Food choices may depend more on blood sugar than body weight

Maybe you shouldn’t always listen to your gut.

Researchers at the Fralin Biomedical Research Institute at VTC recently studied flavor-nutrient learning – how people come to prefer certain food based on how it makes them feel. Flavor-nutrient learning is one factor that influences eating habits and may impact body weight.

“We have to learn what we are going to eat, and one factor that’s less well studied is post-ingestive signals – our gut talking to our brain, teaching us what to eat,” said Alexandra DiFeliceantonio, who led the research. She is a faculty member at the institute and interim co-director of its Center for Health Behaviors Research. 

The team’s findings, recently published and slated for the Nov. 1 issue of Physiology & Behavior, suggest that measures of glycemic control – such as fasting glucose and HbA1C, which measures glucose levels over time – were more closely linked to how much participants’ food preferences changed during the study.

That stood out, because while participants represented a wide range of body-mass index categories, none were diagnosed with diabetes or prediabetes. 

We wanted to know whether the gut-to-brain system for relaying information about nutrient learning might be different for people who have obesity and for those with differences in glycemic control. If it’s different, we should be using different targeted strategies to help them change their diet.”

Alexandra DiFeliceantonio, Study Leader

What they did

Research in animal models points to the importance of signals from the gut to the brain after eating. “They’re actually necessary, beyond just oral taste signals, to guide food preference,” said Mary Elizabeth Baugh, a research scientist at the Fralin Biomedical Research Institute and the study’s first author.

Scientists have argued that flavor-nutrient learning is difficult to show in people because eating history and food preferences vary so widely, while testing conditions and diet in animal studies can be strictly controlled.

To address that challenge, 26 people from Southwest Virginia were introduced to 10 atypical flavors: acerola, bilberry, horchata, lulo, yuzu, papaya, chamomile, aloe vera, mamey, and maqui berry. 

“The best practice is to take something strange, because we want new learning to happen,” said DiFeliceantonio, who also holds an appointment in the Department of Human Nutrition, Foods, and Exercise in Virginia Tech’s College of Agriculture and Life Sciences.

Study participants were asked to rate how familiar the flavors were and how much they liked them. The research proceeded with two flavors that were less familiar and less liked by individual participants. 

Flavored drinks were first experimentally matched for sweetness with sugar and an artificial sweetener; one provided calories and the other did not. Participants consumed the drinks at home at specified times over a period of weeks. Later, artificial sweeteners were used in both, so calories during the testing session couldn’t confound results.

As expected, some participants learned to prefer the flavor that had been paired with calories, even when the sugar was removed. “And that’s because of post-ingestive mechanisms, not anything related to sweetness,” Baugh said.

But the findings weren’t uniform.

The expectation was that participants would prefer the flavor that had nutrients in the form of calories, but those with fasting glucose and A1C at the high end of normal were less likely to prefer the flavors that had been paired with nutrients. 

“One of the most interesting findings was that measures of body weight status – body-mass index, waist-to-hip ratio, and waist circumference – were not related to individual responses,” Baugh said. “We need more data, but this points to potentially impaired learning based on post-ingestive signals. With higher values of glycemic control, even within the normal range, there could potentially be some disruption in gut-brain signaling.”

What comes next

This doesn’t only affect people who meet the criteria for overweight and obesity. “Even if you are a person with a healthy range BMI and a healthy range A1C, fluctuations in your blood glucose are still actually influencing what you eat in a way that you might not be aware of,” DiFeliceantonio said.

Baugh notes that this was a small study and more research is needed. She is recruiting participants with an even wider range of glycemic control and different body weights to better inform public health. 

“Ultimately, understanding the mechanisms that influence food choice and eating behaviors can be really impactful in developing different pharmacological or behavioral strategies for obesity treatment – and even prevention,” Baugh said.

The research was funded in part by the National Center for Advancing Translational Sciences and the National Institute of Diabetes and Digestive and Kidney Diseases, both part of the National Institutes of Health.