Why are obese mice so easy to chase?

Dysfunctional signaling in the brain makes obese mice less active


Obesity is accompanied by a lack of motivation/desire to exercise. This has lead to the idea that lack of exercise leads to obesity. A new study challenges this by showing that both “lazy” and “active” mice gain weight on a fatty diet [1]. All mice on high fat diet become obese and then move around less than mice fed on standard chow. The researchers go on to show that the lack of motivation to exercise that accompanies obesity may well be brought about by neuronal changes in the regions of the mouse brain that respond to movement.


The physical inactivity that accompanies obesity is frustrating for those wanting manage their own weight as well as those who want support a near and dear one who does [2,3]. A better understanding of where this seeming ‘lack of motivation’ to exercise comes from may help design better intervention strategies. Previous studies suggest that obese animals and humans may have defects in dopamine signaling in a region of the brain that controls movement behaviours with the result that they may find physical activity less rewarding [4,5]. However, does lack of exercise cause weight gain?

What did they do and find?

Mice were fed standard chow (lean) or high fat diet (obese) for 18 weeks. Mice become both obese and less active when fed a high fat diet. The researchers of this study wanted to understand if the mice became fat because they were less active. To their surprise they found that low activity and weight gain occurred hand in hand but were not cause and effect. The weight gain however was correlated to the high fat diet.

So what was causing the lower activity in obese mice? There is a bit of the brain called the striatum which is responsible for movement and is disrupted in disorders such as Parkinson’s. There are neurons in this region of the brain that are sensitive to the neurotransmitter dopamine and fire (get activated) during movement. The authors of this study reasoned that perhaps it is this region of the brain that is responsible for inactivity in obese mice.

First they looked for components of dopamine signaling, the levels of dopamine itself and the dopamine receptors which when present on neurons allows them to respond to dopamine. They found that the striatum of obese mice had Dopamine Receptors of a specific kind (D2R receptors) which showed decreased binding while the levels dopamine itself and the other receptor for dopamine was the same in lean and obese mice. This reduction in dopamine 2 Receptor binding did not correlate with weight gain but was correlated with movement loss.

So would lean mice also move less if they had lower binding D2Rs?

Indeed, in genetically modified mice that lacked D2R receptors in the striatal region – lower activity levels were observed even in lean mice. This showed that neuronal changes underlie lower activity levels in obese mice.

To probe this further the researchers measured the activity of neurons in the striatum by inserting an electrode in the brain of live obese and lean mice. These recordings showed that during movement there was less overall firing in the brain of obese mice.

In order to test if these brain regions and neurons were indeed responsible for the lower activity observed in obese mice, the researchers used a special set of mice. These mice are specially modified to express a molecule that is usually produced by active Dopamine signaling via D2R binding (Gi) coupled to an opiod receptor only in the neurons of the striata that naturally express D2R. This allows Gi to be uniquely switched on by use of a synthetic chemical (Salvinorin B). When Gi is artificially produced by the D2R expressing neurons of the striatum both lean mice and obese mice become more active.

Reducing the D2R levels artificially in the neurons of the striatum results in mice with lower activity levels however these mice were not more susceptible to weight gain. Nor are mice with low D2R binding in the beginning of the diet predisposed to weight gain.

Take homes from the study

Experiments on animal behaviour are difficult and sometimes hard to extend beyond specific cases because genetic and environmental effects play a large role in shaping observed behaviour and this study is no different. These data convincingly argue that in mice, obesity is accompanied by and not caused by lack of activity. It also gives us a perspective on how integrated an animal’s body and mind are. At the very least it makes us think that in combating obesity, a role for the mind cannot be ignored.


1. Basal Ganglia Dysfunction Contributes to Physical Inactivity in Obesity. Danielle M. Friend, Kavya Devarakonda, Timothy J. O’Neal, Miguel Skirzewski, Ioannis Papazoglou, Alanna R. Kaplan, Jeih-San Liow, Juen Guo, Sushil G. Rane, Marcelo Rubinstein, Veronica A. Alvarez, Kevin D. Hall, Alexxai V. Kravitz, Cell Metab. 2017 Feb 7

2. The mysterious case of the public health guideline that is (almost) entirely ignored: call for a research agenda on the causes of the extreme avoidance of physical activity in obesity. Ekkekakis P, Vazou S, Bixby WR, Georgiadis E, Obes Rev. 2016 Apr;17(4):313-29

3. Exercise does not feel the same when you are overweight: the impact of self-selected and imposed intensity on affect and exertion, P Ekkekakis and E Lind, International Journal of Obesity (2006) 30, 652–660

4. Reward mechanisms in obesity: new insights and future directions. Kenny PJ. Neuron. 2011 Feb 24;69(4):664-79.

5. Obesity and addiction: neurobiological overlaps (Is food addictive). Volkow ND, Wang GJ, Tomasi D, Baler RD. Obes Rev. 2013 Jan;14(1):2-18.

6. Do Dopaminergic Impairments Underlie Physical Inactivity in People with Obesity? Kravitz AV, O’Neal TJ, Friend DM, Front Hum Neurosci. 2016 Oct 14;10:514. eCollection 2016.

7. Increases in Physical Activity Result in Diminishing Increments in Daily Energy Expenditure in Mice. Timothy J. O’Neal,, Danielle M. Friend, Juen Guo, Kevin D. Hall, Alexxai V. Kravitz Curr Biol. 2017 Feb 6;27(3):423-430.

An Interview with Dr. Alexxai V. Kravitz

1. What is causing the change in dopamine signaling in the neurons responsive to movement in obese mice? Do you have more insights into this from your study of Parkinson’s?

This is a great question, but unfortunately one that we don’t know the answer to. Parkinson’s disease is caused by the death of neurons that make dopamine, and we looked at dopamine neurons in obese mice and learned that they were not dying. So in that way, the mechanism underlying the changes in dopamine signaling in obese mice is very different than with Parkinson’s disease. This is a good thing, as it would frankly be scary if a diet high in fat were causing the death of dopamine neurons! Instead, we observed dysfunction in a specific dopamine receptor (a protein that detects dopamine) in obese mice. We’re looking into what exactly is causing the dysfunction of this receptor, but unfortunately we do not currently know.

2. You data does show that mice become both obese and move less on high fat diet, but which bit convinces you that the “laziness” is because of the obesity? Can they not be two parallel outcomes of a high fat diet? If yes, then would a high fructose or high calorie diet lead to a similar outcome?

Let me clarify here – I don’t think the *weight* of the mice is causing the laziness, I believe dysfunction in their dopamine receptors is causing their laziness [More on this in Ref. 6]. And both this dysfunction and weight gain can be caused by the high fat diet. So in that say, yes, they can be two parallel outcomes of the high fat diet. To answer your second question, I’m not sure if other high calorie diets can cause the same dysfunction. This would be a great follow up experiment!

3. In your paper, you describe the limitations of human studies that have measured Dopamine signalling and its links to obesity. Can you tell us a bit more about what the challenges are?

To date there have been a handful of studies that have compared D2 receptor levels in people with obesity vs. normal weight, and a minority have reported dysfunction in D2Rs in people with obesity. It is not clear why some studies have reported lower levels of D2 receptors, while most have not. However, measuring dopamine receptor levels in humans is difficult. The only technique for measuring receptor levels in humans is PET scanning, a technique where a radioactive tracer is injected and the brain is scanned for the location where the tracer binds. If more tracer binds, it is assumed there are more “available” receptors in that brain area. However, this technique can be affected by many factors, including what other transmitters are bound to that receptor. If internal levels of dopamine are higher during the scan, for instance, the amount of a radio-tracer that binds to a dopamine D2 receptor will be lower. The complexity increases when we consider how many things can alter dopamine levels throughout the day, which include caffeine use, food intake, and sleep. These are some of the challenges that face clinical research. Animal studies are less likely to incur these sources of variance, and have more consistently reported decreases in D2 receptors in association with obesity.

4. Are the changes in the striatum reversible, by forced exercise for example or are there natural molecules that could restore Gi signaling?

There are no known ways to reverse these changes, but there is also very little research on this. There is a small amount of evidence in rats that forced exercise increases D2 receptor levels, but this is very preliminary and has not been replicated, nor studied in humans. This idea of how to alter D2 receptor levels is an extremely important concept for future research!

5. Are there common themes about obesity and lower activity levels that have emerged from animal studies and how would you extend them, if at al, to humans? For instance, you say mice and rats are different, then would you expect people to be more similar to mice than rats? Why?

It is very difficult to extend results from mice to humans, so I will be cautious on this one. However, there are some concepts from animal work that are relevant to humans. Many researchers have noted that animals voluntarily over-eat high fat diets, and that this leads to weight gain and obesity. While the specific macronutrient (fat vs. carbs vs. protein) content of human diets is the subject of a lot of debate when it comes to human obesity, it is fair to say that diets that induce over-eating will lead to obesity. Typically, foods that induce over-eating are highly palatable, such as junk foods that pack large numbers of calories into small volumes. While people are all different from one another, understanding the foods that a specific person overeats will inform what is likely to cause that person to gain weight.

As another concept that I believe is relevant to human s, in our study we reported that physical inactivity did not correlate with weight gain in mice. That is, we examined inactive mice that lacked D2 receptors, and found that they gained weight at the same rate as normal active mice. We also examined the natural variation of activity levels of normal mice and did not note any relationship here either. This seems to counter the conventional wisdom that inactivity should cause weight gain. However, this conventional wisdom is based largely on correlations between obesity and inactivity, rather than causal tests of this hypothesis. We all know that correlation does not imply causation, but it is very easy to get caught in this trap. In fact, in causal tests, the contribution of exercise alone (without changes in diet) to weight loss in humans is fairly small, generally resulting in 3-5 pounds of weight loss over the first year. This is consistent with our conclusions in mice. Studies in mice can help us understand at a mechanistic level why changes in activity (both increases and decreases) don’t translate into large changes in body weight [More on this in Ref.7].

6. In their natural habitats animals such as mice and rat consume high fat diets. Do you think your results would hold in wild rodents instead of lab reared ones, especially if they were allowed to interact freely with each other and the environment?

Wow, what a great question! We use lab mice, which have been bred in captivity for many decades. This is somewhat similar to studying domesticated dogs vs. wild dogs. And in many ways, our laboratory mice are quite different from wild mice. However, I believe that even wild mice would become inactive on a high fat diet. The association between obesity and inactivity has been seen in many species including humans adults, children, non-human primates, domesticated cats and dogs, rats, and mice. When an association occurs across so many species of animals, I think it is likely that it would extend to wild mice as well as laboratory mice. This would be a great student project to find some wild mice and test!

Emulsifiers bring gut bacteria too close for comfort

Feeding mice with artifical emulsifiers impacts their metabolism

Snap-shot of the study

Emulsifiers are used extensively in the food we eat (ice creams, biscuits etc.). This study examines the effect of feeding mice emulsifiers both in their food and drink (5).

Mice fed on emulsifiers (Carboxymethylycellulose CMC and Polysorbate P-80) showed increased appetite. They also showed signs of low-grade inflammation in the gut and increased fat deposition. The authors attribute these effects to changes in the gut microbiota. While the number of microorganisms in the gut was not altered by diet containing emulsifiers, the kinds of microbiota were completely different. The mucus lining was also depleted and the microbes were closer to the cells in the gut, possibly causing the inflammation. While this study has been done on mice, perhaps the quality and quantity of our microbiota and their response to emulsifiers has some bearing for us too.

What did they do?


They added widely used  emulsifiers- Carboxymethyl cellulose (E466) and Polysorbate-80 (E433) to the food and drinking water of young mice at equivalent concentrations commonly used in human food.

They measured the abundance and diversity of the gut microbiota, inflammation of the gut (colitis) and also metabolic disorders (fat accumulation, increase in food intake and fasting blood sugar levels) in the emulsifier-fed mice and compared to control mice (no emulsifier in food or drink)


What did they find?

These mice (treated) had same number of bacteria (in their gut) compared to mice that were not fed emulsifier (control mice). The types of microbes however was quite different. The microbes were also found closer to the gut that in control mice. The treated mice showed increased appetite, followed by  fat deposition and low grade inflammation of the gut. Interestingly, transplantation of the microorganisms of the gut from emulsifier fed animals into germ-free mice also resulted in increased fat deposition and inflammation of the gut. Suggesting that changes in the microbiota caused by the emulsifiers may be sufficient to cause the metabolic dysfunction and observed inflammation.

Erosion of the protective mucosal layer around gut epithelium in emulsifier-fed mice resulting in reducing the separation between the microbiota and the gut epithelium. Emulsifiers caused a marked change in gut microbiota composition – Higher pro-inflammatory microbiota including the bacterial species that are the leading cause of colitis like Bilophila and Helicobacter. Changed gut microbiota in emulsifier-fed mice increased gut inflammation and colitis. Emulsifier-fed mice also show – dysregulation of blood sugar levels (mild diabetes), increased food consumption correlated with increased adiposity(fat deposition) and weight gain. In older mice (4 months old) the changes persisted for more that 6 weeks even after emulsifiers were stopped. The observed effects of emulsifiers are exclusively due to the change in gut microbiota as the emulsifiers did not show any effect in mice having no gut microbiota (germ free mice). Interestingly such germ free mice become labile to the effects of the emulsifiers if the regular gut microbiome is reintroduced in them.


Background to the study

An undisturbed gut flora is emerging as an important factor in health versus disease (1). Multiple different physiological conditions including obesity and type 2 Diabetes are now associated with changes in the gut microflora (2-3). Recent studies have found that artificial sweeteners can cause blood sugar related disorders in humans (4).


Take-home and implications

This necessitates a reevaluation of what goes into our food, how it affects our gut microbiota and our health. Standard food safety tests include toxicity and carcinogenicity (ability to cause cancer), however, the importance not perturbing the natural flora of the intestine is becoming clear only now. These findings suggest the following in mice- intake of food/drink containing emulsifiers leads to weight gain and disorders such as diabetes, by directly increasing food intake. These findings need to be verified in humans.The intriguing  realization that dawns on someone after looking at this study is that not just the quantity, but also the quality of the microorganisms in the mouse gut matters. In humans the importance of gut microbial diversity has been documented in other contexts (1-3, video below – courtesy MinuteEarth)

Limitations and Open Questions

Only 2 synthetic emulsifiers have been tested. We feel that this work makes a strong argument for the development of assay systems that monitor microbial health (especially gut microbes) for compounds added to food, medicines etc. Given that the findings have such strong implications, we hope to see in the future a wider spectrum of compounds (including  the more natural products like lecithin) examined similarly by the authors and others.

The authors have only discussed in brief the possible mechanism underlying the change in the microbial population or how these changes result in increased inflammation. This remains a major open question.

Germ free mice already have a really bad situation in their gut, they are somewhat prone to inflammation. It is important to bear this in mind while interpreting the results of the fecal transplantation into germ free mice.

The study is a mouse study, it remains to be extended to humans.

An interview with Dr. Andrew Gewirtz

Q. From your work, it is clear that altered microbiota could lead to weight gain, fat deposition and the loss of the ability to control blood sugar levels, can this be reversed by altering the microbiota?

Our studies in mice indicate it is reversible but it takes some time.

Q. How do you think the emulsifiers are changing the gut microbiota? Can you elaborate on some potential mechanisms?

They seem to promote bacteria breaching the mucus, which promotes inflammation, which changes bacterial populations, possibly by favoring detrimental bacteria.

Q. Are you suggesting that the normal gut flora under different conditions (presence absence of emulsifiers) could turn pro-inflammatory? Are the other missing microbiota (in the presence of emulsifiers) keeping them in check under normal circumstances?


Q. What according to you are the major caveats of your study?

It is a mouse study.

Q. Did you face challenges in publishing this work, given that it has such strong implications?

Some reviewers suggested a dialog with food industry prior to publication but we argued our tax payer funded research did not require such approval. Nature editors agreed with us.

Q. Do you plan to take this study forward in humans? What would be a suitable cohort for such study?

Yes. Probably start with healthy college students.

Q. Your work clearly has implications for how we decide what to put in our foods.. What Changes would you suggest to the current process by which such compounds are screened, approved and used?

I think major overhaul is needed. Both more tests are needed and more information made readily available to consumers.

Q. Has your study affected your life and food choices?

Yes, my family has cut our consumption of processed foods in general and emulsifiers in particular.

Related Viewing