit specialists, even more so than other primate species (1). Studies of the feeding ecology of wild chimpanzees report varied percentages of the diet coming from fruit, varying from 40-90% (2). The amount and type of fruit consumed is widely variable among different populations of chimps of observed, and particularly effected by seasonality of fruits. Dietary Diversity varies inversely with fruit availability (3). Chimpanzees, getting a large percentage of their calories from fruits, were getting a large percentage of their calories from fructose. It's difficult, as is always the case with plant foods, to understand fruit consumption throughout the rest of the hominid timeline - it's highly likely that fruit/fructose was consumed, but how much and by which geographically distinct hominid species can't be determined. However, we do know that modern hunter-gatherers, like the Hadza, consume not only fruit but honey (16), providing relatively significant amounts of calories depending, again, on the season.
So what's the buzz around fructose, if primates have been consuming it for millennia?
One theory that has arisen in the biological anthropology field, and transferred to the Paleolithic diet community, is that the loss of uricase activity interacted with fructose consumption to be selectively advantageous, promoting a fat storage state (8, 21) - this loss has also become conjectured to have had a combined effect with the loss of endogenous vitamin C synthesis.
Many organisms express uricase, the enzyme that converts uric acid to allantoin to eventually be excreted as urea (5). However, throughout hominid evolution, 2 nonsense mutations arose in the promoter regions of the uricase gene (6). Humans and higher primates do not appear to show any uricase activity, and in other Old/New World Monkeys, uricase activity is moderate, anywhere from 2-4 times lower than mice and rabbits. It's estimated that the loss of activity occurred during a time 14-15MYA in the middle Miocene (5). Given that two distinct mutations have been found in divergent lineages of the primate tree of life (6) - the one in gibbons being different than orangs and chimps - it's not strange to think that there was some selective advantage to losing uricase activity.
The theory posited by some anthropologists/Paleo'ers for this loss of uricase activity goes like this: fructose is special because it enters glycolysis past the rate limiting step of phosphofructokinase, and is processed primarily in the liver ,because most cells don't express Glut-5 transporters (4). Once inside the liver, fructose is driven through glycolysis and can (spoiler) become a source of glycerol-3-p for use in triglyceride formation, or acetyl-CoA, for de novo fatty acid synthesis. Since fructose passes this rate limiting step and must be phosphorylated, the ATP that donates this phosphorous group, ultimately leads to a depletion in Liver ATP and an increase in ADP/AMP. Excess adenosine can be converted to inosine, and drive through the uric acid pathway. Without uricase, increased consumption of fructose would drive increased uric acid production.
From this, it's been suggested that the loss of uricase, leading to an increase in uric acid, combined with high fruit (fructose) consumption would've led to a metabolic state that promoted fat storage, and would've conveyed a survival advantage, for times of food shortage. In rats, oxonic acid, which inhibits uricase activity, combined with sweetened beverage consumption (11% fructose-glucose), induces hypertension, insulin resistance, heightened plasma glucose, and hepatic triglcyeride accumulation (7), whereas just Oxonic Acid induces glomerular hypertension and just sweetened beverage consumption induced insulin resistance. It's also been shown that knockout mice models (xanthine oxidoreductase) that cannot produce uric acid have a 50% reduction in adipose mass relative to wild types (22).
As a recap, the theory goes: high fructose consumption, combined with the loss of uricase, leads to a metabolic state that promotes fat storage, and would've allowed for greater chance of survival during the Miocene climate shift (12), when food availability may have been limited. Some (21) have added the loss of endogenous vitamin C synthesis into this equation, citing that vitamin C supplementation prevents fructose induced hypertension in rats (23) and fruits, at their ripest when fructose is ripest, also have the lowest vitamin C content (24).
These theories, while interesting, have some major flaws:
1. Animal models - Humans have significantly lower rates of de novo lipogenesis (DNL) than rats (17, 18). Much of this theory is based off of models showing this impaired metabolic effect in animal models where DNL is high. Most fructose in human consumption is not shuttled towards DNL (20).
2. Fruit vs Isolated Fructose - studies of fructose overfeeding occur in the context of isolated fructose overfeeding. Unless one is arguing that most of the fructose consumed comes from honey (which may have protective factors of its own), fruit would've been the major provider of fructose. Overfeeding fruit might be quite difficult, due to its higher fiber content. Fruit also contains polyphenols/flavanoids, which have been shown to attenuate the detrimental metabolic effects of fructose (9,10,11).
3. Chimpanzees - Chimpanzees also lack uricase activity - however, there is little evidence to suggest that chimpanzees are regularly insulin resistant; C-peptide levels are more determined by dominance hierarchy (25). Those that do become diabetic are captive and usually obese (26). More data from wild chimpanzees would be needed to back up this theory, though I've never seen data that suggests chimpanzees specifically overeat fruit during times of fruit abundance (28).
4. Human data - Stephen Guyenet cited a study (27) recently that showed Humans consuming 82% of their kcals from fruit improved their metabolic profiles and lost or maintained their weights. I'm extremely skeptical that hominid ancestors would be able to consume enough fruit to induce insulin resistance and store a significant amount of fat to convey a survival advantage.
An alternative theory has been suggested for the loss of uricase - Blood Pressure Maintenance. In the Miocene environment (12), hominids would've had very low salt and high potassium intake d representing a challenge for maintaining blood pressure (13). High uric acid levels are associated with hypertension and CVD risk (14), and acute/chronic hyperuricemia has been shown to maintain blood pressure, by stimulating the renin-angiotensin system/inducing salt sensitivity. This is a more interesting theory, though I don't know enough about comparative herbivore physiology to comment on why this would've occurred in humans and not any other herbivore species.
How is all of this relevant to modern nutritional sciences? Animal models studying fructose may not be representative of the effects of fructose overfeeding in humans. Not only are mice/rat not appropriate models due to uricase, they also have more efficient rates of de novo lipogenesis compared to humans (17,18). Apart from that, we can tell very little - as always, inferring anything about health from evolutionary theorizing is quite difficult and unreliable.
Modern fructose hysteria, leading some Paleo dieters to limit fruit intake, seems to take the context of fructose and dosage out of the conversation. Most studies of fructose are done with isolated fructose and glucose feedings (19), whereas we almost never consume isolated fructose in the food supply. Fructose does not promote weight gain, anymore than other calories, in humans (15). Most fructose does not primarily end up as fat in humans (20, 29). For more up-to-date discussions on fructose, see here and here.
-Side note: These gene mutations also predisposed humans to higher incidence of gout.
1. http://www.indiana.edu/~semliki/PDFs/WranghamEtAlFruit.pdf
2. Feeding Ecology In Apes and Other Primates. 2012. Hohmann et al.
3. http://onlinelibrary.wiley.com/doi/10.1002/ajp.21016/abstract
4. http://www.ncbi.nlm.nih.gov/pubmed/15971409
5. http://www.ncbi.nlm.nih.gov/pubmed/11961098
6. http://link.springer.com/article/10.1007%2FBF00163854
7. http://www.ncbi.nlm.nih.gov/pubmed/23303409
8. http://onlinelibrary.wiley.com/doi/10.1002/evan.20266/abstract
9.http://www.ncbi.nlm.nih.gov/pubmed/23275372
10.http://www.ncbi.nlm.nih.gov/pubmed/21062538
11. http://jn.nutrition.org/content/early/2013/12/11/jn.113.184358.full.pdf
12. http://www.nature.com/nature/journal/v360/n6405/abs/360641a0.html
13. http://hyper.ahajournals.org/content/40/3/355.long
14. http://www.nejm.org/doi/full/10.1056/NEJMra0800885
15. http://annals.org/article.aspx?articleid=1132642
16.http://www.ncbi.nlm.nih.gov/pubmed/19350623
16. http://advances.nutrition.org/content/4/5/527.full
17. http://ajcn.nutrition.org/content/74/6/707.full
18. http://www.ncbi.nlm.nih.gov/pubmed/10831168
19.http://www.ncbi.nlm.nih.gov/pubmed/19381015
20.http://www.ncbi.nlm.nih.gov/pubmed/20086073
21. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2917125/
22.https://www.ncbi.nlm.nih.gov/pubmed/17276354
23. https://www.ncbi.nlm.nih.gov/pubmed/12482032
24. https://www.ncbi.nlm.nih.gov/pubmed/7358939
25. http://dash.harvard.edu/bitstream/handle/1/3693702/C-peptide%20Thompson%202008.pdf?sequence=1
26. http://ilarjournal.oxfordjournals.org/content/47/3/259.full
27.http://www.ncbi.nlm.nih.gov/pubmed/4928686
28.http://www.ncbi.nlm.nih.gov/pubmed/3578495
29.http://www.nutritionandmetabolism.com/content/9/1/89
Kevin,
ReplyDeleteGreat post with good references. I look forward to looking through each one.
Truly relevant to this discussion is my interview with John Sievenpiper (who you cite) of a couple of years ago in "Fate of Fructose," where he makes similar arguments http://evolvinghealth.wordpress.com/2012/05/26/fate-of-fructose-interview-with-dr-john-sievenpiper/
David
Hey David, thanks much. They're pretty interesting, especially the anthropological theories - though i don't think the current state of fructose metabolism literature supports them much. Sorry the refs are out of order - restructured it several times.
ReplyDeleteI'll definitely check it out - I've kept up with your current posts since I found your blog a couple months ago.
If you're interested in more evolution/nutrition, I have a lot of posts starting from the beginning of the blog and then others scattered throughout.
-Best,
Kevin