Project Summary Metabolic syndrome (MetS) is associated with the development of diabetes and cardiovascular disease; however it is also linked with cognitive decline and dementia. The study investigators have shown that MetS is associated with lower cerebral blood flow (CBF) and memory function in late middle-aged adults at increased risk for developing Alzheimer's disease (AD). Insulin resistance (IR) is at the core of MetS, and a hallmark feature of IR is higher fasting blood glucose (FBG) as well as post prandial hyperglycemia.

While the study investigators and others have demonstrated links between IR and CBF as well as cognition from an observational perspective, no studies have investigated CBF and cognition after an intervention involving exercise and a carbohydrate restricted diet (CRD) designed to improve or normalize IR and glucose homeostasis. The study investigators propose to determine the effect of improving or normalizing glucose homeostasis on CBF and cognition, through diet and exercise, in individuals with IR and at risk for the development of AD.

While exercise and a CRD have been shown to improve IR and glycemic control, we have only limited knowledge of the mechanisms behind these improvements. Nutritional metabolomics, the global measurement and interpretation of metabolic profiles, assesses the interaction of diet with the endogenous gene-protein cascade and the gut microbiome. Additionally, exercise has been shown to have an impact on the human metabolome. Finally, numerous metabolites have been specifically linked to IR and impaired fasting glucose (IFG). The study investigators propose to use metabolomics to measure changes in metabolites as individuals normalize or improve IR and glucose homeostasis.

Should this exploratory study reveal increased brain blood flow and improved memory in response to diet and exercise, then early treatment of these individuals at risk might offer new avenues for disease-course modification. Strategies towards early and effective risk factor management could be of value in reducing the risk of metabolic as well as cognitive decline. In addition, should this study reveal changes in metabolic abnormalities consistent with early indications of diabetes, metabolomics could be an effective approach to complement disease risk analysis in our goal toward precision care.

Research Strategy Design. The study investigators propose recruiting 40 middle-aged (45-65 years) participants from the Wisconsin Alzheimer's Disease Research Center's (ADRC) "IMPACT" cohort to participate in a 12 week diet and exercise intervention with a 6-month follow-up. The Wisconsin ADRC has a special emphasis in preclinical AD, and participants are recruited with the expectation of being approached for intervention studies. Since the treatment will be applied to groups of 10 participants each, consisting of a total of 4 groups, the study investigators will not be able simultaneously to roll out the treatment program in all groups. Subsequently, an incomplete stepped wedge cluster randomized trial (SWCRT)48,49 will be considered. The design allows each block (intervention group) to be randomized to the intervention in a stepped fashion, initially providing a baseline measure, as indicate by '0' in the design matrix below, then followed by the 12 week intervention 'X' with '.' indicating no measurement, and '1a' indicating measurement post-12 week diet and exercise intervention and '1b' 6 months after the intervention.

Time 1 Time 2 Time 3 Time 4 Time 5 Time 6 Time 7 Group 1 0 X 1a 1b . . . Group 2 . 0 X 1a 1b . . Group 3 . . 0 X 1a 1b . Group 4 . . . 0 X 1a 1b

Timeline. It is anticipated that after recruitment and random assignment to one of the 4 groups, baseline data collection and the 12 week intervention will begin for the first group in month 3 of the 24 month study. Baseline data collection will be completed for all groups by month 12, and post 12-week diet and exercise intervention data collection will be completed by month 15. Post 6 month follow-up data collection will be completed by month 20. The study investigators will share the results of baseline and follow-up findings in manuscripts and at national meetings, in addition to using the generated data to design a larger intervention trial.

Participants. The study investigators will recruit 40 sedentary (exercise <1 hour/week) participants who are 45-65 years of age, have impaired fasting glucose and are cognitively normal (based on the comprehensive ADRC battery). The cohort has received detailed annual medical and cognitive evaluation since enrolling in the Wisconsin ADRC. The cohort is well-characterized for health status and medication use, has undergone genotyping AD-related risk genes, and has banked fluid samples and DNA. Importantly, participants are characterized on metabolic risk, with slightly more than half meeting criteria for central obesity and/or elevated fasting glucose. Exclusion criteria are: active lifestyle (exercise >1 hour/wk); body mass index > 40; MRI contraindications; history of neurological disease, prior neurosurgery; diagnosed and/or treated type 1 or 2 diabetes; pregnancy; acute or subacute active cardiac disease (ongoing chest pain or myocardial infarction < 3 months); significant orthopedic or musculoskeletal condition that limits weight bearing.

Eligibility, Screening and Testing. Following IRB approval, a study coordinator will send an invitation letter describing the study to eligible participants previously identified via fasting glucose levels as pre-diabetic. Interested participants will be contacted by the study coordinator for further eligibility screening. Exercise testing will take place at the School of Nursing Exercise Physiology Laboratory. After an overnight fast, participants will arrive in the morning at the Clinical Research Unit to have blood samples collected for Metabolomics, A1C, glucose and insulin, receive cognitive testing, and magnetic resonance imaging (MRI). Participants will undergo MRI at the Wisconsin Institutes for Medical Research (WIMR).

12-Week Diet and Exercise Intervention. The intervention is based on the research team's successful prior work in behavioral interventions. During the 12-week intervention, all participants will participate in a diet and exercise program. Exercise: Participants will attend a supervised exercise program, 10 subjects per group meeting 3x/wk. The program will begin with exercise for 20 minutes at a moderate intensity (determined during initial exercise testing). After week 2, exercise will be increased to 30 minutes; and after week 3, exercise will be increased to 50 minutes, resulting in 150 minutes of supervised exercise weekly. Exercise will be preceded by a 10 minute warm-up and a 10 minute cool down following exercise. The exercise classes will be held at the University of Wisconsin Natatorium, which includes stationary bikes, treadmills, elliptical machines, etc. Participants will be encouraged to exercise on their own for 30-50 minutes at least 2 days/week for a total of 5 days/week of exercise. Participants will be provided with heart rate monitors and instruction on how to use them to maintain a target heart rate during exercise. Anaerobic threshold measured during baseline testing will be used initially to determine target heart rate, but participants will be monitored every other week during the exercise sessions using heart rate variability techniques published by the Co-PI, R. Gretebeck to adjust for improvements in exercise capacity. Diet: Participants will be instructed by a dietitian (Co-PI, Dr. R. Gretebeck) to follow a diet designed to reduce post prandial and overall glucose levels. The American Diabetes Association approves the use of a CRD in overweight or obese individuals, but cautions that this approach should be limited to one year. There is no consensus on the definition of a CRD, but most studies have used diets providing between 30-100g of carbohydrates per day accompanied with a moderate amount of protein (15-30% of calories), with fats providing the rest of daily energy requirements. For this study, participants will be instructed to avoid or restrict grains, sugars, legumes, starchy vegetables, and fruits. Carbohydrate from dairy, condiments, nuts and seeds etc. will not be limited, resulting in a diet that is not considered severely restricted and thus easier to follow, but still less than 100g of carbohydrates per day. In concrete terms, most meals will include some source of animal protein including dairy products with non-starchy vegetables and fat from sources such as olive oil, avocado, nuts, butter and cheese. Behavior Change: The dietitian (who is also an exercise physiologist) and Co-I, K. Gretebeck, behavior change expert, will conduct a weekly 30-minute behavior change class with the focus on self-regulation (goal setting, self-monitoring, relapse prevention, overcoming barriers, etc.) to promote diet and exercise adherence. Specific emphasis will be placed on glucose self-monitoring for immediate feedback of glycemic control. Building on the behavior change classes, the study investigators will use glucose self-monitoring as an immediate feedback loop to motivate participants to maintain normalized or improved glucose homeostasis through diet and exercise. Following the 12-week diet and exercise intervention, participants will continue to self-monitor their blood glucose, and continue the diet and exercise program on their own. After the 12 week intervention, investigators will meet with participants monthly to collect blood and urine for metabolomics, and download the glucose history from the glucose monitors. The following objective measures will be used to assess adherence to the lifestyle intervention: history of glycemic control from glucose monitors, HOMA-IR, exercise capacity (V02 max), 6 minute walk, and metabolomic changes.

Glucose Self-Monitoring. Participants will be provided with glucose monitors that maintain a history of results for fasting and post prandial blood glucose. During the first 2 weeks of the 12 week intervention, participants will be instructed to monitor their fasting and post prandial (evening meal) blood glucose daily, and after week 2 every other day, with the goal or reaching a FBG <100 mg/dl (<86 ideal), and 2 hour post-prandial blood glucose <140 mg/dl (<120 ideal). After the goal has been reached, participants will be asked to continue to self-monitor weekly. It is anticipated that many participants will reach this goal within the first 4 weeks.

Primary Outcomes. CBF (measured at baseline, after the 12 week intervention and after 6 month follow-up). All participants will be screened for contraindications to MRI using standard clinical screening questionnaires. Participants will undergo MRI on a 3.0 Tesla X750 GE Discovery scanner with an 8-channel array head coil. Scan time is ~45 minutes and will include scans relevant to the primary outcome measure of micro- and macro-vessel CBF. Additional scans will include a 3D T1-weighted volume, and T2 FLAIR to assess ischemic lesion burden.

Cognition (measured at baseline, after the 12 week intervention and after 6 month follow-up). Memory and executive function will be measured as follows: 1) Memory function as indexed by the California Verbal Learning Test-II: learning slope for trials 1-5 and long delay retention57, and 2) Executive function as indexed by: the Delis-Kaplan Executive Function System Trails Test, and Color-Word Interference Test58. Also, participants will be fully characterized on cognitive function by the ADRC, including the National Alzheimer's Coordinating Center Uniform Dataset.

Metabolomics. Fasting blood and urine (second void of the morning) samples will be collected before and after the 12 week diet and exercise intervention as well as every 2 weeks during the 12 week intervention, and monthly during 6 month follow-up. Our project will use the metabolomics database developed at the NMRFAM at UW-Madison and made freely available through the BMRB59 at (www.bmrb.wisc.edu/metabolomics).

Other Measures at baseline, after the 12 week intervention and after 6 month follow-up.

Exercise Capacity (VO2max) will be measured using treadmill walking at 3.0 mph for 3 minute stages progressing potentially through 0, 3, 6, 9, 12, and 15% grades of incline. Every 3 minutes the grade will be increased until the participants reach volitional fatigue. Heart rate and respiratory gases will be monitored continuously, and blood pressure and blood lactate (finger prick) will be monitored at the end of each stage.

Physical Function (6 minute walk). Participants will be instructed to cover as much ground as possible on a hallway course over 6 minutes. 6 minute walk is measured in feet walked over 6 minutes65.

Metabolic Status. Fasting glucose, insulin, and A1c will be measured by University of WI Clinical Research Unit

Diet and Exercise Self-efficacy and Self-regulation. Validated questionaires for self-regulation66 and self-efficacy67 for physical activity will be utilized, and a questionnaire developed by Anderson et. al.68 for nutrition related behavior will adapted for use with a CRD.

Glucose Monitoring. Participants will receive a ReliOn® Ultima (Abbott Diabetes Care Inc) glucose monitor and test strips, which meets FDA standards for accuracy (+20%). See description above for glucose self-monitoring. Data from the glucose monitors will be downloaded before, every 2 weeks during the 12 week intervention, and monthly for 6 months.

Inclusion Criteria:

Participants from the Wisconsin Alzheimer's Disease Research Center (ADRC) -

Exclusion Criteria:

• . Exclusion criteria are: active lifestyle (exercise >1 hour/wk); body mass index > 40; MRI contraindications; history of neurological disease, prior neurosurgery; diagnosed and/or treated type 1 or 2 diabetes; pregnancy; acute or subacute active cardiac disease (ongoing chest pain or myocardial infarction < 3 months); significant orthopedic or musculoskeletal condition that limits weight bearing.