Presiding Chair: Harry R. Kissileff, Ph.D.
Rapporteur: Kathleen L. Keller, Ph.D.
Date:December 1st, 2005

Title:"Learning to be obese: behavioral and molecular plasticity of food intake"
Speaker:Stephen Benoit, Ph.D., University of Cincinnati, Department of Psychiatry

Attendees and their Affiliation:

Kathleen Keller	Columbia/Obesity Research Center
Harry Kissileff	Columbia/Obesity Research Center
Bina Tejuda	Manhattan Pharma
Carol Maggio	NY Obesity Center
Joseph Lesauter	Barnard College
Nawshin Hogue	Barnard College
Michael Lowe	Drexel University
Eshter VanDe Wall	AECOM
Melanie Ruger	NYU Medical/ Center for Brain Health
Hannah Bruhl	NYU Medical Center
Doug Mook	NYC
Rebecca Gordon	ORC
A. Dunn-Meynell	VA E. Orange NJ
Bonan Irani	UMDNJ
Christa Patterson	UMDNJ
Cheryl Dym	Graduate Center-CUNY
Karen Acroff	Brooklyn College
Tony Sclafani	Brooklyn College
Rich Bodnar	Queens College
Khalid Touzani	Brooklyn College
Gary Schwartz	AECOM
Simon Klebanov	Obesity Research Center
Anthony Azzara	AECOM
Tim Bartness	Georgia State University
Carla Wolper	Obesity Research Center
Janet Guss	Columbia ORC
Allan Geliebter	Columbia ORC
Joe Vasselli	ORC

Summary:

Dr. Benoit's seminar addressed the intriguing inter-relationships between learning and obesity. The first part of the talk was a brief overview of research on how learning affects eating. Many studies support the notion that learning influences food preference and intake. Pre-ingestive cues that influence food intake include taste and palatability; ingestive cues include satiety and hunger; post-ingestive cues include the macronutrient consequences of a food or a meal. The specific studies that support a role for learning in each of these areas were not reviewed in detail in this talk, as the primary focus was to discuss data suggesting that obesity and food intake influence cognitive function.

The sensory cues produced by food deprivation are conditionable stimuli about which animals can learn appropriate responses, including ingestive or physiological responses. Benoit SC et al., (Peptides 2005) have shown that conditioned stimuli can blunt the effect of orexigenic neuropeptides on sucrose intake. This is because animals learn to increase their intake across trials. Other studies also suggest that conditioned meal initiation blunts the effect of homeostatic signals to reduce meal size and intake, meaning that learning could potentially block one target of the pharmaceutical industry.

The second focus of Dr. Benoit's talk provided evidence on how diet, metabolism, and obesity might affect learning and cognitive function. The primary question that was addressed was whether or not eating influences learning when studied in the laboratory. Correlational and epidemiological evidence says that eating does in fact influence learning. To illustrate, insulin insensitivity is correlated with poor memory performance and neuronal survival. Many neurodegenerative disorders also feature impaired insulin-related mechanisms, particularly in diseases such as Alzheimer's, Parkinsons, and Vascular dementia.

In the laboratory, there are many examples of how a high-fat diet can impair learning. In rats, a high-fat, high refined sugar diet reduces hippocampal brain-derived neurotrophic factor (BDNF), neuronal plasticity, and learning. In another experiment, it was hypothesized that an inherited metabolic deficit (Diet-induced obesity) will impair DIO rats' performance in cognitive tasks before the existence of diet-induced obesity. In order to test cognitive performance, a procedure called the Morris Water Maze was used. This classic test of spatial and hippocampal-dependent memory measures the time necessary for an animal to locate a hidden platform of water. A decreased latency time to escape on repeat performances is used as an index of learning. DIO rats have impaired spatial memory, relative to diet-resistant rats, as evidence by an increased latency time to escape the maze. In further testing with spontaneous alternation tasks, no difference was found between DIO and DR in their frequency of investigating novel arms of the maze before reentering arms. These experiments were controlled for differences in body weight and speed of the animals.

Another experiment tested whether DIO rats would be more susceptible to the deleterious effects of high-fat diets than DR rats. Rats in this experiment were given 5-weeks of access to a high-fat diet. Results showed that DIO rats were more susceptible to high-fat, high energy diets, as supported by a lower success rate on the spontaneous alternation task after, but not before, exposure to a high-fat diet. Does this depend on the kind of fat to which animals are exposed (saturated vs. polyunsaturated vs. simple carbohydrate)? In studies using a spatial version of the novel object recognition (NOR) test, in which the percent of time animals spend investigating a novel object is used as a index of memory, saturated but not monounsaturated fats attenuated performance. Saturated fat diets also impair alternation performance when compared to animals on chow or olive oil.

What might be the potential mechanisms that link diet to cognitive performance? Insulin resistance is a likely target, in particular, reduced brain-derived neurotrophic factor (BDNF). Data show that high-fat diets reduce expression of BDNF and other "plasticity" genes, as measured in hippocampal mRNA. Three month exposure to a high-fat diet also reduces insulin receptor mRNA, independent of obesity. In addition, expression of the insulin receptor protein was also reduced in animals pair-fed a high-fat diet, compared to those on a low-fat diet.

In further exploring the mechanism of central insulin sensitivity, Dr. Benoit reviewed recent data on protein kinase-C (PKC). At present, a total of 10 isoforms of PKC have been identified, all of which phosphorylate serine and threonine residues on many target proteins. In the September 2004 Journal of Clinical Investigation, Kim et al. found that PKC- mice were protected against fat-induced insulin resistance. Further, saturated fats increase the amount of membrane bound PKC in the hypothalamus, relative to animals fed chow. It is hypothesized that PKC might be a potential unifying mechanism that links insulin resistance to cognitive performance, as data suggest that it does not only play a role in food intake, but it also is involved in learning and conditioning.

In summary, learned behaviors and experience clearly impact ingestive behaviors. Recently, we have also become aware that eating may influence learning. Ingestion of certain foods, particular saturated fats, can impact the ability to learn and cognitive performance. Thus, it might be possible to one day think of obesity as a learning disorder, in addition to its many metabolic consequences.

Discussion:

Q. Are cues about post-ingestive state unconditioned or conditioned stimuli?
A. In that case, we believe that they are conditionable.

Q. Did you the reciprocal true for anorexic peptides? Or in other words, does sucrose intake decrease following the injection of anorexic peptides?
A. Yes, and I will show you these data next.

Q. Could it be that there is fatty acid fuel substitution in long term exposure to a high-fat diet that is resulting in a decreased insulin sensitivity in the brain?
A. Yes, I'll go back to that.

Q. Are DIOs (Diet Induced Obesity) animals slightly above body weight, compared to controls, even before they are exposed to a high-fat diet?
A. Yes

Q. Are DIOs inherently different from diet resistant strains? Have you pair fed them down to see what happens?
A. No. I'm not sure if Barry Levin has done this.

Q. Is this an effect of being obesity or being on a high fat diet (in reference to data to suggest that animals on a high fat diet have difficulties with a spontaneous alternation task?
A. We don't know this yet.

Q. Were restricted animals more active than fed animals?
A. That is a really interesting question. However, the task doesn't really depend on the level of activity of the animal.

Q. Did you do OGTs on insulin secretion?
A. Yes, and we found that olive oil impairs insulin response much less than butter.

Q. Are there clinical differences in animals maintained on olive oil? Do these effects persist after animals are taken off the diet?
A. Yes, there are, but we don't know what happens to them yet when they are taken off the diet. At this point, it does not look like the effects persist after animals are taken off the diet.

Q. Are you suggesting that animals kept on olive oil fare better than those on chow (in terms of hippocompal PKC)?
A. Yes, but the final story will be much more complicated than this. There are other PKCs and we don't yet know the effect of olive oil on them.

Q. Have you looked for the effects of ketones on cognitive function?
A. No, but other research has shown that some individuals maintained on a high fat diet do show some impairments in cognitive function.

Q. What about the essential fatty acids? What might we expect for animals maintained on these diets?
A. At this point, we can't say. We have not tested that yet.

Q. How much are you attributing the cognitive effects to the obese state, and how much to the high-fat diet?
A. That is a great question, but they are ultimately confounded.

Q. You can leave animals on a high fat diet for long enough so that when you put them back on chow, they don't lose weight? This would allow you to test whether it's the obesity state, or the diet.
A. Yes, but they are still confounded.

Q. If the reason that animals are cognitively impaired is insulin resistance, would you predict that exercising animals would show less impairment in cognition than non-exercising animals?
A. Yes, we have been thinking about conducting those experiments.

Q. In humans, what kind of cognitive impairments are found (related to high-fat diets)?
A. The data in humans are confounded, and there are a variety of tasks used to measure cognition, so it's difficult to interpret what they mean.