Letters to the Editor

Medium-chain triacylglycerols may not raise cholesterol
Dear Sir: Although medium-chain triacylglycerols (MCTs; 8:0 and 10:0) historically have been considered nonlipemic for most individuals, the report by Asakura et al (1) suggests that MCTs may be hypercholesterolemic in subjects with hypertriglyceridemia. Although the authors emphasized the hypercholesterolemia induced when MCTs progressively replaced corn oil in the carefully manipulated diet of these subjects, an alternative conclusion would be that the removal of dietary linoleic acid (18:2) was the primary cause of the plasma cholesterol elevation. Patients were adapted to a low-fat diet (containing 22% of energy as fat) in which 100% corn oil represented about one-half (12%/22%) of the total fat energy, including Ϸ8% of energy from 18:2. Progressive replacement of corn oil with MCTs resulted in < 2% of energy from 18:2 in the final diet period (estimated as 20% of the intrinsic dietary fat) when 100% MCT was the added fat. This exchange progressively increased total cholesterol and triacylglycerol, with VLDL cholesterol and LDL cholesterol contributing equally to the elevation in total cholesterol, even though the increase in triacylglycerol was not significant. Thus, MCTs increased cholesterol in apolipoprotein B–rich lipoproteins as dietary 18:2 was reduced from 8% of energy to < 2% of energy, the rise in total cholesterol being significant only when all the corn oil was removed. The background carbohydrate and other intrinsic fatty acids were constant and tended not to complicate interpretation here, which was a nice aspect of the study design. The failure to alter plasma triacylglycerol values in this study likely reflects the fact that high-carbohydrate diets (22% of energy as fat in this case) are hypertriglyceridemic in their own right. Had the total fat provided 30–40% of energy, MCTs might have had a more favorable effect on triacylglycerol. More important is the fact that 18:2 is the principle dietary fatty acid responsible for reducing hepatic fatty acid and triacylglycerol synthesis as well as VLDL secretion induced by carbohydrate, and presumably, induced by MCTs acting like carbohydrate (2). It is well appreciated that 18:2 also can enhance impaired LDL receptor activity (3), and carefully controlled fatty acid exchanges showed that progressive decreases in 18:2 per se raise total cholesterol when diet saturated fatty acids (SFAs) are high but stable, ie, when 18:1 replaces 18:2 (4). In the relative absence of dietary cholesterol, the resulting increase in LDL reflects an overproduction of LDL more than impaired clearance (5). We described these fatty acid interrelations previously in terms of the 18:2 threshold, wherein a specific amount of 18:2 is required to protect against total cholesterol elevation during consumption of SFAs and cholesterol (6).

Examples of these fatty acid interrelations specifically involving MCTs were shown in normolipemic women (7) and hamsters (8). In fact, even when women were fed a low-18:2 diet (3% of energy), MCTs proved less cholesterolemic than a source of longer-chain SFAs (trilaurin). Futhermore, the hamster study found MCTs to be as cholesterol lowering as safflower oil in a cholesterol-free diet when 18:2 intake was adequate at 5% of energy. Although the current data support the 18:2 threshold concept, they are not definitive because, as is often the case in such experiments, 2 important variables (MCTs and 18:2) were altered simultaneously in opposite directions. Thus, it is not clear which is to blame, rising MCTs or declining 18:2. But substantial evidence would argue the latter is most critical (6). A more definitive design would have kept 18:2 constant at 5–6% of energy and exchanged MCTs for 18:1 or carbohydrate, which are considered neutral. Furthermore, the clinical data cited in support of the cholesterol-raising nature of MCTs (9, 10), like the comparison in normolipemic women (7), suffer the same shortcoming as the present study, ie, 18:2 was lower (and below threshold) in the MCT diet period than in the control diet period. The clinically relevant point is that an adequate source of 18:2 needs to be supplied when MCTs, or even carbohydrate, replace other longchain SFAs and monounsaturated fatty acids (5). In summary, MCTs should not be considered as SFAs that raise total cholesterol and LDL. Nor do they represent a substitute for 18:2 that will effectively reduce circulating apolipoprotein B–rich lipoproteins in the absence of 18:2. KC Hayes Brandeis University, MS-029 Waltham, MA 02454-9110 E-mail: kchayes@brandeis.edu REFERENCES
1. Asakura L, Lottenberg AMP, Neves MQTS, et al. Dietary mediumchain triacylglycerol prevents the postprandial rise of plasma triacylglycerols but induces hypercholesterolemia in primary hypertriglyceridemic subjects. Am J Clin Nutr 2000;71:701–5. 2. Ntambi JM. Regulation of stearoyl-CoA desaturase by polyunsaturated fatty acids and cholesterol. J Lipid Res 1999;40:1549–58. 3. Spady DK, Woolett LA, Dietschy JM. Regulation of plasma LDL by dietary cholesterol and fatty acids. Annu Rev Nutr 1993;13:355–81. 4. Pronczuk A, Khosla P, Hayes KC. Dietary myristic, palmitic, and linoleic acids modulate cholesterolemia in gerbils and hamsters. FASEB J 1994;8:1191–200. 5. Hajri T, Khosla P, Pronczuk A, Hayes KC. Myristic acid-rich fat raises plasma LDL by stimulating LDL production without affecting fractional clearance in gerbils fed a cholesterol-free diet. J Nutr 1998;128:477–84.

Downloaded from ajcn.nutrition.org by guest on May 8, 2013

Am J Clin Nutr 2000;72:1583–93. Printed in USA. © 2000 American Society for Clinical Nutrition


and high oleic acid sunflower oil on plasma triacylglycerol fatty acids and lipid and lipoprotein concentrations in humans. Cater NB. 2013 Reply to KC Hayes Dear Sir: In our study of medium-chain triacylglycerols (MCTs) in hypertriglyceridemic individuals (1). Tsai YH. We argue that the shore-based ecologic niche was uniquely able to stimulate expansion of the primate brain because. Those who cannot choose freely often develop malnutrition or specific nutrient deficiencies. Lean muscle tissue is 2–3% fat. First. Lipids 1999. Neves MQTS. and food availability would have been constraints affecting the daily or seasonal diet of hunter-gatherers. Am J Clin Nutr 1997. In commonly available databases of the macronutrient contents of plant foods other than nuts and seeds. In contrast. Am J Clin Nutr 2000. ed. Dietary mediumchain triacylglycerol prevents the postprandial rise of plasma triacylglycerol but induces hypercholesterolemia in primary hypertriglyceridemic subjects. their reference macronutrient values for plant foods were 62% carbohydrate. IL: AOCS Press. it was the discovery of and adaptation to a high-quality shorebased diet that was a major determinant of the rate and extent of human brain evolution. However. Modern humans in a totally free-choice situation ultimately choose a diet that is complete in energy and nutrients. and other abilities directed at securing these nutrients” (5). therefore. This is an interesting suggestion that should be tested experimentally in humans no matter how convincing the animal data.or MCT-rich) may correct the hypertriglyceridemia provided that an ideal proportion of 18:2 is added to the diet. Mensink RP. Hayes raises the possibility that a low-fat diet (carbohydrate. . in that sense they may not have been totally free to determine their diets. these dietary needs apparently allowed for selection to favor increased brain size in the human lineage and the concomitant development of technologic. Park S. Mechanisms mediating lipoprotein responses to diets with medium-chain triglyceride and lauric acid. Khosla P. However.1584 LETTERS TO THE EDITOR 6. protein toxicity probably did not occur often. not the other way around as implied by Milton (5). this value needs to be corrected downward by Ϸ30% to yield the content of actual fatty acids that are available for energy from plant material other than nuts. but animal organs other than muscle. in addition to being a plentiful supply of dietary energy and protein. Was this correction made? Second. alternatively. Heller HJ. Hunter-gatherer diets—a shore-based perspective Dear Sir: Cordain et al (1) estimated that Paleolithic hunter-gatherers would have consumed as much animal food as possible. plasma cholesterol. and use. Saturated fats and blood lipids: new slant on an old story.nutrition. in my view. Denke MA. Trautwein E. and 14% protein. it Downloaded from ajcn.34:895–905. myristic acid. the fat content rarely seems to exceed 1% by weight. had the inconvenience of raising plasma cholesterol.38:1746–54. Temme EH. Whatever answers one draws from future experiments on these proposed dietary modifications. it makes empirical sense that foods of relatively high nutrient and energy densities would be consumed when available. 1998: 170–81. Hornstra G. J Lipid Res 1997. Comparison of the effects of medium-chain triacylglycerols. 24% fat. oleic acid on serum lipoproteins in healthy subjects. Does this imply that Cordain et al’s macronutrient database does not really represent most plant foods or. Many factors. that nuts and seeds are interpreted to represent most plant foods consumed? Even if plant foods in the Paleolithic period did contain an average of 24% of energy as fat by proximate analysis. Hayes KC. The list of food types for which this reference macronutrient profile was obtained did not include vegetables (see Table 3 of reference 1). Champaign. At the same time. Structurally modified food fats: synthesis. Can J Cardiol 1995.br REFERENCE 1. Eder CR Quintão University of São Paulo Medical School São Paulo 01246-903 Brazil E-mail: lipideq@usp. biochemistry.71:701–5. including climate. and certainly not for extended periods. the bottom line is that MCTs seem useless for the treatment of hyperlipidemia. I support the inclusion of fish and shellfish in Cordain et al’s estimate of animal food intake because I believe that fish. which is a well-known phenomenon that occurs in both healthy persons and in several cases of moderate hypertriglyceridemia. Snook JY. Hayes KC.11:39G–46G. palm oil. Lottenberg AMP. and other shore-based foods were crucial for human brain evolution (2–4). I have 2 comments about the reference values Cordain et al used for plant and animal macronutrient composition. competition. Cordain et al emphasized the risk of protein toxicity by referring extensively to the outcome of the consumption of large amounts of meat containing < 5% fat by weight. 9. Milton says that “Hunter-gatherers were not free to determine their diets. Kovacic J. would also have been consumed. 7. et al. carbohydrate-induced hyperlipidemia. rather it was their predetermined biological requirements for particular nutrients that constrained their evolution. and probably lowering. 8. As Milton’s (5) editorial points out.org by guest on May 8. Effects of medium chain fatty acids (MCFA). 10. In: Christophe A. may benefit from Hayes’s interesting proposal because it would likely be circumvented by adding the right amount of 18:2 to a fat-free diet without raising. and in addition. which tend to be 5–10% fat. social. Lindsey S. Asakura L. Body fat itself would also have been eaten. hominids would have faced fierce competition from carnivores for the copious amounts of meat needed to be eaten to induce protein toxicity. we were disappointed to find that provision of MCTs as the major source of dietary fat did not lower plasma triacylglycerol concentrations. which was the major objective of the study. Furthermore. Fatty acid modulation of lipoprotein metabolism by natural triglycerides in hamsters: lipoprotein turnover and hepatic mRNA abundance. Pronczuk A.65: 41–5. shellfish.

Crawford MA. 3. Am J Clin Nutr 2000. and plants on lakeshores and seacoasts—provided an abundance of this important dietary stimulus for human brain evolution without special effort or substantial competition from predators (5). Hence. weirs.79:3–21. For the same reasons. 24% fat. and fertility (6). Dietary or genetically imposed deficiencies of all of these brain-selective nutrients leaves the modern human brain extremely vulnerable to subnormal development.34(suppl):S39–47. Nutrition 1999. Nutr Health 1993. 1585 Reply to SC Cunnane Dear Sir: Our analysis was based on data derived from historically studied hunter-gatherers (Homo sapiens). 2.LETTERS TO THE EDITOR provided certain brain-selective nutrients. onions (bulbs). Evidence for the unique function of docosahexaenoic acid during evolution of the modern hominid brain. normal brain development. 5. Crawford MA. and iron (2–4). shellfish. Crawford MA. and other shore-based foods likely would have played a minor role in providing nutrients. lack of effective weapons. The hominid fossil record shows that at least fish and shellfish—but probably also eggs. we agree that Early Paleolithic hominids such as Homo habilis—because of their small size (male: height = 132 cm. the high fish consumption (median: 26–35% of energy) we showed for 229 historically studied hunter-gatherers likely would not have been representative of Early (2. iodine. Broadhurst CL. Harbige LS. we clearly included vegetables in our estimates. 3). weight = 37 kg). A common nutritional element of those hominid species that eventually led to anatomically modern humans was the inclusion of more energy-dense animal foods in their diet (2. zinc. Cunnane SC. Bloom M. Milton K. and Homo) that encompassed ≥ 11 separate species were simultaneously present (1). and limited behavioral sophistication—would have been unsuccessful hunters of large herbivores and hence would have had only occasional access to “copious amounts of meat” as well as abdominal organs and depot fat. as we argued (2–4). our weighting of the plantfood database in Table 3 of our article reflects the preferential foraging of these fat-containing plant foods by hunter-gatherers. This is what happened most clearly with iodine deficiency. There is little or no fossil evidence to indicate that animal foods derived from the aquatic environment played a significant role in the diet of either early or later hominids until the Upper Paleolithic (35 000–40 000 y ago) period (4. bulbs. evolved and varied throughout the Paleolithic period. and 14% protein).71:665–7 (editorial). depending on the species and the ecologic niche that was exploited. equally importantly. copper. as Cunnane did. lines. 9:219–35. During the Paleolithic period (a time period extending from roughly 2. People can survive even severe iodine deficiency but they cannot thrive or reproduce. and lacustrine locations probably favored human brain evolution by providing abundant energy and protein but. The fossil record shows that invertebrate shell refuse piles (middens) and fossilized fish remains associated with hominid occupation sites did not appear until the Upper Paleolithic period. these diminutive hominids would also have had little success in confrontational scavenging and stealing prey from large. Metabolic maladaptation: individual and social consequences of medical intervention in correcting endemic hypothyroidism. Lipid-rich seeds and nuts (mean energy density: 13. which affects more than a billion mostly vegetarian people in inland areas of all continents. roots.15:908–32. Lipids 1999. Consequently. Cunnane SC. rather. human brain evolution would have faltered and long-term colonization of those areas would have ceased until the appropriate foods were found or supplements were invented. 4.18 kJ/g. and iron are more abundant and available from seafood than from plants. 3 hominid genuses (Australopithecus. Brand Miller J. and barbed spears (4. concurrent with the technologic advent of hooks. Tubers.14 kJ/g) would have been selected preferentially over vegetable foods when available. carnivo- Downloaded from ajcn. copper. Equally important is the issue of access to reliable sources of foods rich in brain-selective nutrients that required minimal effort to locate and consume. Stephen C Cunnane Department of Nutritional Sciences University of Toronto Toronto M5S 3E2 Canada E-mail: s. and broccoli (flowers).3% of our entire wild plant food database and have a mean energy density of 4.org by guest on May 8. Br J Nutr 1998.cunnane@utoronto. lettuce (leaves). nets. leaves. Rift Valley lakefish and shellfish provided brain-specific nutrition for early Homo. Hence. Hunter-gatherer diets—a different perspective.ca REFERENCES 1. hunter-gatherers collected plant-food species not randomly but in a fashion predicted by optimal foraging theory that would tend to maximize the ratio of energy capture to energy expenditure. that Paleolithic hunter-gatherers would have maintained identical nutritional patterns characteristic of modern hunter-gatherers. hence. Broadhurst CL. Boyd Eaton S. these categories would subsume such modern vegetables as potatoes (tubers). 6. fish. The basis for this hypothesis is that terrestrial foods are deficient in iodine and contain little docosahexaenoate (only in animal tissue). These food categories accounted for 29. In contrast. Geelhoed G. such as docosahexaenoate. 5). Mann N. Zinc. 5). In regard to the physiologic protein ceiling. Plant-animal subsistence ratios and macronutrient estimations in worldwide hunter-gatherer diets. Cordain L. 2013 . If the nutrient and energy supplies were consistently inadequate in some geographic areas over thousands of years. estuarine. Speth JD. Iodine is essential for energy metabolism. that were crucial for the rapid hominid brain expansion that occurred during the Early Paleolithic. et al. Such foods would have to have been available for thousands of years before intelligence had risen sufficiently to conceive of and experiment successfully with true fishing or hunting and trapping of wild animals. The importance of energy and nutrient supply in human brain evolution. There is abundant evidence suggesting that the hominid’s diet was not static but. including essential fatty acids. Am J Clin Nutr 2000. brain-selective nutrients. 71:682–92. it may be inappropriate to assume. Paranthropus. marine. radishes (roots). In regard to our estimation of the mean plant-food macronutrient profile (62% carbohydrate. Holt SHA.6 million y ago until 10 000 y ago).6 million y ago until 250 000 y ago) and Middle (250 000–40 000 y ago) Paleolithic hominids. amphibians. As we mentioned in our article. coastal peoples experience no known nutrient deficiencies affecting brain function.nutrition. Hence. and flowers are plant-food categories and are included in Table 3 of our article.

Sci Am 1992.” In a previous article. The first evidence supporting the systematic use of coastal resources is dated between 127 and 57 thousand years BP (5).colostate. Aiello LC. Carcass consumption sequences and the archaeological distinction of scavenging and hunting. N Engl J Med 1985. because the energyprotein ratio in African ruminant marrow (477 kJ/g protein) is almost 20 times greater than for African freshwater fish (27 kJ/g protein). 8. The human genus. Thus. By 150–100 thousand years BP. The fossil record indicates that the passive scavenging of the abandoned and defleshed long bones and skulls of herbivores with their intact contents of marrow and brain would have represented the primary large animal food source for early ancestral humans (6. the energy return versus the energy expenditure for scavenged marrow bones would have far exceeded that available from the manual capture of freshwater fish. Blood pressure. 2013 Reply to SC Cunnane Modern humans are noted for their large brain but factors related to the brain’s evolution are imperfectly understood.34(suppl):S39–47. 10. 5). The evolution of a large metabolically active brain in our species required food sources that were energetically dense (2. 3) and that contained docosahexaenoic acid (22:6nϪ3) (8). serum lipids. . . rous predators. Well before this. Pauletto P. 7). Lipids 1996. mermaids.6:77–88. Furthermore. 5). Although East African freshwater fish are good sources of 22:6nϪ3 (549 mg/100 g). Robertson ML. Cunnane states that a “shore-based ecologic niche was uniquely able to stimulate expansion of the primate brain. 5. Blumenschine RJ. 9. Am J Hum Biol 1994.21:727–50. Crawford MA.org by guest on May 8. Cunnane and others (1) described “the African savanna ecosystem of large mammals and primates [as] associated with a dramatic decrease in relative brain capacity associated with little docosahexaenoic acid” (DHA. 2). Taken together. 3. Wheeler P. Paleolithic nutrition. 15:639–59. Lipids 1999. Evidence for the unique function of docosahexaenoic acid during the evolution of the modern hominid brain. Maritime hunter-gatherers: ecology and prehistory.31(suppl):S309–12. Wood B. Data suggest that the major increase in encephalization in Homo occurred during the Middle Pleistocene. . For theory building in archaeology: essays on faunal remains. Strandloopers. Osborn AJ. 20:4nϪ6) are regarded as absolute requirements for advanced neural growth in humans and other mammals (1–3).1586 LETTERS TO THE EDITOR REFERENCES 1. Evolutionary perspectives on human nutrition: the influence of brain and body size on diet and metabolism. 2. Cavallo JA. Broadhurst CL. and systemic modeling. they are a poor energy source (498 kJ/100 g) (9) and are less energetically dense than is a mixture of wild. The expensive tissue hypothesis. Curr Anthropol 1995. spatial analysis. particularly DHA and arachidonic acid (AA. whereas marrow consumption would not have been. aquatic resources. Scavenging and human evolution. If consumption of coastal resources underlies expansion of the mod- . 7. Loren Cordain Department of Health and Exercise Science Colorado State University Fort Collins. NSW 2006 Australia S Boyd Eaton Departments of Radiology and Anthropology Emory University Atlanta. Curr Anthropol 1980.284:65–71. Because tropical freshwater fish and shellfish and many marine fish offer plentiful preformed DHA. Science 1999. and other fairy tales: ecological determinants of marine resource utilization.36:199–222. Yesner DR. New York: Academic Press. absolute brain size in Homo appears to have been within the modern range. Puato M. CO 80523 E-mail: cordain@cahs. GA 30322 Neil Mann Department of Food Science Royal Melbourne Institute of Technology University GPO Box 2476V Melbourne. when the option was available.312:283–9.267:90–6. I found little evidence to support this.edu Janette Brand Miller Department of Biochemistry University of Sydney Sydney. Blumenschine RJ. fish consumption would have been constrained by the physiologic protein ceiling. Leonard WR.nutrition. VIC 3001 Australia Downloaded from ajcn. Because scavenged marrow is a more highly concentrated energy source (3289 kJ/100 g) than is freshwater fish (498 kJ/100 g). whether Homo is viewed as a single or multiple species (4. 22:6nϪ3). Bloom M. Eaton SB. and fatty acids in populations on a lake-fish diet or on a vegetarian diet in Tanzania. Scavenged marrow is a rich energy source (3289 kJ/100 g) and scavenged brain is a more concentrated source of 22:6nϪ3 (861 mg/100 g) than is East African freshwater fish. 1977:157–205. Konner M. 6. edible plants (540 kJ/100 g) consumed by hunter-gatherers (10). Angeli MT. scavenged marrow and the brain that was concurrently present in the skull of the defleshed skeleton would almost always have been chosen over active capture of either fish or aquatic invertebrates. the data indicate that scavenged marrow from ruminant long bones would have represented the concentrated energy source required for hominid brain evolution and that the brains of scavenged skulls would have represented the predominant source of 22:6nϪ3. et al. 4. Hominids did not become successful hunters of large game until the Middle to Upper Paleolithic period. A consideration of its nature and current implications. Collard M. Abundant long-chain polyunsaturated fatty acids (LCPUFAs). ed. Members of the genus Homo have always been distinguished by a large brain relative to body size (4. Cunnane et al propose that the lacustrine and marine food chain was being extensively exploited at the time cerebral expansion took place in the ancestral line leading to modern humans (1. et al. J Hum Evol 1986. In: Binford LR. overall body size and degree of sexual dimorphism in Homo had arrived at essentially the modern level (5). 600–150 thousand years before present (BP) (4).

The human career: human biological and cultural origins. the essential fatty acids linoleic acid and ␣-linolenic acid. 1973. Evolution of the brain and intelligence. Br J Nutr 1998. Body mass and encephalization in Pleistocene Homo. In fact. including eicosapentaenoic acid (a precursor of DHA) and DHA (12. Not only are the selective pressures involved in this scenario unspecified. Where do humans get the LCPUFAs that are so critical in brain development? Preformed LCPUFAs can be obtained from foods. 9). 2013 REFERENCES 1. what factors explain the precipitous increase in brain size on 3 different continents by members of the genus Homo well before evidence for the exploitation of shore-based resources? If humans in the African Rift Valley consistently utilized lacustrine resources (2). 1587 the brain is free to enlarge and modern humans result. “elongation and desaturation of ␻3 fatty acids in the human liver is very active and capable of providing the high levels of long ␻3 PUFA required by the developing brain” during the crucial stage of brain development (3). This technologic explosion was not accompanied by any increase in human brain size and thus some other factor. for example. Policy and Management Division of Insect Biology University of California Berkeley. Tissues of the eye and brain can also synthesize DHA if the appropriate precursors are available (7). the savanna baboon ( Papio spp. Although the question of where and when anatomically modern humans originated remains unresolved (5). Nor is it the case that all large savanna species have small brains relative to their body mass.berkeley. Evidence for the unique function of docosahexaenoic acid during the evolution of the modern hominid brain. fetal liver. Although nϪ3 deficiency can be induced in humans by a very poor supply of ␣-linolenic acid or an excessive supply of linoleic acid relative to ␣-linolenic acid. Archaeologic evidence testifies to the increasing technologic proficiency and continuous exploitation of terrestrial mammals by members of the genus Homo over the course of their evolution (5). The idea that the African savanna could not support largebrained species (1. the possibility of nϪ3 fatty acid deficiency in the wild-food diets of evolving humans seems unlikely “because of the abundance of these fatty acids in nature. 2nd ed.edu Downloaded from ajcn. The association of stone tools with the earliest evidence for hominid exploitation of meat and marrow from large terrestrial ungulates strongly suggests that even the earliest humans used extrasomatic (cultural) innovations to help them solve immediate dietary problems (11). . no information is provided as to how these large-brained humans were then able to provide DHA and other brain-specific nutrients for themselves or their developing offspring once they moved away from lacustrine or shore-based environments. a progressive increase in fatty acid length and degree of unsaturation from maternal liver to placenta. the quiescent brain appears to be waiting patiently for humans to discover aquatic foods and then. No evidence suggests that primates in this environment have brains smaller for their body mass or a lower encephalization quotient (6) than do their counterparts in tropical forests. marrow. Rift Valley lake fish and shellfish provided brain-specific nutrition for early Homo.org by guest on May 8. Although the conversion of ␣-linolenic acid to DHA in humans is stated to be weak (2). Mena P. 2. liver. their tissues have relatively high proportions of n Ϫ3 fatty acids. eureka. The direct incorporation of dietary LCPUFAs in the developing brain was also shown (3). Chen H. not only sufficient amounts of DHA but also of AA and docosatetraenoic acid (14). In humans. but such foods seem unnecessary for brain expansion in the human lineage. including LCPUFAs (12. 5. Holliday TW. flesh. and human breast milk contains both linoleic acid and ␣-linolenic acid as well as LCPUFAs. Effects of parenteral nutrition with high doses of linoleate on the developing human liver and brain. “were enlarged even beyond the extent expected for their large bodies”(6). The accretion of docosahexaenoic acid in the retina. 7. has been suggested to underlie it (5). et al. tongue. Arachidonic and docosahexaenoic acids are biosynthesized from their 18-chain carbon precursors in human infants. have brains that. Stinson A. and other parts of wild terrestrial mammals would have served as a concentrated source of many essential nutrients required by early humans. 6. 13) and wild-plant foods would provide ␣-linolenic acid and linoleic acid. Jerison HJ.387:173–6. Katharine Milton Department of Environmental Science. or LCPUFAs can be synthesized in the mammalian liver from dietary precursors. the fossil and archaeologic records suggest that the modern physical form of our species evolved before the modern capacity for culture (5). 8. including DHA (8. Martinez M. and fetal brain has been documented (3).75:124–7. Crawford MA.22:133–8.93: 49–54. The brains. The “aquatic foods argument” also offers no real explanation for why these foods stimulated human brain expansion. Lipids 1999. Nature 1997. Calculations indicate that a diet composed of 35% terrestrial animal matter and 65% terrestrial plant matter would have provided more than adequate raw material for brain-building purposes. possibly the development of fully modern language. New York: Academic Press.LETTERS TO THE EDITOR ern human brain. 2) seems inaccurate. Proc Natl Acad Sci U S A 1996. Crawford MA. ancestral humans likely would have exploited aquatic foods whenever possible. Ballabriga A. their small minimum requirements and the enzyme preference for the linolenate family” (3). Trinkaus E. Broadhurst CL. Salem N Jr. Taken together. why the long period of stasis in human encephalization between 1800 and 600 thousand years BP (4)? Another puzzle concerns the technologic explosion (5)—a burst of creativity in anatomically modern humans that appears to have begun fairly abruptly in the Late Paleolithic period some 40 thousand years BP and involved the dramatic acceleration of cultural evolution. Chicago: University of Chicago Press. 4. ie. World Rev Nutr Diet 1994. 3. Klein RC.79:3–21. Dietary pressures appear to have been a major stimulus in human evolution (10). Bloom M.) and savanna-woodland vervet (Cercopithecus aethiops) have relatively large brains and high encephalization quotients compared with most African forest primates (6). As highly opportunistic foragers. Wang N. Lipids 1987. Full-term infants can synthesize DHA. 1999.34(suppl):S39–47. 13). Anderson RE. Because wild animals consume diets with very low ratios of nϪ6 to nϪ3 fatty acids. Ruff CB. CA 94720-3112 E-mail: kmilton@socrates. Cunnane SC. In this Lamarckian scenario.nutrition. over the course of their evolution. data do not suggest any causal association between the exploitation of aquatic foods and human brain expansion. Elephants. Wegher B. Broadhurst CL. Uauy R.

and 50 g) were replaced by equivalent amounts of isolated soy protein. only the groups who received either 50 g ISP or casein did not receive a mix of the 2 proteins. the results of our study.6% and 2. a strong body of evidence from several laboratories. There is still more to learn about soy.” This appears to be an incorrect conclusion considering the data and discussion put forth by these authors. Therefore. Clark JD.5-million-year-old Bouri hominids. The authors state that non-HDL cholesterol equates to subtracting HDL from total cholesterol (TC). 40. Specifically. daily servings of soy on lipid profiles. we cannot accept the conclusion of the authors as a true measurement of non-HDL cholesterol (2). World Rev Nutr Diet 1998. Evol Anthropol 1999. in fact. However. we must insist on a strong body of evidence. the control group received 50 g casein (ie. 30. Furthermore. A hypothesis to explain the role of meat-eating in human evolution. Effects of feeding 4 levels of soy protein for 3 and 6 wk on blood lipids and apolipoproteins in moderately hypercholesterolemic men. CT 06830 E-mail: peakwell@idt. Milton K. they mistakenly labeled non-HDL cholesterol. CT 06830-4697 REFERENCES 1. ␻3 fatty acids in meat raise plasma levels of eicosapentaenoic and docosapentaenoic acids. before we accept generalized conclusions such as those made in Teixera et al’s article. Potter SM. World Rev Nutr Diet 1994. Teixeira SR. Thus. Therefore. our conclusion about the replace- . soy. with all of the technical machinery available to the authors for firsthand measurement of these various lipids. including a meta-analysis. 2nd ed. For all other groups.2%. Hannum S.83:233–4 (abstract). 2013 Reply to DS Kalman and CM Colker Dear Sir: We reassert our conclusion that “as little as 20 g soy protein/d instead of animal protein for 6 wk reduces concentrations of non-HDL cholesterol and apo B by Ϸ2. lipoprotein(a). 14. Am J Clin Nutr 2000.net Carlon M Colker Department of Internal Medicine Greenwich Hospital 6 Perryridge Road Greenwich. Although it can be said that 65–70% of TC is carried as LDL.1588 LETTERS TO THE EDITOR deserves more independent prospective trials to further determine both the risks and benefits of including this food in the Western diet. Cordain L. Utilization of ␻3 fatty acids in companion animal nutrition. Environment and behavior of 2. 10. 1998:59–62. it is questionable for them to put forth that observed changes in mathematically determined VLDL and LDL are accurate and truly reflective of the non-HDL lipid pool.6% and 2. HDL. 2.83:176–85.75:105–8. 10–15% as VLDL. In keeping with the given protocol of evaluating the effects of various levels of soy protein supplementation (soy with or without casein) in moderately hypercholesterolemic adults. Weigel R. so that the total protein intake remained constant. There is still more to learn about soy Dear Sir: Teixera et al (1) concluded that “consuming as little as 20 g soy protein/d instead of animal protein for 6 wk reduces concentrations of non-HDL cholesterol and apo B by Ϸ2. additionally. We fully agree with Kalman and Colker that there is much more to learn about soy. Mann NJ. NJ: Quest Diagnostics Incorporated.284: 625–9. Fisher DA. Teterboro. respectively. Reinhart GA. no replacement). and apolipoprotein B. respectively” (1). Sinclair AJ. Kaplan B. Such a body of evidence must be in place before we can make recommendations to patients or to the general public. 13. de Heinzelin J. and 20% as HDL and that there are various subfractions of LDL. apolipoprotein A-I. triacylglycerol.2%.8:11–21. the conclusion that ingesting ≥ 20 g ISP/d instead of animal protein for 6 wk reduces non-HDL cholesterol is both misleading and inaccurate. Budowski P. 9. Merlob P. Certainly. provide additional support for the FDA-approved health claim. The qualifying amount of soy protein approved in this health claim (25 g) is similar to and consistent with the amount of soy protein shown to decrease blood lipids in our paper (20 g).org by guest on May 8. Science 1999. Eaton SB. different amounts of casein (20. Eaton SB III. Endocrinology: test selection and interpretation. Simply stated. Douglas S Kalman Peak Wellness.71:1077–84. the authors had the means to measure actual VLDL and LDL because they had measured TC. the group who received 20 g isolated soy protein (ISP) also received 30 g casein. 11. White T. The Food and Drug Administration (FDA) reviewed the extensive literature published on this subject and in October of 1999 approved a health claim for foods containing soy protein (5). Hasler CM.83:12–23. Inc 50 Holly Hill Lane Greenwich. Sinclair AJ. their conclusion does not match the test. Mature milk from Israeli mothers is rich in polyunsaturated fatty acids. Given that only the macronutrient and isoflavone contents of the diet were quantified and that actual meat. Dietary intake of long-chain polyunsaturated fatty acids during the Paleolithic. et al. The comment of Kalman and Colker regarding our conclusion may have resulted from a misunderstanding on their part of our study design. In our study. the role of soy and soy protein as medicinal foods for the treatment of mild to moderate hypercholesterolemia Downloaded from ajcn. the error is within the framework of labeling non-HDL as VLDL + LDL cholesterol. World Rev Nutr Diet 1998. or other lipid-affecting nutrients were not evaluated. already exists on the effects of soy protein on blood lipids (2–4). Druckmann H.nutrition. Mann NJ. 12. the authors’ conclusions become weak. Erdman JW Jr. Hayek MG. World Rev Nutr Diet 1998. However. In the apparent zeal of the authors to show the positive effect of small.

lipoproteins and apolipoproteins. Clin Chem 1972. Meta-analysis of the effects of soy protein intake on serum lipids. Grundy S. it is common knowledge that the quantitative and qualitative lipid composition of domesticated meats is vastly different from that found in wild game. 2.71:1077–84. Hasler CM. The dependence on hunted and fished foods for subsistence was 86–100% (modal value) and 66–75% (median value). Laboratory measurement of lipids. Vega G. J Nutr 1995. Cook-Newell ME. especially saturated fat. and it corresponds to the cholesterol from all apolipoprotein B–containing lipoproteins [ie.10:668–71 (editorial). Our take-home messages were that hunter-gatherer diets were higher in protein and lower in carbohydrate than are current Western diets or dietary guidelines and that this macronutrient balance may provide insight into potentially therapeutic diets. Most patients in our study fell into 1 of the 2 previous categories.5 mmol/L (200–400 mg/dL). and polyunsaturated fat. Washington. IDL. 6. Anderson JW. Levy R. 3. N Engl J Med 1995. McNamara J.333:276–82. 8. Estimation of the concentration of low-density lipoprotein cholesterol in plasma. We labeled non-HDL cholesterol as VLDL + LDL cholesterol because the traditional definition of LDL entails LDL + IDL + lipoprotein(a) (7). 10.LETTERS TO THE EDITOR ment of animal protein with soy protein was indeed consistent with our data and study design.91:820–7. Warnick G. DC: AACC Press. IL E-mail: j-erdman@uiuc. 1994:21–42. Carroll KK. low in nϪ6 fats.18:499–502. Overview of the diagnosis and treatment of lipid disorders. Besides the analysis of macronutrients and isoflavones. and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). lipoproteins and apolipoproteins. Washigton. Warnick G. DC: AACC Press. when triacylglycerol concentrations are between 2. Erdman JW Jr. Belcher J. LDL cholesterol is also commonly determined through use of the Friedewald formula (9).66 mmol/L (400 mg/dL) (10). even on a whole-carcass basis (5). 2013 Macronutrient estimations in hunter-gatherer diets Dear Sir: We disagree with the editorial (1) that accompanied our recent article on hunter-gatherer plant-animal subsistence ratios (2). Food and Drug Administration. We agree that measuring actual concentrations of LDL and VLDL cholesterol separately would provide additional useful information. Milton appears to have misinterpreted our findings as well as Lee’s (3) original analysis of the Ethnographic Atlas (4). Friedewald W. VLDL. McNamara J. Nowhere in our article did we recommend that people should eat high-fat.nutrition. The Expert Panel. Soy intake was also known because the only form of soy consumed was the one provided in the test protein (1). soy protein and coronary heart disease. Sandra R Teixeira Sandra Hannum John W Erdman Jr Division of Nutritional Sciences University of Illinois at Urbana-Champaign Urbana.125:589S–93S. Milton’s editorial repeated the same error that has occurred continually in the anthropologic community since Lee published his work 32 y ago (3). In: Rifai N. Schaefer E. especially because LDL cholesterol is used historically to determine risk of coronary heart disease (8).296:3015–23. than does meat from grain-fed domesticated animals. MO Clare M Hasler The Functional Foods for Health Program University of Illinois at Urbana-Champaign Urbana. Evaluation. In: Rifai N.org by guest on May 8. IL REFERENCES 1. we analyzed Lee’s sample of 58 hunter-gatherer societies as a subset and obtained results almost identical to those of our analysis of the entire sample (n = 229). This is one of the reasons our reanalysis of the Ethnographic Atlas is original and noteworthy. Bachorik P. JAMA 1993. Hannum S. it would be that dietary fat should emulate fat sources found in game meat and organs (high in nϪ3 fats. eds. and high in monounsaturated fats). Measurement of low-density lipoprotein cholesterol concentration. non-HDL cholesterol is by definition any cholesterol that is not associated with HDL particles. Regarding the non-HDL-cholesterol values. monounsaturated.3 and 4. without use of the preparative ultracentrifuge. If any implication were to be inferred. so we chose not to use the Friedewald formula. Game meat contains lower proportions of fat. Teixeira SR. J Am Diet Assoc 1991. 9. Does measurement of apolipoprotein B have a place in cholesterol management? Arteriosclerosis 1990. We chose to report non-HDL cholesterol because it has been shown to be a good indicator of coronary heart disease (6). Frederickson D. Am J Clin Nutr 2000. 1589 4. LDL-cholesterol values obtained by the Friedewald formula show considerable variability as compared with those from ultracentrifugation (10). Effects of feeding 4 levels of soy protein for 3 and 6 wk on blood lipids and apolipoproteins in moderately hypercholesterolemic men. Weigel R. eds. Laboratory measurement of lipids. Grinstead G. Fed Regist 1999. as stated in the paper. Downloaded from ajcn. Kritchevsky D. it is well known that this formula is not accurate if triacylglycerol concentrations are > 4. LDL. Warnick G.edu Susan M Potter Protein Technologies International/DuPont St Louis. 7. Milton’s statement that “emphasis on hunting occurred only in the highest latitudes” is also inaccurate because our analysis of Lee’s . Food labeling: health claims. cholesterol and atherosclerosis: review of the early history. and lipoprotein(a)] (6). Potter SM. Review of clinical studies on cholesterol-lowering response to soy protein. Moreover. we also calculated total cholesterol and saturated. Lee did not report the total food intakes derived from animal sources because he did not sum hunted and fished animal foods. Within the nutritional community. Rifai N. Summary of the second report of the National Cholesterol Education Program (NECP) Expert Panel on Detection.64:57699–733. domesticated livestock. Dietary protein. The comment of Kalman and Colker regarding nutrient analysis is also incorrect. 5. However. Johnstone BM. Although we did not report it in our article. 1994:107–24.

VIC 3001 Australia Downloaded from ajcn. most hunter-gatherer societies derived > 50% of dietary energy from animal foods and suggested that “the universally characteristic macronutrient consumption ratios of hunter-gatherers in which protein is elevated at the expense of carbohydrate” may have therapeutic health effects for modern humans. Humans. carbohydrate. and now in their letter to the Editor. Manson JE. eds. Eur J Clin Nutr 1999. and the ratio of total to HDL cholesterol). only 1 other society maintains a plant-animal subsistence ratio as high as that of the !Kung and only 13% maintain a ratio as high or higher than that of the Hazda. Brand Miller J. Traianedes K. Am J Clin Nutr 1999. both of which have been shown by the Ethnographic Atlas and modern quantitative studies to maintain high plant-animal subsistence ratios (67:33 and 56:44. respectively). intakes of fished food increase and of hunted animal food stay constant—the same conclusion we reached with our original analysis. Again. et al.53:895–9. particularly in subjects with preexisting kidney disease. Randomized trial on protein vs carbohydrate in ad libitum fat reduced diet for the treatment of obesity. Plant-animal subsistence ratios and macronutrient energy estimations in worldwide hunter-gatherer diets. A compilation of the few available quantitative dietary studies in hunter-gatherers showed a plant-animal subsistence ratio of 41:59 (6). Int J Obes Relat Metab Disord 1999. on the contrary: as intakes of plant food decrease with increasing latitude. DeVore I. Leonard WR.23: 528–36. only of lean protein from lean animals. we do not recommend increases in intakes of domesticated animal fat. highprotein animal foods improved blood lipids (LDL. They concluded that. VLDL. Ethnology 1967.89:1076–86.01) between dependence on hunting and latitude. Robertson ML. The effect of diet on plasma homocysteine concentrations in healthy male subjects.6:77–88.71:665–7 (editorial). Dietary protein and risk of ischemic heart disease in women.org by guest on May 8. In: Lee RB.70:221–7. 3. Increases in low-fat dietary protein at the expense of carbohydrate may have therapeutic effects. Lee RB. J Am Diet Assoc 1989. 5. O’Dea K. Replacement of carbohydrate by protein in a conventional-fat diet reduces cholesterol and triglyceride concentrations in healthy normolipidemic subjects. Chicago: Aldine. 9. Skov AR. 7. 2013 Reply to L Cordain et al Dear Sir: In their article in the March 2000 issue of the Journal (1).1590 LETTERS TO THE EDITOR REFERENCES 1. 11. Cordain et al discussed plant-animal subsistence ratios and likely macronutrient intakes (percentage of energy) in recent hunter-gatherer societies. Speth JD. Hunter-gatherer diets—a different perspective. Ireland P. increased dietary protein may reduce the risk of coronary heart disease (8) and reduce serum homocysteine concentrations (9) while facilitating weight loss (10) and improving insulin metabolism (11). worldwide. 1968:30–48. Eaton SB. data showed that there is no correlation (Spearman’s rho = 0. Milton K. the !Kung and the Hazda. Lipids 1992. hunter-gatherer societies. Holm L. Evolutionary perspectives on human nutrition: the influence of brain and body size on diet and metabolism. Li D. Toubro S. 2. which is similar to the aggregate value (45:55) we reported in our article. Because all hunter-gatherer societies are largely free of chronic degenerative disease. What general features of hunter-gatherer diets might contribute to this lack of degenerative disease? One important feature may be that many wild foods consumed by hunter-gatherers are similar or identical to foods consumed by their prehuman .nutrition. Further research is needed before this dietary pattern can be recommended without reservation. Ronn B. illustrating the adaptability of human metabolism to a broad range of energy substrates. The effects of diet differing in fat. Am J Hum Biol 1994. both recent and ancestral. et al. displayed a wide variety of plant-animal subsistence ratios. Of the 229 hunter-gatherer societies listed in the Ethnographic Atlas.edu Janette Brand Miller Department of Biochemistry University of Sydney Sydney.27:814–20. As discussed in my March 2000 editorial on this topic (2). 6. Astrup A. NSW 2006 Australia S Boyd Eaton Departments of Radiology and Anthropology Emory University Atlanta. 8. 71:682–92. Consumption of low-fat dietary protein at the expense of carbohydrate is the nutritional pattern that is consistent with our species’ evolutionary history and represents a viable dietary option for improving health and well-being in modern people. Hu FB. Sinclair AJ. triacylglycerol. Clin Invest Med 1999. preferably protein that may also contain significant amounts of nϪ3 and monounsaturated fat such as that found in game meat. What hunters do for a living. lipids and evolution. Wolfe and Piche (7) showed that the replacement of dietary carbohydrate with low-fat. CO 80523 E-mail: cordain@cahs. Am J Clin Nutr 2000. and fiber on carbohydrate and lipid metabolism in type II diabetes. Stampfer MJ. Furthermore. Wolfe BMJ. et al. Holt SHA. 10. 22:140–8. Man the hunter. Eaton SB. Mann NJ. Piche LA. Am J Clin Nutr 2000. Loren Cordain Department of Health and Exercise Science Colorado State University Fort Collins. The editorial deemphasizes the importance of animal foods in hunter-gatherer diets by citing 2 extreme and nonrepresentative societies.6: 109–236. or how to make out on scarce resources. Cordain L. there seems little justification for advocating the therapeutic merits of one type of hunter-gatherer diet over another. GA 30322 Neil Mann Department of Food Science Royal Melbourne Institute of Technology University GPO Box 2476V Melbourne. 4. total cholesterol. Mann N. Ethnographic atlas: a summary. Murdock GP.colostate.

Lipids 1986. Thus. Potential alternatives—gluconeogenesis or the use of ketones to fuel the brain—represent alternative. Consumption of digestible carbohydrate is the most efficient way for humans to obtain glucose for brain function. contemporary Western populations live surrounded by volumetrically concentrated foods that are high in sugar and fat and that can be ingested in enormous quantities. Hunter-gatherer diets—a different perspective. complex brain. in combination with the slow transit of ingesta characteristic of humans (4). 2. Mann N. Brand Miller J. Holt SHA. Speth JD. such as the Mbuti (Africa) and the Maku (South America). 3.15:488–98. 4. In his 1968 analysis of hunter-gatherer diets. then. 11). the overrepresentation of hunter-gatherer societies from more temperate locales and the differences in classification and data analysis between these authors. rather. few genetic adaptations to diet have been identified in humans. 5). Lee (8) reclassified some Atlas data and also excluded mounted hunters with guns and “casual” agriculturalists from his database. Such features.LETTERS TO THE EDITOR ancestors. this recommendation seems prudent. the Atlas provides. but foraging for a 100-g apple might prove to be the most therapeutic of all. it bears repeating that this Western dietary pattern. Downloaded from ajcn. Despite its general utility. 71:682–92. 8. some hunter-gatherers. To derive their conclusions on hunter-gatherer diets. often. Asia. suggesting that. plant foods contributed the bulk of daily energy intake. The !Kung and Hazda. To secure a dependable source of dietary carbohydrate. regardless of the macronutrient ratio or principal energy source. To date. Milton K. Plant-animal subsistence ratios and macronutrient energy estimations in worldwide hunter-gatherer diets. Sinclair AJ.” Given that most Westerners do not have access to wild game. Cordain et al (1) identified 229 hunter-gatherer societies in the Atlas. Although often stated. 6).edu REFERENCES 1. 2013 . Food transit times in humans are very similar to those of apes and notably different from those of carnivores (2. should make it difficult for hunter-gatherers to digest more than a limited quantity of these wild foods each day (2). For this reason. and wild plant foods tend to have a low glycemic index (5) and. Muscle tissue of wild prey is consistently low in fat and fat depots tend to be very small in most wild animals (3). they also combined 2 of Lee’s discrete categories (hunting and fishing) to estimate the total contribution of animal foods to energy subsistence. Eaton SB. Milton K. Nutritional characteristics of wild primate foods: do the natural diets of our closest living relatives have lessons for us? Nutrition 1999. it could be said that human biology is adapted to characteristics of a wide range of wild plant and animal foods but apparently is less well adapted to characteristics of many contemporary Western foods. in turn. most hunter-gatherers have a natural barrier between themselves and chronic dietary or energy excess. at best. different conclusions seem inevitable and all conclusions appear to merit closer study. whereas North America alone contained > 80% (135) of the 165 “hunting” societies listed in the Atlas. Examination of the literature suggests that huntergatherers throughout the world took full advantage of any dependable sources of dietary energy in their environment (9–11). dismissed by Cordain et al as “unrepresentative. in combination with a largely sedentary lifestyle. a brain that. Such dependable plant foods. Katharine Milton Department of Environmental Science. Policy and Management Division of Insect Biology University of California Berkeley. The gut proportions of humans do not indicate a highly carnivorous diet. inactive American might benefit from reaching for 100 g lean protein rather than a 100-g cheese danish. Past hunter-gatherers did not have unlimited dietary options but had to make the best of whatever was available in a particular habitat. they indicate adaptation to a diet made up of high-quality foods of all types and amenable to digestion primarily in the small intestine (14). O’Dea K. The technologic abilities of humans derive from their unusually large. In effect. However. more costly metabolic solutions. even devising complex technologies to secure energy from potentially toxic plant sources such as acorns and cycads (10. is fueled by a steady supply of glucose. this does not imply that such a dietary pattern is the most appropriate for human metabolism or that it should be emulated today. 14). CA 94720-3112 E-mail: kmilton@socrates. in their analysis. In contrast. Naughton JM. Given the uneven quality of most dietary data in the Atlas.21:684–90. Cordain L.” differ from many hunter-gatherers listed in the Atlas precisely because they have been relatively well studied dietarily—in both cases. It is extremely easy for individuals in Western nations to consume far more energy each day than they expend. under normal conditions. Cordain et al’s comments on the “low carbohydrate content of wild plant 1591 foods” seem largely beside the point—what is key is the steady availability of energy from 1 or 2 reliable wild-plant staples. Cordain et al (1) used Murdock’s Ethnographic Atlas (7). considerable dietary fiber (4. Certainly the average well-nourished. a “quantitative overview” (1) of the dietary behaviors of recent (largely 20th century) hunter-gatherers and “in almost all cases represents subjective approximations by Murdock of the ethnographer’s or anthropologist’s original observations” (1). Am J Clin Nutr 2000. tended to be relied on heavily for dietary energy. 13). In contrast. Most wild foods have a low energy density compared with the refined foods of Western nations. Australia.org by guest on May 8. In Lee’s opinion. and South America could be classified as hunter-gatherers. in their evolution. Gut proportions of carnivorous mammals differ from those of humans (2). only 24 societies from all of Africa. established symbiotic trade relationships with indigenous agriculturalists (12).nutrition.71:665–7. humans tended to resolve dietary problems primarily by using technology rather than biology. Am J Clin Nutr 2000. Most wild fruit is hexose dominated (4). Although Cordain et al noted a neutral or therapeutic effect for high protein intakes in some instances.berkeley. appears to contribute to many chronic degenerative diseases that affect Western nations but are largely or completely absent in huntergatherer and similar societies (2. Animal foods in traditional Australian Aboriginal diets: polyunsaturated and low in fat. Hu and Willard (15) recently cautioned application of their findings on heart disease and a high protein intake to public dietary advice because “a high dietary protein intake is often accompanied by high saturated fat and cholesterol intakes. There seems little doubt that many hunter-gatherer societies had a high intake of animal protein (and animal foods) by presentday standards (1.

Our assumption is that there is a pool at the intestinal mucosal surface from which iron is taken up by special nonheme-iron receptors. maize) and the iron compound to be tested. Reply to TC Campbell. 6. needs to be multiplied by the amounts of iron present in the corresponding labeled pools. This mucosal pool is directly connected with the nonheme. Bovell-Benjamin et al (1) compared the absorption of iron from ferrous sulfate.58 ± 0. In that pool.38:71–99. when the same amounts of iron were added to the same maize meal. An incomplete isotopic exchange between iron in another iron chelate.044 between biosynthetically radioiron-labeled maize and the iron in FeNaEDTA (n = 10). t = Ϫ0. based on tracer methodology. Primate diets and gut morphology: implications for human evolution. Are health and ill-health lessons from hunter-gatherers currently relevant? Am J Clin Nutr (letter.77). respectively. 1878. Ecol Food Nutr 1999. Man the hunter. in press).299. FeNaEDTA. Brand-Miller JC. Kuhnlein HV. P = 0. 14. In unpublished studies in our laboratory we found an absorption ratio of 0. Hu FB.1592 LETTERS TO THE EDITOR absorption studies were done in the same subjects. Protein and carbohydrate resources of the Maku indians of northwestern Amazonia. Holt SHA. This would imply that it is impossible to estimate the total amounts of iron absorbed. the data can be analyzed in a way that is more sensitive and specific by comparing the absorption in the same subjects and not in 2 groups of subjects. In: Harris M. Turner NJ. Actually.8%. They concluded that iron absorption was better from ferrous bisglycinate than from ferrous sulfate or ferric trisglycinate and that ferrous bisglycinate was an effective and safe source of iron that was particularly useful as an iron fortificant in diets rich in phytate. In: Lee RB. Their other main conclusion was that iron from ferrous bisglycinate does not exchange in the intestinal nonhemeiron pool with the iron from maize or ferrous sulfate. We know from study 2A that iron in ferrous bisglycinate is less well absorbed than is ferrous sulfate when given alone.65) when ferrous sulfate was given together with ferrous bisglycinate and 2) that the percentage absorption of iron from a hypothetical chelate pool of ferrous bisglycinate was not influenced. Am Anthropol 1984. Philadelphia: Temple University Press. 1978:93–116. Downloaded from ajcn. These mean values suggest that the absorption was different in the 2 studies. 7. The Aborigines of Victoria.7 times greater absorption of iron from ferrous bisglycinate ( P < 0. This implies 1 ) that the absorption of iron from the nonheme-iron pool dropped by Ϸ40% (1/1. indicating a 4. The mean absorption of iron from the tracer for ferrous sulfate given alone with maize was 1.86:7–27. 8. Vol I. amount of iron is released into the nonheme-iron pool (maize-meal pool). when ferrous sulfate was given together with maize and ferrous bisglycinate (study 1B) the absorption was lower (1. Australian Aboriginal plant foods: a consideration of their nutritional composition and health implications.54:103–12. eds.org by guest on May 8. DeVore I. Ethnology 1967. the only way to correctly analyze the isotopic exchange between an iron compound and iron in a food is by comparing iron absorption from a biosynthetically radioiron-labeled food (eg.7% and 6.” which we know from several previous studies is uniformly labeled by the added ferrous sulfate. and biosynthetically radioiron-labeled maize was observed by several investigators (3–5). 2013 No advantage of using ferrous bisglycinate as an iron fortificant Dear Sir: In the June 2000 issue of the Journal. and that some unknown fraction is absorbed from a kind of possible mucosal-iron pool. iron absorption was 1. intraluminal nonheme-iron pool. absorption of the tracers was 1. Traditional plant foods of Canadian indigenous peoples. respectively. the absorption of iron from ferrous bisglycinate in study 1A cannot be calculated because we do not know 1) how much iron moved from ferrous bisglycinate to the nonheme-iron pool in maize. Bovell-Benjamin et al concluded that iron from ferrous bisglycinate does not exchange with the iron from maize or ferrous sulfate. Walker ARP. Econ Botany 2000. The authors combined the mean percentage absorption in their studies 1A and 1B. the percentage absorption. Chicago: Aldine. What hunters do for a living. It is not clear how the conclusion of “no exchange” was drawn between the labels in the intestinal pool in study 1B. that some iron is dissociated and exchanges with the nonheme-iron pool.nutrition. Iron chelates such as ferrous bisg- 5. 1991. All these results suggest that a fraction of iron chelates may form a separate pool. Nutritional influences in the geographic dispersal of Pleistocene man. eds. Pecan food potential in prehistoric North America. Milton KM. Murdock GP. 13. Lee RB.436. Milton K.0%. Smyth RB. Nutr Res Rev 1998. Am J Clin Nutr 2000. This conclusion was based on observations that. When we compared more correctly the individual ratios in absorption of the ferrous sulfate tracer in studies 1A and 1B. ferric iron is probably reduced to ferrous iron to be absorbable.0%). ferrous bisglycinate. All this implies that the iron absorption from ferrous sulfate given with maize in study 1A was measured correctly. and ferric trisglycinate added to a whole-maize meal.4%. However.11:5–23. 9. London: Trubner and Co. An interesting part of the discussion in the present study (1) addressed the process of absorption of iron from the intestines when strong iron chelates are also present. In comparisons of iron absorption from meals. Hall GD. or. Food and evolution: toward a theory of human food habits. Ross EB. 10. 6:109–236. It may be assumed that ferrous bisglycinate is partly dissociated and that an unknown. A corresponding comparison of ferrous bisglycinate in studies 1A and 1B showed that the absorption was the same when ferrous sulfate was given alone in study 1A and when given together with the same amount of ferrous sulfate in the same meals in study 1B (mean ratio: 0. Flodin NW. because the . The most obvious explanation is that some iron moved from “the ferrous bisglycinate pool” to the “maize pool. which were then 1. 1968:30–48.653 (t = 2. 11. but possibly considerable. P = 0. when the same amounts of iron as ferrous sulfate and ferrous bisglycinate were given separately together with the maize meal. However.3% and 6. this mean ratio was 1. 15.0% and 6. Philadelphia: Gordon and Breach.956. respectively.7% (study 1A). 71:850–1 (letter).05).0375). Ethnographic atlas: a summary. how to make out on scarce resources. An absolute condition in these kinds of tracer studies is to know the specific activity of the iron. 12. Willard WC. and thus 2) how much iron remained in chelate form.

as noted by Hallberg and Hulthén. we cannot accept the main conclusions drawn by Bovell-Benjamin et al. 2. iron status influences the absorption from both the iron in the chelate pool (as reported here) and the iron in the usual intraluminal nonheme-iron pool. Am J Clin Nutr 1977. the absorption from ferrous sulfate differed between studies 1A and 1B (P = 0. Martinez-Torres C. Romano EL. CA 95616 E-mail: lhallen@ucdavis. a significant difference.nutrition. 5. even if the cause of the 41% lower absorption of iron from ferrous sulfate in study 1B was dilution with iron from ferrous bisglycinate. MacPhail AP. Layrisse M. care was taken to provide the same amount of iron as sulfate or bisglycinate. However. Inc. If. so that the size of the intestinal iron pool was similar in the 2 studies.30:1166–74. 3.77. Allen LH. Our data were analyzed correctly because it is appropriate to control for the repeated-measures nature of the design.04). the opposite occurred. However.org by guest on May 8.333–43. Am J Clin Nutr 2000. New York: Alan R Liss. et al.0001). the means became more separated: the ratio of absorption of iron from ferrous bisglycinate to absorption of iron from ferrous sulfate in study 1A was 3. which was not significant. It is therefore appropriate to make a comparison of iron sources between studies. less label from bisglycinate would have been absorbed in study 1B. If the iron from these sources had exchanged in the intestinal pool there would have been a similar trend in iron absorption from both iron sources in studies 1A and 1B. the explanation is that some iron moved from the ferrous bisglycinate into the “maize pool. as Hallberg and Hulthén suggest. because a log transformation was used. This comparison showed that iron absorption from the bisglycinate was the same in each trial (P = 0.5 and increased to 6. Layrisse M.8 in study 1B. it still could not explain this amount of divergence. Leif Hallberg Lena Hulthén Institute of Medicine Department of Clinical Nutrition Sahlgrenska University Hospital Annedalsklinikerna S-41345 Göteborg Sweden REFERENCES 1. Further studies. 4. the results do not support their interpretation that there is an exchange between iron from ferrous bisglycinate and the nonheme-iron pool. The basis of our conclusion is that absorption of iron from ferrous sulfate and ferrous bisglycinate was the same whether the Department of Nutrition University of California Davis. Tukey’s test) but. In: Nutrition in health and disease and international development.71:1563–9. where its iron may be released and absorbed. Iron absorption from ferrous bisglycinate and ferric trisglycinate in whole maize is regulated by iron status.32:809–16. CA 94720 . or both. Importantly. Martinez-Torres C. Br J Nutr 1981. Viteri FE. Bovell-Benjamin A. 2013 Reply to L Hallberg and L Hulthén Dear Sir: Hallberg and Hulthén question the interpretation of our data that showed that ferrous bisglycinate is absorbed better from whole maize than is ferrous sulfate.” which is equilibrated with the ferrous sulfate iron. Bioavailability of iron from different meals. In fact. Torrance JD. Bothwell TH. As stated in our article. it increased to 1. As the geometric mean absorption ratio of iron from ferrous sulfate between studies 1A and 1B dropped to 0. What we did not do was use the repeated-measures procedure to examine the interaction between study and iron source. 1981. in effect the absorption ratios were compared. On the basis of our analysis of the data presented.59. Factors affecting the absorption of iron from Fe(III)EDTA. Lindsay H Allen Downloaded from ajcn. Fe(III)-EDTA complex as iron fortification. We have since used that procedure and found the interaction to be significant (P = 0. Renzi M. the average absorption of iron from ferrous sulfate was about one-fifth that of its absorption from ferrous bisglycinate (P < 0.45:215–27. Specifically. Hallberg L. In this way. Fe(III)EDTA complex as iron fortification. they question our conclusion that there was no exchange in the intestinal pool between iron from ferrous bisglycinate and iron from ferrous sulfate. Am J Clin Nutr 1979. There is no evidence to support the conclusion that ferrous bisglycinate is useful as an iron fortificant.edu Fernando E Viteri Department of Nutritional Sciences University of California Berkeley. 1593 labeled forms of these compounds were given separately or mixed together in the same meal. and there would have been less of a difference between the geometric absorption means. Also. Such a hypothesis might explain many of the seemingly contradictory results. not the absolute values.LETTERS TO THE EDITOR lycinate and FeNaEDTA are present initially in an iron chelate pool that is connected both with the common nonheme intraluminal pool (where an isotopic exchange may take place) and directly with the mucosal nonheme-iron pool.01). In addition.13 for ferrous bisglycinate.

Sign up to vote on this title
UsefulNot useful