Please note, this is recovered content from the former website of the New York Obesity Research Center website.


A discussion of research showing social stress experiments results in obesity


Date: October 6th, 2005
Title: “Effects of social stress on energy compensation and body composition”
Speaker: Randall Sakai, Ph.D., University of Cincinnati Medical Center, Dept of Psychiatry

The October Appetitive Seminar of the 2005-2006 featured accomplished neuroscientist, Dr. Randall Sakai, speaking on the “Effects of social stress on energy compensation and body composition.”

worried-girl-413690_1280In this seminar, Dr. Sakai discussed the effects of chronic stress in rats housed in a visible burrow system (VBS). When males are housed in a VBS in mixed-set groups, they form a dominance hierarchy in which subordinate animals exhibit a series of stress-induced behavioral, endocrine, and physiological changes. The nature of these changes will be discussed in the following summary.

The VBS is a laboratory habitat that provides a series of burrows and chambers, along with a larger open area for socialization of male and female rats. Food and water are also provided, so it is unnecessary for animals to leave the environment. In addition, the chamber is on a light dark cycle, and video recorders are set up to record animal activity. The concept for this system was developed by the Blanchards.

When male rats outnumber female rats, typically in a 4:2 ratio, a dominance hierarchy will form, wherein some male animals become dominant aggressors, while others become subordinates. If this were a natural habitat in the wild, subordinate males would eventually leave the colony after a short time. However, in the VBS, subordinates are unable to leave, so they exhibit a state of chronic stress that results in a series of behavioral and hormonal changes. Within this laboratory environment, Dr. Sakai’s laboratory has investigated the effects of stress on the animals’ behavior, physiology, and body composition.

What are the key behavioral changes that differentiate a subordinate animal from a dominant animal? Subordinates will show increased weight loss and will typically have defensive wounds on their back. They also spend less time in the open area of the chamber. Subordinates have decreased sexual behavior, fewer social contacts, and a heightened sense of risk assessment. In contrast, dominants spend more time in the open area, they have wounds only on their face and head, and they exhibit copulatory priority.

Chronic stress sets off a cascade of events in the subordinate animal that ultimately alter the animal’s body composition and metabolism. In the subordinate, these include an increase in basal glucocorticoids, and a decrease in plasma aldosterone, testosterone, and luteinizing hormone. Body composition changes include a loss in overall body weight and lean body mass, compared to dominant animals.

Follow up studies determined that the changes in body weight are likely due both to a decrease in food intake and an increase in metabolic rate. Studies have shown that subordinate animals eat less when confronted by dominant animals, and their periods of eating at food chambers are shorter than those of dominants.

Some subordinates have been discovered that do not show the expected increase in glucocorticoids. These animals have been termed non-responding subordinates (NRS). In situ hybridization histochemisty was used to quantify corticotrophin releasing factor (CRF) in these NRS, and it was found that nonresponsive subordinates expressed lower average number of CRF mRNA per cell in the PVN, compared to stress-responsive, dominants, or control animals.

Thus, changes in body composition and lean mass in subordinates may in part be a response to these alterations in the quantity of CRF.


Q. How did you settle on the number of rats you used in your visible burrow system (VBS)?
A. We used the standard numbers routinely used by the Blanchards. I’ll also show you the results if we change the number of males and females.

Q. Do you keep the temperature in the VBS constant?
A. Yes

Q. Have you tried giving a sugar solution to see if subordinates drink more?
A. Yes, and I will show that data later.

Q. When do cortisol levels increase in the animals?
A. Before the lights go out.

Q. Are these isolated animals?
A. They are housed with females. But even in isolation, we would expect cortisol levels to be the same.

Q. Are these measurements made in the basal situation?
A. Yes.

Q. Is the non-responding subordinate situation heritable?
A. Yes, possibly, but the profile is not specific to the restraint situation.

Q. Could it be that the dominant animal isn’t actually that dominant? What happens then?
A. If the dominant animal isn’t that dominant, you never get a non-responding subordinate.

Q. If you just use weight matched controls, how do glucocorticoids respond?
A. They increase, but not as much as they do in subordinates.

Q. Is this sexual hierarchy the only model in rodents where you get a dominant and a subordinate?
A. This has also been tried in mice, but it does not work that well because they are not social animals.

Q. Do females form a hierarchy? We’ve seen this with monkeys.
A. We don’t see this. They either get pregnant or pseudopregnant, or they are cycling. We do not get dominants and subordinates in female rats.

Q. If you took the female out of the situation after 2 weeks, would the subordinate males revert back to normal?
A. Not completely.

Q. Are there any biochemical or physiological markers to predict dominance and subordinance?
A. (Did not get response)

Q. What happens if you administer testosterone to females? Will they form a hierarchy?
A. We have not done that yet, but when we clamp male rats at the same testosterone levels, they still form a hierarchy.

Q. Is the dominant animal more active?
A. We don’t see much of a difference in activity.

Q. Do you measure meal patterns of the animals?
A. We have not yet, but we are trying to work on this at the moment.

Q. What happens when dominant and subordinate animals want to eat at the same time?
A. The subordinate animal will leave.

Q. What happens if you block all the chambers?
A. The animals get along better. No dominants and subordinates are formed.

Q. Do you have any idea what the mechanism for increased weight loss in subordinate animals (compared to dominant) might be?
A. I have no idea right now.

Q. Do you have MRI data on your animals?
A. No

Q. Did you take any measures of ulcer formations?
A. We do not see any. Animals need to be food deprived to see this.

Q. Did you ever look at the other animal strains, OB, DB, or Zucker, or others?
A. Not yet, but we want to.

Q. What if you go out of the laboratory into a field situation. To what degree, in terms of survivability, does the VBS laboratory situation represent a stressful situation and not normal everyday living?
A. We have artificially set this at 3 weeks. In a real world situation, a subordinate would eventually leave the colony after about 2 weeks of getting beat up by the other animals.

Q. Did you measure eating rate on your animals?
A. Not yet, but we will calculate that in the future.

Q. In the experiments where the subordinates drank more alcohol, where did you put the alcohol?
A. All over in the VBS.

Q. Does this stress model have any analogy in humans?
A. The subordinates respond differently to amphetamines and often develop cross-tolerance to other substances. Addiction in humans might be a comparable model.

Q. Are there data on serotonin transport genes in rodents?
A. I don’t know off-hand.

Q. What about anti-anxiety agents? Do they have an effect on the formation of the hierarchy?
A. We have not done those experiments yet.

Q. Do you consider this to be a short-term stress response?
A. I think this is a nonhabituation stress response. Cortisol levels in the animals remain high, even after they are taken out of the situation (VBS).