Tuesday, June 11, 2019

Plant Nutrients, Part 1: Vitamin C

If you’ve been anywhere near the nutrition-oriented corner of Twitter or if you follow certain public figures like Jordan Peterson or Joe Rogan, surely you’ve heard of the carnivore diet. The diet is characterized by a strict abstinence from all plant foods, with approaches ranging from diets consisting only of muscle meats to diets including any and all animal foods as they are tolerated. Needless to say, much like an approach such as veganism, certain nutrients are typically scarce on such a restrictive diet. With vitamin C being the most obvious of the lot. However, many advocates of the carnivore diet propose that an all-meat diet lowers one’s vitamin C requirements.

Essentially, this is how the argument goes—the recommended dietary allowance (RDA) for Vitamin C was established using subjects who were eating ostensibly high carbohydrate diets. Vitamin C and glucose are so structurally similar that they share the same transporters across cell membranes. This creates a competition between glucose and vitamin C for uptake into cells from the bloodstream. Therefore, eating a high carbohydrate diet will artificially inflate our vitamin C requirements because glucose will be inhibiting vitamin C’s entry into cells. Sounds pretty persuasive. But, is it true?

First, let’s talk about what vitamin C does in the body. Very broadly speaking vitamin C has two very important jobs [1]. Vitamin C’s first job is to act as a co-factor for collagen synthesis [2]. When this system fails due to insufficient vitamin C intake, a life-threatening condition known as scurvy quickly develops. Scurvy is a condition of poor connective tissue functioning, leading to bruising, bleeding gums, loose teeth, and stunted growth. 

Vitamin C’s second job is to act as an antioxidant [3]. In this capacity vitamin C acts as a single cog in a large network of antioxidant molecules. These molecules pass electrons between each other so that free radicals can be neutralized as cleanly as possible.

First, α-tocopherol (vitamin E) becomes oxidized by free radicals and becomes α-tocopheroxyl radical (vitamin E radical). From there, vitamin E radical accepts electrons from vitamin C, oxidizing vitamin C and turning it into dehydroascorbic acid (DHAA). DHAA then accepts electrons from glutathione (GSH), turning two GSH molecules into a single glutathione disulfide (GSSG). The end of this chain involves reduced nicotinamide adenine dinucleotide phosphate (NADPH) donating its electrons to GSSG. This creates two GSH molecules and one nicotinamide adenine dinucleotide phosphate (NADP+) molecule. NADP+ is a form of co-enzyme vitamin B3 (niacin).

Now that we understand the basics of vitamin C’s roles in the body, we can now address a number of carnivore talking points related to vitamin C:

1) Glucose inhibits the cellular uptake of vitamin C.

As the story goes, vitamin C and glucose share the same transporters on cell membranes. Therefore, an excessive amount of glucose in the bloodstream creates a competition between glucose and vitamin C for cellular uptake. This artificially increases our need for vitamin C. As such, diets that aim to reduce glucose intake as low as possible (such as an all-meat diet) should lower our vitamin C requirement. Again, this sounds persuasive. But, it’s only a half-truth. Maybe even a quarter-truth.

Generally speaking, vitamin C comes in two forms (seen below). One form is ascorbic acid (AA), and the other form is dehydroascorbic acid (DHAA). AA is the reduced (or “active”) form of vitamin C, while DHAA is the oxidized (or “inactive”) form of vitamin C.

As it turns out, both types of vitamin C use different transporters to cross cell membranes. AA uses specialized transporters called sodium-dependent vitamin C transporter 1 (SVCT1) and sodium-dependent vitamin C transporter 2 (SVCT2) [4]. However, DHAA actually does rely on the same transporters as glucose. DHAA uses glucose transporter type 1 (GLUT1), glucose transporter type 3 (GLUT3), and glucose transporter type 4 (GLUT4) [5]. GLUT1 and GLUT3 transporters are virtually always present on eukaryotic cell membranes for passive glucose uptake. However, GLUT4 is an insulin-dependent glucose transporter. It is responsible for transporting the majority of glucose across cell membranes after a high carbohydrate meal [6].

Despite the fact that dietary DHAA does contribute to a cell’s total pool of AA, we probably should not be terribly concerned with DHAA as a significant source of vitamin C. We’re primarily concerned with AA’s transport into cells, because the lion’s share of AA inside of a given cell got there as active AA from the diet, not by having large amounts of dietary DHAA reduced back to AA inside the cell [7][8]. The point is, AA and glucose do not share transporters. DHAA and glucose do share transporters, but DHAA is less much, much less relevant to the argument. Very, very little of our bodies' stores of AA got there first as DHAA [9]. On the basis of this alone, I’m going to assume carnivores are not talking about DHAA when they are referring to vitamin C uptake into cells.

But, does glucose still somehow inhibit the active transport of AA into cells? Actually, yes it does! Glucose does compete with AA for access to the sodium ions needed to properly transport AA via SVCT1 and SVCT2. Despite the fact that glucose does not actually use this transporter itself for its own entry into the cell [10].

It sounds like the carnivore proponents might be right! Well, not so fast. While it is true that glucose inhibits AA’s uptake into cells, it isn’t relevant in my estimation. This is because the rate of competition largely remains constant at physiologically normal concentrations of blood glucose. There are also wide swings in variation between tests.

Here, researchers measured the uptake of AA by human enterocytes in varying concentrations of glucose. As we can see, AA uptake into the cells is superb when there is no glucose present. However, nobody is walking around with no blood glucose. For this reason, we absolutely should not care about this fact. It’s not physiologically relevant. AA uptake decreases steadily as glucose concentrations increase until about 4 mmol/L of glucose. Which is the bottom of the reference range for healthy blood glucose levels. The rate of AA uptake stays relatively constant until around 10 mmol/L of glucose, which is analagous to the blood glucose of a type II diabetic after a high carbohydrate meal. So yes, glucose inhibits AA uptake into cells, but not in any way we should care about if we are non-diabetic. Based on these data, the rate of competition would be the same regardless of whether or not we’re eating an all-meat diet, because even someone eating an all-meat diet has blood glucose that is well within the reference range.

Also, as an added painful irony, even though DHAA shares transporters with glucose, glucose does not meaningfully inhibit DHAA absorption at any concentration. DHAA uptake into cells with no glucose present is virtually identical to DHAA uptake when glucose is at its highest recorded concentration. It's a surprising yet hilarious twist of fate that the type of vitamin C that actually does share transporters with glucose seems to be completely uninhibited by glucose. So, even if the lion's share of vitamin C did enter the body first as DHAA, uptake would still be fantastic.

But just to hammer the point home, let’s stratify the in vitro glucose concentrations in the graph above by what each concentration represents physiologically.

Based on the above graph, we see that there is a ~37.5% reduction in AA uptake into cells from 4 mmol/L of glucose to 8 mmol/L of glucose. Mind you that this means that we'd have to go from the best fasting blood glucose to the worst non-diabetic postprandial blood glucose to even get an effect that large. This should not adequately justify reducing the RDA of vitamin C by 88.9-91.6%, going from 90-120 mg/day down into the neighborhood of 10mg/day. How does a 37.5% decrease in AA availability end up translating to an eightfold to elevenfold increase in AA requirements? Do we just piss AA out that fast? Not according to this study, which found that AA bioavailability and retention was better when AA was taken in a solution containing glucose as opposed to merely being dissolved in water [11].

Personally, I would like to see some AA and glucose-titration kinetics data in humans to look at the area under the curve for excretion and blood concentration over time. That should settle the question definitively, in my opinion. If anybody knows of such data, let me know. 

Above, we see that pretty much the exact same findings were replicated by another group studying human neutrophils [12]. AA uptake into the neutrophil is excellent when no glucose is present, but the rate of AA inhibition by glucose remains relatively constant at physiologically analogous concentrations of glucose. Again, if competition is such that the presence of glucose creates an artificially higher need for vitamin C, as some carnivores claim, it applies to carnivores too. AA uptake into cells is virtually the same regardless of whether our blood glucose is 4 mmol/L as fasting, fat-burning carnivores or 8 mmol/L as well-fed, carb-burning couch potatoes.

But for the sake of argument, let’s actually just pretend that AA can utilize the glucose transporters as well. It can't, but let’s pretend it can. In normal people, blood glucose is always accompanied by the hormone insulin. So, even if it were true that AA and glucose shared transporters, the rate of competition would still be roughly the same regardless of the glucose concentration. This is because the representation of GLUT4 on cell membranes is modulated by insulin. More glucose, more insulin, more GLUT4, and thus more AA transport into cells. No matter which way we cut it, the argument that high carbohydrate diets increases vitamin C requirements doesn’t make any sense.

Not to mention that nearly all of the plant foods highest in AA are typically low-carbohydrate plant foods, and typically contain low-to-negligible DHAA [8]. Virtually none of the food selections listed as the richest sources of AA would even m
ove the needle on blood glucose. So, even if it were true that glucose artificially inflates our vitamin C requirements, a food like broccoli should still be a viable option for vitamin C. But hey, who cares about these sorts of details, right?

2) All-meat diets do not cause scurvy.

At this point in the discussion, someone proposing the first argument is usually keen to eventually propose this second argument. The idea is that a diet consisting of only meat does not reliably produce scurvy, if at all. Therefore, it must be the case that an all-meat diet does indeed lower our vitamin C requirement, at the very least through some unknown mechanism. This is a common rebuttal. But in my opinion it is very weak, as it demonstrates a failure to understand precisely what vitamin C does in the body and how it does what it does.

Remember when I stated that vitamin C has two primary functions in the body? One function is preventing scurvy through promoting healthy collagen synthesis. The other function is supporting the antioxidant defense system. Well, those two different functions require two different amounts of vitamin C. It has been documented in depletion-repletion experiments that the amount of vitamin C needed to support collagen synthesis (and thus prevent scurvy) is around 10mg of vitamin C per day [13]. It was also discovered that for some individuals, body vitamin C stores required to prevent scurvy could be maintained with less than 6.5mg of vitamin C per day. 

Subjects H and P were given 6.5mg of vitamin C per day for the entirety of phase IV (in red) of the experiment. Vitamin C status gradually improved with an indication of a plateau that was broken by complete repletion in phase V.

The RDA for vitamin C is mostly set to accommodate its turnover in relation to its use as an antioxidant [14]. The RDA for vitamin C is not specifically set to accommodate its turnover as it relates to its use as a co-factor for collagen synthesis, because the body requires so little for that particular function (300mg of a total body pool of 2000mg).

When it is proposed that an all-meat diet must reduce absolute vitamin C requirements, I ask: "which requirement?" Remember, vitamin C demands two different requirements for two different functions. One requirement is huge, but there are no obvious symptoms related to deficiency (because oxidative stress is a chronic long term issue). One requirement is very tiny, and the deficiency symptom is, well, scurvy.

This is actually an excellent example of Bruce Ames’ triage theory [15]. Which in my opinion is closer to being a principle rather than a theory or hypothesis at this point. The theory proposes that the body will direct the use of nutrients toward functions that maximize short term survival and away from functions that maximize long term health under conditions of scarcity. For example, say we’re running low on folate. Folate has a few functions, one of which is participating in the production of certain neurotransmitters, creatine, red blood cells, and all sorts of acutely important things. Another function of folate is preventing DNA damage and regulating gene expression. Under conditions of scarcity, the body will actually direct folate to functions that maximize short term survival, and sacrifice functions that maximize long term health [16]

We can observe that sub-optimal folate status causes DNA damage, but acutely important functions are presumably defended first. If they weren't, the subject would be either dead or gravely ill. The same is probably true of vitamin C. Under conditions of scarcity it is likely that collagen synthesis is defended before antioxidant support. We can't feel or see oxidative stress the way we can feel or see bleeding gums or bruising. We also can't feel or see DNA damage the way we can feel or see exercise intolerance or anemia.

Unless the individuals making the argument for reduced vitamin C requirements on all-meat diets are actually having their antioxidant capacity measured, they can’t know for certain. It is entirely possible to get ~6.5mg/day of vitamin C on an all-meat diet. There is absolutely no question about that. Even a diet of pure muscle meats could potentially supply enough vitamin C to prevent scurvy.

But there is another side to this as well. There is an added detail to consider from the depletion-repletion studies. For some of the subjects, it took over 100 days on a zero vitamin C diet to maximally deplete their body’s total pool of vitamin C [13]

People eating a carnivore diet were, at least ostensibly, eating a vitamin C sufficient diet long before they decided to abstain from eating plant foods. Let's say for the sake of argument that an all-meat diet doesn't provide much vitamin C at all. If we were to provide the body with merely less than 6.5mg of vitamin C per day instead of just zero (starting with total body stores that are potentially maximally replete), how long would it take to get scurvy? Who knows. My guess is that it would take longer than 100 days. Maybe it would take years? A decade? Maybe it wouldn't happen at all. It’s uncertain. But, I wouldn’t personally take the absence of scurvy in the short term as an indication that vitamin C needs are actually being met.

3) Vitamin C’s role as an antioxidant is merely theoretical.

The idea sounds cute, and I will admit the justification for the RDA of vitamin C being as high as it is might appear a bit wishy-washy in the official DRI report [14]. However, vitamin C’s usage and turnover in the antioxidant defense system has been well documented, and it is well understood [17][18][19][20][21][22]. There are perfectly legitimate reasons to conclude that the RDA of vitamin C is a reasonable recommendation. Personally, I feel there is ample justification to increase the RDA of vitamin C beyond 200mg/day. As there is no obvious downside, and it is virtually guaranteed to capture all documented variation in vitamin C turnover in humans. Not to mention it has been associated with lower morbidity and mortality in epidemiological studies [23].

Here we see that the total body turnover of vitamin C in 14 non-smoking subjects ranges from 14.4mg/day all the way to 142mg/day [24]. This means that there can be a whole order of magnitude difference in vitamin C turnover from person to person. Given that the turnover of vitamin C in the support collagen synthesis leaves us requiring somewhere south of 10mg/day as a baseline, the above turnover rates give us a picture of just how much vitamin C is required each day to fulfill the rest of its functions outside of collagen synthesis. The average of these turnover rates is around 65.74mg/day. If we minus 10mg/day for collagen synthesis, this leaves us with an average of ~56mg/day worth of vitamin C turnover that probably can’t be adequately explained by collagen synthesis alone. The body is presumably using that vitamin C. It's probably a fair bet that vitamin C is being used by the body in amounts that sometimes exceed what would be satisfied by the RDA alone.

There is an additional complication to measuring turnover accurately, however. One of the primary ways to measure vitamin C turnover since the early 1900s is to measure 2,3-diketogulonic acid (
2,3-DKG) in urine or plasma. This is a metabolite of vitamin C that occurs when vitamin C's carbon ring spontaneously ruptures. Typically, this rupture occurs when vitamin C has exceeded the number of redox cycles it can support. This means we can actually directly measure just how much vitamin C the body is using.

Here we see that 2,3-DKG (DKA in this paper) is maintained at approximately 10-15% of total vitamin C excretion in human subjects on low-plant diets [25]. Free AA excretion only increased with 75mg of vitamin C per day. Interestingly, in one subject 2,3-DKG excretion increased with supplementation while the other subject's 2,3-DKG excretion decreased. 

In an odd contrast, another paper measured plasma 2,3-DKG and found that up to half of human plasma vitamin C could consist of 2,3-DKG [26]. Sometimes up to 15mg or 25mg of plasma vitamin C was 2,3-DKG. This could indicate that urinary excretion is not the most accurate way to measure turnover. But without data to measure the flux of 2,3-DKG generated and excreted, this number doesn't mean much. But, it could mean that vitamin C turns over at a rate that far exceeds what you could reliably get from meat.

I would be super interested to see complex flux or kinetics data on 2,3-DKG, considering that the body tends to retain more 2,3-DKG than it excretes. As we can see below, urinary excretion of 2,3-DKG pretty much never exceeds intake from the diet. This is puzzling to me and I want to understand more about this. I'm guessing that some proportion of the 2,3-DKG in the blood gets converted to another metabolite like oxalate and is excreted that way.

Nevertheless, all of the subjects in the study were excreting far more AA than 2,3-DKG. One might respond to this by saying: "a-ha! See, those fuckers are pissing way more ascorbic acid than they're using!" Right. But remember those depletion-repletion graphs above? We urinate a relatively steady amount of free AA all the way to scurvy, regardless of how depleted or repleted we are. Even after months with an AA intake of zero. So, don't get excited.

Additionally, we have direct evidence of vitamin C supplementation upwards of 1000mg/day reducing levels of oxidative stress relative to placebo [20]. We have other research indicating that 500mg/day can lower markers of oxidative stress in humans when compared to 50mg/day over five years. In fact, the group receiving a mere 50mg/day had net increases in average oxidative stress [27]. Implying that their needs were probably higher than a mere 50mg/day.

Think about it. The body has a maximum storage capacity of 2000mg of vitamin C. Scurvy symptoms only present when total body pool drop below 300mg. We know from the above experiments that certain people who are asymptomatic for scurvy (meaning they have a total body pool of >300mg of vitamin C), can reduce markers oxidative stress with vitamin supplementation ranging from 500-1000mg per day. This implies to me that you need a total body pool of 300mg of vitamin C to prevent scurvy, and that the bulk of your storage capacity (~1700mg) is likely to support other functions, like the antioxidant defense system. 
The exercise data referenced above paints a very important picture that is critical for understanding human vitamin C requirements—you can exercise your way to a higher vitamin C requirement, but you cannot exercise your way to scurvy. 

Some could argue that studies on vitamin C supplementation aren’t always consistent. Some studies show no significant effect of vitamin C supplementation on oxidative stress [28]. I would agree. However, remember that vitamin C doesn’t act alone. Vitamin C requires good vitamin E status, riboflavin status, niacin status, and adequate protein intake in order to properly fulfill its role as an antioxidant. To illustrate this, whether in humans or in animals, effects of vitamin C supplementation are more pronounced when it is complimented with vitamin E supplementation [29][30][31][32].

It is possible that the few vitamin C supplementation studies that have yielded null or near-null results may have been confounded by poorer nutritional status in the subjects. It’s hard to say for sure. Even when we randomize subjects, we're still randomizing subjects belonging to a single population. So, it's possible for population-specific variables to confound a study. I realize this is speculative on my part, but it's not at all impossible. 

Nevertheless, what we do know is that we need enough vitamin E to accept electrons from vitamin C once vitamin E is oxidized. We also need to take in enough protein to make the GSH to pass electrons to vitamin C once it is oxidized by vitamin E radical. We then need adequate riboflavin to oxidize NADPH to NADP+. But before that, we need adequate niacin status to provide the nicotinamide adenine dinucleotide (NAD) to make the NADPH. What I’m saying is that a lot can go wrong here, and merely dosing vitamin C is not a guarantee that we will see a result at all. Null results in vitamin C dosing trials shouldn’t automatically be taken as an indication that vitamin C doesn’t matter.

4) Glutathione can replace vitamin C as an antioxidant.

This is probably the most outlandish claim of the lot, but I’ve heard it many times from many different individuals. I have no idea where this claim comes from or who first proposed it. Naturally, my knee-jerk response is to simply remind those making this claim that it just doesn’t work that way. For reasons I’ve already described vitamin C is an integral part of the antioxidant defense system. We can’t just hotswap reducing agents at our leisure. They all depend on each other because they all require their own unique enzymes to facilitate each reduction-oxidation (redox) reaction [33]. For example, we simply cannot pass electrons from two GSH molecules to DHAA without dehydroascorbate reductase (DHAR). Without this enzyme, there is a break in the chain of redox reactions that carry electrons all the way from NADPH to vitamin E radical.

In fact, at least one group of researchers have attempted to answer this exact question [34]. What happens when the antioxidant defense system within a cell is deprived of vitamin C, and that vitamin C is replaced by GSH? Spoiler alert! The cell straight up dies. Without vitamin C present to pass and accept electrons between the other two reducing agents, GSH and vitamin E, the cell cannot withstand oxidative stress at all. GSH cannot act as a stand-in for vitamin C. Period. When vitamin C status is lower than would be required for optimal antioxidant capacity within a given cell, the cell suffers. If we remove vitamin C entirely, the cell dies. Simple.

5) Dietary collagen can spare vitamin C.

In the body, vitamin C is responsible for the hydroxylation reactions that convert proline and lysine into hydroxyproline and hydroxylysine. It has been proposed that perhaps consuming dietary collagen could spare vitamin C by providing these two pre-hydroxylated amino acids. However, researchers have suggested that they are not very likely to survive digestion [35]. At least two studies have discovered that hydroxyproline peptides are detectable in human blood after ingestion [36][37]. But, these are not the free hydroxylated amino acids necessary for the production of new collagen. However, ingestion of collagen peptides does seem to have an effect on pre-existing collagen structures in humans [38][39]

It is unlikely that the ingestion of collagen peptides turns down the body's production of its own collagen. In fact, one group discovered that ingesting collagen peptides was most effective at augmenting the body's collagen when vitamin C was provided, and that the supplement they provided actually stimulated collagen synthesis [40]

In an interview, researcher Keith Baar remarked that one unpublished experiment failed to yield results because the vitamin C content of the collagen supplement was lost due to overcooking. This is likely because vitamin C is also responsible for activating the enzymes that cross-link collagen [41]. Meaning that upon ingestion of collagen peptides, the effects are most robust when vitamin C status is optimal. It is possible that ingesting collagen for the purpose of augmenting the body's collagen actually taxes our vitamin C rather than sparing it.

But for the sake of argument, let's be generous and pretend that ingesting collagen peptides spares all of the vitamin C necessary for collagen synthesis. At most, we'd be expected to spare less than 10mg/day. This does not leave enough to cover even a fraction of recorded vitamin C turnover in some humans. Again, vitamin C has two primary functions. The requirement for one function is very large, and the other is very small. Collagen synthesis requires very little vitamin C. Antioxidant support requires an enormous amount of vitamin C by comparison. 

6) The guinea pigs.

There is a lot of love in the carnivore community for a vitamin C depletion study performed on guinea pigs [42]. Like us, guinea pigs do not synthesize their own vitamin C, so they are a reasonable choice for designing rodent models of vitamin C deficiency. Essentially, researchers deprived the guinea pigs of vitamin C and administered a form of bioavailable GSH (remember GSH is also an important part of the antioxidant defense system). Doing this, the researchers were able to slow the progression of scurvy. Some in the carnivore community have claimed that this is a vitamin C-sparing effect of GSH that could explain the absence of scurvy on all-meat diets. I can see the logic—diets higher in protein would probably best support GSH status.

However, I look at this a little bit differently. Again, we know that vitamin E is recycled by vitamin C, vitamin C is recycled by GSH, and GSH is recycled by NADPH. It's a system, and the system has many cogs that all need to function together. If you dose GSH and delay the onset of scurvy, this isn't an effect of GSH sparing vitamin C so much as it is just an example of what happens when everything works properly. If there is not enough GSH to reduce DHAA back to AA, a steady supply of AA is required to make up for it. Not only that but you will have more vitamin C wasting due to DHAA spontaneously converting to 2,3-DKG. 

In this context, insufficient GSH creates a higher-than-normal need for AA. When you delay the onset of scurvy by administering GSH to the AA-deprived guinea pigs, what you're seeing is the guinea pigs just needing more GSH. You are not seeing the guinea pigs requiring less vitamin C per se. The guinea pigs are made artificially deficient in vitamin C as a consequence of sub-optimal GSH status. You're just seeing what happens when you correct a GSH deficiency.

  • Ascorbic acid does not share transporters with glucose. Dehydroascorbic acid does share transporters with glucose, but it doesn’t matter.
  • Glucose does inhibit cellular uptake of ascorbic acid, but it is not meaningful at physiologically relevant concentrations of glucose.
  • Less than 10mg of vitamin C per day is required to prevent scurvy.
  • Vitamin C turnover in healthy people far exceeds 10mg per day, presumably as a consequence of its use as an antioxidant.
  • Vitamin C requires a number of other nutrients to do its job correctly, and could explain why vitamin C supplementation has inconsistent effects on markers of oxidative stress.
  • The antioxidant effects of vitamin C are more robust when coupled with either glutathione or vitamin E.
  • Vitamin C is an integral part of the body’s antioxidant defense system, and complete vitamin C deprivation triggers cell death.
  • Consuming collagen is unlikely to spare vitamin C and could perhaps even increase the body's demand for vitamin C.
  • Vitamin E, vitamin C, and glutathione all cooperate together to support the antioxidant defense system. Reducing any of them increases the need for the other two.
  • There is no persuasive reason to believe that an all-meat diet modulates our vitamin C requirements such that the RDA should be reduced from 90-120 mg/day to any amount one could reasonably expect to obtain from eating only animal foods.
PS. If you like what you've read and want me to continue writing, consider supporting me on Patreon. Every little bit helps! Thank you for reading!


1. Nicholas N. DePhillipo, et al. Efficacy of Vitamin C Supplementation on Collagen Synthesis and Oxidative Stress After Musculoskeletal Injuries: A Systematic Review. Orthop J Sports Med. October 2018. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6204628/

2. Juliet M. Pullar, Anitra C. Carr, and Margreet C. M. Vissers. The Roles of Vitamin C in Skin Health. Nutrients. August 2017. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5579659/

3. Tamari Y, et al. Protective roles of ascorbic acid in oxidative stress induced by depletion of superoxide dismutase in vertebrate cells. Free Radic Res. January 2013. https://www.ncbi.nlm.nih.gov/pubmed/23016763

4. Savini I, Rossi A, Pierro C, Avigliano L, and Catani MV. SVCT1 and SVCT2: key proteins for vitamin C uptake. Amino Acids. April 2008. https://www.ncbi.nlm.nih.gov/pubmed/17541511

5. Rumsey SC, Kwon O, Xu GW, Burant CF, Simpson I, and Levine M. Glucose transporter isoforms GLUT1 and GLUT3 transport dehydroascorbic acid. J Biol Chem. July 1997. https://www.ncbi.nlm.nih.gov/pubmed/9228080

6. Watson RT and Pessin JE. Intracellular organization of insulin signaling and GLUT4 translocation. Recent Prog Horm Res. February 2001. https://www.ncbi.nlm.nih.gov/pubmed/11237212

7. Ron BH Wills, Pushparani Wimalasiri, and Heather Greenfield. Dehydroascorbic acid levels in fresh fruit and vegetables in relation to total vitamin C activity. J Agric Food Chem. July 1984. https://pubs.acs.org/doi/abs/10.1021/jf00124a035

8. Joseph T Vanderslice, Darla J Higgs, Jeanne M Hayes, and Gladys Block. Ascorbic acid and dehydroascorbic acid content of foods-as-eaten. J Food Compos Anal. June 1990. https://www.sciencedirect.com/science/article/pii/088915759090018H

9. Corpe CP, Lee JH, Kwon O, Eck P, Narayanan J, Kirk KL, and Levine M. 6-Bromo-6-deoxy-L-ascorbic acid: an ascorbate analog specific for Na+-dependent vitamin C transporter but not glucose transporter pathways. J Biol Chem. February 2005. https://www.ncbi.nlm.nih.gov/pubmed/15590689

10. Malo C, et al. Glucose modulates vitamin C transport in adult human small intestinal brush border membrane vesicles. J Nutr. January 2000. https://www.ncbi.nlm.nih.gov/m/pubmed/10613768/

11. Vinson JA and Bose P. Comparative bioavailability to humans of ascorbic acid alone or in a citrus extract. Am J Clin Nutr. September 1988. https://www.ncbi.nlm.nih.gov/pubmed/3414575

12. Washko P and Levine M. Inhibition of ascorbic acid transport in human neutrophils by glucose. J Biol Chem. November 1992. https://www.ncbi.nlm.nih.gov/pubmed/1429700

13. Baker EM, Hodges RE, Hood J, Sauberlich HE, March SC, and Canham JE. Metabolism of 14C- and 3H-labeled L-ascorbic acid in human scurvy. Am J Clin Nutr. April 1971. https://www.ncbi.nlm.nih.gov/pubmed/5090632

14. Institute of Medicine (US) Panel on Dietary Antioxdants and Related Compounds. Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. 2000. https://www.ncbi.nlm.nih.gov/books/NBK225480/

15. Ames BN. Low micronutrient intake may accelerate the degenerative diseases of aging through allocation of scarce micronutrients by triage. Proc Natl Acad Sci U S A. November 2006. https://www.ncbi.nlm.nih.gov/pubmed/17101959

16. Blount BC, et al. Folate deficiency causes uracil misincorporation into human DNA and chromosome breakage: implications for cancer and neuronal damage. Proc Natl Acad Sci U S A. April 1997. https://www.ncbi.nlm.nih.gov/pubmed/9096386

17. Stankova L, Gerhardt NB, Nagel L, and Bigley RH. Ascorbate and phagocyte function. Infect Immun. August 1975 https://www.ncbi.nlm.nih.gov/pubmed/1150324

18. Parker A, Cuddihy SL, Son TG, Vissers MC, and Winterbourn CC. Roles of superoxide and myeloperoxidase in ascorbate oxidation in stimulated neutrophils and H2O2-treated HL60 cells. Free Radic Biol Med. October 2011. https://www.ncbi.nlm.nih.gov/pubmed/21791243

19. Oberritter H, Glatthaar B, Moser U, and Schmidt KH. Effect of functional stimulation on ascorbate content in phagocytes under physiological and pathological conditions. Int Arch Allergy Appl Immunol. February 1986. https://www.ncbi.nlm.nih.gov/pubmed/3744577

20. Yimcharoen M, Kittikunnathum S, Suknikorn C, Nak-On W, Yeethong P, Anthony TG, and Bunpo P. Effects of ascorbic acid supplementation on oxidative stress markers in healthy women following a single bout of exercise. J Int Soc Sports Nutr. January 2019. https://www.ncbi.nlm.nih.gov/pubmed/30665439

21. Heather C Kuiper, Richard S Bruno, Maret G Traber, and Jan F Stevens. Vitamin C Supplementation Lowers Urinary Levels of 4-Hydroperoxy-2-nonenal Metabolites in Humans. Free Rad Biol Med. April 2011. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3046321/

22. Anitra C Carr and Silvia Maggini. Vitamin C and Immune Function. Nutrients. November 2017. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5707683/

23. Enstrom JE, Kanim LE, and Klein MA. Vitamin C intake and mortality among a sample of the United States population. Epidemiology. May 1992. https://www.ncbi.nlm.nih.gov/pubmed/1591317

24. Kallner AB, Hartmann D, and Hornig DH. On the requirements of ascorbic acid in man: steady-state turnover and body pool in smokers. Am J Clin Nutr. July 1981. https://www.ncbi.nlm.nih.gov/pubmed/7258125

25. Cox BD and Whichelow MJ. The measurement of dehydroascorbic acid and diketogulonic acid in normal and diabetic plasma. Biochem Med. February 1975. https://www.ncbi.nlm.nih.gov/pubmed/1137579

26. SD Chen and C Schuck. Excretion of the two Biologically Active Forms of Ascorbic Acid and of Diketogulonic Acid by Human Subjects. J Nutr. October 1951. https://www.ncbi.nlm.nih.gov/pubmed/14889318

27. Sasazuki S, et al. Protective effect of vitamin C on oxidative stress: a randomized controlled trial. Int J Vitam Nutr Res. May 2008. https://www.ncbi.nlm.nih.gov/pubmed/19003734

28. Bunpo P and Anthony TG. Ascorbic acid supplementation does not alter oxidative stress markers in healthy volunteers engaged in a supervised exercise program. Appl Physiol Nutr Metab. February 2016. https://www.ncbi.nlm.nih.gov/pubmed/26789096

29. Zal F, Mostafavi-Pour Z, Amini F, and Heidari A. Effect of vitamin E and C supplements on lipid peroxidation and GSH-dependent antioxidant enzyme status in the blood of women consuming oral contraceptives. Contraception. July 2012. https://www.ncbi.nlm.nih.gov/pubmed/22494786

30. Maret G Trabera and Jan F Stevens. Vitamins C and E: Beneficial effects from a mechanistic perspective. Free Radic Biol Med. Semptember 2012. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3156342/

31. Maryam Taghiyar, et al. The Effect of Vitamin C and E Supplementation on Muscle Damage and Oxidative Stress in Female Athletes: A Clinical Trial. Int J Prev Med. April 2013. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3665020/

32. Ashor AW, Siervo M, Lara J, Oggioni C, Afshar S, and Mathers JC. Effect of vitamin C and vitamin E supplementation on endothelial function: a systematic review and meta-analysis of randomised controlled trials. Br J Nutr. April 2015. https://www.ncbi.nlm.nih.gov/pubmed/25919436

33. Meister A. Glutathione, Ascorbate, and Cellular Protection. Cancer Res. April 1994. https://www.ncbi.nlm.nih.gov/pubmed/8137322

34. Viviana Montecinos, et al. Vitamin C Is an Essential Antioxidant That Enhances Survival of Oxidatively Stressed Human Vascular Endothelial Cells in the Presence of a Vast Molar Excess of Glutathione. J Biol Chem. May 2007. https://www.ncbi.nlm.nih.gov/pubmed/17403685

35. Guoyao Wu, et al. Proline and hydroxyproline metabolism: implications for animal and human nutrition. Amino Acids. April 2011. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3773366/

36. Shigemura Y, Iwasaki Y, Tateno M, Suzuki A, Kurokawa M, Sato Y, and Sato K. A Pilot Study for the Detection of Cyclic Prolyl-Hydroxyproline (Pro-Hyp) in Human Blood after Ingestion of Collagen Hydrolysate. Nutrients. September 2018. https://www.ncbi.nlm.nih.gov/pubmed/30248982

37. Iwai K, et al. Identification of food-derived collagen peptides in human blood after oral ingestion of gelatin hydrolysates. J Agric Food Chem. August 2005. https://www.ncbi.nlm.nih.gov/pubmed/16076145

38. Kumar S, Sugihara F, Suzuki K, Inoue N, and Venkateswarathirukumara S. A double-blind, placebo-controlled, randomised, clinical study on the effectiveness of collagen peptide on osteoarthritis. J Sci Food Agric. March 2015. https://www.ncbi.nlm.nih.gov/pubmed/24852756

39. Do-Un Kim, Hee-Chul Chung, Jia Choi, Yasuo Sakai, and Boo-Yong Lee. Oral Intake of Low-Molecular-Weight Collagen Peptide Improves Hydration, Elasticity, and Wrinkling in Human Skin: A Randomized, Double-Blind, Placebo-Controlled Study. Nutrients. July 2018. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6073484/

40. Gregory Shaw, Ann Lee-Barthel, Megan LR Ross, Bing Wang, and Keith Baar. Vitamin C–enriched gelatin supplementation before intermittent activity augments collagen synthesis. Am J Clin Nutr. November 2016. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5183725/

41. Levene CI, Shoshan S, and Bates CJ. The effect of ascorbic acid on the cross-linking of collagen during its synthesis by cultured 3 T6 fibroblasts. Biochem Biophys Acta. February 1972. https://www.ncbi.nlm.nih.gov/pubmed/5063249

42. J Mårtensson, J Han, O W Griffith, and A Meister. Glutathione ester delays the onset of scurvy in ascorbate-deficient guinea pigs. Proc Natl Acad Sci U S A. January 1993. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC45651/

Thursday, June 6, 2019

Personalizing Vitamin E Requirements

Anyone who has ever attempted to track their micronutrient intake has likely noticed just how difficult it can sometimes seem to achieve the recommended dietary allowance (RDA) of α-tocopherol (vitamin E). Especially if we're eating certain restrictive diets such as the paleo diet or even just a calorie restricted diet. On some days it may certainly seem impossible—it's baffling! We find ourselves doing everything right, eating a diet of entirely whole foods, and hitting every micronutrient target in excess with the sole exception of vitamin E. Maybe you decide to take a supplement to make up for it. Maybe you chow down on an extra 200 calories worth of almonds. Maybe you decide to cook your eggs in red palm oil instead of coconut oil. Maybe you juice a whole kilogram of carrots. Don't panic, though. Chances are good that your vitamin E intake is actually just fine!

Did you know that everyone’s vitamin E requirement is likely different, and that vitamin E requirements probably vary drastically from day to day? It’s true. This is due to how vitamin E fulfills its role in the body. Vitamin E’s only clearly established role in the body is to protect polyunsaturated fatty acids (PUFAs) in cell membranes from an oxidative process called lipid peroxidation [1]. Lipid peroxidation is a form of oxidative damage that occurs as a consequence of normal metabolism, but is accelerated in certain disease states like type II diabetes. 
The reason they are uniquely vulnerable to lipid peroxidation is because they possess more than one double bond in their carbon chain.

In vivo human trials consistently show reductions in lipid peroxidation with vitamin E supplementation [2]. Monounsaturated fatty acids (MUFAs) are not particularly vulnerable to this process, though vitamin E intake has been shown to protect MUFAs from lipid peroxidation in animal tissues as well [3]. Saturated fatty acids (SFAs) on the other hand are completely resistant to peroxidation at body temperature. This means that vitamin E offers no special protection to SFAs in cell membranes. 

The official RDA for vitamin E is set at 15mg per day for adults, 4-11mg per day for children, and up to 19mg per day for lactating women [4]. In the dietary reference intake (DRI) report for vitamin E, we discover that the RDA for vitamin E was set to accommodate population-wide averages in PUFA intake, with the primary PUFA being linoleic acid, an omega-6 fatty acid found in plants and abundantly in nuts, seeds, and most vegetable oils. Based on a ratio of 0.4mg of vitamin E per gram PUFA (as linoleic acid) per day, they merely calculated the RDA of vitamin E based on how much vitamin E would be required to cover the average person's PUFA intake. We can surmise from this that half of the population likely needs less than the RDA of vitamin E, while the other half very likely needs more than the RDA of vitamin E. But, this actually tells us nothing about our own personal vitamin E requirements as individuals. It only tells us that we have a 50/50 chance of needing more or less than the RDA.

Because the body does not synthesize PUFAs de novo (from scratch), all of the PUFAs in our cell membranes got there because we ate them. This suggests that we have at least some control over how vulnerable our cell membranes are to oxidative damage. If vitamin E's one and only job is protecting PUFAs from damage, what might happen to our vitamin E requirements when we go out of our way to limit dietary PUFAs?

If we indeed can modulate our needs for vitamin E by adjusting our PUFA intake, how might we go about calculating our own personal vitamin E requirements? Luckily, researchers have actually answered this question in amazingly fine detail [5]. As it turns out, each individual dietary PUFA (of which there are about five relevant types) alter vitamin E requirements differently due to each PUFA’s relative degree of unsaturation. In fact, the reason such imprecise methods were used to establish the RDA of vitamin E is that it is just too damn complicated to capture all of the nuances in a single general recommendation. This is partly because our vitamin E requirements not only change depending on how much PUFA we consume, but what kinds of PUFAs as well.

This means that if we had precise knowledge of our own dietary intake of these dietary fatty acids, we could approximate a vitamin E requirement that was personalized and custom tailored to our own dietary habits. Pretty cool, eh? I thought so too.

First we have to use a nutrition tracking tool like Cronometer to log a typical day of eating. Don’t do this while on vacation or over the holidays if your eating behaviour typically wavers during those times. Once you have an accurate accounting of a typical day’s worth of food it should look something like this: 

Below, we can see that this sort of dietary pattern yields ample amounts of all of the vitamins with the exception of vitamin E. Is this really a problem? Let's find out!

At this point, the next task is to separate these foods into animal foods and plant foods. This is because the PUFA that are found in animals are different than the PUFA found in plants. For example, the PUFA found in plant foods like walnuts are alpha-linolenic acid (ALA) and linoleic acid (LA), whereas the PUFA found in animal foods like salmon are eicosapentaenoic acid (EPA), docosahexaenoic acid (DHA), and arachadonic acid (AA). Each of these fatty acids have different degrees of unsaturation and thus alter our vitamin E requirements differently.

In Cronometer, we can shift-click to highlight certain foods together. Doing this we can highlight animal foods and plant foods separately:

Once we have our foods stratified by animal foods and plant foods, we want to scroll down to see the omega-3 and omega-6 content of both groups. 

Record the numbers for both groups separately. Once we have those numbers recorded, the last thing we need to jot down is the MUFA content of all of our foods combined, as shown above.

The next step is to use this handy dandy vitamin E calculator designed by my brother, and input the fatty acid composition of our food selection. Input your plant omega-3 (alpha-linolenic acid), plant omega-6 (linoleic acid), animal omega-3s (EPA and DHA), animal omega-6 (arachadonic acid), and then the total oleic acid (MUFA) for all foods together. For the animal omega-3s, I tend to just divide the total animal omega-3 between the EPA and DHA boxes, as there isn't really a practical way to calculate this precisely. 

Once all your data has been inputted into the calculator, your personalized vitamin E requirement for that day will be displayed in the total vitamin E box!

As you can see, our vitamin E requirement for this particular day (7.5mg) was well below the RDA (15mg), and well below what we consumed (11mg). This means that despite not actually achieving the RDA of vitamin E on this day, we actually overshot our vitamin E requirement for this particular day by 47 percent!

If you want more granular detail, repeat this process for two weeks worth of typical eating. Take every calculated vitamin E requirement for each day, and use them to calculate an average. The more days you add, the more refined the result! Now that you know your own vitamin E requirements, you’ll probably be surprised by how little you need if you structure your diet appropriately. Personally, I’ve calculated my own requirement as being around 6mg of vitamin E per day. Well below the RDA! But I also regularly consume well above the RDA for vitamin E.

Alternatively, we can do a quick and dirty calculation based on a simple ratio proposed in the DRI report for vitamin E. This calculation is much, much less precise. As such, I'm skeptical as to how accurately it will represent our vitamin E requirements. But if we wanted to give it a shot, we would start by taking our average daily PUFA intake in grams and multiplying that number by 0.4mg of vitamin E.

For example, if our average daily PUFA intake over two weeks is 15g per day, our vitamin E requirement is 6mg per day based on this ratio. However, I share the skepticism expressed by the authors of the DRI report. I don't personally believe that this ratio is likely to capture all of our vitamin E requirements, as it fails to factor in a number of variables. It doesn't account for the modest effect of dietary MUFA, and is also based solely on linoleic acid as the primary dietary PUFA. We want to determine our requirements based on the entire breadth of dietary unsaturated fatty acids, not just one variety. 

In my opinion, if we wanted to be extra safe, we could likely capture all of our vitamin E requirements in excess by maintaining a 1g to 0.84mg ratio of PUFA to vitamin E. I arrive at this by averaging the vitamin E required per gram of each of the six relevant dietary fatty acids listed above. The average ends up being 0.84mg of vitamin E per gram of PUFA. This has a greater chance of capturing requirements for both PUFA and MUFA, in addition to considering more than just linoleic acid as the primary dietary PUFA. It is also unlikely to generate a vitamin E requirement as drastic as the official RDA. But at the end of the day, using the handy dandy vitamin E calculator that I have provided above is probably the best bet.

PS. If you like what you've read and want me to continue writing, consider supporting me on Patreon. Every little bit helps! Thank you for reading!


1. Valk EE, Hornstra G. Relationship between vitamin E requirement and polyunsaturated fatty acid intake in man: a review. Int J Vitam Nutr Res. March 2000. https://www.ncbi.nlm.nih.gov/pubmed/10804454

2. Kun Liu, Suyun Ge, Hailing Luo, Dubing Yue, and Leyan Yan. Effects of dietary vitamin E on muscle vitamin E and fatty acid content in Aohan fine-wool sheep. J Anim Sci Biotechnol. June 2013. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3700886/

3. Huang HY, Appel LJ, Croft KD, Miller ER 3rd, Mori TA, Puddey IB. Effects of vitamin C and vitamin E on in vivo lipid peroxidation: results of a randomized controlled trial. Am J Clin Nutr. September 2002. https://www.ncbi.nlm.nih.gov/pubmed/12197998

4. Institute of Medicine (US) Panel on Dietary Antioxidants and Related Compounds. Dietary Reference Intakes for Vitamin C, Vitamin E, Selenium, and Carotenoids. 2000. https://www.ncbi.nlm.nih.gov/books/NBK225461/

5. Daniel Raederstorff, Adrian Wyss, Philip C. Calder, Peter Weber, and Manfred Eggersdorfer. Vitamin E function and requirements in relation to PUFA. Br J Nutr. October 2015. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4594047/