Sunday, September 2, 2012

Circadian Rhythmicity: Retinol (Vitamin A) & Caffeine and Their Effects on the Central & Peripheral Clocks of the Body

Image 1: Is it a bad idea to "wake yourself up" with a pot of coffee in the morning, I mean from a circadian rhythm perspective?
In the last installment of this series we have been dealing with breakfast. Now, if you are following the mainstream advice neither of the two subjects of this installment should actually be a staple of it. Vitamin A, in its active form, retinol, is "bad and dangerous" and only present in such "evil cholesterol laden foods" such as eggs. And since coffee will sure give you a heart attack, you better stick to your calcium fortified orange juice, a minimal amount of white water, ah.. I mean low fat "milk" (learn more about the difference between white water and milk in "Mutant Milk!? New Research Fuels the Flames on Hushed Up Concerns About Ill Health Effects of Homogenized Milk") and - of course - "healthy cereals". And while you will hardly be able to argue that skipping a breakfast like that is probably the best you can do for your health, this was the topic of the last installment, while vitamin A and caffeine, will be what this episode of the Circadian Rhythmicity Series will be all about.

Vitamin A the circadian vitamin?

Only recently (officially, at least; preliminary results have been published ahead of print in March 2012, already; cf. Golini. 2012) a group of researchers from the Multidisciplinary Institute of Biological Research San Luis (IMIBIO-SL), at the National University of San Luis in Argentina found that contrary to the peripheral clock gene expression in the liver, which does not appear to be disturbed by vitamin A deficiency (Shirai. 2006), the superordinate (=master) clock gene expression in the hippocampus of rats housed at a regular 12h-light/dark interval gets profoundly compromised, when the rodents are fed a vitamin A (retinol, not beta carotene!) deficient diet (Navigatore-Fonzo. 2012). According to Navigatore-Fonzo et al. the effects are mediated by modified temporal patterns of the retinoic acid receptor in the hippocampus, which plays an essential role in the activation of a whole set of clock-genes that, in turn, have been implicated - among others in the anti-cancer effects of vitamin A, you've read about at the SuppVersity not too long ago!

In the light of these recent results many previously observed, but not fully understood effects of vitamin A deficiency, such as the permanent memory impairments (Etchamendy. 2003) and its repeatedly suggested involvement as a signaling molecule (and as it now turns out potential zeitgeber) in physiological (synaptic plasticity, learning and memory, sleep), as well as pathological (schizophrenia, depression, Parkinson disease, and Alzheimer disease) neurological conditions (cf. Tafti. 2007).

The fact retinol availability is so tightly regulated alone tells us something about its importance

Is there a fluctuation in serum retinol levels as well or is vitamin A only a prerequesite for the circadian rhythm to function normally? With the activity level of vitamin A depending on both the availability as well as the release and binding of retinoic acid from the stores (mostly) in the liver and to the respective binding proteins, which are also produced in the liver, it is obvious that the liver is the most important regulator of vitamin A metabolism (Buzio. 1989). Maybe this is also why it is protected against circadian disturbances subsequent to vitamin A deficiency.
Unfortunately, our understanding of the exact function of the retinol binding proteins is still very limited, what we do know, however, is that their release and renal clearance show a distinct circadian rhythm which is synchronized to meal ingestion and the excretion of (Buzio. 1989). Our understanding of these mechanisms is yet still too preliminary to make any supplement recommendations besides "don't avoid the full-fat vitamin A rich foods, we have been eating for ages!". This is all the more true, since the range, within which beneficial effects can be seen is not just very narrow, but will also depend on (a) your baseline vitamin A status and (b) the way your body metabolizes dietary and supplemental vitamin A, the latter of which usually comes in the form of retinyl palmitate.

At doses in the <10,000IU/day range vitamin A is regarded as totally benign, but even doubling that dosage, which was basically what Behr et al. did for their recently published paper on the potential anti-oxidant effects of vitamin A on menopausal increases in oxidative brain damage, when they  supplemented the diets of ovariectomized rats with 1,500IU /kg retinol palmitate (human equivalent ~20,000IU) per day, can result in profound increases in cerebral oxidative damage (Behr. 2012 Jul).

In conjunction with vitamin A's beneficial effect on serum markers of oxidative damage Behr et al. had observed in a previous trial with 500IU/kg and 1,500IU/kg per day (human equivalent ~6,700IU / ~20,000IU) in the same ovariectomized rodent model of menopause (Behr. 2012 Apr), the latest results from the laboratories of the Center of Oxidative Stress Research, at the Federal University of Rio Grande do Sul in Rio Grande do Sul, Brazil, only contribute to the emerging image of the hitherto hardly understood "Dr. Jekyll and Mr. Hide nature" of the (imho) most underrated vitamin there is (sorry, for the rant, but I won't get tired of raising the awareness that retinoic acid is, contrary to its overrated cousin, "vitamin D", a "real vitamin", in the sense that it is a substance we must necessarily get from our diet, while "vitamin D" is nothing but a cholesterol metabolite we should actually be able to produce ourselves, if we just got enough dietary cholesterol and sun exposure).
Figure 1: The profound loss of the rhythmicity of clock gene expression (BMAL1, PER1, top)  subsequent to three months of a virtually retinol free diet could not be restored after only 15 days on the regular rodent chow (same as control). These changes coincide with a similar loss of / shift in the expression of the antioxidant enzymatic cascade (shown here is the GPx activity) and subsequent increases shifts (deficiency) and increases in malondeyaldehyde expression (vitamin A refed group; bottom right - based on Fonzo. 2009
A closer analysis of the expression of selected markers of antioxidant activity and oxidative damage in the brain of vitamin A deficient rodents (3 months on a virtually retinol free diet) and vitamin A replete animals, who were fed the control chow for only 15 days after the depletion phase appears to confirm some of these results (Fonzo. 2009):
As expected, temporal patterns of CAT and GPx activities observed in the rat hippocampus were consistent with the rhythm of lipoperoxidation. While the lowest CAT activity occurs during the light period and, at least in part, brings lipid peroxidation into the maximal level, highest CAT and GPx activities, practically concur with the nocturnal peak of lipoperoxidation. Thus, antioxidant enzymes would have a complementary and proper timing for protecting hippocampus against peroxides, maintaining lipoperoxidation at controlled fluctuating levels, with the lowest MDA concentration occurring during the diurnal, anabolic, period in rats [...] the location of enzymes activity peaks during the night-feeding-period, may suggest the influence of feeding cycle, and macro or micronutrients, such as proteins, carbohydrates, aspartate, glutamate or some vitamins, on those rhythms, [...] the nocturnal peaks of CAT and GPx antioxidant activity seen in the hippocampus of our control rats would be in phase with the best time for performing learning and memory tests."
In this context it is interesting to see that the peak of CAT and GPX (in figure 1, only GPx is shown) does still coincide with the nightly (remember, rats eat during the dark period!) drop in GPX activity. The daily (=sleep / low activity phase) steady decline of which Fonzo et al. state that in coincides with the variation in the expression and activity of the BMAL1:CLOCK and the PER1 protein activity with
  • a peak in GPx and Cat activity following the the BMAL1 protein peak at the end-of-the-night/beginning-of-the-day in the control rats, and 
  • a trough of the Cat and GPx experssion after the negative regulator, PER1 protein peaks at the end of the activity phase during the day,
on the other hand, is profoundly disturbed in the vitamin A deficient animals that present with a complete loss of the BMAL1 and PER1 rhythm (figure 1, top). It does therefore appear obvious that we are (once more) dealing with two controlling mechanism:
  1. an "externally" modulated, food (in the widest sense) induced regulatory mechanism and 
  2. a fundamental, time- or rather light-dependent, centrally mediated circadian rhythm 
And while the latter of the two can be partly restored by vitamin A repletion. The 15-day repletion phase in the study at hand was obviously not long enough for the GPx and lipid peroxidation levels (as measured in malondyaldehyde TBARs) to return to their pre-intervention levels. If this is, as the scientists argue a result of transcriptional changes in the vitamin A receptor (RXR) "sensitivity", it is however likely that both the GPx peak activity (which should increase) and the closely related formation of lipid oxidation byproducts (MDA) should return to baseline, as soon as the stores are fully replete and the RXR levels have recovered.

From vitamins to ergogenics, from chronic to acute, from retinol to caffeine

Contrary to the effects of vitamin A which can be stored and released whenever our bodies deem it necessary, the impact of caffeine on the circadian rhythm is by the very nature of its metabolism acute and relatively short lived. This is at least true as long as the caffeine-induced circadian shifts do not lead to permanent deteriorations of the circadian rhythm. Intuitively, we all believe that caffeine can effect the circadian rhythm (or what our mainstream understanding is telling us, the circadian rhythm would be). It's not by chance that millions (ab?)use coffee and caffeine beverages on a regular basis to get going in the morning or keep going in the evening - times when our natural, undisturbed circadian rhythm should be telling us that our bed is the place our body would prefer to be, now.

One of the more exercise specific studies on this matter comes from the Exercise Physiology Laboratory at the University of Castilla-La Mancha in Toledo, Spain, where Mora-Rodríguez and his colleagues investigated the effects of a standardized caffeine containing (6mg/kg) or caffeine-free breakfast (ingested at 9:15AM) on early morning (10:00AM) or late afternoon (18:00PM) workout performance.
Figure 2: Hormone levels, performance and catecholamine levels on AM during AM and PM training sessions with or without caffeine containing breakfast (red = AM breakfast contained 3mg/kg caffeine); * indicates significant difference to AM (Placebo), PM trials were always performed on separate days with regular breakfast (based on Mora-Rodríguez. 2012)
As the data in figure 2 goes to show the whopping dose of 225mg of caffeine (note: in the graphical summary the scientists write 6mg/kg, if this is correct and the 3mg/kg that are repeatedly being mentioned in the text, then the dosage would have been 450mg) the twelve highly resistance trained men (75kg body weight; age 20; body fat 11%) did compensate for the "early morning weakness" of the participants and increased their bench press and squat performance as well as their isokinetic leg extensor strength (not shown in figure 2) to late afternoon levels, without inducing statistically significant changes in any of the measured hormonal parameters (growth hormone, testosterone, cortisol) compared to the placebo trial.

Short-term stimulation is not (yet?) equivalent to changes in circadian rhythmicity

Hack your training, not your rhythm? If the chronic use of caffeine and other stims to increase your performance at times of the day, where your circadian rhythm does not allow for maximal performance, entails possible negative downstream effects on the regular expression of your clock-genes, why don't you just train by the clock, then? Basically this is also what Hayes et al. suggested in their 2010 paper in Chronobiology International, where they state that despite the higher testosterone levels in the morning "an increased resistance exercise-induced T response [...] in the late afternoon [would suggest a] greater responsiveness of the hypothalamo-pituitary-testicular axis" later in the day - that this is bullshit, is something you should be aware by now, as the increased expression of testosterone has, as Hayes et al. have to coincide little to no influence on the hypertrophy response to training. Rather than that, they do therefore suggest to obey to the "individual responsiveness" and train whenever you feel you perform best (without the use of stims).
At times, when this is not possible, the use of stims (esp. caffeine, which is still among the "less damaging" stimulants on the OTC market), can provide temporary relief - as soon as even  3 cups of coffee only make you sleepy it is more than high time to take a break from caffeine and high intensity training (see "Tapering & Detraining - When and How to Take a Break")
For Mora-Rodríguez et al. these observations are a clear-cut sign of "circadian rhythm effects", but are they really related to changes in circadian rhythmicity? They blunt the morning reduction in muscle performance due to circadian rhythm - there is no debating that, but the study does not provide convincing evidence that this is due to changes in the expression of zeitgeber proteins and thus a direct consequence of a shift in circadian rhythmicity. If we take another look at figure 2, we would thus expect to see similar hormonal expressions, as well. After all, both the spike in cortisol in the morning as well as the steady decline of testosterone and even steeper decline in cortisol that occurs in the course of the day are both mediated by the circadian rhythm. The adrenaline spike in response to the ingestion of caffeine, which is unquestionably responsible for the observed performance enhancing effects in the study at hand, on the other hand, has nothing to do with circadian rhythmicity.
Did I mention that results from in-vitro studies suggest that cortisol spikes, esp. the huge spike in the morning, could act as a "reset switch" for the circadian clock? (cf. Balsalobre. 2000)
If anything, we could - based on the acute catecholamine response in the Mora-Rodíguez study, that chronic morning caffeine consumption could lead to subsequent downstream changes in the expression of zeitgeber genes, which would in turn trigger a 12h shift in circadian rhythmicity with low morning and high evening cortisol levels that would basically reverse the natural pattern as it was observed in the AM/PM(Placebo) trials.That this would entail a whole host of negative health effects is something you should by now be familiar and renders the (long-term) use of caffeine to "avoid the morning reduction in muscle performance due to circadian rhythm" at least highly questionable, as it would go- in the most fundamental sense of the word - against our nature. If chronic caffeine consumption did actually induce the aforementioned changes in circadian rhythmicity. So, the next question would be...

Are the effects of caffeine even of circadian origin / does it affect circadian rhythms?

The answer to this question is not exactly easy to find, as most studies follow the flawed assumption that "being more awake" would equal "being able to hack the circadian rhythm", when it could just as well be nothing more (and nothing less) than a highly effective way to outwit the latter. Against that background it's strange that Oike et al. were the only scientists I found that explicitly mention that it "remains unknown" "whether or not [caffeine] affects mammalian circadian clocks remains unknown" (Oike. 2011).

Figure 3: The in-vitro exposure of human osteosarcoma cells (a common model used in gene essays) messes with the previously mentioned clock genes Per2 and  Bmal1 genes (left) and the in vivo ingestion of coffee / administration of caffeine in drinking water did increase the locomotor activity period length of mice after normal lighting conditions (first two weeks lower panel) and constant darkness (upper panel, right; based on Oike. 2011).
Luckily Oike at el. did not just nag at the absence of reliable evidence for / against the effects of caffeine on circadian rhythmicity, but also conducted a couple of in vitro and in vivo studies, in the course of which they were able to show that notwithstanding it's disturbing effects in on clock gene expression in the petri dish (figure 3, left), the "real-world" test with coffee and caffeine did
  • lengthen the circadian rhythm of reporter gene expression in liver explants of the rodents, without affecting the time of the rhythm peak in the liver explants (not shown), while
  • left the period length in the likewise explanted suprachiasmatic nuclei unchanged, but delayed the peak time of the rhythm
the real world results of these somewhat schizophrenic modulatory effect of caffeine on the peripheral (liver) and central (suprachiasmatic nucleus) rhythm is an increased length of the circadian pattern in dark-exposed (=constant day for mice!) mice, without affecting the locomotor activity in the presence of appropriate light cues!
In other words: The effects of caffeine will only mess with your circadian rhythm if they are not overridden by appropriate light cues!
Similar results have been reported by Sherman et al. who made an even more complex experiment which the results of which will be part of the next installment of this series, as the inclusion of a restricted feeding regimen a la intermittent fasting with a minimalist 3h feeding window segues quite nicely into the discussion of the metabolic implications of caffeine and nutrient (esp. glucose) availability, we will take up in the next installment of the Circadian Rhythmicity Series.

Image 2: I admit that all this is not easy to understand and many of the implications on our everyday lives are yet not clear, either. I still hope you don't feel you have wasted your valuable time with this post.
Before I let you go, I do yet still want to give you the elevator pitch on this long and allegedly very complicated post. While much of what we have been studying today must still be considered preliminary (also on the expert level) there are three important and theoretically, as well as experimentally relatively well certain take home messages. The first pertains to the importance of light cues as the main regulators of the central clock gene expression in the brain, the second relates to the vital, hence "vitamin", importance of vitamin A for the integrity of the central clock, and the third relates to the modulatory effect certain molecules, such as caffeine, can have on the peripheral clocks.

The practical implications of these insights, on the other hand are pretty straight forward and for most of you probably no real news, anyway:
  1. stick to the "natural" dark/light cycle - reread episodes one and two of the series for tips on how you can make do so in our "light polluted" world
  2. get adequate amounts of vitamin A in your diet - there is no need to supplement, your body manges the levels of vitamin A very effectively, so that a piece of liver once in a while is a way better choice than a vitamin pill every day
  3. don't be scared of coffee - as long as you still stick to the natural cycle (see first point), your circadian rhythm may exhibit slight shift, it will yet only break if you use caffeine + light as in popping a caffeine pill and surfing on the Internet with your melatonin suppressing iPad (see episode I) in the middle of the night
Now, before you switch off your iPad and go to bed today, I suggest you check out the SuppVersity Facebook Wall, for the latest news - it is no coincidence that an item about the -57% reduced Parkinson's risk in habitual coffee drinkers who consume at least three or more cups per day, as well as a reference to the latest confirmation of the liver protecting effects of coffee there... ah, and by the way, it could be that we will be able to track those back to circadian gene expression (peripherally, obviously ;-) in the next installment, as well...

References:
  • Balsalobre A, Brown SA, Marcacci L, Tronche F, Kellendonk C, Reichardt HM, et al. Resetting of circadian time in peripheral tissues by glucocorticoid signaling. Science 2000;289:2344–7
  • Behr GA, Schnorr CE, Moreira JC. Increased blood oxidative stress in experimental menopause rat model: the effects of vitamin A low-dose supplementation upon antioxidant status in bilateral ovariectomized rats. Fundam Clin Pharmacol. 2012 Apr;26(2):235-49.
  • Behr GA, Schnorr CE, Simões-Pires A, da Motta LL, Frey BN, Moreira JC. Increased cerebral oxidative damage and decreased antioxidant defenses in ovariectomized and sham-operated rats supplemented with vitamin A. Cell Biol Toxicol. 2012 Jul 18.  
  • Buzio C, Mutti A, Capani F, Andrulli S, Perazzoli F, Alinovi R, Negro A, Rustichelli R. Circadian rhythm of proteinuria: effects of an evening meat meal. Nephrol Dial Transplant. 1989;4(4):266-70.
  • Fonzo LS, Golini RS, Delgado SM, Ponce IT, Bonomi MR, Rezza IG, Gimenez MS, Anzulovich AC. Temporal patterns of lipoperoxidation and antioxidant enzymes are modified in the hippocampus of vitamin A-deficient rats. Hippocampus. 2009 Sep;19(9):869-80.
  • Golini RS, Delgado SM, Navigatore Fonzo LS, Ponce IT, Lacoste MG, Anzulovich AC. Daily patterns of clock and cognition-related factors are modified in the hippocampus of vitamin A-deficient rats. Hippocampus. 2012 Aug;22(8):1720-32. 
  • Hammouda O, Chtourou H, Chahed H, Ferchichi S, Chaouachi A, Kallel C, Miled A, Chamari K, Souissi N. High Intensity Exercise Affects Diurnal Variation of Some Biological Markers in Trained Subjects. Int J Sports Med. 2012 Jul 12.
  • Mora-Rodríguez R, García Pallarés J, López-Samanes Á, Ortega JF, Fernández-Elías VE. Caffeine ingestion reverses the circadian rhythm effects on neuromuscular performance in highly resistance-trained men. PLoS One. 2012;7(4):e33807. Epub 2012 Apr 4. 
  • Navigatore-Fonzo LS, Golini RL, Ponce IT, Delgado SM, Plateo-Pignatari MG, Gimenez MS, Anzulovich AC. Retinoic acid receptors move in time with the clock in the hippocampus. Effect of a vitamin-A-deficient diet. J Nutr Biochem. 2012 Aug 16.
  • Oike H, Kobori M, Suzuki T, Ishida N. Caffeine lengthens circadian rhythms in mice. Biochem Biophys Res Commun. 2011 Jul 8;410(3):654-8. Epub 2011 Jun 13.
  • Shirai H, Oishi K, Ishida N. Circadian expression of clock genes is maintained in the liver of Vitamin A-deficient mice. Neurosci Lett. 2006 May 1;398(1-2):69-72.
  • Tafti M, Ghyselinck NB. Functional implication of the vitamin A signaling pathway in the brain. Arch Neurol. 2007 Dec;64(12):1706-11.