Phytic acid: not enough of it to be an issue?


#1

In another thread Rob hinted that phytic acid might not be an issue, simply because the amount of phytic acid present in soylent might not be enough to do much damage. I decided to follow up on this idea. The tl;dr of what follows is that it appears, based on the data I looked at, that a typical soylent recipe doesn’t contain enough phytic acid to create problems, provided that there is a sufficiently generous quantity of minerals (particularly Zinc) in the soylent.

For starters, I calculated the following things. How much phytic acid would a typical soylent recipe contain? And, how much of each mineral could that amount of phytic acid bind? My hypothesis was that it might be the case that we could safely add enough of a given mineral that even if all of the phytic acid bound to that mineral, there would still be more than the RDA of that mineral not bound to phytic acid. This turned out to be the case for some minerals.

Let us assume a prototypical soylent recipe: 200g maltodextrin, 100g oat powder, 85g whey protein, 30g psyllium husk powder, 50g oil, and micronutrients. Here the phytate-containing ingredients are oat powder and psyllium husks. The oats will contain 0.89-2.4% phytic acid by mass (source). I couldn’t find any information about the phytic acid content of psyllium husks, so let’s say 0-3.5%. (3.5% is higher than any food listed in the source). Using these amounts, we obtain a range of 0.89-3.45g of phytic acid.

Phytic acid has a molar mass of about 660g/mol. I’ll use the molar mass of the major minerals to calculate how much of each these minerals could be tied up by 3.45g of phytic acid. Again I am being very pessimistic here, making the totally unrealistic assumption that all of the phytic acid (in my maximally pessimistic quantity estimate) binds to a single mineral.

To calculate this, first note that since I have 3.45g of phytic acid and the molar mass is 660g/mol, I have 3.45/660 = 0.00522 mol of phytic acid. That amount can bind 0.00522 mol of calcium, or of magnesium, etc. If I multiply by the molar mass of the mineral of interest I get the amount of mineral, by mass, that might be bound. The results are:

Calcium: 209mg
Iron: 291mg
Phosphorus: 161mg
Magnesium: 127mg
Zinc: 341mg
Copper: 331mg
Manganese: 287mg

As long as I have an excess of at least this amount of each mineral above the RDA, my phytic acid can’t possibly create shortages of these minerals. In the cases of calcium, phosphorus, and magnesium, it would be easy to add sufficient amounts to make this happen, while remaining well below the upper limits.

In the remaining cases – iron, zinc, copper, and manganese – the phytic acid could theoretically bind the entire supply of the mineral in question, even if we had the maximum safe amount of the mineral. This means that in these cases the issue becomes subtler. But it doesn’t necessarily mean that phytic acid is going to create a problem for these minerals.

Heuristically, we expect that a larger proportion of the phytic acid will bind to the minerals of which there is a lot: phosphorus, magnesium, etc. Only a small amount of phytic acid will bind to the minerals of which there is a small amount.

The reason the phytate isn’t an issue for phosphorus, calcium, and magnesium, is that in these cases there just isn’t much phytate relative to how much mineral there is. Since the phytate is shared by all of the minerals it binds, what is really true is that there isn’t much phytate relative to the overall quantity of minerals. When it comes to a specific mineral, we expect that the amount of phytate binding to that mineral is a proportion of the total phytate which is not too far away from the proportion of the mineral of interest out of the total quantity of phytate-binding minerals. Given that, what we can theoretically expect to see is that no mineral gets knocked down to a really problematic extent by phytic acid, since the overall amount of phytic acid is small relative to the overall amount of mineral.

Does this idea bear out in practice? Let’s see if we can find some empirical data.

Zinc

Zinc is supposed to have an especially high affinity for phytic acid. According to source, “molar ratios [of phytic acid to Zinc] in excess of 15:1 progressively inhibit Zn absorption and have been associated with suboptimal Zn status in humans.” I have 0.00522mol of phytic acid at worst. Say I include 30mg of Zinc: just a ways below the DRI’s upper limit of 40mg. Then I have 0.0004587mol of Zinc, for a molar ratio of 11.38. According to our source, this is safe! However, there isn’t a ton of error margin here, which shows that the balancing act of Zinc and phytic acid is a little bit delicate in our situation.

We haven’t taken into account other factors affecting Zinc absorption. The presence of high amounts of calcium hurts it; the presence of protein helps it. (Source.) Hopefully we are close enough in our estimate anyway.

Iron

Let us try to estimate whether phytic acid will interfere with our iron absorption. According to the same source, vitamin C and protein both reduce the extent to which phytic acid interferes with the absorption of iron. This is good news for us.

One study found that when people ate wheat rolls containing 2mg phytic acid (added artificially after the natural phytate was artificially removed), their iron absorption was inhibited by 18%, relative to their iron absorption when eating rolls with no phytic acid. Subjects ate two rolls, each prepared using 40g wheat flour, and we can estimate that the meal contained roughly 400 calories. A 400-calorie portion of our soylent should contain about 3.45g * (400/2,000) = 690mg of phytic acid, worst case: 35% the phytate content of the mentioned rolls. Moreover, we may conjecture that the absorption of iron from soylent would be ceteris paribus better than it would be from rolls, because soylent has high levels of vitamin C and protein. So, with much margin for error, the phytic acid in our soylent should inhibit iron absorption by less than 18%. This clearly poses no issue.

Other minerals

For copper and manganese, from what I saw there isn’t enough data to draw a lot of meaningful conclusions. We have contradictory data concerning the question of whether phytic acid hurts, helps, or doesn’t affect copper absorption. As far as I can tell there is exactly one study on how phytic acid affects manganese absorption. At least one study found that phytic acid improves selenium absorption. I found nothing about chromium or molybdenum.

The big picture

Overall, I have tried to argue that there is not enough phytic acid in soylent to create major problems, provided that sufficiently generous quantities of minerals are included. Since phytic acid has a large molar mass, the number of phytic acid molecules is much smaller than the number of minerals in soylent. This suggests a priori that the phytic acid shouldn’t create a major issue. For the minerals where empirical data exists, the data supports this conclusion; it suggests that the amount of phytic acid in soylent won’t inhibit the absorption of the minerals in question to a problematic extent.

I don’t consider all of this to close the book on the phytic acid question. I am certain that there are plenty of holes that can be picked in my science, and I didn’t use nearly as much data as might be desired. But I think the data is pointing in a good direction. Please let me know if you have any comments, criticisms, things to add, etc.!


Danger of ingesting too many micronutrients?
Phytic acid and mis-information?
#2

Approaching the same problem from a different angle, it seems like it would be possible to add phytase – the enzyme which disables phytic acid – to soylent. Apparently this is done with animal feed. It transpires that the phytase used in animal feed is outrageously cheap (like $1-2 for the amount used in a ton of animal feed). But this phytase also does not seem to be available in quantities less than a ton or so. There is also phytase sold as part of a generalized dietary enzyme supplement for humans, but I haven’t found any stand-alone phytase sold in small quantities for humans.


#3

Excellent post Nick.

This is a very interesting topic indeed.


#4

I believe that my post contains a substantial error, which is that I assumed that a molecule of phytic acid can bind to only one atom of mineral. I haven’t found anyone explicitly saying how many it can bind, but based on this guy’s comments it looks like a molecule of phytic acid can actually bind six atoms of mineral.

This could affect my conclusions about calcium, phosphorus, and magnesium. My 3.45g of phytic acid can actually bind 1.25g of calcium, 966mg of phosphorus, or 762mg of magnesium – that is, six times what I claimed. Interestingly it’s still safe to add this much “extra” of each of these minerals, which shows just how much slop room we had before. But in the case of magnesium most of it would have to come from food, not supplements, and in the case of calcium we would be pushing the upper limit.

To clarify, I don’t think it’s necessary to actually do this. Based on the examination we did of zinc and iron, the amount of extra minerals needed is probably much less. One reason is that in reality the phytic acid is distributed among the minerals, rather than all being given to one mineral. Also some of the phytic acid gets disabled by our bodies’ natural phytase. Also probably other things of this nature.

@8O8: Thanks!


#5

Is the upper limit you reference relative to food-borne micronutrients or supplement-based? I ask because I didn’t fully understand what you meant about magnesium having to come from food. Great work by the way!


#6

The body absorbs magnesium much more readily from supplements than it does from food, as it is receiving it in the raw mineral form. So while the RDA for magnesium is 400mg, the recommended upper limit for magnesium coming from a supplement (as most Soylent recipes have) is 350mg. And some DIY-ers have even found that to be too much. The salient point being magnesium toxicity is only associated with mineral supplements, not with food. So while you COULD consume an extra 762mg of magnesium, if you try to do it through a supplement things are not going to go well. Instead you should look for magnesium rich foods.


#7

Thank you for the excellent explanation!


#8

Excellent post, Nick! Thank you for writing this up.


#9

Good job, Nick. I agree: I don’t think this closes the book on the phytic acid issue, but it certainly represents progress in the right direction. I’ve been uncomfortable about people just blowing it off without really going into it; so it feels good to see someone make this kind of effort. It seems to be an extraordinarily difficult question (like so many of the critical issues in human nutrition) due to there being too many variables to nail down comfortably.

Obviously those who have hitherto dismissed the issue as of no importance due to the high molar mass of phytic acid have not gone very deeply into the matter; the capacity of the phytic acid molecule to bind multiple molecules of various minerals means that each phytic acid molecule is a sextuple-threat binding agent. Ahother matter that hasn’t yet received serious attention here is the capability of phytic acid to inhibit amylase, pepsin and trypsin, thus interfering significantly with the digestion of both starches and proteins.

My personal hunch is that those who keep repeating that phytates are not a significant issue are indulging in highly unscientific wishful thinking and that a questionable agenda is being pursued in trying to prove that there isn’t enough phytate in oat powder to be a problem. It seems to me that a more sensible way of proceeding would be either to look for a complex carb source that is low in phytic acid and high in phytase, or else to include phytase in the formula. I lean toward the latter option myself, but either would do the job. I wonder why no one has given buckwheat flour a serious look; buckwheat is not a cereal grain, thus avoiding the gluten issue, and it has some nutritionally valuable phytochemicals like rutin.

Anyway, thanks, Nick, for opening this subject and going into it seriously; this wins you some kudos for real scientific attitude.


#10

@J_Jeffrey_Bragg: I was looking forward to hearing from you about this! You raise a lot of good questions; I hadn’t heard anything about the ability of phytic acid to inhibit digestive enzymes. I’ll have to look into that!

My personal hunch is that those who keep repeating that phytates are not a significant issue are indulging in highly unscientific wishful thinking and that a questionable agenda is being pursued in trying to prove that there isn’t enough phytate in oat powder to be a problem.

It’s fair for you to feel that way. I think that the scientific thing to do is to look at the data that exists, and ask whether or not it says that phytates are an issue. If the data says they are an issue, then it is scientific to believe they are an issue. If the data says they are not an issue, then it is scientific to believe they are not an issue.

To be clear, are we in agreement that phytate is definitely a non-issue for calcium, phosphorus and magnesium, given sufficient quantities of those minerals? To me this part of this picture seems clear.

As far as the other minerals are concerned, I see plenty of room to continue doubting the safeness of the phytic acid in soylent. The studies I looked at suggest, in a non-rigorous way, that there will be no issue with zinc and iron given sufficient quantities of those minerals; but other studies might say different things.

However, in all fairness, I think that if you wish to say that a problem exists, you ought to find some studies which suggest that a problem exists! I.e., you ought to find some different numbers wherein the phytic acid comes out creating shortages of these minerals.

There is definitely a lot that we don’t know enough about. There are the minerals for which there is little or no data on their interaction with phytic acid; and there is the digestive enzyme issue that you have just introduced me to. So I definitely think caution is warranted in any case.

It seems to me that a more sensible way of proceeding would be either to look for a complex carb source that is low in phytic acid and high in phytase, or else to include phytase in the formula. I lean toward the latter option myself, but either would do the job.

With this I agree. Using phytase is a cleaner approach, because there are fewer things one has to be right about for it to work. If I had access to phytase then I would use it.

I think it’s useful to distinguish between figuring out what course of action to take, and figuring out what’s true. Maybe we are strongly concerned about getting enough minerals, and even a small probability that we aren’t is worth worrying about to us. Whether the chance is big enough to worry about, however, is distinct from the question of how big the chance is.

It is consistent to say that phytic acid probably isn’t giving us mineral deficiencies, and also that we want to take steps against the possibility of it doing so. To me, it seems like what’s probably true is that the effect of phytic acid in soylent is not very pronounced. As for what to do about it? Well, ideally I’d like to take steps against phytic acid anyway. Phytase is a very cheap enzyme; if I could order it quantities less than a ton, I already would have! Such is life.

Thanks for weighing in on this!


#11

Nick, I agree in general with nearly all that you’ve said here. I haven’t gone looking for studies because my online time is limited and every time I find a promising abstract the study itself is behind an expensive paywall, making online research frustrating for me.

I cope with my own phytate anxieties by (reluctantly) soaking my cereal grains with lemon juice and/or whey from homemade yoghourt. My crock pot won’t hold a low enough temperature to soak at 90-120F and I see no point in soaking too hot because it only denatures the phytase. I always put buckwheat in the mix, which has a surplus of phytase. I can’t soak everything that contains phytic aci d, though – imagine a peanut butter sandwich with soaked peanut butter! :frowning:

My homemade soylent uses ground oats; I also include buckwheat and spelt, though. And I DO soak the cocoa and peanut butter powders along with the grain.

Fair comment about the abundance of calcium – my own soylent has a full cup of high-yield dry skim milk in it, so I’m well covered for calcium. However, I am still concerned about the overall balance of the calcium - phosphorus - vitamin D triangle. Because there should be an optimum ratio of those three items, it’s not that good a solution just to throw more calcium at the problem and hope for the best, yet that is what nearly all of us are doing. We can’t know how much calcium – or phosphorus – is tied up in phytates, and I doubt that any of us actually knows how much vitamin D we ought to be consuming relative to our uncertain amounts of calcium and phosphorus. Do you see the problem here?

Damn right – far too little is known about these non-urgent grey areas of nutrition. Just because people aren’t dropping dead of deficiencies does not mean that everything is beautiful, our dietary intake (soylent or otherwise) is perfectly balanced, and all’s for the best in the best of all possible worlds! Ramiel Nagel does a good job of outlining the concerns associated with diets high in cereal grains and phytate vegetables in countries where nutrition may sometimes be marginal. Soylent consumption needs to be considered as analogous in one way to those countries – because it is a NON-DIVERSE kind of diet, so whatever might be lacking in its ingredients is reinforced by daily repetition and is not covered by diversity, unless perhaps the consumer uses soylent only 50% of the time or less. So I would say, if we’re going to consume 100g of oat powder (I’ve seen some formulae as high as 400g BTW) every single day, we need to be aware of all possible nutritional ramifications entailed by constant consumption of that particular ingredient.

We must not forget that what we are doing here is EXPERIMENTAL. Just because our intention is “perfect nutrition” with all RDIs satisfied does not mean that we shall automatically fulfil that goal. Some things may be missing from our fomulae; absorption of some things may be poor; absorption of some things may be blocked; some ingredients may have extra unintended consequences. Only time and widespread usage – plus some large-scale controlled experiments – will tell whether we have achieved optimal nutrition or not. Wouldn’t you agree?


#12

Thank you for the response!

I haven’t gone looking for studies because my online time is limited and every time I find a promising abstract the study itself is behind an expensive paywall, making online research frustrating for me.

I can feel you there. Not being in university at the moment, I have the same problem.

It’s cool that you’ve got a lot of measures in place to deal with phytic acid. I’d almost certainly be soaking and cooking my soylent, if I was planning on having a kitchen in the foreseeable future. I’m sure it would help with things besides phytic acid – such as the flatulence that I am more or less just hoping will get better over time.

Regarding the peanut butter, if you were sufficiently determined you could possibly make your own peanut butter, using your own soaked peanuts. Nut butters are made simply by putting nuts in a food processor (or similar) for several minutes. I imagine it wouldn’t have great shelf life, due to the added moisture.

However, I am still concerned about the overall balance of the calcium - phosphorus - vitamin D triangle.

This is new to me; can you explain the idea, or point me to somewhere I can read up? Thanks!

Your final two paragraphs make a lot of great points, and I’m in agreement about all of them.


#13

Basically, Calcium and Phosphorus need to be present in the correct ratio for proper calcium absorption and bone formation and maintenance, and correct vitamin D intake must also be there for Ca absorption in the small intestine as that vitamin is needed for hormonal regulation of proteins responsible for Ca absorption. If these relationships are not in proper balance then Ca absorption is shot to hell and you get rickets, osteoporosis and other associated ills.


#14

your links are kinda broken. There’s some random string attached to the end of each URL.
Also the 4th link is the same as the 3rd. Maybe you didn’t copy-paste correctly?
Otherwise, interesting results on Ca:P ratio. I knew about their competition for absorption but wasn’t aware of the effect on bone turnover. But that would make sense.
I’ve been just taking calcium and phosphorus at different times during the day to maximize absorption for each kind, but if their ratio is to determine bone turnover, then I need to be more careful about the amounts as well.

Kept reading and found this to be a good overview


#15

Yeh, sorry about the links – I had a 3-way battle with this dumb laptop and the app, looks like I was the loser there. App refused to post the 4 links nicely on 4 lines and ran them all together – I tried to fix it and only made things dramatically worse.

Tldr = Google “calcium, phosphorus, vitamin D” and read all about it. :frowning:


#16

I am currently attending university, so if you need to look at a study, feel free to PM me.


#17

Great work on OP Nick! It’s encouraging to see some preliminary calculations to suggest this is probably not an issue. Something worth considering I think is that there’s a reasonable chance that the phytic acid present in soylent constituents like oats is probably already bound to a bunch of the oat nutrients. Would there be so much that it could still bind with additional minerals (in the rest of the soylent), or just enough to severely harm the oats’ nutritional value? Is the binding factor of PA considered when the nutrient profile of oats is measured, or not?

Another thought is that there literally must be sources for phytase in smaller quantites. Labs use it for research, ergo it must be availabe via lab suppliers in lab-appropriate quantities. Does anyone have relevant industry knowledge here?


#18

I think everyone may be ignoring the most obvious way to determine if phytic acid is a problem or not: blood testing. Phytic acid is only a problem is you’re actually deficient in something, and a blood test should prove that out. Sorry if I’m being overly simplistic about this, but generally speaking, I find simple solutions to be the ones I’m most likely to actually try.

That said, we’re also assuming that the phytic acid is still present in the oat powder as the primary source of phytic acid. My understanding is that it’s largely in the outer coating of the oat, which is removed in processing. I could see it still being an issue in the flax seed, since the hull is still present, but I suspect the oat powder is not going to have enough phytic acid to be a problem. Regardless, any time you’re experimenting on yourself, blood tests are prudent.


#19

I decided to go further in trying to calculate whether PA would be an issue in a given recipe. We already know that there isn’t much phytic acid in soylent relative to the quantity of minerals. The question becomes, will the phytic acid be bound to one mineral out of proportion to its frequency of binding to other minerals, to such an extent that there is no longer enough of that mineral which PA likes especially well?

We don’t have sufficient empirical data on that question. An approach we can take in the absence of empirical data (what can I say, I like a priori solutions) is to use a prior probability distribution to consider all the possibilities for how the PA distribution could be biased, and see what proportion of them (adjusted for likelihood) cause problems. If the proportion is low, we can say that we are safe.

I wrote all of this up into a paper, ultimately specifying a calculation that you could perform to see whether your recipe was safe according to this standard. I haven’t yet written software to perform the calculation, but if there is interest in this let me know. I’d also be quite happy to hear any feedback on the paper, in the way of comments, questions, errors, doubts about my assumptions, etc.


#20

Very Interesting, I’m not saying I understand everything in there, but I would be interested in the software you mentioned in the last paragraph of your post.