Vitamin C on the keto diet (everything you need to know)
Guest post Written by L. Amber O’Hearn
Table of Contents
The Recommended Daily Allowances for different nutrients were developed on Western diets, and therefore, high-carb diets. Given that a ketogenic metabolism uses different metabolic pathways and induces cascades of drastically different metabolic and physiological effects, it would be astonishing if any of the RDAs are entirely applicable as is.
One micronutrient that seems to be particularly warranting reassessment is vitamin C, because vitamin C is biochemically closely related to glucose.
Most animals synthesize it themselves out of glucose. It shares cellular uptake receptors with glucose. Some argue that because we don’t make vitamin C, we need to ensure a large exogenous supply.
I will argue the opposite: so long as we are eating a low-carb diet, we actually need less. On our way, we’ll briefly re-examine the relationship between vitamin C deficiency and insulin resistance.
There are particular nutrients people need to develop normally and stay healthy, that we can’t make in our own bodies, and so we have to get them from our diets. We only started recognizing this at the end of the 19th century.
Before that, the germ theory of disease was new and exciting, and we wanted to explain all maladies as infections. However, we ultimately learned that some diseases come from malnutrition.
The best exemplifiers of this are when people or other animals die for lack of one specific ingredient, as in pellagra, beriberi, rickets, and scurvy.
These ingredients were named, initially, “vitamines”, meaning vital amines, but when it turned out they weren’t all amines, the name was shortened to “vitamins” .
Most dietary therapies are based on this notion, that significant health improvements can be made by adding enough of missing nutrients.
This is effective when the baseline diet was grossly deficient, but when nutrient issues are not acute, traditional dietary therapies are little better than nothing in the face of diseases of civilization.
This is the crux of what makes a ketogenic diet uniquely powerful. It is not just about changing nutrient intakes. Indeed, even ketogenic diets can be poorly constructed nutritionally.
What makes a ketogenic diet powerful is that it induces a complete change in metabolic strategy from the modern, high-carb diet. (See The medical-grade diet for more on why a ketogenic diet is a class of its own among “diets”.)
But micronutrient needs depend on the metabolic state
It turns out that micronutrient needs depend on whether your metabolic state is glucose based, or fat and ketone based. Not only do the biochemical reactions involved in producing energy rely on different substances, but the downstream effects of this create different environments that deplete nutrients at different rates. This means that in many cases, the established RDA has little bearing on the context of a ketogenic diet.
The RDA for micronutrients can be imprecise. Before supplementing, make sure you’re eating enough nutrient dense foods (e.g. shrimp).
Animals making their own Vitamin C make less when carbs are low
At this point it will be helpful to understand some facts about vitamin C synthesis. The reason we need to consume vitamin C at all is that we are one of the few species that have lost the ability to make it ourselves. This evolutionary change happened to primates before humans emerged, so we share it with other primates, as well guinea pigs, some bats, and there are non-mammalian examples as well. Evolution does not systematically drop functions that are merely no longer useful. Because this genetic mutation is spread across entire species, and has happened in multiple lineages , there must be a selective advantage in not producing it . I will return to this in a later post.
Vitamin C is chemically similar to glucose, and its synthesis is intimately tied to glucose metabolism. You may have already read that vitamin C and glucose compete for uptake in cells. That’s based on the fact they their chemical similarity allows them to use the same cell receptor (Glut1) .
However, the connection is deeper than that. Vitamin C is made out of glucose. In animals that synthesise vitamin C, synthesis is downregulated exactly in fasting or low-carbohydrate conditions , or when glycogen is otherwise low .Notably, this is not the case in hibernation, where the reverse is true .
In other words, even in animals who synthesise their own vitamin C, synthesis is low in otherwise normal low-calorie or low-carbohydrate conditions. Note that these conditions would often also be lower dietary vitamin C conditions. This is interesting, because if optimal vitamin C levels were independent of carbohydrate intake, then we might expect to see the opposite. That is, we might expect to see more vitamin C synthesis in tandem with less carbohydrate intake in order to make up for the missing dietary component.
The fact that vitamin C synthesis is downregulated when food or carbohydrates are low suggests the following hypotheses.
First, it suggests that vitamin C might be more necessary in a glucose based metabolism. Second, it suggests that there are compensatory mechanisms that come into play when vitamin C is low that are also triggered by low-carbohydrate conditions, and therefore, vitamin C requirements are lower in low-carbohydrate conditions. Third, it suggests that high levels of vitamin C may even be detrimental under low-carbohydrate conditions.
I’ll leave the final one for a subsequent post, but let’s look at the first two.
Vitamin C is more necessary when glucose is high
It was proposed in 1975 by Mann and Newton that vitamin C transport across cell membranes may be impaired by glucose and suggested that diabetes may be a form of mild, chronic, localised scurvy . Since then it’s been shown how glucose and vitamin C compete for uptake in cells and how strikingly well the symptoms of diabetes and heart disease can be explained as latent scurvy .
It’s been shown (same paper) that infusions of high levels of vitamin C can mitigate the hyperglycemia-induced deficiency to a degree. So in that sense, vitamin C is definitely “more necessary” when glucose is high. It is plausible then, that a diet very low in carbohydrate would require less vitamin C, for two reasons. First, blood glucose values will be lower on average, meaning that there will be a more favourable ratio of vitamin C to glucose, even at the same vitamin C level. Second, in ketosis, many cells are taking up ketones for fuel, and therefore much less glucose needs to be taken up.
Your essential micronutrients are involved in the chemical reactions happening inside you all the all the time. How much you need can change significantly depending on aspects of your diet (e.g. lots of sugar or only a little?)
Endogenous antioxidants glutathione and uric acid spare Vitamin C
Vitamin C can be spared by something that takes over one of its functions, or by something that increases its effectiveness.
Vitamin C has multiple distinct functional roles
Vitamin C serves many functions; new discoveries and hypotheses are still being made. It is well established that it can serve as an antioxidant in vitro, although its antioxidant action in actual humans has not been confirmed . More particularly, it has not been established that increasing exogenous vitamin C intake has an antioxidant effect.
Perhaps the most important function of vitamin C is that scurvy does not occur in its presence. It is postulated, and widely believed that this is due to its role as a cofactor in hydroxylation reactions, though this is also unclear .
Ketogenic diets increase biosynthesis of glutathione
Even those that can’t synthesise vitamin C, can make more efficient use of it under the right conditions. In fact, scurvy can be substantially delayed in guinea pigs in the absence of dietary vitamin C, if glutathione esters are given. In one such experiment there was no sign of scurvy after 40 days, even though they usually die of it in 21-24 . That’s because one of the functions of glutathione is its essential role in vitamin C recycling .
We know that ketogenic diets upregulate glutathione biosynthesis . It’s unclear to me from the literature whether total levels are increased. In rats, it goes down in liver tissue, but up in hippocampal mitochondria .
It’s clear, however, that in addition to recycling vitamin C, glutathione has overlapping functions with vitamin C as an antioxidant and that they mutually spare each other . I hypothesise that in ketogenic conditions, other antioxidants such as glutathione take over many functions that would be served by vitamin C in synthesisers.
Another candidate for this is uric acid.
Uric acid is similarly sparing of Vitamin C
Another mutation that humans and primates share is a loss of function of the uricase enzyme. Uricase breaks down uric acid, and the result of this mutation is higher uric acid levels in primates. Just like with the loss of vitamin C synthesis, there is good reason to believe that this mutation conferred a selective advantage, but the nature of that advantage is controversial. In an upcoming post, I will review the state of that controversy.
One hypothesis is based on the antioxidant properties of uric acid. This was put forth by Bruce Ames et al. in 1981 . The idea is that because uric acid is a major antioxidant (more potent than vitamin C, for example) , its higher levels might explain the relatively long lifespans that apes have .
The uric acid mutation occurred in primates tens of millions of years after the vitamin C mutation, but it is plausible that they are related, that increased uric acid was of particular advantage in the context of lack of endogenous vitamin C. Regardless of whether its antioxidant role sufficiently explains the selective advantage of high uric acid, those antioxidant properties still stand.
Moreover, ketogenic diets decrease oxidative stress
Producing energy produces free radicals, but this is less true when fat and ketones are the energy source than when glucose is . This is one reason that ketogenic diets improve outcomes in traumatic brain injury .
So, we would expect the role of exogenous antioxidants to be less critical in a metabolic state that endogenously decreases oxidative stress.
Vitamin C isn’t the only anti-oxidant our body uses, in fact many anti-oxidants are made internally. How much Vitamin C you need also depend on how much of the other anti-oxidants there are.
What’s the relationship between insulin resistance and Vitamin C deficiency?
Given the above observations, that, on the one hand, some symptoms of diabetes and heart disease (i.e. of metabolic syndrome or insulin resistance) can be viewed as latent scurvy, and on the other hand, that the antioxidant properties of vitamin C in vivo are replaceable, it suggests another way of looking at the relationship between vitamin C and insulin resistance.
Linus Pauling famously believed that the diseases of civilisation were curable by high doses of vitamin C, but in practice, this has not consistently panned out. What if Pauling was right in the sense that an important underlying mechanism of those diseases is that insulin resistance reduces the access of tissue to vitamin C, causing many of the symptoms we associate with diseases of insulin resistance? There is much suggestive evidence about the relationship between vitamin C deficiency and hypertension, fasting blood glucose, and fatty liver disease, for example. This could explain why vitamin C infusions have inconsistent results. It is a band-aid solution; it doesn’t address the underlying problem, which is glucose overload, and impaired uptake of vitamin C. The popular hypothesis that the (inconsistent) beneficial effect of vitamin C in these diseases is a result of antioxidant properties, could be replaced by the simpler hypothesis that high intake of vitamin C can sometimes compensate for the scorbutic effect of glucose overload and insulin resistance.
Vitamin C requirements are probably much lower on a ketogenic diet:
– The amount of vitamin C required just for preventing scurvy was determined to be 10 mg a day, and that was determined in a high-carb context . Subsequently, a nearly tenfold inflation of this recommendation is based on speculative data about the ability to derive antioxidant properties from vitamin C, and the effect it could have on mitigating blood sugar complications of a high-carb diet .
-Insofar as antioxidant effects are important, these are likely to be met more powerfully by uric acid, glutathione, and the natural antioxidant consequences of low-carb diets, rather than exogenous supplementation.
-The inflated recommendations for vitamin C intake are likely to be completely inapplicable to a person following a ketogenic diet, because that person can use much smaller amounts of vitamin C efficiently.
How much vitamin C you need depends on how much you use. How much you use is significantly determined by your metabolic context, meaning how much anti-oxidant action you require, the availability of other anti-oxidants and crucial co-factors that help maintain adequate vitamin C status. Government RDAs for vitamin C were arrived at in the context of the modern American diet, characterized by its high carbohydrate content, high seed oil content and lack of complete quality protein intake. When considering what adequate vitamin C intake looks like on well-formulated ketogenic diet, it is important not to confuse this metabolic context with the one from the modern American diet. Empirical evidence suggests people on a well-formulated ketogenic (or carnivorous) diet aren’t at increased risk of scurvy, the conspicuous deficiency of vitamin C. Much is yet to be learned in this area.
Evidence type: review
Clinical Chemistry Vol. 43, Issue 4 April 1997
“In 1911, Casimir Funk isolated a concentrate from rice polishings that cured polyneuritis in pigeons. He named the concentrate “vitamine” because it appeared to be vital to life and because it was probably an amine. Although the concentrate and other “accessory food substances” were not amines, the name stuck, but the final “e” was dropped. “
Evidence type: review
Drouin, Guy, Jean-Rémi Godin, and Benoît Pagé.
Current Genomics 12.5 (2011): 371–378. PMC. Web. 19 Dec. 2016.
“Vitamin C (ascorbic acid) plays important roles as an anti-oxidant and in collagen synthesis. These important roles, and the relatively large amounts of vitamin C required daily, likely explain why most vertebrate species are able to synthesize this compound. Surprisingly, many species, such as teleost fishes, anthropoid primates, guinea pigs, as well as some bat and Passeriformes bird species, have lost the capacity to synthesize it. Here, we review the genetic bases behind the repeated losses in the ability to synthesize vitamin C as well as their implications. In all cases so far studied, the inability to synthesize vitamin C is due to mutations in the L-gulono-γ-lactone oxidase (GLO) gene which codes for the enzyme responsible for catalyzing the last step of vitamin C biosynthesis. The bias for mutations in this particular gene is likely due to the fact that losing it only affects vitamin C production. Whereas the GLO gene mutations in fish, anthropoid primates and guinea pigs are irreversible, some of the GLO pseudogenes found in bat species have been shown to be reactivated during evolution. The same phenomenon is thought to have occurred in some Passeriformes bird species. Interestingly, these GLO gene losses and reactivations are unrelated to the diet of the species involved. This suggests that losing the ability to make vitamin C is a neutral trait.”
Evidence type: observation
Bánhegyi Gábor,Csala Miklós,Braun László,Garzó Tamás and Mandl József
FEBS Letters, 381, doi: 10.1016/0014-5793(96)00077-4 (1996)
“Ascorbic acid and glutathione are involved in the antioxidant defense of the cell. Their connections and interactions have been described from several aspects: they can substitute each other , dehydroascorbate can be reduced at the expense of GSH  and glutathione depletion results in the stimulation of ascorbate synthesis . In ascorbate-synthesising animals, the formation of ascorbate from gulonolactone catalysed by microsomal gulonolactone oxidase is accompanied by the stoichiometric consumption of O2 and production of the oxidant hydrogen peroxide . Metabolism of hydrogen peroxide by glutathione peroxidase requires reduced glutathione. Therefore, we supposed that synthesis of ascorbate should decrease the intracellular glutathione level. To prove our hypothesis, experiments were undertaken to investigate the effect of ascorbate synthesis stimulated by the addition of gulonolactone on the oxidation of GSH in isolated mouse hepatocytes and liver microsomal membranes.”
“In this paper, a new connection between ascorbate and GSH metabolism is described. Our data show that the synthesis of ascorbate leads to consumption of GSH, the other main intracellular antioxidant (Fig. 1). We suppose that the formation of hydrogen peroxide is underlying the increased GSH consumption. First, oxidation of GSH caused by increased ascorbate synthesis was prevented by the addition of catalase in microsomal membranes (Table 1). Second, inhibition of glutathione peroxidase by mercaptosuccinate moderated the gulonolactone-dependent glutathione consumption in microsomes (Table 2). Third, the inhibition of catalase by aminotriazole deepened the ascorbate synthesis-dependent GSH depletion in isolated hepatocytes (Table 3). This interaction may be one of the causes why primates and some other species have lost their ascorbate-synthesising ability. This event occurred in the ancestors of primates about 70 million years ago, owing to mutation(s) in the gulonolactone oxidase gene . Despite the well-known benefits  of ascorbate, the mutation(s) had to be advantageous, as this metabolic error did not remain an enzymopathy affecting only a minority of the population, but spread widely amongst the species (and individuals) of primates and became exclusive . There is no explanation for this unexpected outcome. Based on these analytical data, the following conceptual evolutionary hypothesis can be outlined: in the tropical jungle of the Cretaceous Period, when exogenous ascorbate was abundant [17,18], the loss of gulonolactone oxidase activity could have proved to be advantageous. It saved the reduced GSH, the main defence system against oxidants, while the access to ascorbate was not hindered. Later, the evolutionary gains of these periods allowed the conservation of the genetic disorder manifested in the loss of ascorbate synthesis.”
Evidence type: experiment
KC S, Cárcamo JM, Golde DW.
FASEB J. 2005 Oct;19(12):1657-67.
Reactive oxygen species (ROS)-induced mitochondrial abnormalities may have important consequences in the pathogenesis of degenerative diseases and cancer. Vitamin C is an important antioxidant known to quench ROS, but its mitochondrial transport and functions are poorly understood. We found that the oxidized form of vitamin C, dehydroascorbic acid (DHA), enters mitochondria via facilitative glucose transporter 1 (Glut1) and accumulates mitochondrially as ascorbic acid (mtAA). The stereo-selective mitochondrial uptake of D-glucose, with its ability to inhibit mitochondrial DHA uptake, indicated the presence of mitochondrial Glut. Computational analysis of N-termini of human Glut isoforms indicated that Glut1 had the highest probability of mitochondrial localization, which was experimentally verified via mitochondrial expression of Glut1-EGFP. In vitro mitochondrial import of Glut1, immunoblot analysis of mitochondrial proteins, and cellular immunolocalization studies indicated that Glut1 localizes to mitochondria. Loading mitochondria with AA quenched mitochondrial ROS and inhibited oxidative mitochondrial DNA damage. mtAA inhibited oxidative stress resulting from rotenone-induced disruption of the mitochondrial respiratory chain and prevented mitochondrial membrane depolarization in response to a protonophore, CCCP. Our results show that analogous to the cellular uptake, vitamin C enters mitochondria in its oxidized form via Glut1 and protects mitochondria from oxidative injury. Since mitochondria contribute significantly to intracellular ROS, protection of the mitochondrial genome and membrane may have pharmacological implications against a variety of ROS-mediated disorders.
Evidence type: non-human animal experiment
Stirpe F, Comporti M, Caprino G.
Biochem J. 1963 Feb;86:232-6.
“Effect of starvation and subsequent feeding. The effect of starvation was then investigated, and it appeared that a 24 hr. period of starvation was enough to decrease the synthesis of ascorbic acid (Table 2). Since Caputto et al. (1958) had shown that the maximum effect of vitamin-E deficiency on the synthesis of ascorbic acid was reached as shortly as 3-4 days after deprivation, the possibility was considered that the effect of starvation was actually due to lack of vitamin E. This was discounted by giving starved animals enough vitamin E to prevent formation of peroxides; there was no effect on the synthesis of ascorbic acid. The effect of starving was quickly reversed by feeding the rats again for 24 hr.”
“Effect of omission of carbohydrates from the diet and of administration of precursors: The effect of starvation could be attributed either to the stress or to the lack of some dietary components. A strong impairment of the synthesis of ascorbic acid was observed in rats given a carbohydrate-free diet for 24 hr., whereas values significantly higher but still below normal ones were obtained by giving this same diet for 6 days (Table 3). Rats on this ration had a lower content of ascorbic acid in the liver, but showed an enhanced excretion of ascorbic acid in the urine. Since carbohydrates are precursors of ascorbic acid in the rat, this observation led to the hypothesis of an adaptive response of the enzyme system to lack of substrates, and evidence was sought by giving glucuronolactone to rats. Administration of glucuronolactone did not affect the rate of synthesis in normal rats, but caused a moderate but significant enhancement in starved animals. However, a similar enhancement followed the administration of an equal amount of glucose. All rats receiving glucuronolactone had a higher liver content and an enhanced urinary excretion of ascorbic acid.”
Evidence type: non-human animal experiment
Braun L1, Garzó T, Mandl J, Bánhegyi G.
FEBS Lett. 1994 Sep 19;352(1):4-6.
“The role of the hepatic glycogen content in ascorbic acid synthesis was investigated in isolated mouse hepatocytes. The cells were prepared from fed or 48 h-starved mice and the ascorbic acid content was measured in the suspension (cells+medium). After 48 h starvation hepatocytes did not contain measurable amounts of glycogen. The initial concentration of ascorbic acid was lower in the suspension of glycogen-depleted hepatocytes compared to the fed controls (Fig. 1) and only a moderate synthesis could be observed under both nutritional conditions. The effects of dibutyryl CAMP and glucagon on ascorbate synthesis were examined. Glucagon or dibutyryl cyclic AMP caused a stimulation of ascorbic acid synthesis in hepatocytes from fed mice, while in hepatocytes from 48 h starved animals ascorbic acid production was not increased significantly by the two agents (Fig. 1). The addition of glucose and gluconeogenic precursors to the incubation medium did not result in a significant increase in ascorbic acid production (Fig. 1). In another series of experiments glucose and ascorbic acid production of the cells was measured simultaneously. The rate of glucose production (in the absence of gluconeogenic precursors mainly via glycogenolysis) and ascorbic acid synthesis showed a close correlation (r = 0.9091) (Fig. 2). As ascorbic acid synthesis and glycogenolysis seemed to be connected, we examined the effect on ascorbic acid synthesis of various agents known to increase glycogenolysis. The al agonist phenylephrine, the protein phosphatase inhibitor okadaic acid and vasopressin all increased the rate of ascorbic acid production in isolated hepatocytes prepared from fed mice similarly to glucagon (Table 1).
“Glycogenolysis was stimulated by the in vivo addition of glucagon. Glucagon elevated the blood glucose level of mice by 50%; at the same time a more than fifteenfold increase of plasma ascorbic acid concentration could be observed (Table 2). The concentration of ascorbic acid in the liver was also increased, indicating a stimulated hepatic synthesis (Table 2).”
Glycogen content is considered to be a sensitive marker showing the actual metabolic state of the liver. Observations described in this paper suggest that ascorbic acid synthesis in murine liver is tightly connected with the glycogen pool; the source of ascorbic acid is glycogen. The following results gained in isolated hepatocytes support this assumption: first, in hepatocytes isolated from glycogen-depleted animals the ascorbic acid level as well as the rate of synthesis is lower than that in hepatocytes from control fed mice (Fig. 1); second, different glycogen-mobilizing agents acting via different mechanisms enhance ascorbic acid production in hepatocytes from fed but not from fasted animals (Fig. 1, Table 1); third, addition of glucose to hepatocytes prepared from glycogen-depleted mice failed to increase the formation of ascorbic acid (Fig. 1). The results gained under in vitro conditions in isolated hepatocytes were confirmed by in vivo experiments: a single i.p. injection of glucagon elevated both the plasma and liver ascorbic acid levels within 15 min (Table 2). “
“The finding that the source of ascorbate production is glycogenolysis is in according with the fact that liver and kidney -the main sites of glycogen storage – are responsible for the ascorbic acid supply in most animal species . The increased hepatic ascorbic acid production after glucagon administration can be explained as a compensatory mechanism of the missing intake of ascorbate, i.e. adaptation of ascorbic acid supply from external to internal sources. Considering the fifteenfold elevation of plasma ascorbate levels, in the light of recent findings concerning the effect of ascorbate on insulin secretion  and on the calcium channels in pancreatic beta cells  it might be also regarded as a possible intercellular messenger. “
Evidence type: experiment
Drew KL, Tøien Ø, Rivera PM, Smith MA, Perry G, Rice ME.
Comp Biochem Physiol C Toxicol Pharmacol. 2002 Dec;133(4):483-92.
“During hibernation plasma ascorbate concentrations w(Asc)px were found to increase 3–5 fold in two species of ground squirrels, AGS and 13-lined ground squirrels (TLS); S. tridecemlineatus and cerebral spinal fluid (CSF) ascorbate concentration w(Asc)CSFx doubled in AGS (CSF was not sampled in TLS) (Drew et al., 1999). During arousal, however, when oxygen consumption peaks and the generation of reactive oxygen species is thought to be maximal, plasma ascorbate concentrations progressively decrease to levels typical for euthermic animals (Fig. 3).”
Evidence type: observation
The membrane Transport of ascorbic acid
George Mann and Pamela Newton
Ann N Y Acad Sci. 1975 Sep 30;258:243-52.
“We have formulated two hypotheses. The first proposes that the transport of ascorbate across cell membranes may be impaired by glucose. The second proposes that the transport of ascorbate in certain tissues is facilitated by insulin. If either hypothesis is valid, those species requiring exogenous ascorbate would be in double jeopardy if they were also hyperglycemic. Carbohydrate intolerance resulting from either a lack of or a resistance to insulin is common in Western man. Gore et al. have shown with electron microscopy that the vascular lesion of scurvy involves collagenous structures in the basement membranes, and this is also the site of the lesion in diabetic microangiopathy. These hypotheses, which propose that the intracellular availability of dehydroascorbate (DHA), the transportable form of vitamin C, would be impaired in certain tissues by either hyperglycemia or lack of insulin, suggest that diabetic microangiopathy, the main complication of human diabetes, may be a consequence of local ascorbate deficiency. The laboratory investigations described here deal with the first and somewhat simpler of these hypotheses: Glucose will impair the transport of dehydroascorbate into cells. The data collected show that D-glucose does inhibit the transport of dehydroascorbate into human red blood cells, a noninsulin-dependent tissue. Trials with other sugars show a hierarchy of sugars that inhibit transport, suggesting that DHA and D-glucose share a carrier mechanism.”
Evidence type: review
Price KD, Price CS, Reynolds RD.
Atherosclerosis. 2001 Sep;158(1):1-12.
“Hyperglycemia-induced ascorbic acid deficiency
Vitamin C is a derivative of glucose and Mann  proposed that the structural similarity between these two molecules may account for many of the complications of diabetes. Glucose has been shown to inhibit vitamin C transport in several mesenchymal cell types, including endothelial cells , mononuclear leukocytes , neutrophils [141,142], fibroblasts [143,144], and erythrocytes . Facilitative glucose transporters (GLUTs) bind dehydroascorbic acid and are thought to be the primary transporters of vitamin C in mammalian cells . After transport, dehydroascorbic acid is quickly reduced to ascorbic acid. Glucose competitively inhibits the uptake of dehydroascorbic acid but does not affect ascorbic acid transport. Ascorbic acid is transported by a family of membrane-bound proteins that are Na+-dependent and whose function is not directly inhibited by elevated extracellular concentrations of glucose [146,147]. This latter system is prevalent in bulk-transporting epithelia (e.g. kidney and small intestine) and have been recently isolated in both human  and rat  biological systems. Many cell types, of course, [150,151] express both transport systems.
High blood glucose concentrations mimic the conditions of vitamin C deficiency. Acute hyperglycemia, for example, impairs endothelium-dependent vasodilation in healthy humans , an effect which can be reversed by acute administration of vitamin C . Ascorbic acid plays an important role in extracellular matrix regulation and has a stimulatory effect on sulfate incorporation in mesangial cell and matrix proteoglycans; high glucose concentrations have been shown to impair this effect . Endothelial surface proteoglycans help prevent thrombus formation and also inhibit smooth-muscle growth . High glucose concentrations also have been shown to inhibit the stimulatory effect of ascorbic acid on collagen and proteoglycan synthesis in cultured fibroblasts . Moreover, a high concentration of glucose can induce the expression of intercellular adhesion molecule-1 (ICAM-1) in human umbilical vein endothelial cells . Endothelial cells express these and other membrane-bound proteins to enable leukocyte adhesion and transmigration across the endothelium during an inflammatory response. Atherosclerosis is one such inflammatory response.
Experimental and clinical studies suggest that latent scurvy is characterized by IGT [16,24] and diabetes mellitus is a disease complex characterized by impaired glucose and vitamin C metabolism [27,28]. Diabetic patients are prone to hyperglycemia, prolonged wound healing, infection, increased synthesis of cholesterol, decreased liver glycogen, and notably, diffuse vascular disease. All of these findings are consistent with latent scurvy . Diabetic platelets have been shown to have low intracellular ascorbic acid concentrations and display hypercoagulability . Long-term vitamin C administration has beneficial effects on glucose and lipid metabolism in aged NIDDM patients . It has also been suggested that vitamin C consumption above the RDA may provide important health benefits for individuals with IDDM . This latter recommendation is supported by recent evidence. For example, mesenchymal cells from patients with IDDM have an impaired uptake of dehydroascorbic acid that persists in culture  and ascorbic acid has been shown to prevent the inhibition of DNA synthesis induced by high glucose concentrations in cultured endothelial cells . Diabetic patients have been observed to have a lowered ascorbic acid/dehydroascorbic acid plasma ratio, indicating a decreased vitamin C status . Therefore, diabetic patients may benefit from vitamin C supplementation to alleviate multiple physiologic and metabolic impairments in a variety of cell types.”
Evidence type: review
Padayatty SJ, Katz A, Wang Y, Eck P, Kwon O, Lee JH, Chen S, Corpe C, Dutta A, Dutta SK, Levine M.
J Am Coll Nutr. 2003 Feb;22(1):18-35.
“Problems in Demonstrating Antioxidant Benefit of Vitamin C in Clinical
“Studies Despite epidemiological and some experimental studies, it has not been possible to show conclusively that higher than anti-scorbutic intake of vitamin C has antioxidant clinical benefit. This is despite the fact that vitamin C is a powerful antioxidant in vitro. It is of course possible that the lack of antioxidant effect of vitamin C in clinical studies is real. It seems more likely that vitamin C has antioxidant or other benefits. Detection of these benefits has remained elusive due to the vicissitudes of experimental design. Vitamin C may be a weak antioxidant in vivo, or its antioxidant actions may have no physiological role, or its role may be small. The oxidative hypothesis is unproven, and oxidative damage may have a smaller role than anticipated in some diseases. Further, antioxidant actions of vitamin C may occur at relatively low plasma vitamin C concentrations. Thus additional clinical benefits that occur at higher vitamin C concentrations may be difficult to demonstrate. Although all these are possible explanations, it seems unlikely that these are the real reasons for the lack of detectable effects of vitamin C in clinical studies. Many factors may contribute to the failure so far to demonstrate clear antioxidant benefits of vitamin C in clinical studies. The antioxidant actions of vitamin C may be specific to certain reactions or occur only at specific locations. In either case, beneficial effects can be shown only in disorders where such reactions or sites are the focus of disease process. There may be many different antioxidants that are active at the same time. In the face of such redundancy, only multiple antioxidant deficiencies will have detectable clinical effects. Antioxidant deficiency may have to be of long duration for accumulated damage to be noticeable. Antioxidant effects may be of importance only in those with oxidant stress. Thus, normal subjects or those with mild disease may have no need for high antioxidant concentrations. In a way analogous to the effect of acetaminophen on fever, antioxidants may have no effect in the absence of marked oxidant stress. A further problem is presented by the sigmoidal dose concentration curve for vitamin C. Small changes in oral intake of vitamin C produce large changes in plasma vitamin C concentrations. This makes it difficult to conduct controlled studies such that the plasma vitamin C concentrations of the control and study groups differ sufficiently to have physiological meaning.”
Evidence type: review
Padayatty SJ, Levine M
Oral Dis. 2016 Sep;22(6):463-93. doi: 10.1111/odi.12446. Epub 2016 Apr 14.
“Common symptoms of scurvy include wound dehiscence, poor wound healing and loosening of teeth, all pointing to defects in connective tissue (Crandon et al, 1940; Lind, 1953; Hirschmann and Raugi, 1999). Collagen provides connective tissue with structural strength. Vitamin C catalyzes enzymatic (Peterkofsky, 1991) posttranslational modification of procollagen to produce and secrete adequate amounts of structurally normal collagen by collagen producing cells (Kivirikko and Myllyla, 1985; Prockop and Kivirikko, 1995). Precollagen, synthesized in the endoplasmic reticulum, consists of amino acid repeats rich in proline. Specific prolyl and lysyl residues are hydroxylated, proline is converted to either 3-hydroxyproline or 4-hydroxyproline, and lysine is converted to hydroxylysine. The reactions catalyzed by prolyl 3-hydroxylase, prolyl 4- hydroxylase, and lysyl hydroxylase (Peterkofsky, 1991; Prockop and Kivirikko, 1995; Pekkala et al, 2003) require vitamin C as a cofactor. Hydroxylation aids in the formation of the stable triple helical structure of collagen, which is transported to the Golgi apparatus and eventually secreted by secretory granules. In the absence of hydroxylation, secretion of procollagen decreases (Peterkofsky, 1991) an