No More Midnight Snacks? Mice That Eat at Odd Hours Get Fat

By Sarah Zhang | May 18, 2012 8:21 am

obese mouse
FA=high-fat, ab libitum (eat-at-will) diet, FT=high-fat, time-restricted diet, NA=normal ab libitum (eat-at-will) diet, NT=normal diet, time-restricted

Diets tell you what you eat, but a new study suggests when you eat matters too. Of two groups of mice who were fed the same high-fat diet, the mice who could eat around the clock were much heavier than those who had food restricted to eight hours per day, in a new study published in Cell Metabolism. 

Researchers in the study gave the mice a special high-fat chow, 61% of whose calories come from fat (compared to just 13% in normal feed). The mice who chowed down all day and night became, unsurprisingly, obese, but the ones who ate the same amount of hi-fat food in only eight hours per day did not. Their body weight was comparable to mice fed an equivalent amount of calories on normal feed.

This being a study in Cell Metabolism, the researchers didn’t stop with just weighing the mice; they did a lot of molecular experiments to work out the link between timing and weight gain. Mice on high-fat, eat-whenever diets had the insulin problems associated with obesity-induced diabetes and lower expression of genes linked to breaking down fats in the liver, leading to fat accumulating in the liver. The high-fat, time-restricted diets did not have those problems.

This might make sense in light of our circadian rhythms, which are the approximately 24-hour cycles that govern sleep as well as metabolic functions such when the liver secretes bile and the pancreas insulin. Previous research has found that sleep-deprived and jetlagged people, whose circadian rhythms are out of whack, are at risk for weight gain. The current study adds to the link between circadian rhythms and weight, suggesting that eating fat at odd hours disrupts daily metabolic cycles.

Image courtesy of M. Hatori et al / Cell Metabolism

CATEGORIZED UNDER: Health & Medicine, Top Posts
  • Siobhan

    Must confess to having not read the paper yet. However, initial thoughts are that this would only make sense to compare mice that ate exactly the same amount of food in the 8 hours to the unlimited hours. – which they say they did.

    I wonder how they managed to get the two mouse groups to eat the same amount? – ie how do you stop them guzzling it all down in one go regardless of 8 or 24 hour availability. My first thought would be to give more food. However, if you gave loads of food then the 8 hour mice may not eat it all. Then if you give enough food for the 8 hour mice to eat so that they finished it all how do you stop the 24 hour mice from finishing all this food in 8 hours too? – if you give the food periodically then you may be subjecting the experiment to bias as it will be a little and often feeding pattern with the 24 hour mice.

    I wonder if they tested midnight snacking by providing food only during the sleep hours compared to wake hours. This would see if you get fatter from food being digested in your sleep. Although, you could then suggest that during the sleep period mice would eat less – but I’m sure after a week of not getting food during the wake hours they would soon compensate by eating regular amounts at night.

    Having written this I suspect you could do a scatter plot of amount eaten against feeding time given which would give you a trend, but you’d still be missing valuable data points (mice given 8 hours eating lots and mice given 24 hours eating very little).

  • Anu

    Great question, Siobhan, on how the researchers could get the two groups to eat the same amount. Usually this is done by pair-feeding – feeding a mouse in one group the same as a paired mouse in the other group ate the day before. But this method has the problems which you point out. If you feed the ad lib mouse what the time-restricted mouse ate, it might eat it all in 8 hrs too. If you feed the time-restricted mouse what the ad lib mouse ate, it might not finish it all. It seems that the time-restricted mice would have to be trained ahead of time to eat their food in 8 hrs – but that would add another element into the experiment. I hope they explained how they did it in the paper. Maybe someone can read it and tell us.

  • http://Discover Doug

    I was a researcher that did drug research for many years. I used mice for screening studies and did research on monkeys and humans. The humans were terminal cancer patients, so don’t be too quick to judge me on this. I did contribute significantly to establishing a protocol that stood for about 20 years that extended the life of, and sometimes “cured”, patients. The particulars are complicated, so let me cut the chase scene. Often I would stay at the lab until late at night to get some peace and quiet so I could think . The lab was in an abandoned 6-story hospital. I would get up from my desk and walk occasionally up to the animal colony on the 6th floor or take a walk through the rooms with the experiments that were on the 5th floor. We had about 100,000 mice. Ever walk into a room at Midnight that had, oh say, 50,000 mice in it? I have. The noise would surprise you. Mice chow down at night. They didn’t eat that heartily in the daytime. My guess is that this would not be too difficult a study to do. It would be easy to get the mice to consume their calories at night. It’s the daytime I can’t help but, wonder about? However, it’s easy enough to figure out how much they’re eating and if they ate the same quantity, well they should weigh the same amount, right?

    Well, I’m not so sure. You see, mice prefer to eat at night. Making them eat during the day may have made them less healthy? Could that be why there was a weight difference? If this is the case, we should consume all of our calories at night and fast during the day! Also, I have no axe to grind in this. I’m 175 pounds, I’ve been sleep deprived all my life and I have to watch what I eat just like anyone else. But, one more little factoid about mice. If you get the chance to work with them, you’ll be amazed at what they can digest. A mouse’s liver is far more capable than a human’s. In fact, the dosage of drugs (per unit of body weight) for a mouse is several times that of the dosage for the same drug for a human. That’s because the mouse metabolizes the drug far more quickly and clears the “toxin” from its body more quickly, courtesy of the mouse’s “oh-so-capable” liver. So, when you come to the part about liver function and fat metabolism, the mouse is a whole different animal.


  • Craig

    I would like to know how often a mouse sleeps in 24 hours. Is that period to them like 4 days because of their faster metabolism? (2 days of light and 2 of dark). In that case to translate it to humans would mean (approx) a person eating at will for a day then fasting for 2 and a bit. This would promote the same outcome of weight loss and better sugar level control.

    Otherwise and also I now wonder if caloric restriction gives the same health benefits as the 8/16 eating plan not because of reduced calories but because of the light burden on the liver etc.

    I’ve begun the 8/16 method on myself to see what happens in 6 months.


  • Sach

    Mice learn to eat most of their food in 8 h within 3-4 days. If you give them access to food for 8h, on the first day they will eat less, and in the subsequent 2-3 days they will ramp up to eat all their daily food intake in 8-12 h. That is the trick.

    Mice are different from human in many ways – they have higher metabolic rate per unit body weight, sleep little more and are also more active.

    i guess they did the experiment for 8 h for ease. Remember we come to work for 8 h. I dont think 8 h is a magic number – 10 or 12 might also work.

  • Amanda

    that was a great article


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!


80beats is DISCOVER's news aggregator, weaving together the choicest tidbits from the best articles covering the day's most compelling topics.

See More

Collapse bottom bar