Dandona and colleagues performed the surgery on 15 morbidly obese people with type 2 diabetes, who lost an average of 86 pounds each. When the researchers compared blood samples taken before surgery with samples taken six months later, they saw, as expected, that the patients’ insulin resistance index had fallen dramatically, a sign that their diabetes had improved. Also expected was that levels of numerous inflammatory markers, such as endotoxins, Toll-like receptors 2 and 4, NF-κB, C-reactive protein, and IL-6, dropped by an average of 25 to 35 percent, Dandona told ARF. Inflammation has been tied to numerous neurodegenerative diseases (see, e.g., ARF related news story).

“What was really novel and exciting was that the expression of genes related to AD fell,” Dandona said. “For the first time, there is a relationship between dramatic weight loss and reduction of [expression of] Alzheimer’s genes.” The level of APP protein in white blood cells dropped by about one-quarter, and presenilin-2 and glycogen synthase kinase-3β declined by similar amounts. Presenilin-2 is part of the γ-secretase enzyme that releases the Aβ peptide from APP, while the kinase has been implicated in toxicity of tau, the protein that makes neurofibrillary tangles in the AD brain. Levels in blood were used as a surrogate for what might be happening in brain, leaving open the question of whether brain chemistry is also changing. It is unclear if moderate weight loss would have the same effect. Dandona told ARF they previously looked at lower levels of weight loss, but did not see statistically significant changes in AD proteins.

Dandona is particularly intrigued by the connection between weight loss and AD genes because researchers led by John Gunstad at Kent State University, Ohio, found that people who undergo gastric bypass surgery perform better on memory tests three months later (see Gunstad et al., 2010). “If we can put the two things together, we could say that maybe bariatric surgery in the morbidly obese patient may delay or prevent the onset of AD,” Dandona speculated. To this end, Dandona is currently applying for funding to repeat his bariatric surgery study, adding cognitive tests and neuroimaging to the outcome measures. In addition, he will explore whether dramatic weight loss without surgery can lead to similar reductions in AD-related gene expression.

“I was quite excited and surprised by what they found,” said Paul Thompson at the University of California, Los Angeles. Thompson studies the relationship between obesity and AD. “The question has always been whether we can alter our genetic destiny,” Thompson said, adding that this study suggests that, under some circumstances, people can. In addition, Thompson noted that testing blood for changes in the expression of APP, PS2, and GSK-3β could be useful for epidemiologists as a relatively cheap and easy surrogate marker to see if interventions are having an effect on AD genes.

Dandona told ARF he is also pursuing the link between diabetes and AD. Type 2 diabetes, like obesity, more than doubles Alzheimer’s risk. In work published in the June Journal of Clinical Endocrinology and Metabolism, Dandona and colleagues showed that insulin infusion likewise lowers the levels of APP, PS2, and GSK-3β in the blood. The anti-diabetic drug exenatide, trade name Byetta, has the same effect, Dandona told ARF. He will investigate whether taking exenatide, a glucagon-like peptide 1 (GLP-1) analog, can improve cognitive function. Another GLP-1 mimic that improves insulin signaling, liraglutide, was recently shown by researchers in Ireland to improve memory, reduce Aβ, and protect synapses in an AD model mouse (see McClean et al., 2011 and ARF related news story). Previous diabetes drugs, such as the insulin-sensitizing agent rosiglitazone, have failed in clinical trials (see ARF related news story).—Madolyn Bowman Rogers.

Reference:
Dandona P, Mohamed I, Ghanim H, Sia CL, Dhindsa S, Dandona S, Makdissi A, Chaudhuri A. Insulin suppresses the expression of amyloid precursor protein, presenilins, and glycogen synthase kinase-3{beta} in peripheral blood mononuclear cells. J Clin Endocrinol Metab. 2011 Jun;96(6):1783-8. Abstract