HIIT AS A PREVENTATIVE MEASURE FOR AGE-RELATED COGNITIVE DECLINE By Justin Stephenson July, 2024 Director of Thesis: Ted Graber, PhD Major Department: Kinesiology Abstract PURPOSE: The relationship between aging and cognition is multifaceted, and the effects of exercise on age-related cognitive decline seem promising. Exactly how exercise promotes cognitive health, and which form of exercise promotes it the most, is still uncertain. This study's main purpose was to assess the effects of High-Intensity Interval Training on functional and cognitive performance in a group of middle-aged (17-month-old) mice, compared to a sedentary control group. HYPOTHESES: First, we expect the high-intensity interval trained group (HIIT; n = 8) to exhibit greater resistance to age-related cognitive decline, pre- to post-, than the sedentary control group (SED; n = 7). Second, we predict the HIIT mice maintain physical function compared to the SED group. METHODS: Using C57BL/6 mice, the present study utilized a 12- week treadmill HIIT protocol as the intervention, assessing functional and cognitive performance with the Comprehensive Functional Assessment Battery (CFAB) and a Cognitive Assessment Battery (CAB). CFAB consists of tasks including Treadmill, Inverted Cling, Grip Meter, Volitional Wheel Running, and Rotarod. CAB is comprised of tasks including Open Field, Novel Object Recognition, Y-maze, and Puzzle Box. Cognitive and physical function batteries were performed pre- and post-intervention. EchoMRI and in vivo contractile physiology were used to measure body composition and plantar flexor torque, respectively, at baseline and post- intervention. RESULTS: The 12-week HIIT protocol resulted in significant aerobic capacity improvements for the HIIT group, increasing treadmill time by 28%, while the SED group demonstrated a 41.4% decline in treadmill time (within-subjects effects of time*group F=21.381, p<0.001; between-subjects effects of time*groups F=5.572, p=0.035). No significant differences between the groups’ cognitive function tests were observed pre- to post-training. Differences in body mass, fat, and fat% pre- to post-training, measured at the time of EchoMRI, were significantly larger within groups (2x2 Repeated Measures ANOVA: F=20.062, 46.845, and 35.899, respectively; all p<0.001). DISCUSSION: Contrary to previous research, the current study found little effect of HIIT on body composition. Regarding the lack of cognitive maintenance observed, at only 17 months of age the mice may not experience any cognitive deterioration. Thus, the HIIT intervention may not have had an opportunity to influence cognitive function maintenance. Our next step is to assess young adult mice to establish a CAB baseline, and evaluate older adult cognitive response to HITT. HIIT AS A PREVENTATIVE MEASURE FOR AGE-RELATED COGNITIVE DECLINE A Thesis Presented to the Faculty of the Department of Kinesiology East Carolina University In Partial Fulfillment of the Requirements for the Degree The Master of Science in Kinesiology Exercise Physiology Concentration By Justin Curtis Stephenson July, 2024 Director of Thesis: Ted Graber; PhD Thesis Committee Members: Ted Graber, PhD Tuan Tran, PhD Nicholas Broskey, PhD © Justin Stephenson, 2024 Table of Contents LIST OF TABLES ....................................................................................................................... iv LIST OF FIGURES ...................................................................................................................... v LIST OF SYMBOLS/ABBREVIATIONS ................................................................................. vi CHAPTER I: INTRODUCTION ................................................................................................ 1 PURPOSE ...................................................................................................................................... 6 HYPOTHESIS ................................................................................................................................. 6 DELIMITATIONS ........................................................................................................................... 7 CHAPTER II: REVIEW OF LITERATURE ............................................................................ 8 TYPES OF COGNITIVE IMPAIRMENT .............................................................................................. 8 UNDERLYING MECHANISMS ...................................................................................................... 11 IMPACTS OF COGNITIVE DECLINE .............................................................................................. 15 PREVENTATIVE MEASURES ........................................................................................................ 16 CHAPTER III: METHODS ...................................................................................................... 20 SUBJECTS ................................................................................................................................... 20 PERFORMANCE ASSESSMENTS ................................................................................................... 21 PHYSICAL FUNCTION ASSESSMENTS .......................................................................................... 21 COGNITIVE FUNCTION ASSESSMENTS ........................................................................................ 26 INTERVENTION ........................................................................................................................... 32 TISSUE COLLECTION .................................................................................................................. 34 STATISTICAL ANALYSIS ............................................................................................................. 34 CHAPTER IV: RESULTS ......................................................................................................... 36 PHYSICAL FUNCTION ASSESSMENTS .......................................................................................... 36 OTHER PHYSICAL MEASUREMENTS ........................................................................................... 39 COGNITIVE FUNCTION ASSESSMENTS ........................................................................................ 40 CHAPTER V: DISCUSSION .................................................................................................... 45 CHAPTER VI: CONCLUSION ................................................................................................ 52 SUPPLEMENTARY TABLES .................................................................................................. 53 SUPPLEMENTARY FIGURES ................................................................................................ 67 REFERENCES ............................................................................................................................ 69 APPENDIX A: STANDARD OPERATING PROCEDURES FOR COGNITIVE TESTS 74 APPENDIX B: IACUC APPROVAL LETTER ...................................................................... 87 List of Tables TABLE 1: EXAMPLE OF SPONTANEOUS ALTERNATION COUNTS ................................................ 28 TABLE 2: PHYSICAL FUNCTION WITHIN-GROUPS ANALYSIS ..................................................... 53 TABLE 3: PHYSICAL FUNCTION BETWEEN-GROUPS ANALYSIS ................................................. 55 TABLE 4: BODY COMPOSITION WITHIN-GROUPS ANALYSIS ..................................................... 56 TABLE 5: BODY COMPOSITION BETWEEN-GROUPS ANALYSIS .................................................. 57 TABLE 6: MUSCLE MASS BETWEEN-GROUPS ANALYSIS ............................................................ 58 TABLE 7: MUSCLE FUNCTION WITHIN-GROUPS ANALYSIS ....................................................... 59 TABLE 8: MUSCLE FUNCTION BETWEEN-GROUPS ANALYSIS .................................................... 60 TABLE 9: COGNITIVE FUNCTION WITHIN-GROUPS ANALYSIS .................................................. 62 TABLE 10: COGNITIVE FUNCTION BETWEEN-GROUPS ANALYSIS ............................................. 66 List of Figures FIGURE 1: STUDY DESIGN ............................................................................................................ 20 FIGURE 2: ZONE OVERLAY USED TO TRACK OPEN FIELD BEHAVIOR ......................................... 27 FIGURE 3: HIIT SESSION DESIGN ................................................................................................ 33 FIGURE 4: GRIP METER PRE- TO POST-TRAINING ..................................................................... 37 FIGURE 5: VOLITIONAL WHEEL RUNNING (RIGHT) PRE- TO POST-TRAINING ......................... 37 FIGURE 6: INVERTED CLING (LEFT) AND INVERTED CLING LOG10 (RIGHT) PRE- TO POST- TRAINING .............................................................................................................................. 37 FIGURE 7: ROTAROD PRE- TO POST-TRAINING .......................................................................... 38 FIGURE 8: TREADMILL PRE- TO POST-TRAINING ....................................................................... 38 FIGURE 9: AVERAGE BODY MASS PER WEEK ACROSS THE INTERVENTION ............................. 39 FIGURE 10: OPEN FIELD PRE- TO POST-TRAINING .................................................................... 41 FIGURE 11: Y-MAZE PRE- TO POST-TRAINING .......................................................................... 42 FIGURE 12: PUZZLE BOX PRE- TO POST-TRAINING. .................................................................. 43 FIGURE 13: CFAB TOTAL (LEFT) AND OVERALL IAV (RIGHT) PRE- TO POST-TRAINING ....... 67 FIGURE 14: NOVEL OBJECT RECOGNITION PRE- TO POST-TRAINING ...................................... 68 List of Symbols/Abbreviations AD: Alzheimer’s Disease ARCD: Age-Related Cognitive Decline ARCI: Age-Related Cognitive Impairment BBB: Blood-Brain Barrier BDNF: Brain-Derived Neurotrophic Factor CAB: Cognitive Assessment Battery CDC: Centers for Disease Control and Prevention CFAB: Comprehensive Functional Assessment Battery CVD: Cardiovascular Disease HIFT: High-Intensity Functional Training HIIT: High-Intensity Interval Training HTN: Hypertension IACUC: Institutional Animal Care and Use Committee IGF-1: Insulin-like Growth Factor MICT: Mild-Intensity Continuous Training NIA: National Institute on Aging NOR: Novel Object Recognition OF: Open Field PA: Physical Activity %SAP: Percentage of Spontaneous Alternation Performance %Speedmax: Percent of Maximum Speed SAs: Spontaneous Alternations SCD: Subjective Cognitive Decline SD: Standard Deviation SED: Sedentary control group Speedmax: Maximum Speed achieved during treadmill test TBI: Traumatic Brain Injury VaD: Vascular Dementia VCI: Vascular Cognitive Impairment VEGF: Vascular Endothelial Growth Factor VWR: Volitional Wheel Running WHO: World Health Organization Chapter I: Introduction The world population is currently just under 8 billion people (Census, 2023), and is predicted to increase an additional 2 billion over the next 30 years, therefore, by the year 2050, the global population could be around 9.7 billion (Nations, 2020). As modern medicine and technology continue to develop even further, prospective survival rates are also predicted to increase around the world. For perspective, according to Medina et al. (2020), life expectancy for the United States’ population increased 10 years between 1960 and 2017, and more than 6 years between 2000 and 2019 (WHO, 2023). Life expectancy is projected to increase 6 years by 2060, meaning the U.S. could reach a record-high total life expectancy of 85.6 years (Medina et al., 2020). The percentage of adults aged 65 and older grew by 34.2% between 2010 and 2020 (U.S. Census Bureau, 2020), and though birth rates have slowed down a little, the reason the global population continues to rise is due to how long humans are now living. This shift in age distribution towards an older population will increase disability, sedentation, and the need for medical, residential, and home care. As the global population grows older, the risk of age-related health and functional declines are heightened. Some of the changes associated with the normal aging process include decreases in visual acuity, vestibular function, muscle strength, immunosenescence, physical and cognitive function; and increased risk of diseases like cardiovascular disease (CVD), hypertension (HTN), osteoarthritis, diabetes mellitus, osteoporosis, and dementia (Jaul and Barron, 2017). As stated before, decreases in cognitive function are associated with normal aging, but what is normal aging in regard to cognition? As we grow older, mild changes in cognition are expected and considered a normal feature of the aging process (Hugo and Ganguli, 2014). This is 2 normal cognitive decline. Normal cognitive decline is less crude and primarily affects executive function, processing speed and attention span. With abnormal aging, commonly associated with disease (i.e., dementia, Alzheimer’s disease), cognitive decline is less subtle and often includes deficits in other cognitive abilities, such as navigation issues, rapid memory loss, trouble expressing thoughts verbally, and solving common problems (UCSF, 2023). With abnormal cognitive aging, progressive declines in one’s functional capacity also come a lot sooner than what is considered normal, and it is primarily due to the diseased state of the mind rather than just normal functional decline. This study is not concerned with looking at abnormal cognitive decline, but rather, the cognitive changes related to normal aging. Aside from aging, other well-known risk factors for cognitive impairment are traumatic brain injuries, physical inactivity, and family history. Traumatic brain injury (TBI) and concussions (mild TBI) can occur from a variety of stressors on the brain, from a modest blow to a powerful strike across the head, face, or even body, that forces the brain to hit against the inside of the skull. In the United States, alone, the CDC reports an estimated 1.7 million people experience a TBI every year, with older adolescents (15-19 years old) and older adults (≥65 years old) being the most-likely victims of such an incident (Georges and Das, 2022). Amongst trauma patients, traumatic brain injury is the leading cause of injury-induced disability and death, with falls and car crashes being the most common source of these injuries (Vella, Crandall, and Patel, 2017). TBI was also identified as a high-risk factor for developing post-traumatic stress disorder and other psychiatric disorders (Georges and Das, 2022). However, neither TBI’s nor family history are the risk factor of interest for this study. In 1967, Horn and Cattell developed a conceptual framework that distinguishes two types of cognitive abilities: fluid skills and crystallized skills. The fluid skills deemed by Horn and 3 Cattell (1967) are of more interest to the current study as they include cognitive factors such as attention/focus, memory, information processing speed, visuospatial and executive function skills. Crystallized skills are largely unresponsive to the aging process, while fluid skills are more susceptible to age-related decline (Cadwallader et al., 2022). Another categorization system for age-related cognition, which was touched on earlier, has developed since Horn and Cattell’s, where the process of aging has been categorized into three groups based upon the variability in the degree of cognitive decline. There is healthy aging, successful aging, and accelerated aging. Healthy, or “normal,” aging can be defined as the structural, chemical, and functional changes to the brain that occur with age in the absence of any pathological disease. Therefore, those with dementia or Alzheimer’s disease (AD) do not fall under the category of “healthy aging.” Successful aging can be defined as experiencing less age-related cognitive decline (ARCD), and less physical/structural changes to the brain, than what is observed with normal aging. Accelerated aging refers to those who experience ARCD, and changes in the brain’s structure and functionality, earlier in life compared to normal aging. Evidence suggests there are a variety of genetic, environmental, health and lifestyle factors that play a role in the brain’s aging and cognitive capabilities as we grow older (Cadwallader et al. 2022). Many of the health and lifestyle factors that potentiate ARCD are modifiable risk factors that can be helped with exercise. While exercise is well known as being beneficial for physical health (cardiovascular, pulmonary, metabolic, renal, etc.), a less renowned benefit is its positive effects on cognitive and psychiatric health. Cognitive decline can occur with age for a multitude of reasons, in a variety of ways, and can vary/increase significantly in severity. Experiencing a less robust memory after 4 70 years of living (i.e., not remembering every person you graduated high school with, forgetting how you broke a bone as a child, often forgetting where you put your keys) is vastly different from not remembering how to use your own laundry machine, or running into furniture in your own home as you lose spatial memory. Exercise is being pursued as one of the primary preventative measures for cognitive decline because of how beneficial it is for many of the underlying mechanisms suspected to cause mild-to-severe cognitive impairments, including neurodegenerative diseases such as dementia and AD. Neurogenesis, synaptic neurogenesis, and neuroplasticity are three vital processes involved in maintaining cognitive function, which, evidence suggests exercise seemingly assists with these processes (Cadwallader et al. 2022). One of the primary beneficiary factors for overall health is exercise, therefore, it makes sense that exercise would be one of the first interventions applied in researching preventative measures for ARCD. Exercise is capable of holistically assisting/boosting bodily functions by increasing muscular strength, lean mass, metabolism, cardiovascular circulation, oxygen dispersion, cardiorespiratory endurance, as well as decreasing fat mass, inflammation, and sedentary behavior. Additionally, the exercise type, mode, intensity, frequency, duration, and overall volume can make significant differences in the outcomes of the exercise. As science tries to prevent the onset of age-associated declines in cognitive performance via exercise interventions, testing different modes of exercise to find the best results is pivotal. Therefore, the current study focuses on High-intensity Interval Training (HIIT) as an exercise intervention, to assess whether HIIT, alone, can be an effective preventative measure for age- related cognitive decline. Short-term HIIT has been shown to produce physiological improvements similar to, and sometimes better than, endurance training and resistance training, respectively (Callahan et al. 2021). Also, HIIT has been reported to have beneficial effects 5 among several chronic diseases such as diabetes, cancer, stroke, and hypertension (Molmen- Hansen et al. 2012; Askim et al. 2014; Støa et al. 2017; Rose et al. 2020). These chronic diseases are often associated with cognitive decline in older adults (CDC, 2019). According to Calverley et al. 2020, in a topical review of the effects of HIIT on the brain, HIIT is more likely to work as a preventative measure than a post-diagnosis treatment method for dementia patients. With that in mind, some effects of long-term exercise on cognition include: increases in hippocampal volume, coherence of functional brain networks, systematic modulation of task- related brain activity, dopamine release, and insulin sensitivity. Accordingly, Erickson et al. (2011) conducted a study that showed a 2% increase in hippocampal volume after a 1-year exercise intervention, which equates to a reversal of 1-2 years’ worth of age-associated volume loss. Those changes were also believed to be associated with short-term memory improvement. Another area of interest regarding this topic (and study) is exercise-induced changes in vasculature and vascularization believed to benefit overall cognitive function. Recently, HIIT has been manifesting as the safe and time-conscious mode of exercise with a capacity to improve cardiorespiratory fitness and vascular adaptation (Calverley et al. 2020). Many of the primary catalysts (sedentary behavior, motor vehicle accidents, contact sports, cardiovascular/metabolic diseases, etc.) for various cognitive impairments have unfortunately become normal, commonplace features of society, despite being avoidable/manageable with appropriate preventative measures taken. Preventative measures are constantly being implemented to reduce the number of TBI occurrences in sport, vehicle accidents, and falls in the older adult population. However, research aimed at identifying non- pharmacological preventative measures for ARCD is fairly limited, despite continuous advancements in technology and pharmaceuticals increasing global life expectancy by >6 years 6 between 2000 and 2019 (WHO, 2023), exacerbating the need for proactive defenses against cognitive decline in mid- to older-adult populations before they can no longer self-sufficiently live/function. To counter the negative outcomes that stem from each generation’s eventual cognitive decline, more research needs to be conducted focusing on the one ubiquitous and inevitable source of cognitive impairment: age. Therefore, the current study is concerned with addressing the cognitive decline associated with normal aging by investigating the effects of a specific type of exercise (High-intensity Interval Training; HIIT) on cognitive performance in mice. Purpose The study's main purpose was to assess the effects of High-Intensity Interval Training on functional and cognitive performance in middle-aged mice, compared to sedentary middle-aged mice. In future studies, we will investigate vascular and/or endothelial function in the brain – analyzing certain protein expressions (such as VEGF) via Western Immunoblotting methods, and/or qRT-PCR (quantitative real time polymerase chain reaction) – as they may relate to the influence of HIIT training on potential differences observed in cognitive and functional performance. Hypothesis As a result of regular High-Intensity Interval Training, we hypothesized there would be less cognitive and functional decline – as measured by CFAB and CAB testing protocols – in the HIIT-exercised group as compared to the sedentary control group. The exercised group should 7 maintain and/or improve in cognitive performance more than the sedentary group, from pre to post. Delimitations The current study only included middle-aged male C57BL/6 mice as the subjects. The HIIT protocol was designed with specific variables (training frequency, session durations, interval time, interval frequencies, and length of the intervention). If necessary, gentle prodding and shock grid were the only forms of motivation used during treadmill training and testing. CFAB and CAB assessment protocols were used to examine functional and cognitive performance, and the order in which they were conducted was reversed between baseline and end-point testing to negate any learning/adaptation bias that could result in better performances for one battery over the other. All sedentary control mice were subjected to sham treatment every day the HIIT group was subjected to training. Chow and water were provided ad libitum to all mice. Chapter II: Review of Literature Though it is well-established that exercise and physical activity (PA) play a critical role in the improvement and/or maintenance of overall health and physical function – reducing the likelihood of developing undesirable health outcomes – its effects on cognition are less concrete. The existing literature is indicative of exercise, and physical activity (Nuzum et al., 2020; Dougherty et al., 2020), indirectly potentiating positive effects on cognition through various underlying mechanisms. However, the current study was only concerned with utilizing High- Intensity Interval Training (HIIT), alone, as the independent variable, to observe its effects on age-related cognitive decline via cognitive and functional testing. We also hoped to dive further into the underlying mechanisms at play, potentially investigating vascular and/or endothelial dysfunction in the brain as it may relate to cognitive performance but were unable to do so due to time constraints, thus leaving this exploration for future studies. Therefore, mechanisms of action will be left for follow-up studies to investigate. The following review of available literature will examine the following topics: types of cognitive impairment (injuries and diseases) and the underlying mechanisms (speculated and observed) of those impairments, impacts of age-related cognitive decline (ARCD), documented effects of exercise as a preventative measure in managing/improving cognition, and how all of it pertains to the current study. Types of Cognitive Impairment As discussed, cognitive impairment can naturally occur with age, but it can also occur and/or be accelerated from physical trauma to the brain. A traumatic brain injury (TBI) alters how the brain functions due to physical stress on the brain (CDC, 2022). From a critical blow to a simple instantaneous movement of the head that forces brain tissue against the inner skull, the 9 Centers for Disease Control and Prevention reported that there were more than 64,000 TBI- related deaths in 2020 (CDC, 2022). However, while this is of critical concern to the general health of the youth and older adult population, the current study is solely concerned with examining naturally-occurring (age-related) declines in cognition, free of external influence on the acceleration of cognitive deterioration. Aging of the brain does not occur or progress at a standard, universal rate, nor do all functions of the brain degenerate with normal aging. In fact, wisdom, knowledge, empathy, and altruism have all been shown to generally increase within the confines of normal aging (Jaul and Barron, 2017). However, as previously mentioned, the relationship between age and cognition is an intricate connection, varying from person-to-person (Cadwallader et al., 2022). Subjective Cognitive Decline (SCD) is recognized by the CDC (2018) as the self-reported declining of memory or increasing incidence of confusion, is Subjective Cognitive Decline (SCD). According to the CDC (2018), the prevalence of SCD among adults 45-64 years of age is 10.8% and increases to 11.7% in adults 65 years and older. SCD and subjective memory complaints are one of the first clinically recognizable precursors of dementia and Alzheimer’s disease (Barnes, 2015; CDC, 2018). With common associations between ARCD and neurocognitive diseases, such as dementias and AD, it is worth restating that these diseases are not the result of normal cognitive aging (NIA, 2022). Alzheimer’s disease is a distinctive neuropathology, recognized by a build- up of amyloid plaques and tau neurofibrillary entanglements within the brain, that can ultimately lead to dementia (Barnes, 2015). Before, it was believed that arteriosclerosis was the commonest source of dementia, until Alzheimer pathology was later proven to be the most recurrent source of dementia (Graff-Radford, 2019). There are other pathologies that can elicit different forms of 10 dementia, but for the majority of cases, the underlying cause(s) remains unknown. Dementia is defined by the National Institute on Aging (NIA; 2022) as the loss of cognitive function to the point of interference with one’s everyday lifestyle. Some more common forms of dementia, aside from AD, include frontotemporal dementia, Lewy body dementia, and vascular dementia (NIA, 2022). With the existence of numerous pathologically unique forms of dementia, the current review will only detail the ones of interest to the study. Vascular Dementia (VaD), also known as multi-infarct dementia, occurs when blood- oxygen flow to the brain is disrupted, or blood vessels within the brain are damaged (NIA, 2022), leading to cognitive impairment. According to Hébert and Brayne (1995), despite its linear increase in prevalence, only ranging from 1.2-4.2% in adults >65 years of age, it was still the second most-common form of dementia. The study also reported that the disease, on average, only lasted ~5 years, with a higher mortality rate than AD. However, VaD appeared to have similar risk factors to AD and other cognition-impairing sources, including heart disease, stroke, hypertension, and diabetes (Hébert and Brayne, 1995). Again, three of these are modifiable risk factors that can be addressed with proper exercise. To include all forms of cognitive impairment related to vascular disease, the term vascular cognitive impairment (VCI) was created. At the individual level, there appears to be no observable neuropsychological pattern that differentiates VCI from other sources of cognitive decline (Reed et al., 2007; Graff-Radford, 2019). However, when studying groups of vascular dementia (VaD; n=30) and Alzheimer’s disease (AD; n=30) patients, compared to a normal older adult sample (NE; n=30), Mendez et al. (1997) reported that those with vascular cognitive impairments had greater trouble with executive function than memory function. Meaning, the groups that had VaD and AD had lower overall executive function than they did memory recall. 11 The study also stated the VaD group showed slower cognitive speeds than the AD group. According to Rundek et al. (2022), VCI can be attributed from 20% to 40% of dementia diagnosis. In a population-based autopsy study, Knopman and colleagues (2003) used medical records to identify dementia cases in Rochester, Minnesota, over a four-year period. Investigating 89 neuropathological examination results, they found that 51% were diagnosed with AD, 13% with pure VaD, and 12% with a combined AD and VaD diagnosis. Meaning, vascular disease played a critical role in ≥25% of the observed autopsy dementia cases (Knopman et al., 2003). Moreover, as it relates to the current study, of the observed dementia cases in the study performed by Knopman and colleagues (2003), the age range spanned from 50 to 96 years old, suggesting that dementia and vascular cognitive impairment genesis begins in middle age (45-65 years old). With age, exercise and physical activity tend to decrease while sedentary behavior increases, potentially augmenting rates of dementia, AD, VCI, and cognitive decline (Jaul and Barron, 2017; Rundek et al., 2022). This can be attributed to a multitude of genetic, environmental, psychological, and socioeconomic factors such as age-related frailty (sarcopenia), retirement, a lessened desire to look/feel a certain way, longer recovery times, illness, injury, intimidation, time management, parenthood, and other concerns. That is why it is imperative for researchers to further investigate the short- and long-term effects of exercise training on age- related cognitive decline, which may provide better resolution to the currently theorized mechanisms at work. Underlying Mechanisms 12 The amount of information matching cognitive decline and its cohorts (AD, dementia, VCI, etc.) to specific causations is still being reviewed, researched, and added to constantly, as it is a complex and costly etiology to study. As briefly mentioned, the etiology of Alzheimer’s is commonly defined by the appearance of amyloid-b plaques and tau neurofibrillary entanglements within the brain. However, based on the failures of mass immunization trials along with autopsy evidence indicative of ancillary mechanisms at work, amyloid-B and tau aggregation alone, do not explain the pathophysiology of AD in its entirety (Barnes, 2015). Autopsy evidence has indicated people considered to be cognitively normal individuals may possess significant AD pathology and show no noticeable differences in cognitive function, suggesting there could be more underlying mechanisms at work (Barnes, 2015), which can be interpreted that even cognitive impairments with specific pathologies – thought to be understood – may pose more difficult to diagnose, treat and/or prevent than we think. A topic of interest in this area of research is the role of white and gray matter in the brain as it relates to cognitive decline. It has been documented that overall brain volume and gray matter volume tend to decrease with age, starting around age thirty (Colcombe et al., 2003). Established links between cardiovascular fitness and cognitive performance led Colcombe and colleagues (2003) to examine high-resolution MRI (Magnetic Resonance Imaging) scans of older adults (≥55 y/o; n=55). In the study, the researchers divided the MRI images into maps of gray and white matter, and then assessed any systematic variance in tissue density as it related to aerobic fitness and age. Their findings became the first documented evidence suggesting older adults (≥55 y/o) with higher levels of aerobic fitness retained more brain volume with age. Research using animal models has been reasonably efficient in providing evidence for exercise being beneficial for cognitive function. Animal studies have shown positive correlations 13 between aerobic fitness and a variety of cognitive health markers at the cellular and molecular level in the brain. Some of the positive correlations include: 1) increases in capillary density within the cerebellum of rats (Black et al., 1990); 2) increased brain-derived neurotrophic factor (BDNF) expression in rats (Neeper et al., 1995); 3) increases in hippocampal cell formation in mice (van Praag, Kempermann, & Gage, 1999); and 4) accelerated neurogenesis, as well as a reduction of amyloid-b (Calverley et al., 2020). Neurogenesis is the formation of new neurons in the brain, and as previously mentioned, it is a crucial factor in cognitive ability that also declines with age. Neurogenesis only occurs in certain regions of the brain, such as the subgranular zone in the dentate gyrus, which plays an important role in memory formation, spatial navigation, and pattern separation. It has been hypothesized that, with age, the production/activity of chemicals in the brain, such as BDNF and insulin-like growth factor (IGF-1), that sustain neuroplasticity and neurogenesis, become disturbed and eventually lead to cognitive impairment. Both, macro – such as brain atrophy and network disruptions – and micro – like the aforementioned shifts in neurochemical activity – changes, are likely involved in the underlying mechanisms of age- related cognitive decline (Cadwallader et al., 2022). Which, again, sustained aerobic fitness throughout life could potentially counteract these mechanisms. Vascular Endothelial Growth Factor A molecule of interest for this study, which has not been vastly studied in relation to normal cognitive decline – but more so, rather, in its relation to neurocognitive diseases – is a glycoprotein called vascular endothelial growth factor (VEGF), known as a primary signal protein responsible for angiogenesis (the construction of new blood vessels). This glycoprotein is expressed in endothelial cells, astrocytes, and neurons, and is considered a pleiotropic factor. In 14 addition to angiogenesis, neurogenesis, and enhancing hippocampus-dependent memory, VEGF also plays a role in defending against spatial memory impairment, dysfunction within the blood- brain barrier (BBB), loss of brain cells, and post-injury cognitive decline (Zhao et al., 2019). These beneficiary protective aspects and the suggested negative effects of VEGF deficiency are why it is of interest to the current study. Using animal models, Licht et al. (2011) showed VEGF has the capacity to reverse modulate neuronal plasticity in the synapses of the hippocampus, while Cao et al. (2004) showed that hippocampal VEGF expression is responsive to exercise and environmental stimuli. In brief, Licht et al. (2011) investigated different ways VEGF may affect neurogenesis and cognition in the adult brain by overexpressing exogenous VEGF, and blocking endogenous VEGF expression, in the hippocampus of adult mice. The study reported that animals with loss of VEGF function displayed impaired memory compared to the control, despite no decrease in neurogenesis. To further evaluate that finding, the mice with VEGF expression blocked were subjected to a radial arm maze test and found that those animals had lost spatial learning skills, entirely. The animals with induced overexpression of VEGF showed significant cognitive improvement. Additionally, when the researchers stopped supplying the mice with transgenic VEGF, the improved memory from increased VEGF completely fell off (Licht et al., 2011). The purpose of the Cao et al. (2004) study was to identify molecules that mediate the effects of experience on hippocampal plasticity. Using the place version of the Morris Water Maze (MWM), a hippocampus-dependent variant of the task, and housing its subjects (rats) in an enriched environment or in standard housing, the study looked at hippocampal RNA via real- time RT-PCR (reverse transcription-polymerase chain reaction). The study found that, of the growth factors/molecules assayed, only VEGF expression significantly increased from both 15 stimuli (MWM and enriched environment) compared to the control. However, Cao et al. (2004) also found that exercise alone, without learning involved, did not increase VEGF expression in the hippocampus. The researchers observed the significant increases in VEGF when assessing the protein levels immediately after training, but they also ran RT-PCR analysis 5 days after the last training session and saw that VEGF expression had fallen back down to control levels, which is consistent with the literature. However, even with studies like these, the need to further investigate correlates between VEGF expression and cognition still exists, especially regarding normal age-related cognitive decline. Impacts of Cognitive Decline Cognitive decline and related diseases (dementia, AD, VCI) have become a few of the most prominent health concerns to date, and they are particularly concerning because they affect how the mind works. Cognitive impedances can alter the way a person speaks, thinks, learns, remembers information/people, and makes decisions (Calverley et al., 2020). This is a worrisome concept given an estimated 47 million people around the world suffered from dementia in 2015, alone, and that number is projected to reach a staggering 131.5 million over the next 26 years (Prince et al., 2015; Calverley et al., 2020). Based upon the 2010 census, it was estimated that 9 million adults – independent of other dementias – will have Alzheimer’s disease by the year 2030, in the U.S., alone (Barnes, 2015). Though these numbers only include those individuals that will develop one of the neurodegenerative diseases, these numbers will indirectly affect even more people (i.e., family, friends, colleagues, health care workers, etc.) in numerous ways (mentally, emotionally, physically, financially, etc.). 16 With each occurrence and worsening of cognitive decline, comes the need for treatment and caregiving. In 2011, over 17 million Americans provided aid as a family caregiver to an older adult, with demented adults requiring the most caregiver time (Committee on Family Caregiving for Older Adults et al., 2016). The caregiver position can be outsourced to nurses, certified nursing assistants and personal care aides, or family members can get formal training and carry out the task themselves to save money on outsourcing. However, the latter requires a lot of time and energy that most working adults do not have, therefore, the job is still often outsourced. Globally, in 2015, the estimated cost of dementia was around 818 billion U.S. dollars, and by 2030, it is projected to reach $2 trillion (Jaul and Barron, 2017). In America, alone, total payments (health care, long-term care, and hospice) for people with dementia, 65 years and older, was an estimated $321 billion just in 2022. For AD and other forms of dementia, an estimated 16 billion combined hours of care was provided to those with by more than 11 million unpaid caregivers/family members in 2021 (Alzheimer’s Association, 2022). While some scientists continue researching a viable cure for these neurodegenerative diseases, others must focus on prevention methods, and the same must be done for normal cognitive decline. Until a safe and reliable solution is found, the number of people suffering from different cognitive impairments will continue to increase with the global population. It is likely that, due to their differing pathologies, each neurodegenerative disease will require different treatment. However, normal age-related cognitive decline may pose an easier problem to address. Preventative Measures 17 Thus far, medications for cognitive impairments and neurodegenerative diseases are only marginally effective and tend to treat an already-existing condition. There is an association linking exercise to a reduced risk of developing neurodegenerative diseases shown in the literature (Colcombe and Kramer, 2003), and epidemiological studies have demonstrated greater risk reduction of cognitive decline later in life with higher rates of exercise in early and mid- adulthood (Cadwallader et al., 2022). However, establishing causality between exercise and improved cognitive performance is difficult to do given the potential multitude of factors involved in the relationship. Chronic exercise throughout life, from childhood through old age, seems to be favorable on multiple fronts, but if someone were to not start exercising until later in life, would it even be beneficial? Based on the literature, it appears so. A meta-analysis conducted by Colcombe and Kramer (2003) revealed that aerobic fitness training in sedentary older adults can result in robust cognitive benefits. In fact, even with the multitude of studies they cross-examined using different cognitive tasks, training methods, and participant characteristics, Colcombe and Kramer (2003) found fitness training to increase cognitive performance 0.5 SD on average. A couple of other note-worthy findings from this meta-analysis were: 1) if the study population was >50% female, the group as a whole demonstrated better cognitive improvements than if it were ≥50% male; 2) subjects in the “mid-old” category appeared to benefit the most from exercise; 3) similar improvements from exercise were demonstrated in clinical and nonclinical populations; 4) participants in combined exercise cohorts, performing both strength and aerobic training protocols, improved significantly more than aerobic-only participants; 5) executive function seemed to benefit most from fitness improvements (Colcombe and Kramer, 2003). Since then, the literature suggests greater levels of cardiorespiratory fitness and frequent engagement in 18 exercise can result in better cognitive performance, decreased neural changes (Cadwallader et al. 2022), as well as lower rates of aging decline in gray matter (Barnes, 2015). Still, the pathology/causality of the connection between exercise and cognitive longevity is not well- established. In 2006, a longitudinal study, following cognitively normal older adults for six years, found that those who exercised ≥3 times per week had a greater likelihood of staying dementia- free, independent of other dementia risk factors (Barnes, 2015). Barnes and Yaffe (2011) revealed that, given the modifiable risk factors for AD (diabetes, obesity, hypertension, smoking, physical and cognitive inactivity), the most statistically effective way to combat AD would be to increase the percentage of the population that is physical active by 25%. They predicted this 25% increase of physically active people could potentially prevent 230,000 cases of AD in the United States alone. Though that specific suggestion by Barnes and Yaffe (2011) only pertains to Alzheimer’s disease, which has specific pathological indices, it still illustrates the belief among researchers in the power of exercise/PA. The literature shows benefits for both, but majority results indicate exercise to be the quickest and most efficient of the two. With exercise, blood flow in and throughout the brain increases as cardiac output increases, and exercise creates changes in the brain’s metabolism. As previously mentioned, studies have shown vascular risk correlated to dementia and AD, but similarly, so is cardiac disease, hypertension, obesity, and diabetes (Barnes, 2015). On that note, exercise-induced VEGF secretion and angiogenesis promotes vasodilation (Calverley et al., 2020), which could assist in countering hypertension, vascular risk and cardiac disease. When comparing high- intensity interval training (HIIT) and high-intensity functional training (HIFT), there was a component of HIIT shown to induce VEGF secretion (Morland et al., 2017) during exercise. 19 HIIT has also been found to assist in O2 homeostasis of the brain’s blood vessels (Calverley et al., 2020). However, while few studies have been specifically devoted to understanding the beneficial impact of HIIT on cerebrovascular function, the physiological benefits of HIIT in combating diseases such as stroke (which has a high correlation to the development of neurodegenerative diseases), cancer, diabetes and hypertension have been well-established. With cardiovascular risks being so closely related to cognitive impairment and dementia, HIIT possesses the potential to further improve brain health in adults via its effects on cardiorespiratory fitness (Calverley et al., 2020). Chapter III: Methods Subjects C57BL/6 male mice (n = 16) – commonly used in aging studies – were received from The Jackson Laboratory in Sacramento, California. Mice were received at the age of 10-months old (10m), equivalent to the start of middle age in humans (Catherine Hagan, 2017; Pajski et al. 2024), and were tested/trained through the age of 18-months old (18m). Within a week of receiving the mice, one mouse died of natural causes before pre-testing began, therefore n = 15 for the study. Under the approval of the ECU Institutional Animal Care and Use Committee (IACUC) protocols, the mice were humanely handled and treated. Mice were housed and cared for on a 12-hour light/dark cycle at 22°C, with ad libitum access to food and water. The mice were acclimated for 7 days prior to pre-intervention functional testing. After performance assessments (physical and cognitive) were complete, the mice were randomly assigned to one of two groups: sedentary control (SED; n=7) and High-intensity Interval Training (HIIT; n=8). Figure 1: Study Design 21 Performance Assessments All functional and cognitive assessments were performed by the same 2 investigators. Dr. Ted Graber conducted the functional assessments and contractile physiology to ensure consistency with previous studies, while Justin Stephenson performed all CAB and body composition testing. Baseline and end-point assessments were performed in the same order and at the same time of day. Maintenance exercise is performed during the post-testing period as it takes ~2 weeks to get through all functional and cognitive testing. Physical Function Assessments Comprehensive functional assessment battery (CFAB) score determination was evaluated via functional tests that include Grip Meter (forelimb strength), Inverted Cling (full-body strength/endurance), Volitional Wheel Running (VWR; individual activity levels), Rotarod (overall motor function, gait speed, balance/coordination, power generation), and Treadmill (endurance/aerobic exercise capacity). CFAB testing is performed pre- and post-intervention, with an additional treadmill test at the intervention halfway point to keep exercise intensities relative to the subjects’ current fitness levels. Data results from each functional test are converted into a numerical score. The test scores are then combined and converted into a single composite CFAB score. Summaries of each test involved in CFAB have been previously described (Graber et al., 2015; Graber et al., 2020), but a brief overview of each one is provided below. Grip Meter To assess forelimb strength, the Grip Meter utilizes the animals’ innate reaction to grab ahold and cling to anything in reach when being lifted by their tail. Using a grip strength test 22 (BIOSEP In Vivo Research Instruments, Model GT3), the mice are picked up by the base of their tail and gently swung forward over the grip bar running horizontally in front of them. When the mice instinctively grab the bar, we begin gently pulling straight back away from the device, which measures the maximum amount of force (newtons; N) applied until the mice release their grip. This is all performed in one swift motion and repeated 5 times in succession. Only when a mouse grips the bar incorrectly (monkey grip, grabs the center connecting bar, only grasps with one paw, etc.) or does not achieve a quality grip (paws slip off too quick) is the motion repeated more than 5 times. Each value is recorded in N, and when all 5 attempts are completed, the mouse is returned to its home cage. The outcome measure is the best score of 5 normalized to the mass of the mouse (mN/gbm, milliNewton divided by grams of body mass). Inverted Cling While grip meter tests forelimb strength, inverted cling assesses four limb strength and stamina. The test requires the mice cling to a metal grid cage while hanging upside down. Using a plexiglass box with a metal cage door as the top of the box, the mice are placed on the cage door while it is open in the vertical position. Once they latch onto it, the door is swiftly shut, inverting the mice as they hang onto the metal caging. Again, utilizing the animals’ instinctive nature of not wanting to fall, they cling to the metal grid as long as they can using all four of their limbs. The data value recorded is the amount of time (seconds; s) spent clinging to the cage. Inverted cling data is reported as seconds (s) and Log10 of these values. This is only repeated twice unless the mice fall before reaching the 10 second mark. The outcome measure is the best time (longest) of the two trials. 23 Volitional Wheel Running Volitional wheel running (VWR) is a measure of each mouse’s individual voluntary physical activity. Eight cages are equipped with a running wheel connected to a computer and mice are individually placed in a cage – with home-cage bedding and food added to help them accommodate – where they will reside for 7 days (1 week). Over these 7 days, mouse activity, in total wheel revolutions, is recorded every 10 minutes. Total revolutions are recorded at the end of the week and converted to meters (m), where one revolution is equal to 0.4 meters. Once the 7 days are up, the mice are taken from their VWR cage and returned to their home cage. The data value recorded is the number of wheel rotations completed throughout that week. The outcome measure is meters per day of VWR. Rotarod For assessment of balance, coordination and overall motor function, the rotarod (PanLab LE8205 ROTA-ROD, Harvard Apparatus) test is administered. The mice are placed on a grooved horizontal rod (30 mm), separated by dividers (50 mm wide lanes), where their only objective is to stay on the rotating rod. The test starts when the rod begins to turn over, and the mice must stay on while walking at increasing speeds. When mice fall off the rotating rod, they land on a lane-specific trigger plate that starts and stops the timer measuring time spent on the rotarod. Due to task difficulty, this test is a three-day process, where the first two days are acclimation/learning days, and day 3 is the actual test day. Each day consists of three trials with differing speeds and time. The first trial on day 1 is ran at 4 rotations/min for 2 minutes max, and in trials 2 and 3, the rotation speed accelerates at the rate of 4 to 40 rpm over 600 seconds (10 minutes) for 2 minutes max. For day 2, the first and 24 second trials are the same as day 1, but trial 3 increases the difficulty by accelerating at the rate of 4 to 40 rpm over 300s for 1 minute max. Test day has 3 trials accelerating at the rate of 4 to 40 over 300s until the mice fall off the device. There is a minimum 15-minute break between each trial. On test day, the data recorded is the amount of time (s) taken for each mouse to fall off the rotarod. The outcome measure is the best time of the three trials. Treadmill A PanLab LE 8710 treadmill is used to administer the last functional test, which consists of two acclimation days and one max speed test day. Day 1, the mice are placed on the treadmill and walked at 4 meters/min for 3 minutes. The number of shocks and visits to the shock grid are recorded, along with the speed (m/min), time (mm:ss), and distance (m) covered. Day 2, the mice are walked again at 4 m/min for 2 minutes, and then they are introduced to acceleration. The speed starts at 4 m/min and accelerates 0.6 m/min for 2 minutes max. The same variables recorded as day 1. For test day, the mice start at 4 m/min and accelerate 0.6 m/min with a 30- minute maximum. They are only allowed 4 shocks and 6 visits to the shock grid, once they reach those limits, they are removed from the treadmill unless otherwise noted. For example, if a mouse gets an early shock during the walking phase of the test because they were trying to escape out of the back but are still running fine as the speed becomes more intense, when they technically “shock out,” they are allotted one more visit/shock before removal. This is because we are trying to get as true a max speed as possible from them. The same variables are recorded as day 1 and 2, and the fastest speed they reached before shocking out is the max speed used to design the training sessions for the HIIT group later in the study. The outcome measurement is the time in seconds (s) on the max speed test. 25 CFAB Data Analysis Traditionally, CFAB data analysis consists of using a reference group of 6-month-old mice (mean and standard deviation; SD), and test values are standardized by assessing the distance of each individual mouse’s score from the average 6m value. Distance from the 6m mean is measured in units of the reference group’s SD. The standardized scores for each functional test are then added together to form the CFAB composite score, a single numeric value representative of total physical function capacity (Graber et al., 2020). However, an Intervention Assessment Value (IAV) was used for analysis in the current study. Rather than using the reference group of 6m mice, the baseline for standardization was the pre-test mean and standard deviation of the entire sample (n=15), prior to randomization into individual groups. This is similar to the FIAV previously explained and will be conducted in the same manner (Graber et al., 2015). Other Physical Measures To examine and compare body composition from pre-intervention to post-intervention, EchoMRI was utilized. The EchoMRI-700TM is a quantitative nuclear magnetic resonance system that allows precise, whole-body composition measurements to be taken in live mice from 10-130 grams. One at a time, the mice are placed into an appropriately sized tube with an adjustable barrier used to confine and minimize movement (though a little wiggle room does not compromise measurements). The system automatically analyzes the animal’s Lean Mass, Fat Mass, F. Water and T. Water in units of grams (g), and then stores the numeric results in a 26 database. All of this is accomplished without any need for, or use of, anesthesia or sedation of the animals. In vivo contractile physiology was performed to assess muscle performance via dorsiflexor torque, as detailed previously (Graber et al., 2021). Cognitive Function Assessments A comprehensive cognitive assessment battery (CAB) was used for cognitive assessment. Cognitive performance was determined through the application of memory, behavioral and executive function tasks. The tests involved in CAB included Y-maze (exploratory behavior and spatial working memory), Open Field (anxiety, locomotor and exploratory behavior), Novel Object Recognition (exploratory behavior and long-term memory), and Puzzle Box (short-term memory and executive function), in that order. Scoring for the cognitive tests was performed in the same manner as outlined for the Frailty Intervention Assessment Value or CFAB previously outlined (Graber, et al., 2015; Graber, et al., 2021). CAB testing was performed pre- and pos- intervention, with maintenance exercise performed during post-testing period. All cognitive/behavioral assessments were video recorded with a GoPro HERO6 Black for later analysis and data quantification. Open Field Open Field (OF) is a commonly used behavioral test for assessing general locomotor activity, anxiety levels, and exploratory behavior in mice (Creighton et al. 2019; Szatmari et al. 2021). The testing arena was a 58.1x58.1x40cm box, made of a non-abrasive plastic, with an open top for proper lighting and video recording. Before testing, light distribution was assessed 27 using Light Meter LM-3000 to ensure uniformity of brightness (750 ± 10 lux) across the testing arena. Direct lighting was applied via a single LED lamp over the center of the arena. Each animal is individually placed in the OF box and is allotted 8 minutes to freely move and explore the arena. After 8 minutes, the subject is moved back to its home cage. For this assessment, video recording is taken from above and saved for later analysis. Data points of interest for the OF assessment include: number of line crossings, entries into the center, defecations, urinations, and the percentage of total time spent in the center and perimeter. The data is measured and quantified manually by imposing a 4x4 grid on top of the video footage (as shown in Figure 2) and tracking desired variables based on movements throughout the grid. That data is then recorded in the data table. The OF arena is cleaned with 70% ethanol before testing begins so that the first mouse gets the same experience as the rest, between each trial, and at the end of testing. The outcome measures for OF are the number of entries into the center and time spent in the perimeter. Y-maze To assess spatial working memory, as has been done before (Sabaghi et al., 2019), this task required the animals be individually placed into a symmetrical Y-shaped arena with a non- reflective, neutral-colored surface (i.e., white or beige). The arms of the maze were 41.5cm (about 16.34in) long, 8.5cm (about 3.35in) wide, with 12.4cm (about 4.88in) high walls, and evenly dispersed around the triangular center connection (120° apart). Before testing, light Figure 2: Zone overlay used to track Open Field behavior 28 distribution is assessed using Light Meter LM-3000 to ensure light uniformity (700 ± 50 Lux) throughout the testing arena. Each mouse is positioned into any one arm (designated Arm A) facing towards the maze's center. They are then allotted 8 minutes of uninterrupted exploration time while being recorded from above to allow live monitoring throughout. The video recording is later used for data collection. The inside of the Y-maze was cleaned with 70% ethanol after each mouse. For this experiment, entry into an arm is defined by the animal having all four paws inside the arm. Any delay/hesitation from the mouse to leave its start arm is also measured and is recorded in the Latency column of the data table. The criteria required to be considered a spontaneous alternation (SA) in this experiment is defined as sequential entries into all three arms in overlapping triplet sets (i.e., ABC, BCA, CAB, or vice vera) as shown in Table 1. Total entries are counted as an indicator of locomotor activity, whilst any latency of a mouse to leave its starting arm is an indication of emotionality-related behavior. The percentage of spontaneous alternation performance (%SAP) equals the ratio of total alternations (ToA) to possible alternations (PoA; SA/(total arm entries–2)x100). The outcome measure is the percentage of spontaneous alternation performance (%SAP). See Table 1 for more details. Table 1: Example of Spontaneous Alternation Counts. The first row shows arm entries/total alternations made throughout the test duration, not including the start arm. Triplets are counted in the rows below. Spontaneous alternations (SAs) made in triplets, as seen as in Triplet 2–4 and Triplet 6, are shown in bold. In this example, there are 4 SAs and 16 arm entries. Therefore, the %SAP = (4/(16-2))x100 = 28.57%. Arm Entry A B A C B A B C C A B A B C B B Triplet 1 A B A Triplet 2 B A C Triplet 3 A C B Triplet 4 C B A Triplet 5 B A B Triplet 6 A B C 29 Novel Object Recognition The Novel Object Recognition (NOR) test was administered to assess long-term memory and exploratory behavior in mice, as has been done before (e.g., see Creighton et al. 2019; Szatmari et al. 2021). For this test, each mouse is habituated to the open field (OF) arena for 8 minutes via the OF test (day 1). Again, video recording will take place from above the arena for later data collection and quantification. For the familiarization phase, 24 hours after habituation to the space, the mice are returned to the OF arena where they are then exposed to two identical, non-toxic objects (made of metal or rigid plastic) for 5 minutes (day 2). The amount of time spent exploring each object is measured and recorded by the researcher, where nose entries into the surrounding area (2cm2) of each object is considered “time exploring” the object. After familiarization, the animal is placed back in its home cage, where they reside for a 24-hour retention interval. A 24-hour interval is used for assessing long-term memory. After the retention interval, the test, or “choice phase,” is administered on day 3 by returning the animal to the OF box where two objects – one familiar object (FO) from day 2, and one novel object (NO) – reside. The mouse is allotted 2 minutes to explore the objects, and the amount of time spent exploring each object is recorded with the same criteria as the training session. The OF arena and all objects used are cleaned with 70% ethanol before and after each session. The NOR assessment is a type of assay used to assess human amnesia with rodents. Discrimination of novel versus familiar stimuli requires intact perceptual systems. Therefore, if the mouse spends more time exploring a novel object compared to a familiar object, it is indicative of an intact memory (Creighton et al. 2019). To measure this discrimination with NOR, a discrimination ratio (DR) is calculated to quantify novelty preference by subtracting the time spent exploring the familiar object from time spent exploring the novel object and dividing 30 that by the total object exploration time (DR = (NO – FO) / total exploration). The outcome measure for this test is the discrimination ratio. Puzzle Box The puzzle box test was designed to assess executive function skills in the animals via working memory and problem-solving. The test was adapted from previously used versions (Ben Abdallah et al., 2011; Shepard et al., 2017). For this adaptation of the puzzle box assessment, we are using the OF arena again, except it now has a hole in one of its walls, 2.2cm (about 0.87in) off the floor and 6cm (about 2.36in) in diameter. A PVC pipe connected the OF arena to a much smaller “puzzle box” (17.3x21x17.6cm). The OF box was the wide-open, brightly illuminated starting area, while the smaller, covered puzzle box was the objective area. To access the puzzle box, the mice had to climb into the tunnel and make their way across. All Puzzle Box tests began with each mouse being individually positioned in the center of the wall directly across from the puzzle box access, and assessed on how long it took them to complete their objectives. A treat/prize (i.e., unsalted walnut, almond, or plain Cheerio) was used as an additional incentive. With each test, modifications were made to increase the difficulty of reaching the puzzle box. The mice were faced with various challenges such as an upward-bent entrance (day 3), a blocked exit (day 3), and differing entrance points (day 4). Days 1 and 2 had no obstacles for entering or exiting the tunnel to the puzzle box. For the first day, they were given 5 minutes to acclimate to the new arena, tunnel and box. The second day, they were allotted 4 minutes to further investigate on their own, but if they did not willingly enter the tunnel and/or box, they were placed in the tunnel. This was done so that all of the animals understood it was safe to enter the tunnel/box if they had not willingly done so yet. 31 Day 3 consisted of the Bent Entrance test and the Blocked Exit test. Data from the Bent Entrance task was not counted toward final CAB scores nor repeated in post-testing, as it wound up being too difficult for the mice to complete due to physical limitations. The Blocked Exit test consisted of the normal entrance, but with a piece of plastic located at the end of the tunnel that the mice had to knock out of place to access the puzzle box/incentive. Analysis of this test consisted of a 0-1 scoring system based on completion of certain tasks. Subjects were allotted a 0 for failing to complete each task or a 1 for completing each task. The tasks were scored upon whether the mice entered the tunnel and knocked out the block to access the puzzle box. Therefore, if the mice did not complete either task, they received a score of 0, and if they completed both, they received a score of 2. The outcome measure for this test was the total score achieved. The Differing Entrance Points test was held on day 4, which had two 5-min trials, where time to enter the tunnel was recorded. In the first trial (T1), connected to the tunnel’s entrance was a PVC connection piece, bent at a 90° angle, facing the right side of the arena. The mice were placed near the center of the wall opposite the tunnel and allotted 5 minutes to explore. Should they have entered the tunnel, that time was recorded. After a 5-min retention interval, the mice were placed back in the OF arena for the second trial (T2), where the entrance to tunnel was now facing left. The amount of time taken to figure out where to enter the tunnel was recorded. We also noted if the mice visited the right side of the tunnel first, remembering the entrance was on that side, before going to the left side and discovering the entrance there. The outcome measure for this test was the difference in time (s) to enter between trials 1 and 2. CAB Data Analysis 32 The CAB outcome measures were analyzed individually rather than with a compound scoring system. Intervention After baseline testing and data analysis, the mice are randomly assigned to one of two groups: high-intensity interval training (HIIT, n=8) or sedentary control (SED, n=7). Mice were group-housed in cages with environmental enhancements, such as a block or tunnel to play with, but no means of exercise (i.e., running wheels) available aside from general PA. Some mice needed to be individually housed due to over-aggression and fighting. The intervention consists of one acclimation week, followed by 12 weeks of HIIT. Training is held 3x/week, one session every other day (i.e., Monday, Wednesday, Friday) to allow for rest and recovery between exercise sessions. SED mice are brought up to the lab at the same time as the HIIT group for enrichment purposes and placed on the treadmill as sham treatment for the same duration as training, but they are not exercised. High-Intensity Interval Training The average max speed (Speedmax) of the mice, as measured by the baseline treadmill test, was used to determine percent max speed (%Speedmax) for the HIIT intervals. Based on their fastest recorded speeds, mice were grouped into exercise cohorts, where cohorts of ≥2 animals with a similar Speedmax were exercised together. Subjects were exercised three days per week with one rest day between exercise days (i.e., Monday, Wednesday and Friday), and two rest days over the weekend. Each HIIT session began with a 2-min warm-up at base speed (4 m/min), followed by 1-min intervals at sprint speed (with 30-second ramp-ups and 30-second ramp- 33 downs, for a total of 2 minutes per interval) interspersed with a 1-min relative rest (walking speed). After the final HIIT interval, each session concluded with a 2-min cool-down at base speed. See Figure 3. For the acclimation week, one interval was run on the first day, two intervals on the second day, and three intervals on the third. By the first week of training, subjects were running three HIIT intervals at 75% of their predetermined average Speedmax and transitioned to higher percentages as the intervention progressed. Additional intervals were added as tolerated, up to a maximum of five, as the intensity/speed was also increased to the tolerance of the mice. If a mouse could not complete its scheduled training at the expected intensity, it was relegated to the next slowest group until improvements were shown. After the 6th week (mid-point) of the intervention, mice were retested for a new max speed and their relative interval speeds were adjusted accordingly. The SED group was brought up to the lab and placed on the treadmill every day that the HIIT group was trained as a sham treatment. SED mice were not exercised, but they were left on the treadmill for the same amount of time as the HIIT session that day. For example, if training lasted 15 minutes for a 4-interval training session (see Figure 3), the SED mice were left on the treadmill for at least 15 minutes. Additionally, because the HIIT mice were exposed to an active shock grid, the grid was also active when the SED mice were on the treadmill for sham treatment. This was pertinent Figure 3: HIIT Session Design for Group 1, Week 4. The lane for each mouse is indicated beside its identification (L1 = Lane 1, L2 = Lane 2, etc.). For each step, the start and end speeds (m/min), and duration (seconds) are entered. The treadmill automatically acceler- ates/decelerates at a steady rate from the start to end speed, based on the duration of each step. The HIIT-interval intensities are listed in the top right corner (2x75% means there are 2 intervals at 75% max speed). 34 due to the cognitive aspect of the study; since the HIIT group had an opportunity to gain experience from handling and shock exposure, we felt it important for the SED group to have that same opportunity. The purpose of the sham treatment was to ensure both groups had as similar an experience as possible, minus the variable of training. That way, any potential cognitive differences observed between the groups, pre- to post-, could be a supposed result of the HIIT training, alone. Tissue Collection Following the exercise training period and post-intervention performance testing, the animals underwent non-survival surgery under deep anesthesia (100 mg/ml ketamine mixed with 100mg/ml xylazine) for tissue collection. Anesthesia depth was assessed by checking for a toe pinch reflex, followed by euthanasia via exsanguination and removal of the heart. Brain and muscle tissue was collected and flash-frozen in liquid nitrogen before being stored at -80ºC (VWR -80oC Freezer) for future use in data collection operations (Homogenizations and Western Blots). Subjects in the HIIT group underwent maintenance training during this process (2 days/week, at subjective %Speedmax max speed). Statistical Analysis All baseline characteristics (age, weight, and body composition) and scores/IAVs (CFAB and CAB assessments) were compiled and reported in Tables 2-10, along with skew, kurtosis, normality and means testing. Independent samples t-tests were used to compare mean differences between the HIIT and SED groups performance scores on CFAB and CAB assessments. Paired samples t-tests compared changes within the groups. CFAB functional determinants were 35 assessed using 2x2 mixed-design ANOVA (see results for more details). Differences were deemed significant at p<0.05. Data were expressed in units of mean ± SE (standard error), unless otherwise indicated. Chapter IV: Results We used Student’s Independent Samples and Paired T-Tests to compare the dependent variables with the results reported in the appropriate tables, alongside mean, SD, SEM, skew, kurtosis and the results of the Kolmogorov-Smirnov and Shapiro-Wilks tests for normality. Note that we have previously determined and published normality results for the CFAB determinants; all are normally distributed with the exception of inverted cling (Graber, et al. 2015; 2016; 2021) which is log10 transformed to normality. For CFAB functional measurements, a 2x2 mixed ANOVA (2 groups: SED and HIIT; and 2 time points: pre- and post-intervention) was conducted and the results are detailed in the text below. Physical Function Assessments Physical function was appraised using five well-established determinants, pre- and post- intervention, and used them to measure overall function with the IAV (intervention assessment value). See Tables 2-3 and Figures 4-8 and 13 for more details. Grip Meter, Inverted Cling, VWR, and Rotarod There were no significant changes in the IAV determinants of grip meter, inverted cling, voluntary wheel running, or rotarod: normalized grip meter (within-subjects effects of time*groups F=0.849, p=0.374, partial η2=0.061; between-subjects effects of time*groups F=0.941, p=0.350, partial η2=0.10), inverted cling (within-subjects effects of time*groups F=0.002, p=0.963, partial η2=0.000; between-subjects effects of time*groups F=0.578, p=0.461, partial η2=0.043), voluntary wheel running (within-subjects effects of time*groups F=0.134, p=0.720, partial η2=0.100; between-subjects effects of time*groups F=3.202, p=0.097, partial 37 η2=0.198), rotarod (within-subjects effects of time*groups F=0.013, p=0.909, partial η2=0.001; between-subjects effects of time*groups F=0.937, p=0.351, partial η2=0.067), or in overall IAV (within-subjects effects of time*groups F=0.070, p=0.795, partial η2=0.005; between-subjects effects of time*groups F=0.037, p=0.851, partial η2=0.003). See Figures 4–7 and 13 for more details. Figure 4: Grip Meter Pre- to Post-Training Figure 6: Inverted Cling (left) and Inverted Cling Log10 (right) Pre- to Post-Training Figure 5: Volitional Wheel Running (right) Pre- to Post- Training 38 Treadmill (See Figure 8) Treadmill time was significantly increased by 28.0% post-training in the HIIT group while the SED group declined in performance by 41.4% (within-subjects effects of time*groups F=21.381, p<0.001, partial η2=0.622; between-subjects effects of time*groups F=5.572, p=0.035, partial η2=0.300). HIIT increased a mean of 138.1 seconds and SED declined a mean of 217.7 seconds from pre- to post-intervention (post-hoc testing by independent samples t-test t=5.572, p<0.001; within-subjects changes were both significant and had large effect sizes: paired samples t-test SED t=3.901, p<0.008, Cohen’s d=-1.475; HIIT t=2.612, p=0.035, Cohen’s d=0.923). Figure 8: Treadmill Pre- to Post-Training Figure 7: Rotarod Pre- to Post-Training 39 Other Physical Measurements Body Composition Body mass of all mice was measured weekly (See Figure 9), at euthanasia and prior to each EchoMRI. There was no difference in any of the measurements prior to training. The difference between body mass, fat, and fat% measured at the time of EchoMRI was significantly larger within groups from pre- to post-training (2x2 Repeated Measures ANOVA: F=20.062, 46.845, and 35.899, respectively; all p<0.001). Between subjects’ lean mass tended to decrease in SED and remain the same in HIIT from pre- to post-training (2x2 Repeated Measures ANOVA: F=3.551, p=0.082); although, while the lean mass difference between groups had a strong effect size (-0.746), it was not significantly different (independent samples t-test, and p=0.173). See Tables 4-5 and Figure 9 for more details. Muscle Mass Muscles were collected at euthanasia, where the gastrocnemius (GAS, p=0.444), plantaris (Plant, p=0.107), tibialis anterior (TA, p=0.868), extensor digitorum longus (EDL, p=0.732), soleus Figure 9: Average Body Mass per Week Across the Intervention 40 (SOL, p=0.477), and heart (p=0.293) muscle masses were measured. Between-groups analysis showed no significant difference for each muscle mass or total muscle mass (p=0.797). See Table 6 for more details. in vivo Contractile Physiology Maximum isometric plantar flexor torque was measured pre- and post-training. There was no difference in groups prior to training. We found no significant change in either maximum torque (mN*m) or normalized torque (mN*m/grams of body mass) following training. See Tables 7-8 for more details. Cognitive Function Assessments We measured cognitive function with previously described testing protocols pre- and post- intervention and performed means testing with independent samples t-tests between groups and paired samples t-tests within groups. See Tables 9-10 for further details. Open Field For the OF test, entries into the center and time spent in the perimeter were measured, with the HIIT group spending significantly more time in the perimeter (p=0.016), which is indicative of greater anxiety-related behavior. The SED group showed trends toward a similar result, with fewer center entries and more time spent in the perimeter zone, but nothing of significance. There was no association observed between defecations and anxiety. See Tables 9-10 and Figure 10 for more details. 41 Figure 10: Open Field Pre- to Post-Training Y-Maze The Y-maze test assessed the subjects’ spatial reference memory by measuring the number of Spontaneous Alternations (SAs) made in the allotted test time. This quantification was then made into a percentage of the total number of alternations made within the maze, providing us with the Percentage of Spontaneous Alternation Performance (%SAP; =SAs/[total arm entries–2]). Both groups significantly increased in the number of spontaneous alternations (HIIT, p=0.002; SED, p=0.013) from pre to post. However, with the increase in SAs, both groups also significantly increased in total arm entries (HIIT, p=0.001; SED, p=0.007), therefore showing no significant improvements in %SAP. Though both groups significantly increased in total arm entries, the 42 HIIT group still increased more, which is indicative of higher levels of anxiety. This increase in anxiety, from pre to post, was also seen in the HIIT group’s open field tests. Novel Object Recognition There were no significant changes in NOR (p=0.757) from pre- to post-training. See Table 9 and Figure 14 for more details. Puzzle Box There were no significant changes in the puzzle box tests (Blocked Exit, p=0.689; Different Entrances, p=0.378). However, for the puzzle box blocked exit task, between-groups analysis showed the SED group was significantly faster at entering the tunnel (p=0.036) pre-intervention but could not replicate that significance in post-testing (p=0.306; HIIT increased a mean of 17 seconds; SED declined a mean of 39.8 seconds from pre- to post-intervention), indicating a potential equalization in working memory between the two groups. Though between-groups analysis did show significance for other test variables collected, none fit normality. For example, * * Figure 11: Y-Maze Pre- to Post-Training. Spontaneous alternations (left) were counted to assess spatial working memory, which increased more for the HIIT group (+49 total) than the SED group (+33 total), but the change in percentage of SA performance (right) show. 43 between-groups analysis showed the SED group did significantly better finding their way into the second entrance of the different entrances task (p=0.005) pre-intervention but did not prove to be significantly better in the post assessment (p=0.904). There was also no significance observed in the blocked exit task pre-intervention, but the SED group was significantly faster than HIIT at removing the obstruction (p<0.001) to the puzzle box during post-testing. See Tables 9-10 and Figure 12. Exercise Volume/Intensity Throughout the intervention, the HIIT group’s exercise intensity and work performed increased relative to how each mouse was responding to their current load. Over the course of the HIIT intervention, exercise intensity (%Speedmax) increased an average of 14.36%. Between the midway point (week 6) treadmill retest and the last day of training (week 12), exercise intensity increased an average of 15.27%. The amount of work ([m*g]/min) was calculated for each HIIT session performed, then the intervention (12 weeks) was divided into quarters to obtain the mean Figure 12: Puzzle Box Pre- to Post-Training. Total scores on the Blocked Exit task are shown on the left. The difference of the Trial 1 to Trial 2 differences for the Differing Entrances task are shown on the right. 44 quarterly work (MQW) performed by each mouse. The average MQW difference was equal to 38; meaning, when combined, the HIIT mice increased work performed by 38 units on average. Chapter V: Discussion As medicine and technology advance, so, too, does the average human life expectancy. From 1900 to 2021, the average life expectancy of a newborn increased from 32 years to 71 years (Dattani et al., 2023). One concern of this evolution is that we may be starting to physically outlive our mental capacity, resulting in millions developing different age-related cognitive impairments, requiring even more family caregivers, and hundreds of billions in U.S. dollars spent on medical care for such impairments, annually (Calverley et al., 2020; Jaul and Barron, 2017). According to the literature, the effects of exercise on cognitive function seem promising (Cotman & Berchtold, 2007; Ahlskog et al., 2011; Kirk-Sanchez & McGough, 2014), though there is some disagreement (Gates et al., 2013; Sanders et al., 2020). Precisely how exercise promotes cognitive health in middle- to older-ages, and which exercise modalities provide the most benefit is still uncertain. The current study was designed to determine the effects of High-Intensity Interval Training (HIIT) on functional and cognitive performance in a group of middle-aged C57BL/6 male mice, compared to a sedentary control group (SED). We hypothesized the HIIT group (n = 8) would exhibit greater resistance to age-related cognitive decline and maintain greater physical function, pre- to post-intervention, than the SED group (n = 7). We utilized a 12-week treadmill HIIT protocol for the intervention, assessing physical and cognitive performance with the Comprehensive Functional Assessment Battery (CFAB) and a Cognitive Assessment Battery (CAB), respectively. For CFAB, max speed and aerobic capacity (Treadmill), four-limb grip strength and muscular endurance (Inverted Cling), forelimb strength (Grip Meter), voluntary exercise capacity (VWR), and overall motor function (Rotarod) were tested. CAB assessed anxiety and exploratory behaviors (Open Field), long-term memory and exploratory behavior 46 (NOR), spatial working memory (Y-maze), as well as short-term memory and executive function (Puzzle Box). EchoMRI and in vivo contractile physiology were used to measure body composition and maximum plantar flexor torque, respectively. All measures were assessed pre- and post-intervention. Exercise and Physical Function The 12-week treadmill HIIT protocol was adapted from previous studies (Seldeen et al., 2018; Graber et al., 2021; Pajski et al., 2021). The comprehensive functional assessment battery is a well-validated means of measuring physical function across the life span of a mouse. The Graber et al. (2021) study that validated the use of CFAB, assessed frailty in male mice at 6m, 24m, and 28+m of age, and found overall physical function decreased with age. Seldeen et al. (2018) used mice 22m of age, while findings from Pajski et al. (2024) were based on mice aged 6m (start)– 10m (end), and 22m (start)–26m (end). Therefore, because there are no CFAB data for middle- aged (10m–22m) male mice, the CFAB data collected in this study adds to the body of literature assessing physical function over middle-ages in mice. CFAB, administered pre- and post-intervention (10m and 17m old, respectively), showed no significant changes in the intervention assessment value (IAV) determinants of grip meter, inverted cling, VWR, rotarod, or in overall IAV. Therefore, there was no statistically significant improvement in overall physical function. This contrasts with the findings of Seldeen et al. (2018; 2019) and Pajski et al. (2024), who observed improvements in, or preservation of, physical function following similar HIIT interventions. However, Seldeen et al. did not observe improvements in rotarod for the HIIT group in their 2018 study, nor did they observe significant changes in rotarod or inverted cling for either group (SED or HIIT) in their 2019 study. Again, it 47 should be noted that these studies investigated different age groups from the current study, which could potentially explain some of the differences in physical function observed. Though the previously mentioned physical function assessments used for CFAB did not reveal any significant improvements, the 12wk HIIT intervention did result in marked improvements in our HIIT group’s aerobic capacity and treadmill speeds, while the sedentary control mice declined. This finding is consistent with the literature (Seldeen et al., 2018; 2019; Pajski et al., 2024). Exercise and Body Composition For body composition, we observed significant increases between body mass, fat, and fat % within groups from pre- to post-training, and a strong effect size for lean mass difference between groups but it was not significant. Using older-adult (24m) mice, Seldeen et al. (2018) reported significant declines in fat % for their SED mice, while their HIIT group exhibited no such decline, maintaining fat % more than the sedentary control group. Pajski et al. (2024) reported marked fat % declines in both of their 26m exercise groups (RUN and HIIT) and fat % increases in all of their 10m groups (RUN, HIIT, and control). However, though the 10m exercise groups increased in fat % (RUN, +51.7%; HIIT, +100.9%), the 10m control group (CON) increased 160.5%, which is indicative of exercise’s potential to mitigate increases in body fat %. The natural body composition patterns seen in humans and rodents may explain the findings of Seldeen et al. (2018), Pajski et al. (2024), and the current study. Nagy & Pappas (2019) found, on average, peak fat mass in mice – provided food ad libitum – is between 12 and 24m (40-80 years in humans), and fat mass decline is observed between 17 and 24m (57-80 48 years in humans). As our mice were last assessed for body composition at 16m old – just before fat mass decline begins, on average – it was reasonable for our mice to continue gaining fat, regardless of training or sedentation. Similarly, the review from Nagy & Pappas (2019) explains the drop in fat % for the 24m-old and 26m-old mice in the Seldeen et al. (2018) and Pajski et al. (2024) studies, respectively, as well as the fat % increase in the 10m-old mice in the Pajski et al. (2024) study. Exercise and Cognitive Function The battery of cognitive assessments developed for the current study consisted of well-validated tests for analyzing cognitive performance in rodents and are used throughout the literature. Though the mice did significantly improve in certain areas of the battery from pre- to post- intervention, we observed no cognitive changes of any significance between the two groups. If one group significantly outperformed their baseline results, the other group appeared to do so as well. Therefore, based on what we observed, neither group declined in cognitive function more than the other, as we hypothesized the SED group would. While these findings may seem to be in contrast to the literature, as many studies report exercise to benefit cognitive function and/or aid cognitive impairment (Cotman & Berchtold, 2007; Ahlskog et al., 2011; Kirk-Sanchez & McGough, 2014; Sabaghi et al., 2019), this is not the case for all (Gates et al., 2013; Sanders et al., 2020). In human research, there are mixed reviews on whether exercise, or specific types of exercise, have any significant effect on cognition. Gates et al. (2013) wrote a review on the effect of exercise on cognitive function in older adults (65-95 years old) with mild cognitive impairment (MCI). From the fourteen randomized controlled trials reviewed in the study, the 49 meta-analysis revealed there to be little evidence to support exercise improves cognitive function in older adults with MCI. However, in 2022, de Lima et al. compared the effects of HIIT and moderate-intensity continuous training (MICT) on cognitive function, and the results indicated exercise alone could promote cognitive function, independent of the exercise type. The current study specifically utilized high-intensity aerobic exercise as the training intervention, but it is currently unclear whether aerobic training alone significantly benefits cognitive function. In a meta-analysis by Cammisuli et al. (2017), the authors reported aerobic exercise was beneficial for patients with mild cognitive impairment (MCI), but they could not conclude aerobic exercise promoted a discerning effect upon cognition. Based on nine randomized controlled trials in humans from 2005-2013, the researchers observed only moderate effects of aerobic training on global cognition, inhibitory control, logical memory, and divided attention. Patients with age-related cognitive decline were excluded. Then, in 2018, Cammisuli et al. performed another meta-analysis on the effects of aerobic exercise in patients with Alzheimer’s disease (AD), where they report finding little evidence for improvements in AD patients’ cognition due to aerobic exercise. There is also the element of exercise intensity to consider. We performed high-intensity training based on the subjects’ maximum speeds achieved in the baseline and mid-intervention treadmill test, but we also implemented individualized training. Each mouse was accommodated for and moved to a slower- or faster-running group based on how subjectively easy or difficult each training session was for them. Therefore, should they have appeared to be struggling with their current speeds, whether due to injury, illness, or aging, they were motivated but not forced to continue training at that intensity when they clearly could not keep training at that pace. Kovacevic et al. (2020) reported higher intensity exercise to improve memory in sedentary older 50 adults (>60 years) after a 12-week intervention, suggesting that exercise intensity may be a critical factor in maintaining/improving memory. However, our mice were only the equivalent to that of early 50’s human years at the study’s end. Contrary to our findings, where significant aerobic capacity improvements in the HIIT group showed no cognitive improvements, Kovacevic et al. (2020) found significant correlation between increased cardiorespiratory fitness and memory improvements in humans >60 years of age. The study’s findings also contradict what was reported by de Lima et al. (2022), by suggesting exercise intensity could differentially affect cognitive functions. Caveats A limitation of the current study is that, though all mice started in group-housing, due to within- house fighting between certain cage mates, mice in both groups had to be individually housed while some remained group-housed. According to Hatchard et al. (2014), who published a meta- analysis on common methodological issues seen in animal research investigating the effects of exercise on cognition, singly-housed rodents may suffer from social isolation. They found that socially isolated (singly-housed) rodents were associated with a greater effect of exercise on cognitive performance; meaning, rodents living in social isolation demonstrated better cognitive performance than those who are socially active. However, the analysis only looked at the Morris water maze task as the cognitive measure. Due to randomization and the within-house fighting incidents near the current study’s start, 42.85% of the SED mice were singly-housed, while just 25% of the HIIT mice were singly-housed. This could have potentiated the lack of cognitive differences observed between groups. 51 Another potential limitation of the current study could be the length of the exercise intervention and/or sessions. The HIIT intervention was 12 weeks long (3 months), but Hatchard et al. (2014) observed exercise durations longer than 2 months proved to be ineffective in promoting cognitive function in the studies they reviewed. We designed our HIIT training to mimic that of human HIIT sessions, so the mice were never running for lengthy periods of time (3-5 intervals was equal to 12-18 minutes of exercise, respectively). Therefore, the mice were only exercising 36-54 minutes/week, which could potentially be too little time exercising to induce or observe cognitive differences from 12 weeks prior in middle ages. Another limitation is that we used only male mice of a single breed and age. Future studies are needed to establish baseline measures for cognition, effect of HIIT in older cohorts, and to determine any sexual dimorphisms. In the literature, Pettan-Brewer et al. (2013) developed the radial water tread (RWT) maze to test cognitive ability in male C57BL/6 and CB6F1 mice, from 4 to 28 months of age. The outcome measure of the test was latency to find the correct exit. For the C57BL/6 mice, increasing age was accompanied by increasing latency times. For the CB6F1 mice, a correlation was difficult to establish as the 20 and 28m groups demonstrated a lack of memory retention and learning ability, unlike the C57BL/6 mice. This RWT maze test was later used by Daneshjoo et al. (2022), with C57BL/6 male mice again, and they observed very similar results to the Pettan-Brewer et al. study. Between the two studies, there appears to be a correlation between age and cognitive performance in C57BL/6 male mice, with statistical significance observed between the younger and older age groups in the Pettan- Brewer et al. (2013) study. However, neither study specified whether there was significance between the 12m and 20m age groups, specifically. Therefore, age-related cognitive function patterns still need to be further explored. Chapter VI: Conclusion To investigate exercise as a potential preventative measure for age-related cognitive decline, the current study performed a 12-week HIIT protocol with middle-aged male C57BL/6 mice. Physical and cognitive function performance was assessed at baseline and endpoint, using the Comprehensive Functional Assessment Battery (CFAB) and a Cognitive Assessment Battery (CAB), respectively. Mice were randomized into the HIIT group (n=8) performed high-intensity interval training on a treadmill three times per week, with exercise intensities based on a percentage of their individual maximum speeds reached on the treadmill test. The SED group (n=7) was subjected to sham treatment every day the HIIT group was subjected to training. The current study observed no statistically significant between-group changes pre- to post- in the IAV determinants of grip meter, inverted cling, VWR, rotarod, or overall IAV. The HIIT group significantly increased in aerobic capacity and treadmill time, while the SED significantly declined. No significant between-group differences were observed for any of the cognitive measures assessed, pre- to post-. The lack of cognitive changes observed could be due to the age of the mice, as they may not have yet started experiencing age-related cognitive decline, which would mitigate any potential effects of the HIIT intervention. It could also be due to the mice’s strain, with certain strains potentially being more genetically predisposed to cognitive impairments while others are more genetically resistant. There are also envi