Graduation Year


Document Type




Degree Granting Department

Medical Sciences

Major Professor

David G. Morgan


Alzheimer's Disease, amyloid, calorie restriction, ketogenic diet, ketones, tau


Dietary manipulations and their pharmacological outcomes have been increasingly studied in neurodegenerative diseases. However, a systematic comparison among different methods in validated animal models of Alzheimer's disease is made necessary due to several different approaches applied in recent studies. Moreover, despite the large body of evidence on the effects of calorie restriction (CR) and ketogenic diets (KDs) on amyloid pathology, no consistent data is available on the effects of calorie restriction, ketogenic diet or ketone supplements on tau pathology in transgenic models of AD. Moreover, the ketogenic diet used in our studies was custom made with low carbohydrate content and rich in medium chain triglyceride (MCT) oils, known to be rapidly metabolized in the liver, resulting in sustained peripheral ketosis.

Chapter 1 tested the ability of KD to induce significant ketosis in a mouse model of amyloid deposition. We showed that, despite the mild ketosis induced, KD fed APP mice presented subtle behavioral improvement shown as faster learning in the radial arm water maze, making less errors than APP mice kept on a control diet. Additionally, we observed decreased Aβ immunoreactivity in the anterior cortex of KD fed versus control fed APP mice, despite the lack of changes in congophilic deposits. Due to the mild ketosis induced, a modified ketogenic diet was devised with decreased maltodextrin content and showed greater peripheral levels of β-hydroxybutyrate.

Chapter 2 investigated the effects of a ketogenic diet in two transgenic mouse models of Alzheimer's pathology. Interestingly, we found that both transgenic lines, regardless of diet, weighed less than nontransgenic mice, despite their elevated food intake. The reduced body weight may, in part, be explained by the increased locomotor activity shown by both transgenic lines in both the open field and y-maze. Moreover, KD fed mice performed significantly better on the rotarod compared to mice on the control diet independent of genotype. We did not observed KD-induced changes in spatial or associative memory in the radial arm water maze or contextual fear conditioning, respectively. Furthermore, immunohistochemical levels of amyloid, tau, astrocytic and microglial markers showed no differences between animals fed KD or the control diet.

Chapter 3 studied the effects of calorie restriction on a mouse model of tau deposition. We show here that 35% body weight reduction in Tg4510 mice did not prevent increased locomotor activity in the open field, previously reported in chapter 2. Similarly, CR did not affect motor performance or spatial memory assessed by the rotarod and radial arm water maze, respectively. Interestingly, CR Tg4510 mice showed improved short-term memory tested by the novel object recognition despite spending a minimal percentage of the trial time interacting with the objects presented. However, this improvement was not observed when the test was modified to replace the objects with mice. In this case, we noticed that nontransgenic mice spent most of the trial time interacting with the novel mouse whereas Tg4510 mice spent roughly the same amount of time at any of the areas in the test chamber. Moreover, no changes in histopathological or biochemical levels of tau, astrocytic, microglial or synaptic markers were observed.

Chapter 4 sought to investigate alternative approaches to inducing ketosis in the brain by either administering BHB intracerebroventricularly (i.c.v.) or by using the acetoacetate (AcAc) diester as a dietary supplement in mice. We observed that i.c.v administration of BHB in 20 months old APP mice did not affect body weight or food intake. Consistent with the lack of effects on behavioral performance, amyloid and congophilic load were not different between APP mice infused with either saline or BHB. We also found that enteral administration of AcAc diester was well tolerated and induced peripheral ketosis for at least 3 hours. Acute ketosis, however, was not sufficient to attenuate behavioral deficits in old APP mice. Chronic dietary supplementation with AcAc was tested in control tet mice and was shown to effectively induce ketosis in mice fed a diet with normal contents of carbohydrates. Nonetheless, we observed that AcAc-induced ketosis was not significantly greater than levels induced by the ketogenic diet tested in our lab. Considering that KD did not rescue behavioral or histopathological features of either amyloid or tau depositing mouse models, we anticipated that dietary supplementation with AcAc would not likely modify the phenotype of the same mouse models tested previously.

Taken together, our findings show that our custom made ketogenic diet was effective in inducing and sustaining ketosis and may play an important role in enhancing motor performance in mice. However, the lack of changes on the cognitive and histopathological phenotype of the models studied suggests that KD may not be a disease modifying therapeutic approach to AD. Moreover, calorie restriction showed inconsistent effects on behavioral and histopathological outcomes of a mouse model of tauopathies. Furthermore, dietary supplementation with acetoacetate diester was successful in inducing peripheral ketosis to the same extend as a ketogenic diet even in the context of normal carbohydrate intake, suggesting that it may be of therapeutic interest for diseases of hypometabolism but not a disease modifying therapy in mouse models of Alzheimer's pathology.

Included in

Neurosciences Commons