Photo by Robb Hannawacker, while working for Joshua Tree National Park, via Wikimedia Commons
Mesquite is a different subfamily of the "legumes" the Mimosoideae not the fabaceae, and the food part derived from mesquite is not the seed, but rather the pulpy part of the pod wall. Ethnobotanist Richard Felger claimed indigenous populations could obtain their 50g of protein per day from birds, lizards, snakes etc., but preferred the energy for their bodies, provided by the flour milled from the mesquite pods, not seeds. Using a stone gyratory crusher, they would grind off the pulpy mesocarp, moisten the high sugar content flour into "cakes," and take them on hunting trips. The leathery endocarp containing the hard seed was discarded.
While the article Beans and Legumes: Are They Paleo is very thorough and excellent, the following studies and papers suggest eliminating the seeds eliminates trypsin inhibitors in mesquite bean flour, flatulence producing triglycerides, phytates by ten times less, and a host of other issues that are present in the seeds. Furthermore, carbon 14 data supports mesquite flour consumption approximately 10,000 years before present day.
As part of my Ph.D. in 1977, I worked on the protein and amino acid composition of mesquite pods and seeds and have published more than 100 papers since then, in addition to giving copious talks to international audiences in North and South America, Africa and India/Pakistan.
Certainly fresh fruits, veggies and meats are better nutritionally than mesquite flour, but for conventional diets, mesquite is used at only 12-15% to add flavor and aroma in baked goods at low concentrations.
Shouldn’t mesquite bean flour be approved for Paleo Diets?
Thank you for this courtesy,
Peter Felker, Ph.D.
Dr. Cordain’s Response:
Dear Peter,
Many thanks for sending the papers on mesquite beans. You clearly are the international expert on this topic, and I respect your knowledge of a fascinating topic. I appreciate you getting me up to speed on the nutritional aspects of this traditional food which clearly has been consumed in the Americas for tens of thousands of years. I read the papers carefully, and cross checked your voluminous references. Jennie Brand Miller is a close colleague and co-author on a number of papers -- she did the glycemic index experiments with this food. Given its reported high sucrose content, I am a bit surprised that it did not have a higher GI.
Clearly, traditional agricultural societies in the Americas and Asian utilized this plant as food on a regular basis, but this evidence does not necessarily mean that habitual consumption is necessarily healthful. I see that the phytic acid concentration is high in mesquite bean flour. This characteristic will necessarily bind (in vivo) all the divalent ions (iron, zinc, calcium, magnesium) you have reported in a dose dependent manner making them of low biological availability in the human gut.
It is also a bit problematic that phytohemoagglutinin (PHA) has been detected in Prosopis species when this lectin is normally only present in Phaseolus species. Hence, I suspect that there likely is a specific lectin, yet to be molecularly classified that is present in Prosopis which agglutinates RBCs but which probably is not PHA. To date, animal experiments have shown that PHA breeches the gut barrier and interacts unfavorably with the immune and GI systems. It would be interesting to determine if the agglutinating factor in Prosopsis also has similar in vivo physiological characteristics as Phaseolus PHA.
Note that heating/cooking does not necessarily destroy all of the antinutrient factors in legumes, particularly saponins. Unless, I missed something, I did not see the saponin content specifically of the pod of Prosopis species reported anywhere in the references you reported. It is almost certain that the saponin content of Prosopsis is high. The combination of lectin, saponin, phytate and trypsin inhibitors is an evolutionary strategy virtually all legumes have evolved to prevent predation by insects, microorganisms and birds and mammals. The degree of toxicity ranges from mild to lethal and generally produce adverse physiological effects in a chronic and dose dependent manner (see Arpad Pustzai's life's work).
Although consumption of Prosopsis species products dates back to at least 10,000 years, generally consumption of this legume or any legume cannot be done in their raw state, and requires cooking (eg. the advent of fire production at will). As I have pointed out in an extensive publication, “Ancestral Fire Production: Implications for Contemporary Paleo Diets,” the ability of produce fire is a very recent invention in the 2.5 million history of our genus Homo. Hence, all legumes would not have been a part of the dietary repertoire that shaped the current human genome. Accordingly, humans are not well adapted to legumes, even with cooking.
One of the nutritional obstacles that pre-agricultural people faced was the physiological protein ceiling which our group has extensively described in a paper we published in 2000 in the American Journal of Clinical Nutrition. Basically, protein becomes toxic for a variety of physiological mechanisms when ingested at about 35-40% of the total eucaloric intake. Wild game worldwide is typically very lean, and muscle meat of wild animals averages about 80% protein and 20% fat by energy. Accordingly, if you or any human were only to have wild game muscle meat as your only food source, you would rapidly (within a few days) develop protein toxicity which ultimately causes death -- you would be better off starving or fasting. The solution to the physiological protein ceiling (which all pre-agricultural people must have inferred) was to dilute the protein portion of game animals with either fat or carbohydrate. Since animal carcasses only contain tiny amounts of carbohydrate, then you are dealing with a food mixture of protein and fat. But the problem is that wild game have little fat. However, with selective butchering and carcass consumption, the fattier portions (brain, marrow, tongue, perinephral fat, mesenteric fat etc.) can be consumed and the leaner portions eaten at physiologically tolerable levels. A caveat to this problem is that larger mammals contain more body fat than smaller mammals. Hence bison would be preferred to squirrels or field mice to negate the effects of excessive protein.
A final solution is to dilute the high protein content of wild animals with carbohydrate from plant foods. The problem here is that most wild plant foods are generally inedible to humans unless cooked and processed, and most edible plant foods are only edible seasonally. There are obviously notable exceptions, but until the ability to start fire at will was developed, the carbohydrate from plant foods would have contributed only a small percentage of the total yearly diet (see our papers on the topic available at my website for the references)
So how does this concept relate to Prosopis? Any plant food which is a good source of starch, sugar and or fat would have been exploited by indigenous people during the Neolithic or slightly earlier to offset the physiological protein ceiling. By cooking and processing formerly inedible foods (eg. legumes) using recently invented technology (fire production at will) previously unexploited foods could now be consumed. The problem is that cooking and processing still does not fully remove antinutrients. Apparently, more work needs to be done to fully and completely analyze the antinutrients in Prosopis and then do both in vitro and in vivo testing in animal and human models.
There clearly are better food choices from a nutritional perspective, and that is my point. It's not that we can’t eat cooked Prosopis, but rather fresh meats, fish, seafood, eggs, organ meats, fresh fruits and veggies are better choices to mesquite beans.
Cordially,
Loren Cordain, Ph.D., Professor Emeritus
