Although I have long been aware of vitamin A’s essential role in creating vision, I only recently discovered that vitamin A plays an analogous role in setting our circadian rhythm. When blue light from sunshine enters our eyes, vitamin A translates it into a signal that tells our brain it is daytime. When this signal wanes, our brain knows that it is nighttime. This means that vitamin A plays an essential role in helping us fall asleep on time, get high quality sleep, sleep long enough, wake up feeling rested, and staying alert and energetic throughout the day.
If geeking out on nutritional science is not your thing, skip to the practical recommendations. If you’re all about the geeking out, read the whole thing, and for some extra fun click on the plus signs as you go along to read the technical notes.
Vitamin A plays an essential role in creating vision.+ Since impaired night vision is considered the earliest reliable sign of vitamin A deficiency, this role has been most studied in rods, the cells of the retina that allow us to see outlines of shapes in dim light, but it also takes place in cones, the cells that allow us to see colors in brighter light.
In all cases, a metabolite of vitamin A is bound to a protein of the opsin family. Light enters the eye, makes its way to the retina, and strikes the vitamin A. In a domino effect, the energy from the light causes vitamin A to change shape, which causes the opsin protein to change shape, which then causes an electrical signal to travel through the optic nerve to our brain, which, finally, integrates many such signals to form what we experience as vision.
In rods, the opsin protein is called rhodopsin (rod-opsin). In cones, there are three opsins, specific to red, green, or blue light. In more casual language, these proteins are named after the colors they absorb (“red opsin,” “green opsin,” and “blue opsin”); in more jargony language, they are named after the specific wavelengths of light they absorb.
Apart from rods and cones, there are cells in our retina that are responsible for setting our circadian rhythm, known as intrinsically photosensitive retinal ganglion cells (ipRGCs). In these cells the opsin protein is known as melanopsin, and its function is to communicate to our brain that it is daytime in response to blue light entering the eye.
Vitamin A plays the same role in melanopsin as it does in the other opsins. Thus, without vitamin A, our brains cannot know when it is daytime and cannot set our circadian rhythm. If this system fails, we are less likely to feel alert in the daytime, less likely to feel relaxed in the night time, and less likely to get adequate, regular, restful sleep.
The Scientific Evidence for Vitamin A’s Role in Setting the Circadian Rhythm
The scientific evidence supporting the role of vitamin A in the circadian rhythm includes biochemistry experiments mapping out the melanopsin pathway (e.g., 1 and 2) and an observational study in humans showing that people with lower vitamin A status are more likely to have sleep disorders (3).
The observational study selected just shy of 2,500 people from the 2005-2006 wave of the National Health and Nutrition Examination Survey (NHANES) for whom sufficient data on nutrition biomarkers, sleep quantity, and sleep quality were available. The researchers used serum retinol as a marker of vitamin A status. Retinol is the form of vitamin A our bodies require and serum retinol is the concentration of vitamin A that circulates in our blood.
Each 0.63 umol/L (18 ug/dL) increase in serum retinol was associated with a 22% lower risk of sleep disorders.++
The study also found that higher nutritional status for vitamin D, folate, vitamin B12, a collection of carotenoids, and vitamins E and C were associated with better sleep. I recommend reading the full study for a better appreciation of all of the potential explanations for these findings,+++ and I caution that these are correlations that do not by themselves show causation, but here I will discuss a simple interpretation that fits all the findings into the theme of this blog post:
- 25(OH)D, the marker of vitamin D status, rises with increasing exposure to sunlight. More exposure to sunlight means more blue light enters the eye during the daytime.
- Higher intakes of vitamin A in the form of retinol from animal foods and provitamin A carotenoids from plant foods mean that more retinol will be available to allow the melanopsin pathway in ipRGCs to translate that blue light into the signal that tells the brain it is daytime.
- Vitamin A is very vulnerable to irreversible degradation in response to high-intensity blue and ultraviolet light. Lutein and zeaxanthin are non-provitamin A carotenoids that accumulate in the retina to protect against such damage. Better antioxidant status in general and especially better lutein and zeaxanthin status could reflect a better capacity to preserve vitamin A from degradation.
- With a stronger blue light/vitamin A-mediated daytime signal, it becomes much clearer to the brain when it is nighttime. In response to clearly perceived darkness, serotonin is converted to melatonin, the final step of which is methylation, which requires vitamin B12 and folate.
How “Deficient” in Vitamin A Do You Need to Be to Wreck Your Circadian Rhythm?
Night blindness is conventionally thought to be the most sensitive sign of vitamin A deficiency. Although vitamin A plays analogous and equally important roles in night and day vision, night vision is the first to go in deficiency because day vision gets priority over what precious vitamin A is available. From an evolutionary perspective, this makes sense because, until the recent advent of artificial lighting, we engaged in most of our survival-critical work in the daytime. Vitamin A-mediated gene expression also gets priority, which is needed to keep the eyes lubricated and protect them from infection and from being scratched by debris.
One way to get a sense of whether rhodopsin-mediated night vision or melanopsin-mediated circadian rhythym setting would be more sensitive to deficiency would be to experimentally flood rods, cones, and ipRGCs with high-intensity blue and ultraviolet light, which destroys vitamin A. One such experiment (1) using mouse tissue found that rods and cones were much more sensitive to this than ipRGCs. This would suggest that, at least in mice, vision would be compromised at an earlier stage of deficiency than melanopsin-mediated circadian rhythm setting.
The case may be different in humans (2). Mouse melanopsin binds to vitamin A very tightly, whereas human melanopsin binds it very loosely. In a larger comparison including various non-mammals and non-human primates, humans stand out as having melanopsin that binds vitamin A very loosely. The one exception tested was the galago, a nocturnal primate that does not seem to rely on blue light detection for its circadian rhythm. The loose vitamin A-binding in human melanopsin is thought to be an adaptation to the bright light environments we have inhabited throughout most of our evolutionary history, making our eyes less sensitive to blue light and giving our eyes the ability to more easily adjust our sensitivity to blue light in different environments.
The loose binding could also make the vitamin A in human ipRGCs more vulnerable to degradation. If that is the case, loss of circadian rhythm-setting could in fact be a more sensitive sign of vitamin A deficiency than loss of night vision.
Nevertheless, there are too many variables that would affect this to say for sure. In a live human experiencing a deficiency of vitamin A, degradation by blue and UV light, enzymatic degradation, and supply of circulating vitamin A to ipRGCs will determine how melanopsin is prioritized relative to rhodopsin, cone opsins, and gene expression. The only way to generate firm conclusions would be to study the clinical effects of deficiency and repletion in humans.
It is certainly possible that disruption of the circadian rhythm is more sensitive to vitamin A deficiency than loss of night vision, so if someone is experiencing a disrupted circadian rhythm, vitamin A should not be ignored on the grounds that the person is not also experiencing a loss of night vision or other signs of vitamin A deficiency.
Practical Recommendations for Using Vitamin A to Support Good Sleep
If you have trouble falling asleep or staying asleep, feeling alert during the day and relaxed at night, or feeling fully rested upon waking, you probably have a disruption in your circadian rhythm. The primary sign that it is vitamin A-related should be that efforts to control your exposure to blue light do not work the way they should.
A recent study (4) of two hunter-gatherer groups, the Hadza and the San, and a hunter-horticulturalist group, the Tsimane, gives us some insight into how our sleep was likely coordinated with the rhythms of nature throughout most of our history as a species. Moreover, these groups have no word for insomnia and report very low rates of trouble falling asleep (1.5%) or staying asleep (2.5%) compared to people in industrialized societies (10-30%), which is consistent with the hypothesis that sleep troubles are common in our society because we have divorced ourselves from these natural rhythms.
These groups sleep during the period of declining ambient temperature, and wake up when the ambient temperature reaches its nadir. We are probably adapted to sleeping during declining temperature because that is when it is easiest for us to lower our own body temperature, a natural part of restful sleep.
They go to bed between 2.5 and 4.4 hours after sunset, wake up soon after sunrise, and since they seek refuge from the hot midday sun, their light exposure peaks around 9:00 AM. The several hours of darkness before sleep allows sufficient melatonin to be synthesized to induce sleep around the onset of declining temperature, which typically occurs several hours after sunset. Peak exposure to light in the morning gives the brain a quick and strong message that it is daytime, and time to feel awake and alert.
Our ability to control our own light and temperature is both a strength and a weakness.
Many of us in modern society live in parts of the world where it would be extremely uncomfortable and sometimes dangerous to always be exposed to the natural outside temperature. Artificial lighting gives us control over our productivity and lifestyle.
I do not think these things are bad, in and of themselves. The problem is that we work indoors with artificial lights that are nowhere near as bright as the outside sun, and keep them on during the night, giving us far more blue light than the moon and a campfire. Thus, we get too much blue light at night and not enough during the day. We often use air conditioning in the day and heat at night, flattening or even reversing the natural temperature rhythm.
If we use technology smartly, we can replicate the natural rhythms while also having control over our comfort and lifestyle. Practical ways to approach this include the following (I am not the first to suggest any of these):
- Do what you can to make sure your environmental temperature peaks in the day time and declines while you are sleeping.
- Spend a half hour to an hour in the outside sun in the morning soon after waking.
- Use dim light at night, and if you are using screen-based entertainment (TV, phone, computer) or light that is bright enough to allow you to read a book, use blue light-blocking technology. This includes blue-blocking glasses and apps like f.lux.
If these strategies do not work, poor vitamin A status should be considered a possible explanation. This interpretation is strengthened if you have poor night vision (for example, if you strain your eyes or have trouble seeing when driving down an unlit road with your headlights alone but can see fine without any eyestrain during the day) or dry eyes. This interpretation is also strengthened if you have serum retinol near or below the bottom of the reference range, or if you track your vitamin A intake and your daily average is below the RDA.
It is important to note that any of these additional pieces of evidence strengthen the interpretation of poor vitamin A status, but their absence does not rule it out. If circadian rhythm disruption is more sensitive to deficiency than night blindness, then it may occur in the absence of other clinical signs, at serum retinol concentrations within the reference range, and at vitamin A intakes above the RDA.
The best way to ensure basic adequacy of vitamin A is to eat a serving of liver once per week, and/or a serving of cod liver oil once per day, with regular use of pastured egg yolks and butter, and daily use of red palm oil and/or of red, orange, yellow, or green vegetables. Vitamin A sources should always be eaten in the context of a fat-inclusive meal.
The animal sources above supply vitamin A as retinol, the form that our body requires, while the plant sources supply provitamin A carotenoids that can be converted to retinol. There is a lot of variation between individuals in the ability to convert the provitamin A carotenoids in plant foods to retinol. Therefore, the plant foods listed above are not substitutes for the animal foods, and the surest guarantee of adequate vitamin A is the use of liver and/or cod liver oil.
Some people may need unusually high vitamin A intakes. One well-documented reason for this is poor absorption from the gut as a result of disorders that cause fat malabsorption. Although I have not found any clear investigation of the role of eye color, I suspect that people with lightly colored eyes have a higher requirement for vitamin A because lightly colored eyes let in more blue and UV light that can irreversibly degrade it. I have very lightly colored eyes and a health history suggesting a need for vitamin A intakes much higher than the RDA. There may be many other factors, such as genetic variations in the enzymes that degrade vitamin A, health issues that affect the rate of vitamin A utilization, status of nutrients that interact with vitamin A, and so on.
The best way to add vitamin A to the diet is to consume more of the vitamin A-rich foods listed above, especially liver. An alternative is to use vitamin A supplements. When experimenting with unusually high doses, it is wise to pay close attention to signs and symptoms, work with an ancestral health-minded health care professional, and monitor fasting serum retinyl esters (often estimated as retinyl palmitate) after a few months, which elevate when you are taking more vitamin A than your liver can handle.
It is important to remember that vitamin A, like any other nutrient, does not operate in a vacuum, and it is always best to make sure that nutrient-dense foods are the main source of your vitamins and minerals and that any attempt to strengthen one nutrient with foods or supplements be made in the context of a well rounded, nutrient-dense diet.
+ The description of vitamin A’s role in creating vision herein is simplified and leaves out a lot of detail and technical language. For a free introduction to the topic with more detail from 2009, see this link; for something more up-to-date and much more detail, see the Wikipedia page on opsin proteins. [Back]
++ The association between serum retinol was attenuated or eliminated in mixed statistical models that adjusted for socioeconomic demographics or intakes of other nutrients. I cited the odds ratio, which was calculated from a model adjusted for socioeconomic demographics and was not statistically significant at p=0.08. The unadjusted association was significant at p=0.039, but an odds ratio was not calculated for the unadjusted model.
Several things are worthy of note. This was conducted as an exploratory study of many nutrients with no specific a priori hypothesis. Thus, all the p values are two-tailed. I found this study when searching for support or refutation for my a priori hypothesis that higher vitamin A status would be associated with a lower risk of sleep disorders for the reasons described above, and I think it is fair to view the data from this vantage point according to the corresponding one-tailed p values, in which case the odds ratio would be considered significant. Significance cutoffs are arbitrary, in any case, so I don’t think missing the cutoff is grounds for dismissing the results when there is strong biological plausibility for a causal effect.
Further, the interpretation of statistical adjustments depends on an understanding of the causal pathways involved. If the causal pathways are unknown, statistical adjustments are as likely to obscure real associations as they are to eliminate spurious associations or elucidate hidden but real associations. For example, serum retinol was higher among those who were lean, white, and male, and was lower in those with a higher BMI, females, and non-whites. If BMI, race, and sex affect sleep independently of serum retinol, the adjusted model may better reflect the true associations. But if retinol is the causal factor that protects against sleep disorders, then the lower serum retinol in females, those with high BMI, and non-whites may contribute to more sleep disorders in these populations and adjusting for BMI, sex, and race would obscure this association. I discussed this with the corresponding author, who agreed with me that, since this is an exploratory study, all of the models are equally worthy of attention, including the unadjusted associations. [Back]
+++ Methylation and the active metabolites of vitamins A and D regulate gene expression in the parts of the brain that regulate the sleep-wake cycle, for example, and animal experiments show that disrupting nutrient-related gene expression interferes with proper sleep. See the discussion section in the original study for more details and for references. [Back]
- Sexton TJ, Golczak M, Palczewski K, Van Gelder RN. Melanopsin is highly resistant to light and chemical bleaching in vivo. J Biol Chem. 2012;287(25):20888-97.
- Tsukamoto H, Kubo Y, Farrens DL, Koyanagi M, Terakita A, Furutani Y. Retinal Attachment Instability Is Diversified among Mammalian Melanopsins. J Biol Chem. 2015;290(45):27178-87.
- Beydoun MA, Gamaldo AA, Canas JA, Beydoun HA, Shah MT, McNeely JM, Zonderman AB. Serum nutritional biomarkers and their associations with sleep among US adults in recent national surveys. PLoS One 2014;9(8):e103490.
- Yetish G, Kaplan H, Gurven M, Wood B, Pontzer H, Manger PR, Wilson C, McGregor R, Siegel JM. Natural Sleep and Its Seasonal Variations in Three Pre-Industrial Societies. Curr Biol. 2015;25(21):2862-8.