A PDK-1 allosteric agonist improves spatial learning and memory in a βAPP/PS-1 transgenic mouse-high fat diet intervention model of Alzheimer's disease.

Diabetes mellitus (DM), peripheral insulin resistance (IR) and obesity are clear risk factors for Alzheimer's disease. Several anti-diabetic drugs and insulin have been tested in rodents and humans with MCI or AD, yielding promising but inconclusive results. The PDK-1/Akt axis, essential to the action of insulin, has not however been pharmacologically interrogated to a similar degree. Our previous cell culture and in vitro studies point to such an approach. Double transgenic APPsw/PSENdE9 mice, a model for Alzheimer's disease, were used to test the oral administration of PS48, a PDK-1 agonist, on preventing the expected decline in learning and memory in the Morris Water Maze (MWM). Mice were raised on either standard (SD) or high fat (HFD) diets, dosed beginning 10 months age and tested at an advanced age of 14 months. PS48 had positive effects on learning the spatial location of a hidden platform in the TG animals, on either SD or HFD, compared to vehicle diet and WT animals. On several measures of spatial memory following successful acquisition (probe trials), the drug also proved significantly beneficial to animals on either diet. The PS48 treatment-effect size was more pronounced in the TG animals on HFD compared to on SD in several of the probe measures. HFD produced some of the intended metabolic effects of weight gain and hyperglycemia, as well as accelerating cognitive impairment in the TG animals. PS48 was found to have added value in modestly reducing body weights and improving OGTT responses in TG groups although results were not definitive. PS48 was well tolerated without obvious clinical signs or symptoms and did not itself affect longevity. These results recommend a larger preclinical study before human trial.