Alzheimer's Disease Exercise


AD is considered the most globally prevalent neurodegenerative disease accounting for approximately 60% of all dementia cases. [1] With life expectancy increasing worldwide it is expected that by 2050 nearly 1:85 individuals will be affect by AD. [1] Fortunately, increased physical activity and exercise have been identified as being able to positively influence AD, through altering a variety of its proposed pathogenic pathways discussed previously in the AD cell biology page.[1] Exercise has also been thought to have lasting beneficial effects on both the functional capacity and cognitive abilities of individuals with AD.[1] In this page we elucidate these hypothesized effects by delving into the cellular mechanisms in which exercise positively influences AD, as well as, their functional and cognitive correlations. The goal is to provide the reader with the most current and relevant research regarding the topic and offer our best recommendations based on the findings.


Figure 1. Exercise and AD (Figure Source)[2]

General Overview of Exercise in the Elderly Population

Given that individuals aged 65 and older are more likely to develop or already have AD,[3] it is important to first consider how physical activity affects the older adult and examine the current recommendations for this age group. Numerous improvements in health, psychological status, and functional ability have been associated with regular physical activity in the elderly population.[4][5] These changes may ultimately lead to increases in both quality of life and life expectancy.[4] With regards to health, physical activity has been associated with a decreased risk for many chronic conditions including cardiovascular disease, CVA, hypertension, type II DM, osteoporosis, obesity, and cancer.[4][5][6] It may also be used as a form of treatment for individuals with these conditions, as well as peripheral vascular disease, hypercholesterolemia, chronic obstructive pulmonary disease, and back pain.[4][6] Additionally, it has been found that physical activity improves psychological status and cognition.[4][6] Based on these findings, physical activity has been used to prevent and treat anxiety, depression, and dementia in the elderly population.[4] Finally, physical activity may enhance functional ability and independence in older adults by improving their balance, strength, ADL performance, and endurance.[4] These widespread benefits of physical activity have been attributed to its effect on various body systems including the cardiovascular, metabolic, endocrine, and musculoskeletal systems.[4] For a thorough description of the cellular effects associated with physical activity in the aging population, please refer to the Exercise and Aging page of this website.

Unfortunately, few older adults actually take advantage of the benefits associated with physical activity.[5] It has been reported that the older population engages in the least amount of physical activity in comparison to all other age groups.[6] According to the Department of Health and Human Services, older adults should perform ≥150 minutes of moderate-intensity activity each week for optimal health benefits.[7] If the older individual is capable of performing high-intensity physical activity, than 75 minutes each week is sufficient.[7] Weekly activity levels that surpass these guidelines may facilitate additional health benefits.[7] The American College of Sports Medicine (ACSM), in collaboration with the American Heart Association (AHA), has developed physical activity recommendations specifically for adults ≥ 65 years of age or those between the ages of 50 to 64 with “clinically significant chronic conditions and/or functional limitations”.[6] These recommendations are very similar to those for adults in general with the exception of different methods to measure intensity for aerobic activity and the addition of flexibility and balance training.[6] The recommendations for adults in general may be more appropriate for younger individuals with the rarer form of AD, known as early-onset familial AD,[3] if their mental and physical symptoms are mild. A summary of the recommendations for older adults can be found in the table below. Individuals with chronic conditions, such as AD, should combine the ACSM recommendations with those based on their specific condition to prevent further declines in their health.[6]

Table 1. ACSM Recommendations for Older Adults[6][8]

Parameter Aerobic Exercise Strengthening Flexibility
Frequency ≥ 5 days/week for moderate-intensity
≥ 3 days/week for high-intensity
This is in addition to ADLs and bouts of activity that are <10 min
≥ 2 days/week
Preferably on nonconsecutive days
≥ 2 days/week
Preferably on days when aerobic exercise or strengthening is performed
Duration ≥ 30 min/day for moderate-intensity
≥ 20 min/day for high-intensity
Not Applicable 10 min/day
Based on a 0-10 physical exertion scale where 0 is resting and 10 is maximal effort
5-6/10 for moderate-intensity
Should elicit increases in HR and breathing
7-8/10 for high-intensity
Should elicit large increases in HR and breathing
5-6/10 for moderate-intensity
7-8/10 for high-intensity
5-6/10 for moderate-intensity
Type Any type of activity that does not produce high levels of orthopedic stress such as walking, swimming, and cycling. 8-10 exercises of 10-15 repetitions working the major muscle groups.
Progressive weight training, calisthenics, and stair stepping are recommended.
Each major muscle group should be stretched for 10-30 seconds at a time. Repeat each stretch 3-4 times.

In addition to aerobic exercise, strengthening, and flexibility training, older individuals should also perform balance exercises to reduce their risk of falling.[6] Although specific guidelines for balance training have not been developed, the ACSM does recommend it be performed on 2 to 3 days each week.[8] Suggestions for this component of the exercise program include standing with a narrow base of support, strengthening postural muscles, performing dynamic movements, and eliminating sensory input.[8] The recommendations for each component of the exercise program must be based on the baseline fitness level of the elderly individual as well as their specific needs.[5] When initiating an exercise program, the intensity should begin low and progress gradually over time based on the individual’s tolerance.[5][8] Older adults with chronic health conditions should first be cleared by their physician to ensure that it is safe to participate in exercise.[5] The remainder of this web page will focus on how exercise affects key cellular components of AD pathology and will provide specific exercise recommendations for individuals with the condition based on current research.

Systemic Effects of Exercise

Brain size


Figure 2. Brain Changes with Aerobic Exercise (Figure Source)

Physical activity, in the form of aerobic fitness, has been found capable of preventing and slowing the decay of white and gray matter in the brain, specifically the hippocampus, prefrontal, lateral temporal, parietal, and corpus callosum regions[9],[10],[11],[12],[13],[14]. To date, the majority of research has focused on the brain region regularly affected by AD, the hippocampus. Studies have found the natural decay of the hippocampus to be non-linear[9],[10]; however, after the age of 50 years there is an annual decline in hippocampus size. For non-demented individuals the decline is a 1 - 2 % a year and for individuals with AD the decline is 3-5%[9],[10]. Other factors that influence hippocampal size include depression, hypertension, chronic heavy drinking, hormone therapy, and chronic stress[9].

The majority of studies on exercise and brain volume have been performed on older adults without AD. Aerobic exercise in these individuals has been shown to increase cognition, executive-control, attention processes, and decrease brain atrophy[10],[12],[15],[16],[17],[18]. The exact process of these changes are still being studied but what research has shown is aerobic exercise begins to increase cerebral blood volume, via angiogenesis, to the hippocampus after 3 months of exerise[9]. What is fascinating is the length of time performing aerobic exercise has been linked to greater brain volume. One study found that both the gray and white matter increase after 6 months of exercise, specially the prefrontal, parietal, lateral temporal, and corpus callosum[16].

As the length of chronic aerobic exercise rises to one year, an increase in the neural network between the frontal lobe and hippocampus has been found[17]. In addition, a 2% increase in volume was found in the hippocampus after a yearlong aerobic exercise program[13]. What’s important to note from this study is the anterior hippocampus increased in size, which is the location of cell proliferation[13]. Areas of the brain that do not seem to increase with aerobic exercise include the posterior hippocampus, thalamus, or caudate nucleus[13]. Even sedentary individuals have reported benefits after starting a yearlong walking program late in life, increasing the size of their hippocampus as well as improving their spatial memory function[15]. These reported benefits of aerobic exercise, in non-demented and AD subjects, seem to reverse the GM atrophy and neural deterioration reported reported in our Cell-Bio section on Brain Atrophy. It may be suggested, from a thirteen-year study, that individuals should walk about 1 mile per day to receive the brain volume fighting benefits of aerobic exercise[18].

Research on aerobic exercise and its effects on AD have mainly been performed in animal models; however, human studies are beginning to be documented, most of which are of cross-sectional design. Cognitive improvements have been found after 6 months of moderate-intensity aerobic and non-aerobic physical activity in individuals at risk for developing AD[19]. Honea et al.[12] previously studied a positive relationship between cardiorespiratory fitness (VO2peak) and white matter volume in subject with early AD[20]. Their imaging studies found exercise slows atrophy in the temporal and parietal regions while increasing blood volume in the hippocampus[20].

A more recent study of theirs found no association between cardiorespiratory fitness and brain volume loss in non-demented subjects; however, in subjects with mild AD, a positive correlation between cardiorespiratory fitness and brain loss in the medial and parietal regions were found[12]. A similar cross-sectional study by Burns et al.[11] studied the relationship between brain atrophy and cardiorespiratory (VO2peak) fitness in individuals with early-onset AD. Their findings showed AD subjects had less cardiorespiratory fitness than non-demented subjects and AD subjects with higher cardiorespiratory fitness levels had less brain atrophy[11].

The cellular-biology and histology of how aerobic exercise impacts brain size is still being studied. Animal and human models have found some endothelial growth factors are stimulated through aerobic exercise that facilitates angiogenesis and neurogenesis[14]. Growth factors like IFG-1 and VEGF have been found to cross the blood brain barrier after aerobic exercise[14], which may be linked cell proliferation and blood vessel size in the brain; however, further research is needed on the matter.

Table 2. Brain Changes with Exercise Studies

Author Type of Study Subjects Exercise Intervention Main Outcome Measure Intervention Effects/ Results
Erickson et al. 2009[9] Cross-sectional design 165 non-demented or AD individuals between the age of 59 and 81 years Cardiorespiratory fitness assessment for VO2peak Main Outcome Measure: MRI of brain volume and spatial memory performance Higher cardiorespiratory fitness is linked with larger hippocampus size, which is also associated with better spatial memory performance
Burns et al. 2008[11] Cross-sectional design 64 non-demented and 75 early-stage AD subjects over the age of 60 years Cardiorespiratory fitness (VO2peak) Psychometric evaluations, cardiorespiratory fitness, and MRI of brain volume In early AD, cardiorespiratory fitness was linked to brain size; higher fitness had less atrophy
Honea et al. 2009[12] Cross-sectional study 56 Non-demented and 61 early-stage AD subjects over the age of 65 were studied Physical activity was assessed using peak oxygen consumption MRI and VBM changes of brain size Higher fitness levels are related with less volume loss in the parietal and temporal regions in subjects with AD; however, this was not found in non-demented subjects. No significant ApoE4 relationships were found between fitness levels in either AD or non-demented subjects
Erickson et al. 2011[13] Randomized Control Trial 120 right-handed, non-demented or AD individuals between the age of 55 and 80 years A 12-month supervised walking program that was built up to 40 minutes in a heart rate zone that was between 50-75% of the heart rate reserve VO2Max and MRI images of brain volume changes between baseline 6 months and 12 months 12-months of aerobic exercise increased anterior hippocampus size by 2% and is associated with improvements in spatial memory
Colcombe et al. 2003[15] Meta-analysis Older adults between the ages of 55-80 years from eighteen different RTC’s between 1966-2001 A combination of strength and/or aerobic based exercise interventions Cognitive speed, visuospatial controlled processing, and executive control Long-term (46-60min for more than 6 months) interventions had greater improvement in all cognitive outcomes than shorter-term (15-30min for 1 to 3 months); however, moderate-term (31-45min for 4 to 6 months) exercise still showed more change than control groups and short-term exercisers. There findings also showed a combination of aerobic and strength training was more beneficial than aerobic training alone
Cocombe et al. 2006[16] Randomized control trial 59 sedentary older adults between the age of 60-79 years without AD Three one-hour trainings sessions per week at a level between 40%-70% of their HR reserve was completed over 6 months and compared to a control group of stretching and toning MRI changes of brain volume between baseline and 6 months Increases in the brain volume in both the gray and white matter after 6 months of exercise, specially the prefrontal, parietal, lateral temporal, and corpus callosum
Voss et al. 2010[17] Randomized control trial Non-diagnosed AD or demented younger adults, between 18 – 35 years, and older adults, between 55-80 years, who had less than 30 minutes of physical activity two days a week for more than 6 months A supervised walking program three-times a week for 12months was compared between a control of stretching and toning. Walking duration and intensity increased to 40 minutes by week seven at a heart rate zone between 60 – 75% of maximum HR reserve fMRI/ MRI’s of brain volume change between baseline, 6 months, and 12 months By12 months, statistical increases in the neural network between the frontal, posterior, and temporal lobes were seen in the brains of walking group participants
Erickson et al. 2010[18] Longitudinal design Non-demented or AD older adults began the study at the age of 65 years or older. Subjects were removed from the study if they demonstrated signs of dementia or mild cognitive impairments at either the 9-year or 13-year follow-up form baseline A non-normalized walking program was used, logging the number of blocks walked each week over 13 years MRI’s and VBM’s of brain volume were compared from data collected 13 years apart Walking 72 blocks a week detected an increase in gray matter volume in the frontal, temporal, and occipital lobes with additional sparing the hippocampus. Walking more than 72 blocks a week did not show any additional sparing
Lautenschlager et al. 2008[19] Randomized control study 138 individuals over the age of 50 who were at risk for developing dementia or AD A 6 month intervention consisting of moderate-intensity, aerobic and non-aerobic, physical activity was performed three times a week for 50-minutes Changes in their cognition based on their ADAS-Cog scores The subjects in the intervention group improved by 1.3 points on their ADAS-Cog scores when compared to the control group who did not perform any physical activity
Burns et al. 2008[20] Cross-sectional study 64 non-demented and 57 early-stage AD subjects over the age of 60 were used Peak oxygen consumption was compared with whole brain volume MRI images of brain volume Fitness levels are linked to the level or brain volume loss in AD subjects, the higher the level of fitness the less volume lost. In individuals without AD, no relationship was found between brain volume lost and fitness levels

Effect of Exercise on Cellular Components in AD



Figure 3. BDNF Molecular Model (Figure Source)

Brain-derived neurotrophic factor (BDNF) is considered one of the most important cellular factors involved with plasticity and long-term potentiation in the brain.[21][22] It can help nerves to grow as well as protect them from damage.[22] Unfortunately, hippocampal BDNF levels tend to decrease with aging[23] and decrease significantly as AD progresses.[24] In fact, this drop in BDNF levels may be what allows NFTs to develop. Research has discovered that, in a brain with AD, neurons that contain healthy amounts of BDNF have few NFTs, and neurons that contain large amounts of NFTs generally have low levels of BDNF.[25] BDNF has been found to effectively shield neurons from Aβ toxicity in vitro.[25]

Fortunately, exercise has been shown in a large number of studies to be an effective means of increasing serum and brain BDNF levels.[21][22][26][27] [28]

The following tables present research studies involving exercise and its impact on BDNF.

Table 3. Exercise and BDNF Animal Studies

Authors Type of Study Subjects Exercise Intervention Cellular and Cognitive Outcome Measures Intervention Effects/Results
Hopkins ME and Bucci DJ. 2010[26] Randomized controlled trial 32 male Long-Evans rats were randomly assigned into exercise group (X) or nonexercise group (NX) X rats were placed in a running wheel every other day for four weeks and compared to the NX group which remained sedentary. BDNF concentration in the perirhinal cortex (see Figure 4), elevated plus maze, novel object recognition, locomotor activity X rats demonstrated increased perirhinal BDNF concentrations immediately post intervention compared with NX group and also demonstrated better object recognition and lower anxiety levels. The results did not persist 2 weeks post intervention.
Nichol et al. 2009[27] Randomized controlled trial 16 Apoε3 and 15 Apoε4 sedentary targeted replacement mice were divided into an exercise group (RUN) and a control group (SED). Exercise mice were placed in isolated cages containing running wheels which activity was monitored by a computer. Control group mice were placed in isolated cages without running wheels. Hippocampal BDNF concentration, TrkB, PAK, and citrate synthase concentrations, object recognition, place recognition, radial-arm water maze Both ε3 and ε4 mice in the RUN group demonstrated significantly increased BDNF levels compared with SED. These increases correlated with increases in citrate synthase. TrkB and PAK increased in ε4 mice in the RUN group and PAK significantly decreased in ε3 mice of the RUN group. Mice in the RUN group demonstrated improvements in place recognition and water maze testing with ε4 mice initially performing worse than ε3 but proving similar post intervention.
Baj et al. 2012[28] Randomized controlled trial 30 male C57BL/6J mice ages 8-9 weeks were equally divided into either an exercise group with continuous access to a running wheel, a lock-wheel group exposed to a locked running wheel, or a sedentary group not exposed to a running wheel. Mice were housed individually for 28 days and running activity was monitored by a computer. Expression of BDNF at CA1, dentate gyrus, and CA3 regions of the hippocampus (see Figure 4) The exercise group demonstrated increases in BDNF expression in CA1 regions and dentate gyrus and demonstrated significantly increased BDNF expression in CA3 region compared with lock-wheel and sedentary groups.
Berchtold et al. 2005[22] Randomized controlled trial Adult male Sprague-Cawley rats ages 7-8 weeks old were randomly assigned to either a regular exercise group with continuous access to a running wheel, an intermittent exercise group with access to a running wheel every other day, and a control group with no wheel access. Each group consisted of 7-8 rats. Voluntary exercise was monitored for up to 90 days. At this point some of the rats remained sedentary awaiting analysis at 0, 1, 3, 7, or 14 days after exercise. Others remained sedentary for 1-2 weeks after which another two days of exercise was induced. Hippocampal BDNF levels were evaluated at either 0,1,3,7, or 14 days post-intervention or immediately after the 2-day return to exercise. Regular exercise lead to faster initial increases in BDNF levels but by day 28 both the regular and intermittent exercise groups showed comparable BDNF levels. When exercise was stopped the BDNF levels in the regular exercise group remained significantly higher than control for seven days whereas the intermittent group’s levels only remained significantly elevated for 3 days. Animals who returned to exercise program required 2 days of exercise for BDNF to significantly increase compared with 14 days to reach significance in the original exercise period.
Radak et al. 2006[29] Randomized controlled trial Twenty-one 13-month old male Wistar rats were randomly divided into either an exercise group (ET), a detrained group (DT), or a control group (C). ET and DT groups completed 8 weeks of swimming exercises. The first four weeks consisted of five days of swimming per week for 60 minutes each day. The second four weeks consisted of 120 minutes of swimming each day for five days each week. The DT weeks became sedentary for 8 weeks after the intervention. The control group remained sedentary. Hippocampal BDNF and NGF concentrations, electrol paramagnetic resonance of cerebellum (measures free radical concentration), proteasome activity, mitochondrial electron transport chain (ETC) content, passive avoidance test ET group performed significantly better on passive avoidance test with the DT group performing similar to control. The ET group displayed higher BDNF levels and DT group displayed decreased BDNF levels compared with control. Both ET and DT rats displayed reduced free radical concentration and NGF levels were reduced in DT group compared with ET and control. No change in ETC or proteasome activity was observed.
Wolf et al. 2006[30] Randomized controlled trial 30 females transgenic APP23 mice were randomly assigned to either enriched cages (ENR), regular cages with access to running wheel (RUN), or regular cages without a running wheel (CTR) Mice were observed in their respective cages for 34 weeks and wheel running was measured with a computer. Β-amyloid deposition and plaque load, hippocampal neurogenesis, gene expression of IGF-1, BDNF, FGF-2, VEGF, and NGF, Morris water maze Voluntary exercise had no effect on Aβ plaque load, NGF, BDNF, IGF-1, VEGF, and FGF-2, and mice in the RUN group showed no improvements with cognitive testing.

Table 4. Exercise and BDNF Human Studies

Authors Type of Study Subjects Exercise Intervention Cellular and Cognitive Outcome Measures Intervention Effects/Results
Ferris et al. 2007[21] One-group pretest-posttest design 15 college-age students (11 males and 4 females) All subjects performed a graded exercise test on a stationary bike followed by two 30-minute endurance rides at a fixed intensity. The intensities for the endurance rides were set at either 10% above ventilatory threshold or 20% below. The three bouts were performed on separate days. Serum BDNF levels immediately pre and post exercise, Stroop color and word tests Exercise sessions performed at or above 10% above the ventilatory threshold resulted in a significant increase in serum BDNF compared with baseline. These exercise bouts also resulted in improvements in color-word scores. Exercises below ventilatory threshold did not produce significant changes in serum BDNF or color-word scores.
Currie et al. 2009[31] Cross-sectional design 16 women and 28 men all in good health and regular exercisers Exercise habits and basic demographic information was obtained using surveys serum BDNF concentration Those subjects who reported higher exercise levels were found to have lower serum BDNF concentrations.
Chan et al. 2010[32] Cross-sectional design 85 healthy adults age 20-50 Exercise habits, lifestyle factors such as smoking and alcohol intake, and basic demographic information was obtained using surveys serum BDNF concentration Those subjects reporting higher levels of exercise were found to have lower serum BDNF concentrations and those reporting higher amounts of sedentary behavior had significantly higher serum BDNF concentrations.
Erickson et al. 2011[13] Randomized controlled trial 120 community-dwelling adults (mean age 66.6 years) without dementia were included in the analysis. Sixty were randomly to either an aerobic exercise group and 60 were assigned to a control group. The control group performed a simple stretching exercise routine. The exercise group performed moderate intensity aerobic exercise 3 days per week that began with 10-minute walks that were increased each week by five minutes until 40 minutes was achieved. The target heart rate for the walks was 50-60% of their max heart rate during weeks 1-7 and 60-75% after week 7. Both groups continued with their interventions for one year. Serum BDNF concentration, hippocampal volume as measured by MRI, spatial memory tests One year of aerobic exercise increased anterior hippocampal volume compared with baseline. The stretching group showed a decrease in anterior hippocampal volume compared with baseline. The aerobic exercise group did not demonstrate greater changes in serum BDNF than the stretching group after one year of exercise. However, within the aerobic exercise group, greater changes in serum BDNF were correlated with greater increases in hippocampal volume. Both groups showed similar improvements in cognitive testing.

Many of the studies that have been performed on exercise and its influence on brain BDNF levels have been performed on animals due to limited methods of analyzing brain BDNF levels. Large numbers of animal studies have been performed and the majority conclude that aerobic exercise increases brain BDNF concentration especially in the hippocampus[22][26] [27] [28][33] with a few exceptions reporting that aerobic exercise had no more effect on BDNF levels than control interventions. [30] [13] Most studies also show that along with the increases in BDNF levels, exercise also increases hippocampal volume through neurogenesis [22][24] [13]and improves cognition. [26][27] [29] In fact, low BDNF levels have been correlated with decreased hippocampal volume,[23] and walking distances have been found to be correlated with hippocampal volume up to nine years later.[24]

Figure 4. Hippocampus and Perirhinal cortex in Human Brain (Figure Source)

Some studies measured serum BDNF levels instead of specifically measuring brain BDNF concentration.[21][31][32][13] Serum BDNF is measured instead of brain levels because it can be measured without sacrificing the animal and is also acceptable for human studies. Acute changes in serum BDNF levels are thought to directly reflect changes in brain BDNF concentration[21][24]. There are several reasons for this: rat studies show correlations between serum and brain BDNF levels,[21] BDNF easily travels in both directions across the blood-brain barrier,[21][24] and both serum and brain BDNF levels are significantly reduced in AD.[24] There is conflicting evidence concerning exercise and its impact on serum BDNF.[24] Acute bouts of exercise have been found to increase serum BDNF levels,[21][24] but several studies have concluded that regular (chronic) exercise may lead to decreased serum BDNF levels compared to sedentary individuals.[24][31][32] The reason for this inverse relationship between exercise and serum BDNF is unclear but one hypothesis is that regular exercise leads to a more efficient BDNF uptake system into the central nervous system.[31]

With all the evidence that exercise increases brain BDNF levels, hippocampal volume, cognitive performance, neurogenesis, and reduces oxidative stress the next questions to answer involve type, frequency, intensity, and duration of exercise. Most studies that have been performed on BDNF levels have demonstrated benefits achieved through aerobic, or cardiovascular, exercise[24]. In fact, all of the studies reviewed in the table above used aerobic exercise as the control intervention. There have been a few studies using resistance exercise and stretching but neither were found to be superior to aerobic and aerobic exercise was shown to promote larger increases in serum BDNF and hippocampal volume.[24] However, due to the lack of research in this area, it is difficult to draw hard conclusions that other types of exercise are not beneficial. It remains that cardiovascular exercise has been shown to improve BDNF concentrations in the hippocampus. [22] [26] [27] [28] [33] Types of aerobic exercises studied have been swimming,[29] biking,[21] and running. [22][26][27][28]

Generally, the frequency of aerobic exercise that increases BDNF is daily exercise 4-7 days per week, [22][27][28] [29] although BDNF levels also temporarily rise with acute bouts of exercise.[21] In one study, exercise frequency of three days per week was not sufficient to produce increases in BDNF over a one-year period.[13] Long-term daily exercise was shown to be more beneficial than intermittent long-term exercise by Berchtold et al.[22] because BDNF levels returned to baseline with rest much more quickly in the intermittent exercise group.

Max heart rate, VO2 max, and rate of perceived exertion have all been used to grade exercise intensity in studies on BDNF and exercise.[21][24][13] All of the studies agree that, in order to achieve benefits, the exercise session must be intense enough to increase the heart rate and oxygen consumption.[24] Studies that use VO2 max report that exercise greater than 60% VO2 max is sufficient. [21][24] Ferris et al.[21] report that exercising at 10% above the ventilator threshold (75% VO2 max)or above will increase serum BDNF in acute trials. Participants rated this exertion level as “somewhat hard” or “hard” when using the Borg scale of perceived exertion.[21] Probably the easiest way to monitor exercise intensity is heart rate. Beginning at an intensity of 50% max heart rate and slowly increasing the intensity to 75% max heart rate is sufficient to achieve increases in BDNF levels.[13]

Exercise bouts lasting from 20-150 minutes have been used in research and found to be beneficial to BDNF levels. [21] [24][29] Generally, it is best to begin with low durations and gradually increase them as fitness improves. [24] [13] It is not necessary to reach extreme bout lengths (>60 minutes) to achieve results. Participants in a study by Ferris et al.[21] achieved increases in serum BDNF with only 30 minute exercise bouts. Long term aerobic exercise seems to be the key achieving cognitive benefits and increasing hippocampal BDNF levels.[24] Most studies lasted anywhere from one to twelve months and increases in BDNF levels were evident at all time periods.[24][((bibcite hopkins))][27] BDNF and cognitive benefits quickly diminished in all cases when exercise was discontinued falling below significant levels with only 1-2 weeks of rest in most cases.[22][26] [33]

Based on the results of these studies a recommendation can be made that, in order to promote increases in hippocampal BDNF and cognitive improvements, people with AD should perform greater than 30 minutes of aerobic exercise at least 4 times per week[21][22] at an intensity of 60-75% heart rate max[13] or above 60% VO2 max[21][24] and that long-term exercise is most beneficial.[24]

Amyloid-beta plaques and Tau pathology

In the brain of AD patients, the major pathology that has been identified is amyloid plaques which consist of amyloid-β (Aβ) peptides and neurofibrillary tangles, formed from hyperphosphorylated tau protein.[3] These plaques can be found in regions of the brain associated with learning and memory including the hippocampus, amygdala, and the association cortexes of the frontal, parietal and temporal lobes.[3] The cellular biology of these plaques and the mechanisms in which they affect those with AD can be found in the AD cellular biology page.

With Aβ plaques being the main pathological event in AD, it has become a focus in many studies looking at ways to prevent and protect against declines associated with AD. Exercise has been shown to have effects on levels of Aβ and Aβ plaques in the cortex and hippocampus in some mice studies[34][35][38][39]; however other studies indicate that exercise has no effect on Aβ pathology.[36][37][30] The difference in results from these studies may be a resultant of the use of different genetic mouse models, the age and stage of disease when exercise was begun, and the type, intensity or duration of the exercise intervention.

The following table identifies the findings of recent mice studies and the effect exercise has on Aβ plaques.

Table 5. Exercise and Amyloid-beta plaques

Authors Type of Study Subjects (Mouse model) Exercise Intervention (type/intensity/duration) Cellular and Cognitive Outcome Measures Intervention Effects/Results regarding Aβ
Adlard PA, Perreau VM, Pop V, and Cotman CW (2005) [34] Cohort study TgCRND8 mouse line, with a double mutant form of APP Voluntary exercise for 1 month or 5 months in a cage with a running wheel Extracellular amyloid –β load, secretase activity, mRNA and protein, and Morris water maze experiments Mice with 5 months of voluntary exercise had decreased extracellular amyloid-β plaques in frontal cortex (38%), cortex at the level of the hippocampus (53%), and hippocampus (40%). ELISA measured a decrease in the total formic acid-extractable Aβ and exercising animals had reduced cortical Aβ1-40(35%) and Aβ1-42 (22%) compared to sedentary. No significant decrease in levels of APP, CTFs, or α-, β-, and γ-secretase in 5 month exercise group. In the 1 month exercise group APP protein and secretase activity was unchanged however α-CTFs(54%) and β-CTFs(35%) were reduced and a nonsign. decrease in Aβ1-40(51%) and Aβ1-42 (53%)
Ke HC, Huang HJ, Liang KC, and Hsieh-Li HM (2011) [35] Randomized controlled trial APP/PS1 double mutant transgenic mice compared with wild type (control) Continuous, non-shock treadmill exercise at 10 m/min for 5 days a week with treatment time beginning at 10 minutes and increased 10 minutes each day in the 2nd week until reaching 60 minutes a day for 4 total weeks. Two-day passive avoidance test, open field test, morris water maze, local inflammation, Aβ plaque loading, BDNF level, cholinergic and serotonergic neurons Aβ level in hippocampus was only looked at in adult Tg mice and Aβ1-42 and Aβ1-40 levels were significantly reduced by exercise. Exercise, however did not decrease plaque loading in adult or aged mice.
Nichol KE, Poon WW, Parachikova AI, Cribbs DH, Glabe CG, and Cotman CW (2008) [36] Randomized controlled trial Tg2576 mice compared to wild type Voluntary running wheel for 3 weeks monitoring running Cytokines (IL-1B, TNF-alpha, IFN-y and CD22C), antigen-presenting cells (MIP-Iα and CD40), and Aβ levels No significant differences in Aβ aggregates in the hippocampus between sedentary and exercised mice. Levels of soluble Aβ fibrils were 40% lower in exercise mice as well as significant decreases in soluble Aβ40.
Parachikova A, Nichol KE, and Cotman CW (2008) [37] Randomized controlled trial Tg2576 mice Voluntary wheel running for 3 weeks Morris water maze, Aβ load, APP levels, inflammatory markers, and CXCL1 and CXCL12 proteins No difference in presence of Aβ plaques or levels of APP between sedentary and exercise in Tg mice.
Um HS, Kang EB, Koo JH, Kim HT, Lee J, Kim EJ, Yang CH, An GY, Cho IH, and Cho JY (2011) [38] Randomized controlled trial Tg-NSE/hPS2m, with human PS2 mutation Treadmill training at 12 m/min, 60 min per day, 5 days a week on no gradient for 3 months Aβ levels; tau phosphorylation levels; Cox-2 expression; cytochrome c, Bax protein and Bcl-2 protein; apoptosis; NGF, BDNF, and CREB levels; SOD-1, SOD-2, and HSP-70 expression; and metabolic parameters Tg mice after treadmill exercise had reduced levels of Aβ-42 in the hippocampus
Wolf SA, Kronenberg G, Lehmann K, Blakenship A, Overall R, StaufenbielM, and Kemperman G (2006) [30] Randomized controlled trial Transgenic APP23 mice with Swedish double mutation Mice had enriched housing which consisted of tubes, wire ladders, a house, and a crawling ball or a cage with unlimited access to a running wheel and compared to those housed in a standard cage for 11 months. Aβ plaque load, Morris water maze, mRNA levels hippocampal neurogenesis, and BDNF No significant difference in plaque loads in cortex or hippocampus across groups
Yuede CM, Zimmerman SD, Dong H, Kling MJ, Bero AW, Holtzman DM, Timson BF, and Csernansky JG (2009) [39] Randomized controlled trial Tg2576 mouse strain Voluntary runners were placed in a cage equipped with a running wheel for 1 hour per day, 5 days a week for 16 weeks. The forced exercise group ran on a treadmill with a mild foot shock for an hour at the velocity calculated to be an average of what the wheel running mice performed that day also for 16 weeks. Open field activity, soluble Aβ-40 and Aβ-42, amyloid plaques, and hippocampal volume No differences were found in the Aβ-40 and Aβ-42 levels in cortex or hippocampus. Fewer amyloid plaques were found in cortex and hippocampus of the voluntary exercise group than sedentary or forced exercise groups and forced exercise had fewer amyloid plaques than sedentary.

Several conclusions can be drawn regarding the effect of exercise on Aβ pathology in mice even though these studies have revealed some conflicting evidence. When looking at the studies that have found exercise to affect Aβ aggregation, these studies generally have longer periods of exercise intervention. Adlard, Perreau, Pop, and Cotman[34] found decreased extracellular amyloid-β plaques in frontal cortex (38%), cortex at the level of the hippocampus (53%), and hippocampus (40%) in mice who participated in voluntary wheel running exercise for 5 months. This study also showed that mice who participated in this same voluntary wheel running exercise for 1 month did have some change in the Aβ-40 and Aβ-42 levels even though the amount was not significant.[34] This indicates that exercise may be impacting the AD pathology before 5 months just not as significantly as it does when it is continued for longer periods of time. Ke et al.[35] and Um et al.[38] also support this hypothesis showing Aβ levels were reduced in the hippocampus of mice even though plaque loading was not affected when the exercise intervention lasted for 4 weeks and 3 months, respectively. Nichol et al.[36] and Parachikova et al.[37] both found no differences in Aβ aggregates, plaques, or APP levels when looking at mice who participated in only 3 weeks of exercise.

Yuede and colleagues[39] included another important factor when looking at exercise by comparing mice that participated in voluntary exercise of wheel running in comparison to forced exercise, where a mild shock on the feet of mice was used to increase running velocity on a treadmill. This study found fewer amyloid plaques in the cortex and hippocampus of the voluntary exercise group than in the sedentary or forced exercise groups.[39] In addition the forced exercise mice had fewer plaques when just compared to the sedentary mice.[39] This indicates the importance of the patients view on the exercise program and emphasizes that stress in the form of an exercise program is not as effective as a voluntary exercise program.

In summary, exercise interventions that are voluntary and last for 3-5 months can impact the Aβ levels and plaques in the cortex and hippocampus of mice. The exercise can be in different forms including treadmill or wheel running for mice and can have some effects after shorter periods of time, perhaps 1 month but have more significant impacts when continued for 5 months. Therefore, this can be translated to show long-term, voluntary, non-stressful exercise can impact the amyloid pathology in individuals with AD.

Tau functions to promote the assembly and stabilization of microtubules and plays a role in vesicle transport.[3] When tau is hyperphosphorylated it aggregates into paired helical filaments, which form neurofibrillary tangles leading to Aβ plaques.[3] Further details can be found in the AD cellular biology page. Tau phosphorylation has been shown to be impacted by exercise in a few mouse studies.[40][41][38]

The following table summarizes the findings from a few mice studies looking at the impact of exercise on tau pathology.

Table 6. Exercise and Tau pathology

Authors Type of Study Subjects (Mouse model) Exercise Intervention (type/intensity/duration) Cellular and Cogntive Outcome Measures Intervention Effects/Results regarding Tau
Belarbi K, Burnouf S, Fernandez-Gomez FJ, et al. (2011) [40] Randomized Controlled Trial THY-Tau22 mice, with the overexpression of human 4-repeat Tau mutated at G272V and P301S sites compared with wild type litter Standard housing supplemented with a running wheel to be used for voluntary exercise for 9 months Y-maze test, BDNF levels, tau and phosphorylated tau levels, inflammatory cytokines and microglial markers, cholinergic neurons, and cholesterol related genes There was no change between the total tau and phosphorylated tau levels between sedentary and running THY-Tau22 mice. Exercise was found to reduce the level of pathological Tau species.
Leem YH, Lim HJ, Shim SB, Cho JY, Kim BS, and Han PL (2009)[41] Randomized Controlled Trial Tg-NSE/htau23 mice expression human tau23 under control of neuron-specific enolase (NSE) promoter Exercise was performed on a mouse treadmill at 12 m/min (moderate intensity) or 19 m/min (high intensity) 60 min per day for 5 days a week for 12 weeks Antioxidant enzymes in the brain, kinases that mediate tau hyperphosphorylation, tau hyperphorylation, and signaling pathways Chronic exercise resulted in suppression of tau hyperphosphorylation in the brain due to the effect exercise had on tau-regulating kinases (GSK3β, PKA, MAPK (p38, ERK1/2, JNK), P13K/AKT, and PKC and the Wnt signaling pathways.
Um HS, Kang EB, Koo JH, Kim HT, Lee J, Kim EJ, Yang CH, An GY, Cho IH, and Cho JY (2011)[38] Randomized Controlled trial Tg-NSE/hPS2m, with human PS2 mutation Treadmill training at 12 m/min, 60 min per day, 5 days a week on no gradient for 3 months Aβ levels; tau phosphorylation levels; Cox-2 expression; cytochrome c, Bax protein and Bcl-2 protein; apoptosis; NGF, BDNF, and CREB levels; SOD-1, SOD-2, and HSP-70 expression; and metabolic parameters Tau phosphorylation levels at Ser404, Ser202, and Thr231 residues in the hippocampus was suppressed with exercise. Exercise decreased JNK and p38MAPK and increased ERK phosphorylation in the hippocampus. Exercise also had positive changes in P13K, Akt, and GSK-3α/β signal transductions in the hippocampus.

Again, supporting the conclusions found from the Aβ studies, chronic exercise lasting for 3 months have been shown to impact tau phosphorylation in the brain. Leem et al.[41] and Um et al.[38] both found exercise has led to suppression of tau hyperphosphorylation in the brain, and in particular the hippocampus. These studies both had mice running on a treadmill 12 m/min for 60 minutes a day, 5 days per week for 3 months which is of moderate intensity for a mouse.[41][38] Belarbi and colleagues[40] found no change between the total tau and phosphorylated tau levels between an exercise group participating in voluntary wheel running and a sedentary group house in a standard cage. Exercise was however found to reduce the level of pathological tau species.

Using the tau pathology studies to add to exercise recommendations we can again support the importance of a long-term exercise program, at least 3 months that specifically uses moderate intensity exercise for an hour a day for 5 days a week. These exercise parameters were also used to reduce Aβ levels and can therefore impact both tau and amyloid pathology.

These types of studies have not been translated into human models at this time. Therefore, it is difficult to prescribe exercise programs for humans with AD using studies that are conducted on mice. However, there have been some imaging and neurochemical measures that could be used to assess accumulation of amyloid plaques and neurofibrillary tangles that could potentially determine how exercise impacts AD pathology.[42]

Amyloid-β plaques have been identified in vivo in human brains using positron emission tomography (PET) with Pittsburg Compound-B (PIB) as a radiotracer to bind and thus identify amyloid plaques.[43] A high PIB binding has been found to have an inverse relationship with cerebrospinal fluid (CSF) of Aβ-42, indicating a low level of CSF Aβ-42 taken by a lumbar puncture may be another useful measurement of amyloid plaques in humans.[43] P-tau (phosphorylated) levels in the CSF have also been found to be a good diagnostic biomarker of AD.[44] A multiplexed bead-based xMAP technology has been used to measure CSF levels of Aβ-42, P-tau, and T-tau (total) levels simultaneously to further advance in ways to identity AD in humans.[45] These biomarkers associated with AD pathology are all potentially useful in measuring and assessing levels of amyloid plaques or tau phosphorylation to associate the effects of exercise on AD.

PET-PIB and CSF neurochemical biomarkers as measurements of human AD pathology and exercise have not been explored. One study however has assessed these biomarkers in healthy older adults to compare levels of exercise with biomarker levels.[42] Healthy individuals ages 55-88 were given an exercise engagement questionnaire, had CSF levels collected, and amyloid imaging with PIB performed.[42] Individuals who exercised less had elevated PIB, CSF tau, and p-tau and decreased CSF Aβ-42.[42] Individuals who met exceeded the AHA guidelines of 7.5 MET-hours per week of exercise had lower PIB uptake and after removing one outlier had higher Aβ-42.[42] This suggests that exercise may have beneficial effects on AD pathology particularly when following professional recommendations.


As discussed in the Alzheimer’s Cell Bio page, mitochondrial dysfunction is thought to play a crucial role in both the development and progression of AD. However, the exact cellular mechanism for which this happens is still a highly debated topic.[46] Damaged mitochondrial DNA (mtDNA), altered mtDNA expression, and altered mitochondrial biogenesis are some of the currently proposed reasons for mitochondrial dysfunction.[47] Studies have found dysfunctional mitochondria to have decreased electron transport and membrane potential, as well as, increased oxidative products.[48] It’s encouraging to know that current research suggest that exercise can induce molecular level changes that promote optimal mitochondrial function and development.[48]

The table below provides an overview of the current research discussing the effects of exercise on brain mitochondria and the underlying cellular processes that relate to aging and AD.

Table 7. Exercise effects on mitochondria

Authors Type of Study Subjects Exercise Intervention Cellular Measure Intervention Effects/Results
Navarro A, Gomez C, Lopez-Cepero J, Boveris A. 2004[49] Randomized Controlled Trial CD-1 Strain Mice Once the mice were 28 weeks of age moderate exercise consisting of treadmill training at speeds of 10, 15, 20 cm/s were performed for 5 minutes each, 7 days a week, until they were 78 weeks of age. Oxidative Markers (TBARS, protein carbonyls) ETC Indicators (NADH-cytochrome-c reductase, succinate-cytochrome-c reductase, and cytochrome oxidase) Antioxidant enzymes (Mn-SOD, Mn-SOD, and SOD) Moderate exercise between weeks 28-52; increased mice survival by a median of 19 weeks, attenuated the rise of TBARS and protein carbonyls in the brain, attenuated the decline of both NADH-cytochrome-c reductase and cytochrome oxidase brain activity, and significantly reduce the loss of antioxidant enzyme activites in the brain. Neither aging nor moderate exercise had an affect on succinate-cytochrome-c reductase brain activity. Moderate exercise demonstrated little to no positive effect at 78 weeks.
Kirchner L, Chen WQ, Afjehi-Sadat L, Viidik A, Skalicky M, Hoger H, Lubec G. 2008[50] Randomized Controlled Trial Male Sprague-Dawley Rats Treadmill training at a speed of 20 m/min; for 20 minutes; 2 times a day; 5days/week; from 5 months of age until 23 months old Metabolic Proteins (Isocitrate dehydrogenase activity, Malate dehydrogenase activity, Ubiquinol-cytochrome-c reductase (Uqcrc1) complex core protein 1 activity, Ubiquitin carboxyl-terminal hydrolase isozyme L1 (UCH-L1) activity, Ornithine aminotransferase (OAT) activity) Treadmill training increased the presents of Isocitrate dehydrogenase, Malate dehydrogenase, and Uqcrc1 in hippocampal tissue. Treadmill training reduced UCH-L1 and OAT levels in the hippocampus.
Steiner JL, Murphy EA, McClellan JL, Carmichael MD, Davis JM. 2011[51] Randomized Controlled Trial ICR Mice Exercise training lasted 8 weeks and consisted of treadmill training at 25 m/min and a 5% incline; for 1 h/day; for 6 days/week. A brief 15-min warm-up period was incorporated into the 1 h of training each day. mtDNA, PGC-1α mRNA, Sirtuin 1 (SIRT1) mRNA, and citrate synthase (CS) mRNA Exercise resulted in a 2-fold increase in cortex (CX), frontal lobe (FL), hippocampus (HC), hypothalamus (HY), and midbrain (MB) expression of SIRT1 mRNA. Exercise induced a 3-fold increased expression of PGC-1α mRNA in the brainstem (BS), CX, FL, HC, HY, and MB. CS mRNA was significantly elevated in all brain tissues examined. Exercise also promoted a significant increase in mtDNA copy numbers in the BS, CB, CX, FL, and HC.
Aguiar AS, Tuon T, Pinho CA, Silva LA, Andreazza AC. 2008[52] Randomized Controlled Trial Male CF1 Mice Both the intermittent (3x/day for 15mins) and continuous (1x/day 45mins) exercise groups performed intense treadmill training for 8 weeks. For the first 4 weeks treadmill speed was kept at13.5 m/min, the following 4 weeks speeds were increased to 16.5 m/min. Training occurred 5 days/week. TBARS, Cytochrome c oxidase (COX), and Brain-derived neurotrophic factor (BDNF) Both intermittent and continuous intense exercise significantly increased TBARS levels in the brain. Intense exercise, especially when intermittent, significantly reduction both COX activity and BDNF levels in the frontal cortex of the brain.

Hypothesized cellular mechanisms for which exercise provided favorable effects include: decreasing the amount of oxidative stress markers located in the mitochondrial membrane,[49],[53] blunting the age-related decline in brain antioxidant enzyme activities,[49],[53] maintain the function of mitochondrial electron transfer enzymes,[49],[53] and through promoting increased biogenesis of brain mitochondria.[51]

Boveris and Navarro,[49],[53] found that after 24 weeks of moderate exercise complex IV, the electron transfer enzyme associated with mitochondrial O2 uptake, had a 12-32% increase in activity levels in the brain. This increase in complex IV activity was also correlated to neuromuscular response. The authors,[49],[53] were also able to demonstrate that moderate exercise of 24 weeks could reduce the amount of oxidative content found in the mitochondria, specifically, protein carbonyls which were linearly related to the function of mitochondrial membrane enzymes. This reduction was partly attributed to the increase in antioxidant enzymes, such as, SOD seen after routinely performing moderate activity for 24 weeks.[49],[53] Steiner et al.,[51] were able to provide evidence that 8 weeks of moderate exercise can increase both mtDNA and PGC-1α levels in the brain, suggesting increases in mitochondrial biogenesis. This is significant because mitochondrial biogenesis associated with PGC-1α is thought to result in improved oxygen uptake due to PGC-1α’s link to ATP production and intracellular calcium levels.[51]

Unfortunately, the positive effects of exercise don’t appear to be exponentially linear. In fact, a study conducted by Aguiar et al.,[52] propose that the positive effects of exercise may be dependent on the intensity of the activity. They found that intense exercise inhibited brain mitochondrial function by increasing oxidative stress to a toxic level, decreasing ETC function via COX inhibition, and decreasing BDNF levels in the frontal cortex.[52]


The research investigating the impacts of exercise on the mitochondria as it relates specifically to the brain is still limited. (all references) The studies outlined above only offer a representation of the most current relevant material associated with the topic. Based on the literature results,[49],[50],[51],[52],[53] two general premises can be gathered about exercise and brain mitochondria. Moderate exercise appears to have a beneficial effect on brain mitochondria function, especially when performed habitually throughout ones lifespan.[49],[50],[51] On the other hand, intense exercise could potentially pose as a threat to mitochondria function.[52] So as clinicians its important to use sound scientific and objective measures when assessing the exercise intensity of aging patients, more specifically those with known neurodegenerative impairments as with AD.


The severity of AD pathology has been associated with the microglia activated inflammatory response as discussed in the AD cellular biology page. Research has been focused on determining possible treatments for AD as well as prevention techniques. The lack of understanding of the exact cellular mechanisms makes this goal very challenging and complex.[37] Exercise may impact the inflammatory response associated with AD leading to changes in cognitive function.[37] Other possible preventative measures include specific diet, vitamins, stress levels and general health.[54]

The following tables discuss recent research studies involving acute exercise (Table 8) and chronic exercise (Table 9) and the effect on the inflammatory response associated with AD.

Table 8: Acute exercise and inflammation

Authors Type of Study Subjects Exercise Intervention Cellular Measure Intervention Effects/Results
Nybo L, Nielsen B, Pedersen BK, Moller K, Secher NH. (2002)[55] Pre-post control study 8 young, healthy males (27 yrs +/- 2) Subjects performed 60 minute period of exercise twice with a 60 minute recovery period. Exercise was done on cycle ergometer at ~50% VO2max. Plasma and brain IL-6 levels. There was a 5 fold increase in IL-6 levels in the brain during the second period of exercise compared to initial amount in the brain.
Steensberg A, Dalsgaard MK, Secher NH, Pedersen BK. (2006)[56] Randomized controlled trial 24 young,healthy males (20-29 yrs) randomized to 3 groups: exercise with placebo carbohydrate, exercise with carbohydrate and resting controls. The exercise groups cycled for 2 hours at 60% of VO2max. IL-6, TNFα and hsp72 in plasma and brain. Exercise did not alter IL-6 levels in the brain, however it did increase IL-6 levels in the plasma. There was no change in TNFα in the brain with exercise.

Multiple research studies have focused on the impact of exercise and inflammation.[36][37][38][55][56][57][58] There are several general conclusions that can be made based on the studies presented in the tables above. Acute exercise may cause an increase in pro-inflammatory markers including IL-6.[55] Nybo et al. found elevated IL-6 levels in the brain after two sixty minute periods of exercise.[55] There are several possible cellular mechanisms that could result in this increase of inflammatory markers. In response to exercise, the activated regions of the brain could have increased the total amount of IL-6 production.[55] This release of IL-6 may be used as a stimulus to regulate systemic glucose homeostasis between the brain and the muscles.[55] Another mechanism may be that elevated adrenaline concentrations resulted in activation of astrocytes with consequent release of IL-6.[55] However, Steensberg et al. found no change in IL-6 levels with acute exercise indicating no impact on the inflammatory process.[56] The presented studies offer conflicting evidence in which case further research is required for stronger evidence.

With this information, it can be hypothesized that the response to acute exercise in adults with AD could elicit elevated pro-inflammatory markers.[55] As discussed in the cellular biology page, elevated levels of inflammatory markers are already present with the aging process and AD pathology. In response to a rapid change in stress levels associated with acute exercise, the stress can be viewed as toxic or harmful to the cell resulting in activation of the inflammatory response.[55] The increased inflammatory response would be evident by elevated IL-6, TNF-α and IL-10 levels in the hippocampus. Also, the previously mentioned cellular mechanisms may also have a role in the inflammatory response in adults with AD.

Table 9: Chronic Exercise and inflammation

Authors Type of Study Subjects Exercise Intervention Cellular Measure Intervention Effects/Results
Parachikova A, Nichol KE and Cotman CW. (2008)[37] Randomized Controlled Trial Tg 2576 transgenic and age-matched non-transgenic mice Compared sedentary condition to a voluntary running condition for transgenic and non-transgenic mice for a 3 week period. The running condition included a running wheel placed in the cages for voluntary running, whereas the sedentary condition had no running wheel in the cages. 84 inflammatory markers in the hippocampus, amyloid beta load and CXCL1/12 proteins Sedentary Tg mice had increased gene expression for elevated inflammatory markers (including IL1α and IL1 receptor) and IL1, IL10 and TNF levels compared to non-transgenic mice. Transgenic mice in the exercise group did not demonstrate a return of inflammatory markers back to baseline levels, however there was an increase in 5 other chemokines mRNA levels (IL11, Spp1, CXCL12, CXCL1 and Ccl24).
Nichol KE, Poon WW, Parachikova AI, Cribbs DH, Glabe CG and Cotman CW (2008)[36] Randomized controlled trial Tg2576 mice (Tg group) and C57B16/SJL mice (WT group). Mice were divided into exercise (RUN) group and sedentary (SED) group RUN mice were placed in cages with a running wheel while SED mice were in cages without a running wheel for three weeks. Cytokines (IL-1B, TNF-alpha, IFN-y and CD22C) and antigen-presenting cells (MIP-Iα and CD40) IL-1β and TNF-α were significantly increased in SED Tg mice than in SED WT mice. The exercise protocol lead to significantly lowered levels of IL-1β and TNF-α to a level similar to the WT sedentary mice. The RUN Tg mice had increases in adaptive inflammatory markers including IFN-γ, CD40, MHC II, CD22C, and MIP-Iα.
Leem YH, Lee YI, Son HJ, Lee SH. (2011)[57] Randomized controlled trial Tg-NSE/htau23 mice and control non-Tg mice. Divided into non-Tg, Tg, Tg with low intensity exercise and Tg with high intensity exercise. Low intensity group performed treadmill training at 12m/min and high intensity group at 19m/min for 60 min/day, 5 days/week for 3 months. MAC-1 and GFAP levels (amount of astrocyte and microglia) and pro-inflammatory marker levels including iNOS and NF-kB binding activity. Exercise groups had lower MAC-1 and GFAP levels than Tg group but still higher levels than non-Tg. Exercise lead to decreased levels of inflammatory markers compared to Tg control mice.
Um HS, Kang EB, Koo JH, et al. (2011)[38] Randomized controlled trial Tg-NSE/hPS2m mice and control non-TG mice. Divided into SED and RUN groups RUN group did treadmill training for 12 weeks. (12m/min, 60 min/day, 5 days/week at 0% incline). SED group remained in their cages for the 12 weeks. AB levels, tau levels, Cox-2 expression and apoptosis marker levels The RUN Tg group had decreased levels of Cox-2 after the exercise protocol.
Pervaiz N, Hoffman-Goetz L. (2011)[58] Randomized controlled trial 42 healthy C57BL/6 female mice. They were randomized into running (RUN) or control (SED) groups. The RUN group had a running wheel available 24/7 for a 16 week period. The SED group did not have a running wheel and remained in their cages for the 16 week period. Pro-inflammatory markers, anti-inflammatory markers and apoptotic markers in the hippocampus. The RUN group demonstrated decreased TNFα levels and increased IL-6, IL1ra and IL-12 levels after the 16 week exercise period compared to the SED group. There was no difference in IL-1B and IL-10 levels.
Nicklas BJ, Hsu FC, Brinkley TJ, Church T, Goodpaster BH, Kritchevsky SB, Pahor M. (2008)[59] Single blind, randomized controlled trial 424 community dwelling elderly men and women (70-89 y.o.). They were randomly divided into exercise and successful aging (SA) groups. Exercise group participated in a 12 month moderate intensity physical exercise program involving aerobic, strength, balance and flexibility exercise. Successful aging group participated in weekly health education meetings and functioned as the control group. Plasma C-reactive protein (CRP) and IL-6 levels The exercise group demonstrated a significant decrease in IL-6 levels after completion of the exercise program compared to the SA group. No significant change was found in CRP levels.
Kohut ML, McCann DA, Russell DW, Konopka DN, Cunnick JE, Franke WD, et al. (2006)[60] Pre-/post- intervention study 87 elderly males and females were randomly divided into CARDIO or FLEX groups CARDIO group participated in aerobic classes for the 10 month period. Initially intensity started at 45-60% VO2max and progressed to 65-80%. FLEX group participated in flexibility, strength and balance exercises for the 10 month period. Both groups exercised 3 days a week for 45 min sessions. IL-6, IL-18, CRP and TNFα serum levels. After the 10 month period, the CARDIO group demonstrated decreased IL-6, IL-18 and CRF levels and the FLEX group did not. There was no significant change in TNFα levels for either group.

Chronic exercise can have differing effects on the cellular mechanisms associated with AD pathology.[36][37][38][57] It has been shown to result in the opposite effect on inflammation than acute exercise. Based on several transgenic mice models, aerobic exercise for greater than 3 weeks can result in decreased IL-1β and TNFα levels in the hippocampus.[36][37] The difference in exercise impact could be due to the body’s adaptation to exercise and achievement of homeostasis requiring less inflammatory markers. There may also be a feedback mechanism between the brain and muscles.[55] Research has found that systemic inflammation can enhance CNS inflammation with resultant cognitive decline.[61] It is important to consider the impact of peripheral inflammation in response to stress as well.

After three weeks of voluntary exercise in Tg2576 mice, there was improved cognitive function but no impact on AB levels and inflammatory cytokine levels in the brain.[36] [37] However, there was an increase in chemokine levels of CXCL1 and CXCL12 in the brain after exercise which could be a contributory factor for the change in cognitive function.[37] Another mechanism leading to improved cognition may be an increase in antigen presenting cells including IFNy, MIP1 gamma and CD40.[36] These cells may represent an antigen-presenting phenotype allowing for increased AB clearance.[36] After three months of forced exercise, a change was seen in pro-inflammatory marker levels including IL-6, IL-1B and TNFα.[57] The amount of activated astrocytes and microglia also decreased in the hippocampus after this exercise protocol.[57] Another change with chronic exercise was a decrease of Cox-2 levels.[38] These mechanisms could be the reason for the positive impact on cognitive function demonstrated in these studies. Similar results were found with a 16 week voluntary exercise protocol with decreased levels of pro-inflammatory markers including TNFα in the hippocampus.[58] There was also an increase in anti-inflammatory marker, IL-1ra and pro-inflammatory marker, IL-12.[58] With progressively longer exercise protocols, the impact on inflammatory markers is more significant which may have a greater effect on cognition.

There is minimal evidence of chronic exercise and brain inflammation in human models due to the assessment technique. In mice models, inflammatory markers are evaluated in the hippocampal tissue.[36][37] This is not realistically plausible in human subjects. Therefore, several studies assess these markers in the CSF for a reflection of the response in the CNS with acute exercise.[55][56] Numerous studies focus on the inflammatory response in muscle tissue after chronic exercise.[59][60] A 12 month aerobic exercise program in older adults resulted in lower plasma IL-6 levels with no impact on CRP levels.[59] Kohut el al. found that after 10 months of aerobic exercise, serum pro-inflammatory markers were decrease in older adults; however there was no change in pro-inflammatory levels after 10 months of flexibility/resistance exercise.[60] TNF-α levels were lower in both groups after exercise completion.[60] The proposed mechanisms that resulted in these changes in inflammatory markers included the impact of adipose tissue releasing increased amounts of IL-6 as well as muscle mass altering the uptake of TNF-α.[59][60] With the knowledge that peripheral inflammation can impact CNS inflammation, exercise studies focusing on peripheral inflammation markers may provide a foundation for possible impact on the brain.

Chronic exercise can be hypothesized to attenuate inflammation in the brain associated with AD pathology. Several transgenic animal models demonstrated chronic exercise can result in decreased pro-inflammatory levels in the hippocampus. [36][37] Based on the human studies and inflammatory markers in the periphery, a positive correlation may exist with a decrease in plasma inflammatory markers leading to a decrease in hippocampal inflammatory markers.[59][60] This hypothesis is founded in the fact that there is a feedback pathway between the brain and muscles that allow the correlation to exist. Based on the presented information, chronic exercise can result in a decrease in the inflammatory response in the brain associated with AD.


In summary, the impact of acute exercise on inflammatory process remains unclear.[55][56] Acute exercise in adults with AD may lead to an increase in pro-inflammatory markers due to the initial stress of exercise and the body initiating protective mechanisms.[55] However, chronic exercise may result in a decrease in pro-inflammatory markers as the body adapts to the impact of exercise and achieves a homeostatic level. [36][37][38][57][58] The impact of chronic exercise on the inflammatory response could be a possible mechanism for a reduced amount of cognitive decline in people with AD. The impact of inflammation on cognitive decline needs to be further researched for the exact mechanisms as well as an optimal exercise protocol for maximum benefits.


The effect of exercise on apoptotic mechanisms in the brain can vary based on presence of disease pathology, acute or chronic exercise conditions and the specific exercise protocol (frequency and intensity). The key apoptotic markers in the hippocampus studied after completion of an exercise protocol included pro-apoptotic markers (Bax, cytochrome c, Bid, caspase 3 expression) and anti-apoptotic marker (Bcl-2).[38][62][63][65][66] An increase in the pro-apoptotic markers indicates elevated apoptosis with eventual hippocampus damage and alterations in cognition.[38] Whereas, elevated Bcl-2 markers indicates inhibition of activation of caspase-dependent apoptosis in the brain.[38] These are the most common markers assessed in the following research studies.

Table 10: Acute exercise and apoptosis

Authors Type of Study Subjects Exercise Intervention Cellular Measure Intervention Effects/Results
Kerr AL, Swain RA. (2011)[62] Longitudinal Study 60 Long-Evans male hooded rats were randomly divided into the following groups:Control, Voluntary exercise (VE) at 12 hrs,VE at 24 hrs, VE at 48 hrs, VE at 4 days, VE at 6 days The control group had no access to a running wheel. The voluntary exercise groups had a running wheel in their cages. The time frame stated was used to indicate when the rats were sacrificed and cellular mechanisms assessed Apoptotic marker (cleaved caspase 3) and angiogenesis marker (CD61) in the hippocampus and cerebellum. After completion of the exercise protocol, there was increased levels of cleaved caspase 3 in the hippocampus at 12 hours and peaked at 24 hours. After 24 hours, there was a gradual decline back to baseline by 48 hours. Similar results were found in the cerebellum.
Kocturk S, Kayatekin BM, Resmi H, Acikgoz O, Kaynak C, Ozer E. (2008)[63] Longitudinal design 49 male L. Wistar albino rats were randomly divided into the following groups: control, exercise (EX) 0 hours, EX 3 hours, EX 6 hours, EX 12 hours, EX 24 hours and EX 48 hours. Time frame indicating time of sacrifice for assessment. The exercise group performed a single bout of exercise on a treadmill at 25 m/min at 5 degree incline until exhaustion. The control group did not perform an exercise intervention. Caspase 9, 8, and 3, apoptotic nuclei and cytochrome c in the muscle tissue of gastrocnemius and soleus. TNFα and IL-6 levels in the plasma. A significant increase in apoptotic nuclei was seen in the soleus compared to control and gastrocnemius. Cytochrome c increased after exercise and was at peak levels at 6 hours after exercise for both muscles. Caspase 3 was elevated during all time frames assessed. Based on all measures, soleus muscle experienced more apoptosis after acute strenuous exercise than gastrocnemius muscle.
Quadrilatero J, Bombardier E, Norris SM, Talanian JL, Palmer MS, Logan HM, et al. (2010)[64] One group, pre-/post- intervention design 8 healthy, active human subjects (7 males, 1 female) All subjects performed a single bout of cycling for a 2 hour period at ~60% VO2 max. Anti-apoptotic proteins, fiber type, pro-apoptotic proteins, caspase 3,8 and 9 activity, DNA fragmentation, Hsp70 expression, anti-oxidant proteins based on muscle biopsies and blood samples during exercise. There was no change in pro-apoptotic proteins, anti-apoptotic proteins, DNA fragmentation or caspase expression during the exercise protocol.

Acute aerobic exercise may lead to negative effect on cellular mechanisms initially following exercise completion.[62] Acute exercise has been shown to have an initial impact on apoptosis that attenuates with prolonged exercise.[62] A study by Kerr and Swain assessed the impact of acute exercise (from 12 hours to 6 days after exercise) on apoptosis in the CNS.[62] Within the first 24 hours of moderate aerobic exercise, an increase in apoptosis in the CNS occurred based on elevated cleaved caspase 3 levels.[62] By 48 hours, these levels returned to their initial levels prior to exercise and the caspase 3 expression remained stable until the completion of the study at 6 days.[62] Therefore, once the initial exercise response is passed, there was no evidence of increased detrimental apoptotic mechanisms in the brain associated with exercise. The initial increase in caspase 3 may be due to the body's response to the stress of acute exercise. This is similar to the inflammatory response to acute exercise and the apoptotic pathways can be initated by pro-inflammatory markers. [63] Apoptosis and inflammation may be directly related in response to acute exercise.


Figure 5. Impact of acute exercise on apoptosis and angiogenesis. (Figure Source)

Acute exercise at high intensity levels has shown to result in elevated apoptosis in muscle fibers.[63] A possible mechanism is that with increased muscle metabolism after exercise, increased reactive oxygen species are released and act as oxidative damage to the muscle.[63] The oxidative damage leads to activation of the intrinsic apoptotic pathway with ultimate cell death.[63] The extrinsic pathway may also be involved with acute stress levels after exercise. The stress of exercise can trigger than inflammatory response including IL-6 and TNFα.[63] These inflammatory markers are able to initiate the extrinsic apoptotic pathway resulting in elevated apoptotic markers.[63] This study further supports a relationship between apoptosis and inflammation.A study by Quadrilatero et al. that focused on acute exercise at moderate level intensity in healthy human subjects did not elicit the same change in apoptosis in muscle fibers.[64] The moderate level intensity may not have been perceived as stress at the cellular level in these young, active subjects. Therefore, no cellular changes were required for the body to maintain its homeostatic levels resulting in no changes in apoptotic proteins. [64] Based on this data, moderate intensity exercise does not result in changes in apoptosis but high intensity exercise can lead to elevated apoptotic marker levels in muscle fibers after a single bout of acute exercise. [63][64]

The impact of acute exercise in relation to people with AD can be hypothesized based on this current research.[62][63][64] Strenuous, high intensity exercise can initiate apoptotic mechanisms due to the stress of the exercise with subsequent elevated caspase 3, cytochrome c and apoptotic nuclei.[63] Since elevated apoptotic levels is not desirable in AD pathology, strenuous, high intensity exercise should be avoided in this population. Moderate intensity exercise may result in slightly elevated apoptotic levels initially but return to baseline levels within 48 hours.[62] This level of intensity may not result in a change in apoptosis based on the body’s response.[64] It is recommended that exercise be initiated at low-moderate intensity levels (~60% VO2 max or less) to prevent any initial harmful effects.[62][63][64]

Table 11: Chronic exercise and apoptosis

Authors Type of Study Subjects Exercise Intervention Cellular Measure Intervention Effects/Results
Kim SE, Ko IG, Kim BK, Shin MS, Cho S, Kim CJ et al. (2010)[65] Randomized controlled trial Young (15 month old) Sprague- Dawley rats and old (24 month old) Sprague-Dawley rats. Randomly divided into the following groups:young control, young exercise, old control, old exercise The exercise groups were forced to run on a motorized treadmill for 30 min/day for 6 weeks. Intensity and duration were: 2m/min for 1st 5 minutes, 5m/min for next 5 minutes then 8m/min for the last 20 minutes.The control groups were placed on the treadmill but not forced to run. Apoptotic markers (Bax, Bcl-2 and Bid), caspase 3 expression and neurogenesis markers. Prior to the exercise intervention, the aging groups demonstrated increased levels of caspase 3, increased Bcl-2, Bax and Bid and no change in Bcl-s/Bax ratio compared to the young groups. After the exercise intervention, the aging group had decreased caspase 3 levels, increased Bcl-2, decreased Bax, Bid and Bcl-2/Bax ratio.The Tg RUN group demonstrated decreased expression of caspase 3 after exercise. They also had lower levels of cytochrome c and Bax and higher levels of Bcl-2 in the hippocampus.
Um HS, Kang EB, Koo JH, et al. (2011)[38] Randomized controlled trial Tg-NSE/hPS2m mice and control non-TG mice. Divided into SED and RUN groups RUN group did treadmill training for 12 weeks. (12m/min, 60 min/day, 5 days/week at 0% incline). SED group remained in their cages for the 12 weeks AB levels, tau levels, Cox-2 expression and apoptosis marker levels The Tg RUN group demonstrated decreased expression of caspase 3 after exercise. They also had lower levels of cytochrome c and Bax and higher levels of Bcl-2 in the hippocampus.
Haack D, Luu H, Cho J, Chen MJ, Russo-Neustadt A. (2008)[66] Randomized controlled trial 35 Sprague-Dawley male rates were randomly divided into the following groups:sedentary (SED)/ no stress, SED/stress, exercise(EX)/ no stress, EX/stress. The sedentary groups had no access to a running wheel. The exercise groups did have a running wheel in their cages. The stress groups were restrained for 6 hours/day for 21 consecutive days. The “no stress” groups were not restrained throughout the study. Cortical Bax levels After three weeks, chronic stresses lead to elevated levels of Bax oligomers. However, the combination of exercise and chronic stress demonstrated no increase in Bax oligomer levels. The amount of Bax was significantly reduced in the EX/stress group compared to SED/no stress and EX/no stress.

Chronic exercise conditions have been shown to have the opposite effect on apoptosis compared to acute conditions. The following studies described the impact of chronic exercise associated with aging[65], alzheimer’s disease[38] and chronic stress[66].

Previous research studies have found elevated apoptotic markers in the brain associated with aging. (Please refer to the AD cellular biology page) The randomized controlled trial by Kim et al. found similar results with the young group compared to the old group prior to the exercise protocol.[65] The aging group had an increase in caspase 3 expression, increase in Bcl-2, Bax and Bid levels and elevated Bcl-2/Bax ratio compared to the young group.[65] With 6 weeks of forced light-moderate aerobic exercise on the treadmill, the aging group had decreased levels of caspase 3, increased Bcl-2 and decreased Bax and Bid.[65] This specific exercise protocol had no significant impact on apoptosis in the young group.[65] Exercise may have a neuroprotective function in the aging population. Moderate exercise was also performed by AD Tg mice in a randomized controlled trial by Um et al.[38] After 12 weeks of exercise, the Tg mice had significant improvements in expression of apoptotic markers with decreased caspase 3 expression, decreased cytochrome c and Bax (pro-apoptotic) and elevated Bcl-2 (anti-apoptotic) levels in the hippocampus compared to the sedentary control Tg group.[38] The balance between the amount of Bax and Bcl-2 is crucial to maintain cellular homeostasis and prevent cellular damage with resulting impairments.[38] This evidence supports the use of moderate exercise to reduce apoptotic levels in AD brain with a resulting positive impact on cognitive function. The impact of chronic stress may alter the neuroprotective effect of exercise. Haack et al. focused on the impact of chronic stress and chronic exercise (21 days) on the aging brain.[66] The greatest decrease in Bax formation across all groups was seen in the chronic stress/exercise group.[66] This leads to further evidence that exercise can have a protective function even with the addition of chronic stress.[66] Exercise may be able to reverse damage created by chronic stress levels.


Based on the current research, apoptotic markers progressively elevate with the normal aging process.[65] Exercise can attenuate apoptotic markers in the aging and AD brains for various animal models.[38][65] After the initial increase in apoptotic markers due to stress during acute exercise, all levels of pro-apoptotic markers decreased and anti-apoptotic markers increased in response to chronic exercise.[38][62][65][66] This impact at the cellular level has been associated with improvement at the functional level in cognition. Moderate aerobic exercise for greater than 3 week (chronic) period of time is recommended for prevention of neuronal apoptosis with improvement in cognition in the AD population.

Heat Shock Proteins

Heat shock proteins (Hsps) serve various functions within the cell including folding new proteins, refolding misfolded proteins, preventing protein aggregation, and transporting proteins.[67] Given these functions, Hsps play an important role in AD because they help prevent the accumulation and aggregation of Aβ and tau proteins in the brain which are key features of the disease.[68] It has been found that Hsp expression increases in response to cellular stress such as extreme temperature changes, infection, oxidative stress, or ischemia.[69][70] Additionally, it has been found that cells with higher levels of Hsps demonstrate improved tolerance to these stressors that would normally cause damage.[67][70] Exercise is another type of stressor that has also been associated with increased expression of Hsps.[69] Although the exact mechanisms that lead to increased Hsp expression with exercise are not fully understood, the changes in temperature, oxidative stress, energy availability, and oxygen levels associated with exercise are thought to play a role.[69] Prolonged or more intense bouts of exercise have been shown to correspond with greater increases in Hsps compared to shorter or less intense bouts of exercise.[69]

Hsps are separated into different “families” based on their size, location, and function.[69] These families include the small Hsps (sHsps), Hsp60, Hsp70, and Hsp90.[67] Small Hsps, such as Hsp27 and αB-crystallin, are mainly found in muscle cells but have also been detected in the brain.[68] These proteins are involved in many functions including stabilizing microfilaments and assisting larger Hsps with protein refolding.[67] [68] Of all the Hsp families, Hsp70 has been reported to be the most prevalent and easily expressed.[69] Given these characteristics, Hsp70 is the most commonly studied type of Hsp.[67] These proteins also serve a wide variety of functions in the cell to protect against damage during both normal and stressful cellular conditions.[69]

In order to determine whether exercise would be beneficial for individuals with AD through improvements in Hsp expression, a summary of various animal and human studies has been provided in the tables below.

Table 12. Exercise and Hsp Expression Animal Studies

Authors Type of Study Subjects Exercise Intervention Cellular Measure Intervention Effects/Results
Um HS, Kang EB, Leem YH, et al. (2008) [71] Randomized controlled trial 10 transgenic mice (Tg) with a Swedish mutation of APP (NSE/APPsw) and 10 non-transgenic (Non-Tg) mice that were 13 months old. Mice were divided into exercise and sedentary groups. Exercise mice performed treadmill training at 22 cm/sec (0% grade) for 60 min/day, 5 days/week for 16 weeks Hsp-70 protein levels in the brain (location in brain not specified) In the sedentary group, Tg mice had much lower levels of HSP-70 than non-Tg mice. A significant increase in HSP-70 was found in the brains of Non-Tg and Tg mice in the exercise group compared to their counterparts in the sedentary group.
Um HS, Kang EB, Koo JH, et al. (2011) [38] Randomized controlled trial 16 transgenic mice (Tg) with a PS2 mutation (Tg-NSE/hPS2m) and 16 non-transgenic (Non-Tg) mice that were 24 months old. Mice were divided into exercise and sedentary groups. Exercise mice performed treadmill training at 12 m/min (0% grade) for 60 min/day, 5 days/week for 3 months Expression of Hsp-70 in the hippocampus Among the sedentary mice, Tg mice had lower HSP-70 levels than non-TG mice. Increased HSP-70 expression in the hippocampus was found in Tg and non-Tg exercise mice compared to sedentary mice.
Hu S, Ying Z, Gomez-Pinilla F, Frautschy SA (2009) [72] Posttest-only control group design 2 month old male rats that were sacrificed at baseline (sedentary rats), after 3 days of exercise, or after 7 days of exercise (exercise rats). A wheel was placed in the cage of each exercise rat to allow for voluntary running for 3 or 7 days. Levels of HSP27, α-B-crystallin, and HSP70 in the hippocampus. After 3 days of exercise, higher levels of α-B-crystallin were found in the hippocampi of exercise rats compared to sedentary rats. After 7 days of exercise, significantly higher levels of HSP27 and α-B-crystallin were found in the hippocami of exercise rats compared to sedentary rats. HSP70 levels did not change.
Campisi J, Leem TH, Greenwood BN, et al. (2003) [70] Randomized controlled trial Adult male rats were divided into active and sedentary groups. Following 8 weeks, rats from each group were further divided in no stress, inescapable tail-shock stress (IS), and exhaustive exercise stress (EXS) groups. A functional wheel was placed in the cages of active rats to allow for voluntary running and a locked, nonfunctional wheel was placed in the cages of sedentary rats for 8 weeks. Hsp-72 levels in the brain (hypothalamus , hippocampus, dorsal vagal complex, prefrontal cortex), peripheral tissues, and immune system tissues After undergoing a stressful situation (IS or EXS), active rats displayed higher Hsp-72 levels in nearly all brain areas examined (except the hypothalamus) compared to sedentary mice exposed to the same stress. Similar trends were observed in the peripheral and immune tissues. Additionally, the increased expression of Hsp-72 after IS occurred more quickly in active rats compared to sedentary rats.
Liebelt B, Papapetrou P, Ali A, et al. (2010) [73] Randomized controlled trial 3 month old rats were divided into treadmill exercise or no exercise (control) groups. Stroke was induced in control rats without exercise preconditioning and exercise rats after exercise preconditioning. Exercise rats performed treadmill running at 30 m/min for 30 min/session, 5 days/week for 1, 2, or 3 weeks Levels of HSP-70 after 1, 2, or 3 weeks of exercise. Additional measures included brain infarct size, cerebral apoptosis, and apoptotic protein levels after stroke. After 1 and 2 weeks, HSP-70 levels did not change in exercise rats compared to control rats. After 3 weeks, HSP-70 levels were significantly greater (70%) in exercise rats compared to control rats. After stroke, rats that had previously exercised for 3 weeks demonstrated significantly less brain damage, in terms of infarct size, apoptosis, and apoptotic protein levels, compared to control rats.

These rodent studies demonstrate that aerobic exercise, in the form of treadmill or wheel running, elicits increased expression of Hsp70 and sHsps in the brain.[38][70][71][72][73] The 2008 and 2011 studies conducted by Um and colleagues found that an hour of treadmill running on 5 days/week for 3 to 4 months leads to increased expression of Hsp70 in transgenic mouse models of AD.[38][71] These findings were also supported by studies involving non-transgenic mice.[70][73] In the 2003 study by Campisi et al., Hsp72 levels increased after 8 weeks of voluntary wheel running.[70] Additionally, the 2010 study by Liebelt et al. reported increased levels of Hsp70 after 3 weeks of forced treadmill running but not after 1 or 2 weeks.[73] Conversely, the 2009 study by Hu et al. found no increase in Hsp70 after 7 days of wheel running, however, they did report increased expression of sHsps including α-B-crystallin and Hsp-27.[72] A possible explanation for these contradictory findings is that Hsp70 responds to chronic exercise while sHsps respond to acute exercise.[73]

In addition to showing the effect of exercise on Hsp expression, the Campisi et al. and Liebelt et al. studies also demonstrate that prior participation in exercise can improve the brain’s ability to respond to stress.[70][73] In the Campisi study, rats with previous access to a running wheel for 8 weeks demonstrated higher levels of Hsp-72 in the brain, peripheral, and immune tissues in response to stress induced by electrical shock and exhaustive treadmill running than their counterparts who did not have access to a running wheel.[70] Additionally, the researchers reported that exercise rats increased their Hsp levels more quickly than sedentary rats.[70] In the Liebelt study, rats with higher levels of Hsp-70 in their brains resulting from exercise preconditioning demonstrated less severe brain damage, in terms of brain infarct size, number of cells undergoing apoptosis, and apoptotic protein expression, following ischemia and reperfusion.[73] Taken together, these results show that aerobic exercise can improve the brain’s ability to produce Hsps in response to stress and that these proteins can protect the cells against damage.[70][73] Based on these findings, one can hypothesize that older individuals who regularly participate in exercise may be at a lower risk for AD because their brain responds more quickly and effectively to stress by increasing levels of Hsps. It is also possible that individuals with AD may reduce further damage to their brain by adding exercise to their treatment plan.

Table 13. Exercise and Hsp Expression Human Studies

Authors Type of Study Subjects Exercise Intervention Cellular Measure Intervention Effects/Results
Lancaster GI, Moller K, Nielson B, Secher NH, Febbraio MA, Nybo L (2004) [74] One-group pretest-posttest design 6 males with a mean age of 26 years who were accustomed to endurance exercise Subjects completed a single cycling session for 3 hours at 60% of their VO2max. Every 15 minutes, they drank 250mL of a liquid containing 6% carbohydrate. Serum levels of Hsp72 protein in samples of cerebral venous and peripheral arterial blood Arterial Hsp72 levels increased in 4 subjects and did not change in 2 subjects. Cerebral Hsp72 levels increased in 3 subjects and did not change in 3 subjects.
Liu Y, Lormes W, Wang L, Reissnecker S, Steinacker JM (2004) [75] One-way repeated measures design 6 males with a mean age of 19 years who were trained in rowing All subjects completed 2 training programs that each lasted 3 weeks with a 1 week break in between. The first program consisted primarily of high-intensity strength training with minimal rowing. The second program consisted primarily of low-intensity rowing with minimal strength training. Levels of HSP70 protein and mRNA in the vastus lateralis muscle After the high-intensity program, HSP70 protein and mRNA levels increased significantly compared to baseline. HSP70 protein levels decreased during the 1 week break and did not change in response to the low-intensity program. HSP70 mRNA levels progressively decreased during the 1 week break and the low-intensity program.
Morton JP, Holloway K, Woods P, et al. (2009) [76] One-way repeated measures design 5 active males with a mean age of 21 years and 5 active females with a mean age of 20 years All subjects completed an interval and continuous training program. Each program was performed 3 days/week for 6 weeks with a 6 week break in between. Interval training consisted of running at high- (100% VO2max) and moderate- (50% VO2max) intensity for 30 min. Continuous training consisted of running at 70% VO2max for the same distance as interval training. Concentration of αB-crystallin, HSP27, HSP60, HSP70 in the vastus lateralis muscle In males, αB-crystallin and HSP60 were significantly increased after interval training while only αB-crystallin was significantly increased after continuous training. In females, no significant changes in Hsps were observed following the completion of either program.
Konopka AR, Douglass MD, Kaminsky LA, et al. (2010) [77] One-group pretest-posttest design Nine elderly females with a mean age of 70 years All subjects completed a 12 week aerobic training program consisting of cycling on a cycle ergometer for 20-45 min/session at 60-80% heart rate reserve on 3-4 days/week. Levels of HSP70 protein and mRNA in the vastus lateralis muscle After 12 weeks of aerobic training, HSP70 mRNA levels decreased in the vastus lateralis while HSP70 protein levels remained the same compared to baseline.

The majority of human studies have investigated the relationship between exercise and Hsp expression in peripheral tissues, primarily skeletal muscle, without consideration of central effects.[75][76][77] Therefore, research regarding the effect of exercise on Hsp expression in the human brain is lacking. Only one study was found that examined this relationship in human subjects.[74] This study assessed the effect of a single bout of prolonged cycling (3 hours) at moderate-intensity on extracellular Hsp72 protein levels in blood samples from a cerebral vein and peripheral artery in 6 young male subjects.[74] After the exercise intervention, Hsp72 levels increased in the arterial blood of 4 subjects and in the cerebral venous blood of 3 subjects.[74] These findings show that acute, moderate-intensity aerobic exercise facilitates the release of Hsp72 from cells in the brain in some individuals.[74] While it is known that Hsps inside the cell prevent damage and maintain homeostasis, their exact function outside the cell is not fully understood.[74] It has been suggested that extracellular Hsps interact with and activate cells of the immune system.[74] Additionally, it has been proposed that extracellular Hsps can be taken in by motor neurons and used for protection against damage from stress.[78]

The study conducted by Liu et al. suggests that Hsp expression is affected by exercise intensity.[75] These researchers examined the effect of a 3 week high-intensity strength program and a 3 week low-intensity endurance program on Hsp70 protein and mRNA levels in the skeletal muscle of healthy, young males.[75] They found that protein and mRNA levels increased in response to the high-intensity program but not in response to the low-intensity program.[75] The researchers suggest that high-intensity exercise creates more pronounced physiological changes and, therefore, has a greater effect on Hsp expression.[75] The study by Morton et al. also reported increased Hsp expression in response to a 6 week moderate- to high-intensity aerobic exercise program in young, healthy males.[76] In this study, however, levels of αB-crystallin and HSP60 increased rather than Hsp70.[76]

In addition to exercise intensity, the results from the study by Morton et al. suggest that Hsp expression is also affected by gender.[76] This study found that Hsp levels in skeletal muscle increased in men following 6 weeks of interval (moderate- and high-intensity) or continuous (moderate-intensity) aerobic training but did not change in women.[76] These findings were supported by a randomized controlled trial involving rats that examined Hsp70 protein and mRNA levels in myocardial tissue following moderate-intensity treadmill running for one hour.[79] This study reported that male rats and female rats without ovaries demonstrated greater increases in cardiac Hsp70 expression following exercise than female rats with ovaries or those without ovaries that were treated with estrogen.[79] Additionally, the 2010 study by Konopka et al. reported a decrease in Hsp70 mRNA levels and no change in Hsp70 protein levels in the skeletal muscle of elderly women after completing a 12 week moderate-intensity aerobic training program.[77] This study, however, did not make comparisons with male subjects.[77] Various hypotheses were offered among these studies to explain why females did not produce Hsps in response to exercise. The researchers who conducted the Morton study proposed that estrogen protects against cell damage and, because of this, females must perform higher-intensity exercise to create enough stress to trigger Hsp expression.[76] Similarly, those involved in the Paroo study postulated that estrogen inhibits Hsp70 production in cardiac cells of females in response to stress.[79] Conversely, the Konopka study researchers suggested that the muscle of their elderly female subjects became accustomed to the stress associated with aerobic training and, therefore, decreased production of Hsp70 mRNA.[77] These findings illustrate the need for further research to examine whether these gender differences apply to Hsp expression in the brain. If so, than males with AD would be more likely to benefit from exercise, with regards to Hsp expression, than females with the disease.


In summary, animal and human studies demonstrate that exercise can facilitate increased Hsp expression in the brain and skeletal muscle. The effect of exercise on Hsp expression in the brain has primarily been investigated in animals.[38][70][71][72][73] These studies show that chronic aerobic exercise can lead to significant increases in Hsp70 and acute aerobic exercise can lead to significant increases in sHsps.[38][70][71][72][73] In the human brain, moderate-intensity acute aerobic exercise has been shown to increase extracellular levels of Hsp72 in healthy, active males.[74] The effect of exercise on intracellular Hsp concentrations in the brain has yet to be studied in humans. In human skeletal muscle, chronic aerobic exercise or strength training of at least moderate-intensity facilitated increased Hsp levels in healthy, young males but not in young or old females.[75][76][77] Further research is necessary to determine the effect of exercise on Hsp expression in the human brain and whether gender differences also occur in this tissue.

Effect of Exercise on Cognition


Now that we have examined the effects of exercise on the cellular components related to AD, we will consider how they translate into changes in cognition in humans. Clinical trials have produced conflicting results regarding the effect of exercise on cognition.[80] Some studies have reported that exercise helps maintain and improve cognition while others have failed to find a relationship between these variables.[80] A recent meta-analysis of 29 randomized controlled trials revealed that aerobic exercise facilitates “modest” improvements in attention, speed of information processing, executive functions, and long-term memory in healthy adults and those with mild cognitive impairments.[80] Many of these cognitive improvements were greater when strengthening was performed in addition to aerobic exercise.[80] A positive relationship between exercise and cognition was also reported in a literature review of studies involving older adults with AD.[81] This review also found evidence that aerobic exercise, alone or as part of a comprehensive program, improves fitness levels, functional ability, ADL performance, psychological status, and behavior in individuals with AD.[81]

Refer to the tables below for a summary of specific research studies investigating the impact of exercise on cognition in older adults without AD (Table 14) and older adults with AD (Table 15).

Table 14. Effect of Exercise on Older Adults without AD

Authors Type of Study Subjects Exercise Intervention Cognitive Outcome Measures Intervention Effects/Results
Cassilhas RC, Viana VA, Grassmann V, et al. (2007) [82] Randomized controlled trial 62 male subjects between the ages of 65-75 that did not have dementia and were previously sedentary. Subjects were randomly assigned to a control group, experimental moderate group (EMOD), or an experimental high (EHIGH) group. Subjects in the experimental groups completed a 24 week resistance training program that was performed 3 days/week for 1 hour. Six exercises using the major muscle groups were performed (2 sets of 8). EMOD subjects worked at 50% of 1 repetition maximum (RM) and EHIGH subjects worked at 80% of 1 RM. Control subjects attended weekly warm-up and stretching classes. Wechsler Adult Intelligence Scale III (WAIS III), Wechsler Memory Scale-Revised (WSM-R), Toulouse-Pieron’s concentration attention test, and Rey-Osterrieth complex figure After 24 weeks, subjects in both experimental groups performed better on cognitive tasks requiring short- and long-term memory, executive function, and attention than control subjects. These improvements were not significantly different between EMOD and EHIGH groups.
Lautenschlager NT, Cox KL, Flicker L, et al. (2008) [83] Randomized controlled trial 170 subjects ≥50 years of age with subjective reports of decreased memory. Subjects were excluded if they had severe cognitive impairments, dementia, depression, mental illness, or certain medical conditions. Subjects were randomly assigned to an intervention or control group. Subjects in the intervention group completed a 24 week exercise program that was performed 3 days/week in their home. Each session was 50 minutes in duration and consisted of aerobic exercise and/or strength training. A staff member helped devise individualized programs for each subject and provided them with information to improve their exercise behavior. Cognitive portion of the Alzheimer Disease Assessment Scale (ADAS-cog), Cognitive Battery of the Consortium to Establish a Registry for Alzheimer Disease, Digit Symbol-Coding Test, Delis-Kaplin Executive Function Battery, and Clinical Dementia Rating scale After 24 weeks, subjects in the intervention group achieved better scores on the ADAS-cog compared to the control group. Intervention subjects also performed better on delayed recall and received lower scores on the Clinical Dementia Rating than control subjects. At an 18 month follow-up, the intervention group continued to have better scores on these outcome measures compared to the control group.
Jedrziewski MK, Ewbank DC, Wang H, Trojanowski JQ. (2010) [84] Cohort study 5,280 community-dwelling subjects ≥ 65 years of age were assessed at baseline. Subjects with moderate or severe cognitive deficits were excluded following the initial assessment. The remaining subjects (3,863) were followed from 1994 to 2004 to monitor changes in cognition. No intervention. At baseline, subjects were asked to describe their level of physical activity in terms of frequency, types, and duration. Short Portable Mental Status Questionnaire (SPMSQ) and Mini-Mental State Examination (MMSE) At baseline, subjects who engaged in physical activity were more likely to have intact or mildly impaired cognition. At the 10 year follow-up, subjects who engaged in a greater variety of activities and participated in bouts of activity lasting at least 20 minutes were significantly less likely to have impaired cognition.

Table 15. Effect of Exercise on Older Adults with AD

Authors Type of Study Subjects Exercise Intervention Cognitive Outcome Measures Intervention Effects/Results
Venturelli M, Scarsini E, Schena F. (2011) [85] Randomized controlled trial 21 subjects in the advanced stages of AD who were living in a nursing home. All subjects were ≥65 years of age, required assistance with ADLs, had an MMSE score between 5 - 15, and lacked mobility limitations. Subjects were randomly assigned to a walking or control group. Subjects in the walking group completed a 24 week walking program that was performed 4 days/week in the hallways of the nursing home. Each walking session was 30 min in duration and was completed in the presence of a caregiver who encouraged the subject to walk as fast as possible to facilitate moderate-intensity exercise. Mini-Mental State Examination (MMSE) After 24 weeks, the mean score on the MMSE did not significantly change in the walking group (-1 point) while it declined in the control group (-6 points).
Vreugdenhil A, Cannell J, Davies A, Razay G. (2012) [86] Randomized controlled trial 40 subjects with a diagnosis of AD who lived in the community. At baseline, the mean age and MMSE score of the subjects were 74.1 years and 22.0 points, respectively. Subjects were randomly assigned to an exercise or control group. Subjects in the exercise group completed a 4 month exercise program that was performed daily in their home under the supervision of a trained caregiver. The program consisted of 30 minutes of “brisk” walking and 10 exercises to improve strength and balance. MMSE and cognitive portion of the Alzheimer’s Disease Assessment Scale (ADAS-cog) After 4 months, mean scores on the MMSE and ADAS-cog improved in the exercise group while they declined in the control group. (MMSE: + 1.0 versus -2.0; ADAS: -4.2 versus +4.0)
Kemoun G, Thibaud M, Roumagne N, et al. (2010) [87] Randomized controlled trial 31 subjects >75 years of age diagnosed with “Alzheimer-type dementia” (MMSE score <23) and living in a nursing home. Subjects were randomly assigned to an intervention or control group. Subjects in the intervention group completed a 15 week physical activity program performed 3 days/week for 1 hour. The first 2 weeks consisted of ROM and muscle stimulation exercises. The next 13 weeks consisted of a 10 min warm-up, 40 min of exercises to improve gait mechanics, balance, and endurance (60-70% reserve cardiac frequency), and a 10 min cool-down. Rapid Evaluation of Cognitive Functions test (ERFC) After 15 weeks, the mean score on the ERFC improved in the intervention group (+3.57) and declined in the control group (-5.1 points).
Yu F & Kolanowski A. (2009) [88] Case Study (Pretest - posttest design) 2 subjects diagnosed with mild to moderate AD. The first subject was a 75 year old male who lived at home with his wife. The second subject was an 86 year old female who lived in a retirement community. Both subjects completed a 2 month aerobic exercise program that was performed 3 days/week under the supervision of a trainer. The program consisted of a 5 minute warm-up, 10-30 minutes of cycling on a recumbent bicycle at moderate-intensity (60-65% of maximal HR), and a 10 minute cool-down. MMSE and Stroop Neurophysiological test After 2 months, both subjects demonstrated improvements on the Stroop test. Neither subject could perform this test at baseline due to confusion but were successfully able to complete the test after the intervention. Scores on the MMSE decreased by 3 points for subject 1 and remained the same for subject 2.

In the studies involving older adults without AD, both chronic aerobic exercise and resistance training facilitated improvements in cognition.[82][83] Additionally, participation in exercise was associated with a decreased likelihood of having cognitive impairments at baseline or developing them over a 10 year period.[84] A positive relationship between exercise and cognition was also observed in the studies involving older adults diagnosed with AD.[85][86][87][88] Two studies, both of which used an intervention consisting of aerobic exercise in addition to strength and/or balance training, reported that exercise subjects showed improvements on cognitive outcome measures while control subjects showed declines.[86][87] The other two studies, both of which used aerobic interventions consisting of either walking or cycling, reported that exercise did not elicit improvements in cognition but did prevent a decline.[82][87] These conflicting results, however, may be attributed to the subject population in one study (more advanced cases of AD)[87] and involvement of only two subjects in the other.[88]

All of the studies involving adults diagnosed with AD also examined the effect of exercise on physical function.[85][86][87][88] This is important because individuals with AD are likely to experience disturbances in mobility which can increase their risk for falling and decrease their independence.[87] Two studies examined the effect of exercise on gait.[85][87] Both studies reported that exercise subjects demonstrated significant improvements in gait mechanics while control subjects showed declines.[85][87] Three studies examined the effect of exercise on ADL performance.[85][86][88] Of these studies, two reported that exercise subjects demonstrated improvements in ADL performance while control subjects showed declines.[85][86] The other study reported a decline in ADL performance after subjects completed the exercise intervention, however, this study included only 2 subjects.[87] Additional improvements in depression, balance, behavior, and caregiver burden were also reported in two of these studies.[86][88]


Figure 6. Potential Mechanisms for Improved Cognition with Exercise (Figure Source)


Improvements in cognition associated with exercise have also been reported in studies involving transgenic mouse models of AD.[37][89] Two studies, both of which used Tg2576 mice, found that mice with access to a running wheel performed significantly better on the radial-arm water maze (RAWM) than those without access to a running wheel.[37][89] The RAWM involves a 5-arm maze that has been submerged in water and contains a single platform that mice must locate in order to escape.[89] In the 2007 study by Nichol, Parachikova, and Cotman, transgenic and wild type mice were divided into exercise, wheel-locked, and sedentary groups.[89] Exercise mice had access to functional wheel for running, wheel-locked mice had access to a nonfunctional wheel, and sedentary mice did not have access to any wheel.[89] After 3 weeks, transgenic mice in the exercise group were comparable to wild type mice with regards to both short- and long-term memory as well as number of failures to find the platform.[89] Transgenic mice in the wheel-locked group were also comparable to wild type mice during all trials except 11-15 on days 1 and 2 of testing during which they displayed a poorer performance than exercise mice.[89] Transgenic mice in the sedentary group performed significantly worse than all other mice.[89] These mice failed to find the platform more often and showed greater impairments in long- and short-term memory.[89] These results were confirmed by a similar study conducted by the same researchers in 2008.[37]


In summary, studies show that chronic exercise can elicit either improvements or prevent declines in cognition in older adults without cognitive impairments, older adults diagnosed with AD, and transgenic mouse models of AD.[37][82][83][84][85][86][87][88][89] Moderate-intensity aerobic activity, primarily in the form of walking, for at least 30 minutes on 3 or more days each week was the main component of the exercise interventions used for individuals with AD.[85][86][87][88] Among these studies, greater benefits, with regards to cognition and physical function, were achieved when aerobic exercise was combined with other components such as strength and balance training.[86][87] It is possible that additional benefits could also be achieved through increasing the exercise duration and frequency. Therefore, the activity recommendations for older adults provided by the ACSM may be a useful reference when devising exercise programs for patients with this disease.[6] While the effect of exercise on cognition in healthy older adults has been studied extensively, research is lacking regarding individuals with AD.[87] Further research is necessary to gain a more thorough understanding of the relationship between exercise and cognition in the AD population using larger sample sizes and longer intervention durations. Researchers should also compare aerobic training, resistance training, and combination training to determine which type and intensity of exercise is most effective for these individuals.


There are several limitations regarding evidence on the impact of exercise on the cellular mechanisms associated with AD pathology. First, a majority of the studies used animal models including mice and rats.[37][36] The results found in animal models may not provide a direct connection to the response found in human subjects. Human studies are limited due to assessment techniques requiring access to various regions of the brain which hinders evaluation at the cellular level. Those studies that do use human subjects often use healthy young adults or healthy older adults.[55][56] Second, the sample sizes selected in these animal studies are relatively small which hinders the ability to generalize the results (power). [38] Third, the exercise intensity, duration, frequency and type varied significant across studies.[37][59] In order to establish an evidence-based recommended exercise regimen for people with AD, consistency is required in the research designs and findings. Finally, the cellular measures utilized lack homogeneity which impedes comparison across studies.[37][38][65][66] For accurate conclusions to be made regarding cellular response to exercise, cellular measures need to be comparable. Further research is warranted specifically on AD pathology and the impact of exercise on a cellular level.


In conclusion, exercise has the ability to positively affect AD. Aerobic exercise, specifically, has been studied and found to positively affect Aβ plaques, tau, apoptosis, inflammation, heat shock proteins, BDNF, mitochondria, brain size, and cognition.[2][34][41][38][36][65][69][81] The majority of studies have concluded that moderate intensity aerobic exercise is beneficial at the cellular level.[38][41][42][63][76] High intensity exercise has had mixed reviews demonstrating positive effects on BDNF and heat shock proteins, but being detrimental to mitochondria possibly increasing apoptosis.[52][63][75][76] Resistance exercise (strength training) and balance exercises have also been found to positively affect heat shock proteins and improve cognition.[75][82][83][86] Performing resistance and balance exercises two to three days per week is recommended.[82][87] Most studies agree that 30-60 minutes of voluntary, moderate-intensity exercise on most days of the week is enough to create positive cellular effects.[13][71][73] Chronic exercise is recommended over limited exercise bouts. In fact, creating a lifestyle habit of exercise is the best way to promote long-term health and improve the health of those with Alzheimer’s Disease.[6]


Figure 7. Elderly Individuals Participating in Aerobic Exercise (Figure Source)

Go to Alzheimer's Disease Cell Bio Page
Return To Welcome Page

1. Lucia A and Ruiz JR. Exercise is beneficial for patients with Alzheimer’s disease: a call for action. Br J Sports Med. 2011; 45(6): 468-469.
2. Foster PP, Rosenblatt KP, and Kuljis RO. Exercise-induced cognitive plasticity, implications for mild
cognitive impairment and Alzheimer’s disease. Frontiers in Neurology. 2011; 2(28): 2-15.
3. Finder VH. Alzheimer’s disease: a general introduction and pathomechanism. Journal of Alzheimer’s Disease.2010;22:S5-S19.
4. Chodzko-Zajko WJ, Proctor DN, Singh MA, Minson CT, Nigg CR, Salem GJ, Skinner JS. American college of sports medicine position stand. Exercise and physical activity for older adults. Medicine & Science in Sports & Exercise. 2009;41(7):1510-30.
5. Elsawy B, Higgins KE. Physical activity guidelines for older adults. American Family Physician. 2010;81(1):55-59.
6. Nelson ME, Rejeski J, Blair SN. Physical activity and public health in older adults: recommendation from the American College of Sports Medicine and the American Heart Association. Medicine & Science in Sports & Exercise. 2007;39(8):1435-45.
7. Department of Health and Human Services. 2008 Physical Activity Guidelines for Americans. Rockville (MD): U.S. Department of Health and Human Services; 2008.
8. American College of Sports Medicine. ACSM’s Guidelines for Exercise Testing and Prescription. 8th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2010.
9. Erickson KI, Prakash RS, Voss MW, et al. Aerobic fitness is associated with hippocampal volume in elderly humans. Hippocampus. 2009.
10. Erickson KI. Augmenting Brain and Cognition by Aerobic Exercise. Lecture Notes in Compute Science. 2011; 6780, 30-38.
11. Burns JM, Cronk BB, Anderson HS, et al. Cardiorespiratory Fitness and Brain Atrophy in Early Alzheimer’s Disease. Neurology. 2008; 71(3): 210–216.
12. Honea R, Thomas GP, Harsha A, et al. Cardiorespiratory fitness and preserved medial temporal lobe volume in Alzheimer’s disease. Alzheimer Dis Assoc Disord. 2009; 23(3): 188–197.
13. Erickson KI, Voss MW, Prakash RS, Basak C, Szabo A, Chaddock L, Kim JS, Heo S, Alves H, White SM, Wojcicki TR, Mailey E, Vieira VJ, Martin SA, Pence BD, Woods JA, McAuley E, and Kramer AF. Exercise training increases size of hippocampus and improves memory. Proc Nat Acad Sci USA. 2011; 108(7): 3017-22.
14. Cotman CW, Berchtold NC, Christie LA. Exercise builds brain health: key roles of growth factor cascades and inflammation. TRENDS in Neuroscience. 2007; 30:9.
15. Colcombe SJ, Kramer AF. Fitness effects on the cognitive function of older adults: a 
meta-analytic study. Psychological Science. 2003; 14: 125–130.
16. Colcombe SJ, Erickson KI, Scalf PE, et al. Aerobic exercise training increases brain volume in aging humans: evidence from a randomized clinical trial. Journal of Gerontology: Bio and Med Sci. 2006; 61: 1166–1170.
17. Voss MW, Prakash RS, Erickson KI, Basak C, Chaddock L, Kim JS, et al.: Plasticity of brain networks in a randomized intervention trial of exercise training in older adults. Frontiers in Aging Neuroscience. 2010; 2: 1–17.
18. Erickson KI, Raji CA, Lopez OL, Becker JT, et al. Physical activity predicts gray matter volume in late adulthood: The cardiovascular health study. Neurology. 2010; 75: 1415–1422.
19. Lautenschlager NT, Cox KL, Flicker L, et al. Effect of physical activity on cognitive function in older adults at risk for Alzheimer disease: a randomized trial. JAMA 2008; 300: 1027–1037.
20. Burns JM, Donnelly JE, A HS, et al. Cardiorespiratory Fitness and Brain Atrophy in Early Alzheimer's Disease. Neurology. 2008.
21. Ferris LT, Williams JS, and Chwan-Li S. The effect of acute exercise on serum brain-derived neurotrophic factor levels and cognitive function. Med. Sci. Sports Exercise. 2007; 39(4): 728-34.
22. Berchtold NC, Chinn G, Chou M, Kesslak JP, and Cotman CW. Exercise primes a molecular memory for brain-derived neurotrophic factor protein induction in the rat. Neuroscience. 2005; 133: 853-61.
23. Erickson KI, Prakash RS, Voss MW, Chaddock L, Heo S, McLaren M, Pence BD, Martin SA, Vieira VJ, Woods JA, McAuley E, Kramer AF. Brain-derived neurotrophic factor is associated with age-related decline in hippocampal volume. J Neurosci. 2010; 30(15): 5368-75.
24. Ahlskog JE, Geda YE, Graff-Radford NR, and Petersen RC. Physical exercise as a preventative or disease-modifying treatment of dementia and brain aging. Mayo Clin Proc. 2011; 86(9): 876-84.
25. Arancibia S, Silhol M, Mouliere F, Meffre J, Hollinger I, Maurice T, and Tapia-Arancibia L. Protective effect of BDNF against beta-amyloid induced neurotoxicity in vitro and in vivo in rats. Neurobiol Dis. 2008; 31: 316-26.
26. Hopkins ME, and Bucci DJ. Neurobiology of learning and memory. Neurobiol Learn Mem. 2010; 94: 278-84.
27. Nichol K, Deeny SP, Seil J, Camaclang K, and Cotman CW. Exercise improves cognition and hippocampal plasticity in APOE ε4 mice. Alz & Dementia. 2009; 5: 287-94.
28. Baj G, D’Alessandro V, Musazzi L, Mallei A, Sartori CR, Sciancalepore M, Tardito D, Langone F, Popoli M, Tongiorgi E. Physical exercise and antidepressants enhance BDNF targeting in hippocampal CA3 dendrites: further evidence of a spatial code for BDNF splice variants. Neuropharmacology. 2012; 1-12.
29. Radak Z, Toldy A, Szabo Z, Siamilis S, Nyakas C, Silye G, Jakus J, Goto S. The effects of training and detraining on memory, neurotrophins and oxidative stress markers in rat brain. Neurochem Int. 2006; 49: 387-92.
30. Wolf SA, Kronenberg G, Lehmann K, Blankenship A, Overall R, Staufenbiel M, and Kempermann, G. Cognitive and physical activity differently modulate disease progression in the amyloid precursor protein (APP)-23 model of Alzheimer’s Disease. Biol Psychoiatry. 2006; 60: 1314-23.
31. Currie J, Ramsbottom R, Ludlow H, Nevill A, and Gilder M. Cardio-respiratory fitness, habitual physical activity and serum brain derived neurotrophic factor (BDNF) in men and women. Neurosci Lett. 2009; 451(2): 152-5.
32. Chan KL, Tong, KY, and Yip SP. Relationship of serum brain derived neurotrophic factor (BDNF) and health related lifestyle in healthy human subjects. Neurosci Lett. 2008: 124-8.
33. Berchtold NC, Castello N, and Cotman CW. Exercise and time-dependent benefits to learning and memory. Neuroscience. 2010; 167(3): 588-97.
34. Adlard, P., Perreau, V., Pop, V., Cotman, C. 2005. Voluntary exercise decreases amyloid load in a transgenic model of Alzheimer’s disease. The Journal of Neuroscience; 25(17):4217-4221.
35. Ke, H., Huang, H., Liang, K., Hsieh-Li, H. 2011. Selective improvement of cognitive function in adult and aged APP/PS1 transgenic mice by continuous non-shock treadmill exercise. Brain research; 1403: 1-11.
36. Nichol KE, Poon WW, Parachikova AI, Cribbs DH, Glabe CG, Cotman CW. Exercise alters the immune profile in Tg2576 Alzheimer mice toward a response coincident with improved cognitive performance and decreased amyloid. Journal of Neuroinflammation. 2008;5:13.
37. Parachikova A, Nichol KE, Cotman CW. Short-term exercise in aged Tg2576 mice alters neuroinflammation and improves cognition. Neurobiology of Disease. 2008;30:121-129.
38. Um HS, Kang EB, Goo JH, Kim HT, Lee J, Kim EJ, Yang CH et al. Treadmill exercise represses neuronal cell death in an aged transgenic mouse model of alzheimer’s disease. Neuroscience Research. 2011;69:161-173.
39. Yuede, C., Zimmerman, S., Dong, H., Kling, M., Bero, A., Holtzman, D., Timson, B., & Csernansky, J. 2009. Effects of voluntary and forced exercise on plaque deposition, hippocampal volume, and behavior in the Tg2576 mouse model of Alzheimer’s disease. Neurobiology of Disease; 35: 426-432.
40. Belarbi K, Burnouf S, Fernandez-Gomez FJ, Laurent C, Lestavel S, Figeac M, Sultan A, Troquier L, Leboucher A, Caillierez R, Grosjean ME, Demeyer D, Obrit H, Brion I, Barbot B, Galas MC, Staels B, Humez S, Sergeant N, Schraen-Maschke S, Muhr-Taileux A, Hamdane M, Bucee L, and Blum D. 2011. Beneficial effects of exercise in a transgenic mouse of Alzheimer’s disease-like Tau pathology. Neurobiology of Disease; 43: 486-494.
41. Leem YH, Lim HJ, Shim SB, Cho JY, Kim BS, and Han PL. (2009). Repression of tau hyperphosphorylation of chronic endurance exercise in aged transgenic mouse model of tauopathies. Journal of Neuroscience Research; 87: 2561-2570.
42. Liang, K., Mintun, M., Fagan, A., Goate, A., Bugg, J., Holtzman, D., Morris, J., & Head, D. 2010. Exercise and Alzheimer’s disease biomarkers in cognitively normal older adults. Ann Neurol; 68(3); 311-318.
43. Fagan, A., Mintun, M., Mach, R., Lee, S., Dence, C., Shah, A., LaRossa, G., Spinner, M., Klunk, W., Mathis, C., DoKosky, S., Morris, J., & Holtzman, D. 2006. Inverse relation between in vivo amyloid imaging load and cerebrospinal fluid Aβ-42 in humans. Ann Neurol; 59: 512-519.
44. Mitchell, A. CSF phosphorylated tau in the diagnosis and prognosis of mild cognitive impairment and Alzheimer’s disease: a meta-analysis of 51 studies. 2009. J Neurol Neurosurg Psychiatry; 80: 966-975.
45. Olsson, A., Vanderstichele, H., Andreasen, N., Meyer, G., Wallin, A., Holmberg, B., Rosengren, L., Vanmechelen, E., & Blennow, K. 2005. Simultaneous measurement of beta-amyloid (1-42), total tau, and phosphorylated tau (Thr181) in cerebrospinal fluid by the xMAP technology. Clin Chem; 51: 336-345.
46. Hauptmann S, Keil U, Scherpig I, Bonert A. Mitochondrial dysfunction in sporadic and genenetic Alzheimer’s disease. Experimental Gerontology. 2006; 41(7): 668-673
47. Sheng B, Wang X, Su B, Lee HG, Casadesus G, Perry G, Zhu X. Impaired mitochondrial biogenesis contributes to mitochondrial dysfunction in Alzheeimer’s disease. J. Neurochem. 2012; 120: 419-429
48. Boveris A and Navarro A. Brain mitochondrial dysfunction in aging. Life. 2008; 60(5): 308-314
49. Navarro A, Gomez C, Lopez-Cepero J, Boveris A. Beneficial effects of moderate exercise on mice aging: survival. Behavior, oxidative stress, and mitochondrial electron transfer. Am J Physiol Regul Integr Comp Physiol. 2004; 286: R505-511
50. Kirchner L, Chen WQ, Afjehi-Sadat L, Viidik A, Skalicky M, Hoger H, Lubec G. Hippocampal metabolic proteins are modulated in voluntary and treadmill exercise rats. Experimental Neurology. 2008; 212: 145-151
51. Steiner JL, Murphy EA, McClellan JL, Carmichael MD, Davis JM. Exercise training increases mitochondrial biogenesis in the brain. J Appl Physiol. 2011; 111: 1066–1071.
52. Aguiar AS, Tuon T, Pinho CA, Silva LA, Andreazza AC. Intense exercise induces mitochondrial dysfunction in mice brain. Neurochemical Research. 2008; 33(1)51-58
53. Boveris A and Navarro A. Systemic and mitochondrial adaptive responses to moderate exercise in rodents. Free Radical Biology & Medicine. 2008; 44: 224-229
54. Scarmeas N, Luchsinger JA, Schupf N, Brickman AM, Cosentino S, Tang MX, et al. Physical activity, diet and risk of Alzheimer’s disease. JAMA. 2009;302(6): 627-637.
55. Nybo L, Nielsen B, Pedersen BK, Moller K, Secher NH. Interleukin-6 release from the human brain during prolonged exercise. Journal of Physiology. 2002;542(3): 991-995.
56. Steensberg A, Dalsgaard MK, Secher NH, Pedersen BK. Cerebrospinal fluid IL-6, HSP72, and TNF-α in exercising humans. Brain, Behavior and Immunity. 2006;20: 585-589.
57. Leem YH, Lee YI, Son HJ, Lee SH. Chronic exercise ameliorates the neuroinflammation in mice carrying NSE/htau23. Biochemical and Biophysical Research Communications. 2011;406:359-365.
58. Pervaiz N, Hoffman-Goetz L. Freewheel training alters mouse hippocampal cytokines. Int J Sports Med. 2011;32: 889-895.
59. Nicklas BJ, Hsu FC, Brinkley TJ, Church T, Goodpaster BH, Kritchevsky SB, Pahor M. Exercise training and plasma c-reactive protein and interleukin-6 in elderly people. JAGS. 2008;56: 2045-2052.
60. Kohut ML, McCann DA, Russell DW, Konopka DN, Cunnick JE, Franke WD, et al. Aerobic exercise, but not flexibility/resistance exercise, reduces serum IL-18, CRP, and IL-6 independent of B-blockers, BMI, and psychosocial factors in older adults. Brain, Behavior, and Immunity. 2006;20:201-209.
61. Lista I, Sorrentino G. Biological mechanisms of physical activity in preventing cognitive decline. Cell Mol Neurobiol. 2010;30: 493-503.
62. Kerr Al, Swain RA. Rapid cellular genesis and apoptosis: effects of exercise in the adult rat. Behavioral Neuroscience. 2011;125(1):1-9.
63. Kocturk S, Kayatekin BM, Resmi H, Acikgoz O, Kaynak C, Ozer E. The apoptotic response to strenuous exercise of the gastrocnemius and soleus muscle fibers in rats. Eur J Appl Physiol. 2008;102: 515-524.
64. Quadrilatero J, Bombardier E, Norris SM, Talanian JL, Palmer MS, Logan HM, et al. Prolonged moderate-intensity aerobic exercise does not alter apoptotic signaling and DNA fragmentation in human skeletal muscle. Am J Physiol Endocrinol Metab. 2010;298: E534-E547.
65. Kim SE, Ko IG, Kim BK, Shim MS, Cho S, Kim CJ et al. Treadmill exercise prevents aging-induced failure of memory through an increase in neurogenesis and suppression of apoptosis in rate hippocampus. Experimental Gerontology. 2010;45: 357-365.
66. Haack D, Luu H, Cho J, Chen MJ, Russo-Neustadt A. Exercise reverses chronic stress-induced bax oligomer formation in the cerebral cortex. Neuroscience Letters. 2008;438:290-294.
67. Morton JP, Kayani AC, McArdle A, Drust B. The exercise-induced stress response of skeletal muscle, with specific emphasis on humans. Sports Med. 2009;39(8):643-662.
68. Wilhelmus MM, de Waal RM, Verbeek MM. Heat shock proteins and amateur chaperones in amyloid-beta accumulation and clearance in Alzheimer’s disease. Mol Neurobiol.2007;35:203-216.
69. Madden LA, Sandstrom ME, Lovell RJ, McNaughton L. Inducible heat shock protein 70 and its role in preconditioning and exercise. Amino Acids. 2008;34:511-516.
70. Campisi J, Leem TH, Greenwood BN. Habitual physical activity facilitates stress-induced HSP72 induction in brain, peripheral, and immune tissues. Am J Physiol Regul Integr Comp Physiol.2003;284:R520-R530.
71. Um HS, Kang EB, Leem YH, et al. Exercise training acts as a therapeutic strategy for reduction of the pathogenic phenotypes for Alzheimer’s disease in an NSE/APPsw-transgenic model. International Journal of Molecular Medicine. 2008;22:529-539.
72. Hu S, Ying Z, Gomez-Pinilla F, Frautschy SA. Exercise can increase small heat shock proteins (sHSP) and pre- and post-synaptic proteins in the hippocampus. Brain Research.2009;1249: 191-201.
73. Liebelt B, Papapetrou P, Ali A, et al. Exercise preconditioning reduces neuronal apoptosis in stroke by up-regulating heat shock protein-70 and extracellular signal-regulated-kinase 1/2. Neuroscience. 2010;166:1091-1100.
74. Lancaster GI, Moller K, Nielsen B, Secher NH, Febbraio MA, Nybo L. Exercise induces the release of heat shock protein 72 from the human brain in vivo. Cell Stress & Chaperones. 2004;9(3):276-280.
75. Liu Y, Lormes W, Wang L, Reissnecker S, Steinacker JM. Different skeletal muscle HSP70 responses to high-intensity strength training and low-intensity endurance training. Eur J Appl Physiol. 2004;91:330-335.
76. Morton JP, Holloway K, Woods P, et al. Exercise training-induced gender-specific heat shock protein adaptations in human skeletal mucle. Muscle Nerve.2009;39:230-233.
77. Konopka AR, Douglass MD, Kaminsky LA, et al. Molecular adaptations to aerobic exercise training in skeletal muscle of older women. Journal of Gerontology. 2010;65A(1):1201-1207.
78. Krause M, Rodrigues-Krause JC. Extracellular heat shock proteins (eHSP70) in exercise: possible targets outside the immune system abd thier role for neurodegenerative disorders treatment. Medical Hypotheses. 2011;76:286-290.
79. Paroo Z, Haist JV, Karmazyn M, Noble EG. Exercise improves postischemic cardiac function in males but not females: consequences of a novel sex-specific heat shock protein 70 response. Circ Res. 2002;90:911-917.
80. Smith PJ, Blumenthal JA, Hoffman BM, et al. Aerobic exercise and neurocognitive performance: a meta-analytic review of randomized controlled trials. Psychosom Med. 2010;72(3):239-252.
81. Yu F. Guiding research and practice: a conceptual model for aerobic exercise training in Alzheimer’s disease. American Journal of Alzheimer’s Disease and Other Dementias.2011;26(3):184-194.
82. Cassilhas RC, Viana VA, Grassmann V, et al. The impact of resistance exercise on the cognitive function of the eldery. Medicine & Science in Sports & Exercise. 2007;39(8):1401-1407.
83. Lautenschalger NT, Cox KL, Flicker L. Effect of physical activity on cognitive function in older adults at risk for Alzheimer disease: a randomized trial. JAMA. 2008;300(9):1027-1037.
84. Jedrziewski MK, Ewbank DC, Wang H, Trojanowski JQ. Exercise and cognition: results from the national long term care survey. Alzheimers Dement.2010;6(6):448-455.
85. Venturelli M, Scarsini R, Schena F. Six-month walking program changes cognitive and ADL performance in patients with Alzheimer. American Journal of Alzheimer’s Disease and Other Dementias. 2011;26(5):381-388.
86. Vreugdenhil A, Cannell J, Davies A, Razay G. A community-based exercise programme to improve functional ability in people with Alzheimer’s disease: a randomized controlled trial. Scand J Caring Sci. 2012;26:12-19.
87. Kemoun G, Thibaud M, Roumagne N, et al. Effects of a physical training program on cognitive function and walking efficiency in elderly persons with dementia. Dement Geriatr Cogn Disord.2010;29:109-114.
88. Yu F, Kolanowski A. Facilitating exercise training in older adults with Alzheimer’s disease. Geriatric Nursing. 2009;30(4):250-259.
89. Nichol KE, Parachikova AI, Cotman CW. Three weeks of running wheel exposure improves cognitive performance in the aged Tg2576 mouse. Behav Brain Res. 2007;184(2):124-132.


Add a New Comment
Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License